NICE CG 127 Hypertension

Update of clinical guidelines 18 and 34
Hypertension
The clinical management of primary hypertension in adults
Clinical Guideline 127

Methods, evidence, and recommendations
August 2011
Commissioned by the National Institute for Health and Clinical Excellence
Hypertension (partial update) Contents
Published by the National Clinical Guideline Centre at The Royal College of Physicians, 11 St Andrews Place, Regents Park, London, NW1 4BT First published 2004, republished 2006 National Clinical Guideline Centre - 2011 Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page. The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use. The rights of National Clinical Guideline Centre to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.
Hypertension (partial update) Contents
Contents
Rationale for update .................................................................................................................... 7 Guideline development group members ....................................................................................... 8 Acknowledgments ....................................................................................................................... 9 Acronyms and abbreviations .......................................................................................................10 1 2 Introduction ........................................................................................................................12 Development of the guideline ..............................................................................................14 2.1 2.2 2.3 2.4 2.5 3 What is a NICE clinical guideline? ....................................................................................... 14 Who developed this guideline? .......................................................................................... 15 What this guideline covers .................................................................................................. 15 What this guideline does not cover .................................................................................... 15 Relationships between the guideline and other NICE guidance ......................................... 16 2.5.1 3.1 3.2 Related guidance ................................................................................................. 16
2011 Methods .....................................................................................................................17 Developing the review questions and outcomes................................................................ 17 Searching for evidence ........................................................................................................ 17 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7 3.2.8 3.2.9 3.2.10 3.2.11 3.2.12 3.3 3.3.1 3.3.2 3.3.3 3.4 3.4.1 3.4.2 3.4.3 3.4.4 Clinical literature search ...................................................................................... 17 Health economic literature search ...................................................................... 18 Evidence of effectiveness .................................................................................... 18 Inclusion/exclusion .............................................................................................. 18 Methods of combining clinical studies ................................................................ 19 Appraising the quality of evidence by outcomes ................................................ 19 Grading the quality of clinical evidence............................................................... 20 Study limitations .................................................................................................. 21 Inconsistency ....................................................................................................... 21 Indirectness ......................................................................................................... 22 Imprecision .......................................................................................................... 22 Prognostic studies................................................................................................ 24 Literature review ................................................................................................. 25 Undertaking new health economic analysis ........................................................ 26 Cost-effectiveness criteria ................................................................................... 27 Research recommendations ................................................................................ 27 Validation process ............................................................................................... 27 Updating the guideline ........................................................................................ 28 Disclaimer ............................................................................................................ 28
Evidence of cost-effectiveness ............................................................................................ 24
Developing recommendations ............................................................................................ 27
Hypertension (partial update) Contents 3.4.5 4 4.1.1 4.1.2 4.1.3 4.1.4 4.2 4.2.1 4.2.2 5 5.1 5.2 5.3 5.4 6 6.1 6.2 6.3 6.4 6.5 Funding ................................................................................................................ 28 Review methods .................................................................................................. 29 Group process ...................................................................................................... 31 Evidence statements and recommendations ...................................................... 31 Costs and consequences ...................................................................................... 33 Clinical evidence .................................................................................................. 34 Cost-effectiveness evidence ................................................................................ 36
2004 Methods .....................................................................................................................29
2006 methods ..................................................................................................................... 34
Guideline summary..............................................................................................................37 Algorithms ........................................................................................................................... 37 Key priorities for implementation....................................................................................... 39 Full list of recommendations .............................................................................................. 40 Key research recommendations ......................................................................................... 45 Techniques for measuring blood pressure ......................................................................... 46 6.1.1 Manual blood pressure measurement ................................................................ 46 Cuffs .................................................................................................................................... 47 Conditions and environment .............................................................................................. 47 White Coat Hypertension.................................................................................................... 48 Blood pressure measurement devices ................................................................................ 49 6.5.1 6.5.2 6.5.3 6.6 6.7 6.8 6.9 Mercury sphygmomanometer............................................................................. 49 Aneroid sphygmomanometers ............................................................................ 49 Automated devices .............................................................................................. 49
Measuring blood pressure....................................................................................................46
Ambulatory blood pressure monitors ................................................................................. 49 Home blood pressure monitors .......................................................................................... 50 Recommendations .............................................................................................................. 50 Research recommendation ................................................................................................. 51 Predicting outcome using clinic, home and ambulatory measurements ........................... 52 7.1.1 7.1.2 7.1.3 Clinical evidence 2004 ......................................................................................... 52 Clinical evidence 2011 ......................................................................................... 53 Evidence statements clinical ............................................................................. 54 Clinical evidence .................................................................................................. 58 Evidence statements clinical ............................................................................. 60 Economic evidence literature review ............................................................... 61
Diagnosis of Hypertension ...................................................................................................52 7.1
7.2
Sensitivity and specificity of clinic, home and ambulatory measurements ........................ 58 7.2.1 7.2.2
7.3
Cost-effectiveness of clinic, home and ambulatory measurements................................... 61 7.3.1
Hypertension (partial update) Contents 7.3.2 7.3.3 7.4 7.4.1 7.4.2 7.5 7.6 8 Economic evidence - original economic analysis................................................. 62 Evidence statements economic ........................................................................ 67 Ambulatory blood pressure measurement ......................................................... 68 Home blood pressure measurement ................................................................... 96
Measurement protocols for diagnosing hypertension ....................................................... 68
Link from evidence to recommendations ......................................................................... 102 Recommendations ............................................................................................................ 107 8.1.1 Hypertension and cardiovascular disease ......................................................... 109 Urine testing for proteinuria.............................................................................. 110 Blood electrolyte, urea, creatinine, glucose and total/HDL cholesterol levels . 110
Assessing cardiovascular risk, target organ damage and secondary causes of hypertension . 109 8.2 Routine clinical investigations ........................................................................................... 110 8.2.1 8.2.2 8.3 8.4
Cardiovascular Risk Assessment ....................................................................................... 111 Secondary Hypertension ................................................................................................... 111 8.4.1 8.4.2 8.4.3 8.4.4 Renal and renovascular disease ........................................................................ 111 Pheochromocytoma .......................................................................................... 112 Hyperaldosteronism (primary aldosteronism) .................................................. 112 Cushing's syndrome ........................................................................................... 112 Hypothyroidism ................................................................................................. 113 Hyperthyroidism ................................................................................................ 113 Obstructive sleep apnoea .................................................................................. 113 Coarctation of aorta........................................................................................... 113 Acromegaly ........................................................................................................ 113 Drugs .................................................................................................................. 114
8.5
Other identifiable causes of hypertension........................................................................ 113 8.5.1 8.5.2 8.5.3 8.5.4 8.5.5 8.5.6
8.6 8.7 9 9.1
Recommendations ............................................................................................................ 114 Research recommendations ............................................................................................. 114 Blood pressure thresholds for initiating pharmacological treatment .............................. 115 9.1.1 9.1.2 9.1.3 Clinical evidence ................................................................................................ 115 Evidence statements - clinical ........................................................................... 136 Evidence statements economic ...................................................................... 136 Clinical evidence ................................................................................................ 136 Economic evidence ............................................................................................ 140 Evidence statements Clinical .......................................................................... 140 Evidence statements Health economic .......................................................... 141
Initiating and monitoring treatment, including blood pressure targets ................................ 115
9.2
Treatment of people aged 80 years and greater .............................................................. 136 9.2.1 9.2.2 9.2.3 9.2.4
9.3
Link from evidence to recommendations ......................................................................... 141
Hypertension (partial update) Contents 9.4 9.5 9.6 Recommendations ............................................................................................................ 143 Recommendations for research ........................................................................................ 144 Monitoring treatment efficacy.......................................................................................... 144 9.6.1 9.6.2 9.6.3 9.6.4 9.6.5 9.6.6 9.6.7 9.7 9.7.1 9.7.2 9.7.3 9.7.4 9.7.5 9.8 9.9 Clinical evidence ................................................................................................ 144 Economic evidence ............................................................................................ 151 Evidence statements clinical ........................................................................... 152 Evidence statements health economic ........................................................... 153 Link from evidence to recommendations.......................................................... 153 Recommendations ............................................................................................. 154 Research recommendations .............................................................................. 154 Clinical evidence ................................................................................................ 154 Health economic evidence ................................................................................ 168 Evidence statements clinical ........................................................................... 168 Evidence statements economic ...................................................................... 169 Link from evidence to recommendations: blood pressure treatment targets.. 169
Blood pressure targets for treatment ............................................................................... 154
Recommendations ............................................................................................................ 171 Research Recommendation .............................................................................................. 172
9.10 Frequency of review.......................................................................................................... 172 10 Integrating the assessment of blood pressure, target organ damage and cardiovascular risk assessment and clinical decision making regarding treatment initiation, treatment and targets ........................................................................................................................ 173 11 Lifestyle interventions ....................................................................................................... 175 11.1 Overview ........................................................................................................................... 175 11.1.1 11.1.2 11.1.3 11.1.4 11.1.5 11.1.6 11.1.7 11.1.8 11.1.9 Managing changes in lifestyle............................................................................ 177 Diet .................................................................................................................... 177 Exercise .............................................................................................................. 179 Relaxation therapies .......................................................................................... 180 Multiple lifestyle interventions.......................................................................... 182 Alcohol ............................................................................................................... 184 Coffee................................................................................................................. 185 Reducing sodium (salt) intake ........................................................................... 185 Calcium supplements......................................................................................... 187
11.1.10 Magnesium supplements .................................................................................. 188 11.1.11 Potassium supplementation .............................................................................. 188 11.1.12 Combined salt supplements .............................................................................. 190 11.1.13 Drug therapy versus lifestyle change................................................................. 190 11.1.14 Smoking cessation ............................................................................................. 192 11.1.15 Recommendations ............................................................................................. 192
Hypertension (partial update) Contents 12 Pharmacological interventions ........................................................................................... 193 12.1 2004 guidance: pharmacological interventions ................................................................ 196 12.1.1 12.2.1 12.2.2 12.3.1 12.3.2 12.4.1 12.4.2 12.4.3 Placebo controlled trials .................................................................................... 196 Clinical evidence statements: head-to-head drug comparisons ....................... 204 Meta-analysis results summary ......................................................................... 206 Angiotensin-converting enzyme inhibitors (ACEi) versus Angiotensin Receptor Blockers (ARB) .................................................................................... 208 Diuretics ............................................................................................................. 212 Methodological introduction ............................................................................. 238 Results of the health economic model .............................................................. 240 Conclusions ........................................................................................................ 241 12.2 2006 rapid pharmacological update: head to head trials ................................................. 203
12.3 2011 update: Pharmacological therapy for hypertension ................................................ 208
12.4 Cost-effectiveness analysis ............................................................................................... 238
12.5 Step two therapy............................................................................................................... 242 12.6 Resistant hypertension ..................................................................................................... 245 12.7 Special groups for consideration....................................................................................... 248 12.7.1 12.7.2 12.7.3 12.7.4 12.7.5 12.7.6 People aged over 80 years................................................................................. 248 Younger people.................................................................................................. 248 Ethnicity ............................................................................................................. 248 Chronic kidney disease ...................................................................................... 254 Type 1 and Type 2 diabetes ............................................................................... 255 Women who are pregnant or breast-feeding.................................................... 255
12.8 Stopping treatment ........................................................................................................... 255 12.9 Link from evidence to recommendations- Pharmacological treatment of hypertension 257 12.10 Recommendations ............................................................................................................ 264 12.11 Research recommendations ............................................................................................. 266 13 Patients perspectives ........................................................................................................ 267 13.1 Introduction ...................................................................................................................... 267 13.2 Discovering hypertension ................................................................................................. 267 13.3 Treatment ......................................................................................................................... 267 13.4 Living with hypertension ................................................................................................... 268 13.5 Education and adherence ................................................................................................. 268 13.5.1 13.5.2 13.5.3 Compliance with Prescribed Antihypertensive Medication .............................. 268 Implementing lifestyle measures ...................................................................... 270 Recommendations ............................................................................................. 271
14 Reference list..................................................................................................................... 272 15 Glossary ............................................................................................................................ 317
Hypertension (partial update) Contents Appendices ............................................................................................................................... 326 Appendix A: Scope....................................................................................................................... 326 Appendix B: Declarations of Interest .......................................................................................... 334 Appendix C: Review questions .................................................................................................... 337 Appendix D: Literature search strategies .................................................................................... 338 Appendix E: Review protocols .................................................................................................... 363 Appendix F: Clinical evidence tables ........................................................................................... 378 Appendix G: Evidence tables health economic studies (2011 update) .................................... 387 Appendix H: Forest plots ............................................................................................................. 392 Appendix I: Cost-effectiveness analysis pharmacological treatment (updated 2011) ........... 421 Appendix J: Cost-effectiveness analysis blood pressure monitoring for confirming a diagnosis of hypertension (new 2011) ................................................................... 446 Appendix K: Research recommendations (2011)........................................................................ 503 Appendix L: CG18 Essential Hypertension: managing adult patients in primary care, 2004 ..... 505 Appendix M: CG34 Hypertension: management in adults in primary care: pharmacological update, 2006 ............................................................................... 506
Hypertension (partial update)

Introduction
Rationale for update

This document is a partial update of Clinical Guideline 18 (2004) and Clinical Guideline 34 (2006) on Essential Hypertension in adults. The sections that have not been amended are integrated with the updated guidance in this document. Both guidelines are available in full in the appendices of the document. Improvements in methodology since 2006 mean the way information is presented may, at times, be inconsistent (for example, the style of review write-up, and 2011 recommendations are not graded according to the strength of evidence, unlike those in the 2006). New or amended sections of the guideline are indicated with an update panel in the right hand margin.
Update 2011

Introduction
Guideline development group members

Name Bernard Higgins Bryan Williams Helen Williams Jane Northedge John Crimmins Kate Lovibond Mark Caulfield Michaela Watts Naomi Stetson Paul Miller Rachel OMahony Richard McManus Shelley Mason Terry McCormack Taryn Krause Role Clinical Director, National Clinical Guideline Centre Professor of Medicine, Guideline Development Group Chair Consultant Pharmacist for cardiovascular disease, Southwark Health and Social Care Patient and care representative General Practitioner, Vale of Glamorgan Senior Health Economist, National Clinical Guideline Centre Professor of Clinical Pharmacology, Barts and the London School of Medicine Hypertension Nurse Specialist, Addenbrookes Hospital, Cambridge Primary Care Nurse, Watling Medical Centre, Burnt Oak Senior Information Scientist, National Clinical Guideline Centre Senior Research Fellow, National Clinical Guideline Centre Professor of Primary Care Cardiovascular Research, University of Birmingham Patient and carer representative General Practitioner, Spring Vale Medical Centre, North Yorkshire, Primary Care Cardiovascular Society Senior Project Manager/Research Fellow, National Clinical Guideline Centre
Update 2011

Introduction
Acknowledgments
The development of this guideline was greatly assisted by the following people: Jill Cobb, Information Scientist, National Clinical Guideline Centre Ralph Hughes, Health Economist, National Clinical Guideline Centre Fatema Limbada, Project Coordinator, National Clinical Guideline Centre Jill Parnham, Director of Operations, National Clinical Guideline Centre David Wondering, Health Economic Lead, National Clinical Guideline Centre Jacoby Patterson, Systematic Reviewer Julie Brown, Systematic Reviewer Sue Jowett, Senior Lecturer in Health Economics, University of Birmingham James Hodgkinson, Research Fellow, University of Birmingham Jonathan Mant, Professor of Primary Care Research, University of Cambridge Una Martin, Reader in Clinical Pharmacology, University of Birmingham Carl Heneghan, Reader in Evidence-Based Medicine, University of Oxford Richard Hobbs, Head of Primary Care Clinical Sciences, University of Birmingham.
Update 2011

Introduction
Acronyms and abbreviations

ABPM ACEi ANOVA ARB BNF CBPM CCA CCB CEA c.f. CI / 95% CI CUA DH DSA ED EQ-5D GDG GP GRADE HBPM HES HR HRQoL HT HTA ICD-10 ICER ICH ISH IQR Ambulatory blood pressure measurement Angiotensin-converting enzyme inhibitors Analysis of variance Angiotensin receptor blocker British National Formulary Clinic blood pressure measurement Cost-consequences analysis Calcium channel blocker Cost-effectiveness analysis Confer (refer to) Confidence interval / 95% confidence interval Cost-utility analysis Department of Health Deterministic Sensitivity Analysis Emergency Department EuroQol-5D Guideline Development Group General Practitioner Grading of Recommendations Assessment, Development and Evaluation Home blood pressure measurement Hospital Episode Statistics Hazard Ratio Health-related quality of life Hypertensive / hypertension Health technology assessment International Classification of Diseases, 10th edition Incremental cost-effectiveness ratio Isolated clinic hypertension Ischemia Interquartile range
10

Introduction
INMB IRR ITT LOS LR+ LY MD NCGC NHS NHSEED NICE NNT NPV NS NT OR PICO PPP PPV p.r.n PSA QALY QUADAS RCT ROC RRK RR SD SE SPC SR SS WCH
Incremental Net Monetary Benefit Inter-rater reliability Intention to treat Length of Stay Positive likelihood ratio Life-year Mean difference National Clinical Guideline Centre National Health Service The NHS Economic Evaluation Database National Institute for Health and Clinical Excellence Number needed to treat Negative predictive value Non-significant (not statistically significant) Normotensive Odds ratio Framework incorporating patients, interventions, comparison and outcome Purchasing Power Parity Positive predictive value Pro re nata Probabilistic sensitivity analysis Quality-adjusted life year Quality assessment tool for diagnostic accuracy studies Randomised controlled trial Receiver operating characteristic Riva-Rocci Korotkoff Relative risk Standard deviation Standard error Summary of product characteristics Systematic review Statistically significant White coat hypertension
11

Introduction
1 Introduction
This guideline is for the clinical management of primary hypertension in adults (aged greater than 18 years). Hypertension (high blood pressure) is one of the most preventable causes of premature morbidity and mortality world-wide. Hypertension is a major risk factor for stroke (ischaemic and haemorrhagic), myocardial infarction, heart failure, chronic kidney disease, peripheral vasculardisease, cognitive decline and premature death. Untreated hypertension is associated a progressive rise in blood pressure, often culminating in a treatment resistant state due to associated vascular and renal damage. Blood pressure is quantified as diastolic and systolic pressures measured in millimetres of mercury (mmHg). The diastolic pressure represents the pressure during ventricular relaxation in diastole whereas the systolic pressure represents the peak pressure due to ventricular contraction during systole. Either or both pressures have specified upper limits of normal and elevation in either or both pressures are used to define hypertension. Blood pressure is normally distributed in the population and there is no natural cut-point above which "hypertension" definitively exists and below which, it does not. Epidemiological studies demonstrate that the aforementioned disease risk associated with blood pressure is a continuous relationship and above blood pressures of 115/70mmHg, the risk of cardiovascular events doubles for every 20/10mmHg rise in blood pressure. The threshold blood pressure determining the presence of hypertension is defined as the level of blood pressure above which treatment has been shown to reduce the development or progression of disease. Primary hypertension was previously termed essential hypertension because of a long-standing view that high blood pressure was sometimes essential to perfuse diseased and sclerotic arteries. It is now recognised that the diseased and sclerotic arteries were most often the consequence of the hypertension and thus the term essential hypertension is redundant and the primary hypertension is preferred. Primary hypertension refers to the majority of people with sustained high blood pressure (approximately 90%) encountered in clinical practice, for which there is no obvious, identifiable cause. The remaining 10% are termed "secondary hypertension" for which specific causes for the blood pressure elevation can be determined (for example, Conn's adenoma, renovascular disease, or phaeochromocytoma). Primary hypertension is remarkably common in the UK population and the prevalence is strongly influenced by age and lifestyle factors. Systolic and/or diastolic blood pressures may be elevated. Systolic pressure elevation is the more dominant feature of hypertension in older patients and diastolic pressure more commonly elevated in younger patients, (those less than 50 years of age). At least one quarter of the adult population of the UK have hypertension, (blood pressure 140/90mmHg) and more than half of those over the age of 60 years. As the demographics of the UK shifts towards an older, more sedentary and obese population, the prevalence of hypertension and its requirement for treatment will continue to rise. Routine periodic screening for high blood pressure is now commonplace in the UK as part of National Service Frameworks for cardiovascular disease prevention. Consequently, the diagnosis, treatment and follow-up of patients with hypertension is one of the most common interventions in primary care, accounting for approximately 12% of Primary Care consultation episodes and approximately 1 billion in drug costs in 2006 . NICE first issued guidance for the management of hypertension in primary care in 2004. This was followed by a rapid update of the pharmacological treatment chapter of the guideline in 2006. The current partial update of the hypertension guideline is in response to the regular five year review cycle of existing NICE guidance. It began with a scoping exercise which identified key areas of the existing guideline for which new evidence had emerged that was likely to influence or change existing guideline recommendations.
Update 2011
12

Introduction
Sections of the guideline that have not been updated continue to stand, however, wherever NICE has subsequently issued new and related guidance relevant to existing recommendations, these have been identified and cross-referred to in this partial update, examples include interventions on lifestyle factors and public health policy recommendations such as smoking cessation, dietary salt restriction, alcohol intake and cardiovascular disease prevention and cardiovascular disease risk assessment. In addition, new NICE guidance developed in areas relevant to hypertension are also highlighted and cross referenced (for example, chronic kidney disease, stroke, diabetes and hypertension in pregnancy). The recommendations that have been reviewed in this partial update of the guideline for the clinical management of primary hypertension in adults, include; blood pressure measurement for the diagnosis of hypertension; blood pressure thresholds for intervention with drug therapy and blood pressure targets for treatment; specific aspects of the recommendations for the pharmacological treatment of hypertension; the treatment of hypertension in the very elderly (people aged greater than 80 years); dilemmas surrounding decision making for treatment of hypertension in younger adults (less than 40 years); the treatment of drug resistant hypertension; and wherever appropriate, the impact of age and ethnicity on treatment recommendations. Finally, despite the fact that the treatment of hypertension has a large clinical trial evidence base to inform recommendations, an important aspect of the evidence review for guideline development is to identify where gaps in knowledge remain. In so doing, research questions have been identified to prompt the gathering of further evidence to continue the evolution of guidance and clinical practice.
Update 2011
13

Development of the guideline
2 Development of the guideline

2.1 What is a NICE clinical guideline?
NICE clinical guidelines are recommendations for the care of individuals in specific clinical conditions or circumstances within the NHS from prevention and self-care through primary and secondary care to more specialised services. We base our clinical guidelines on the best available research evidence, with the aim of improving the quality of health care. We use predetermined and systematic methods to identify and evaluate the evidence relating to specific review questions. NICE clinical guidelines can: provide recommendations for the treatment and care of people by health professionals be used to develop standards to assess the clinical practice of individual health professionals be used in the education and training of health professionals help patients to make informed decisions improve communication between patient and health professional While guidelines assist the practice of healthcare professionals, they do not replace their knowledge and skills. We produce our guidelines using the following steps: Guideline topic is referred to NICE from the Department of Health Stakeholders register an interest in the guideline and are consulted throughout the development process. The scope is prepared by the National Clinical Guideline Centre (NCGC) The NCGC establishes a guideline development group A draft guideline is produced after the group assesses the available evidence and makes recommendations There is a consultation on the draft guideline. The final guideline is produced. The NCGC and NICE produce a number of versions of this guideline: The full guideline contains all the recommendations, plus details of the methods used and the underpinning evidence The NICE guideline lists the recommendations the Quick Reference Guide (QRG) presents recommendations in a suitable format for health professionals Information for the public - understanding NICE guidance or UNG - is written using suitable language for people without specialist medical knowledge Clinical Pathway www.pathways.nice.org.uk/pathways/hypertension This version is the full guideline. The other documents can be downloaded from NICE at www.nice.org.uk
Update 2011
14

2.2 Who developed this guideline?

A multidisciplinary Guideline Development Group (GDG) comprising professional group members and consumer representatives of the main stakeholders developed this guideline (see section on Guideline Development Group Membership and acknowledgements). The National Institute for Health and Clinical Excellence funds the National Clinical Guideline Centre (NCGC) and thus supported the development of this guideline. The GDG was convened by the NCGC and chaired by Professor Bryan Williams in accordance with guidance from the National Institute for Health and Clinical Excellence (NICE). The group met every four weeks during the development of the guideline. At the start of the guideline development process all GDG members declared interests including consultancies, fee-paid work, share-holdings, fellowships and support from the healthcare industry. At all subsequent GDG meetings, members declared arising conflicts of interest, which were also recorded in Appendix B: Declarations of Interest. Members were either required to withdraw completely or for part of the discussion if their declared interest made it appropriate. The details of declared interests and the actions taken are shown in Appendix B: Declarations of Interest. Staff from the NCGC provided methodological support and guidance for the development process. The team working on the guideline included a project manager, systematic reviewers, health economists and information scientists. They undertook systematic searches of the literature, appraised the evidence, conducted meta analysis and cost effectiveness analysis where appropriate and drafted the guideline in collaboration with the GDG.
Update 2011
2.3 What this guideline covers

Adults with hypertension (18 years and older). Particular consideration will be given to the needs of black people of African and Caribbean descent and minority ethnic groups where these differ from the needs of the general population. People aged 80 years or older. Ambulatory monitoring. Home blood pressure monitoring. Blood pressure thresholds for intervention and targets for treatment. First-line therapy options, for example angiotensin-converting enzyme inhibitors versus angiotension receptors blockers. Calcium-channel blockers versus diuretics as preferred components in step two of the treatment algorithm, for example, combination therapy. Adherence to medication. Provision of appropriate information and support. Resistant hypertension (that is, fourth-line therapy). Response to blood pressure lowering drugs according to age and ethnicity. For further details please refer to Appendix A: Scope and Appendix C: Review questions.
2.4 What this guideline does not cover

People with diabetes. Children and young people (younger than 18 years).
15

Pregnant women. Secondary causes of hypertension (for example, Conn's adenoma, phaeochromocytoma and renovascular hypertension). People with accelerated hypertension (that is, severe acute hypertension associated grade III retinopathy and encephalopathy). People with acute hypertension or high blood pressure in emergency care settings. Prevention of hypertension. Screening for hypertension. Specialist management of secondary hypertension (that is, hypertension arising from other medical conditions). Non-pharmacological interventions.
2.5 Relationships between the guideline and other NICE guidance

2.5.1 Related guidance
Prevention of cardiovascular disease at the population level. NICE Public Health Guidance 25/ www.nice.org.uk/PH25 Medicines adherence. NICE clinical guideline 76 (2009). Available from www.nice.org.uk/guidance/CG76 Chronic kidney disease. NICE clinical guideline 73 (2008). Available from www.nice.org.uk/guidance/CG73 Stroke. NICE clinical guideline 68 (2008). Available from www.nice.org.uk/guidance/CG68 Lipid modification. NICE clinical guideline 67 (2008). Available from www.nice.org.uk/guidance/CG67 Type II diabetes. NICE clinical guideline 66 (2008). Available from www.nice.org.uk/guidance/CG66 Sleep apnoea continuous positive airway pressure (CPAP). NICE technology appraisal guidance 139 (2008). Available from www.nice.org.uk/guidance/TA139 MI: secondary prevention. NICE clinical guideline 48 (2007). Available from www.nice.org.uk/guidance/CG48 Obesity. NICE clinical guideline 43 (2006). Available from www.nice.org.uk/guidance/CG43 Atrial fibrillation. NICE clinical guideline 36 (2006). Available from www.nice.org.uk/CG36 Nutrition support in adults. NICE clinical guideline 32 (2006). Available from www.nice.org.uk/guidance/CG32 Chronic heart failure. NICE clinical guideline 5 (2003). Available from www.nice.org.uk/guidance/CG5
Update 2011
16

2011 Methods
3 2011 Methods
This guidance was developed in accordance with the methods outlined in the NICE Guidelines Manual 2009.430
3.1 Developing the review questions and outcomes

Review questions were developed in a PICO framework (patient, intervention, comparison and outcome) for intervention reviews, and with a framework of population, index tests, reference standard and target condition for reviews of diagnostic test accuracy. This was to guide the literature searching process and to facilitate the development of recommendations by the guideline development group (GDG). They were drafted by the NCGC technical team and refined and validated by the GDG. The questions were based on the key clinical areas identified in the scope (Appendix A: Scope) and a list can be found in Appendix C: Review Questions. Further information on the outcome measures examined follows this section.
3.2 Searching for evidence

3.2.1 Clinical literature search
Systematic literature searches were undertaken to identify evidence within published literature in order to answer the review questions as per The Guidelines Manual (2009).430 Clinical databases were searched using relevant medical subject headings, free-text terms and study type filters where appropriate. Studies published in languages other than English were not reviewed. All searches were conducted on core databases, MEDLINE, Embase, Cinahl and The Cochrane Library. All searches were updated on 29th November 2010 and no papers were included beyond this date. Search strategies were checked by looking at reference lists of relevant key papers, checking search strategies in other systematic reviews and asking the GDG for known studies. The questions, the study types applied, the databases searched and the years covered can be found in Appendix C: Literature search strategies. During the scoping stage, a search was conducted for guidelines and reports on the websites listed below and via organisations relevant to the topic. Searching for grey literature or unpublished literature was not undertaken. All references sent by stakeholders were considered. Guidelines International Network database (www.g-i-n.net) National Guideline Clearing House (www.guideline.gov/) National Institute for Health and Clinical Excellence (NICE) (www.nice.org.uk) National Institutes of Health Consensus Development Program (consensus.nih.gov/) National Library for Health (www.library.nhs.uk/) 3.2.1.1 Call for evidence The GDG decided to initiate a call for evidence for meta-analyses, based on a systematic review, that include studies that use ambulatory blood pressure measurement as the reference standard and report sensitivity and specificity of home and/or clinic blood pressure measurement, as they believed that important evidence existed that would not be identified by the standard searches. The NCGC contacted all registered stakeholders and asked them to submit any relevant published or unpublished evidence.
Update 2011
17

2011 Methods
3.2.2
Health economic literature search

Systematic literature searches were also undertaken to identify health economic evidence within published literature relevant to the review questions. The evidence was identified by conducting a broad search relating to the guideline population in the NHS economic evaluation database (NHS EED), the Health Economic Evaluations Database (HEED) and health technology assessment (HTA) databases from 2003 onwards to find anything published since the original guideline. There were two questions not covered in either the original guideline or the previous rapid update, for which additional searches with no date restrictions were carried out. Additionally, the search was run on MEDLINE and Embase, with a specific economic filter, from 2009, to ensure recent publications that had not yet been indexed by these databases were identified. Studies published in languages other than English were not reviewed. Where possible, searches were restricted to articles published in English language.The search strategies for health economics are included in Appendix D: Literature search strategies. All searches were updated on 29th November 2010. No papers published after this date were considered.
3.2.2.1
Call for evidence

The GDG decided to initiate a call for evidence for cost-effectiveness analyses from a UK perspective, using methods in line with the NICE reference case, comparing ambulatory, home and clinic blood pressure measurement in the diagnosis of hypertension, as they believed that important evidence existed that would not be identified by the standard searches. The NCGC contacted all registered stakeholders and asked them to submit any relevant published or unpublished evidence.
3.2.3
Evidence of effectiveness
The Research Fellow: Identified potentially relevant studies for each review question from the relevant search results by reviewing titles and abstracts full papers were then obtained. Reviewed full papers against pre-specified inclusion / exclusion criteria to identify studies that addressed the review question in the appropriate population and reported on outcomes of interest (review protocols are included in Appendix E:Review protocols). Critically appraised relevant studies using the appropriate checklist as specified in The Guidelines Manual 430 Extracted key information about the studys methods and results into evidence tables (evidence tables are included in Appendix D: Evidence tables clinical studies and Appendix G: Evidence tables health economic studies. Generated summaries of the evidence by outcome (included in the relevant chapter write-ups): o Randomised studies: meta analysed, where appropriate and reported in GRADE profiles (for clinical studies) see below for details o Observational studies: data has been presented for individual studies narratively or in summary tables (GRADE profiles have not been generated) o Diagnostic studies: data has been presented for individual studies narratively or in summary tables (GRADE profiles have not been generated) o Qualitative studies: each study summarised in a table where possible, otherwise presented in a narrative.
Update 2011
3.2.4
Inclusion/exclusion
See the review protocols in Appendix E: Review Protocols for full details.
18

2011 Methods
3.2.5
Methods of combining clinical studies

Data synthesis for intervention reviews Where possible, meta-analyses were conducted to combine the results of studies for each review question using Cochrane Review Manager (RevMan5) software. Fixed-effects (Mantel -Haenszel) techniques were used to calculate risk ratios (relative risk) for the following binary outcomes: angioedema. Where reported, time-to-event data was presented as a hazard ratio for the following binary outcomes: mortality, stroke, MI, heart failure, new onset diabetes, vascular procedures, angina requiring hospitalisation, study drug withdrawal. The continuous outcome blood pressure (mmHg)] was analysed using an inverse variance method for pooling weighted mean differences and where the studies had different scales, standardised mean differences were used. No quality of life outcome data was reported by any of the studies included in the 2012 update reviews Statistical heterogeneity was assessed by considering the chi-squared test for significance at p<0.1 or an I-squared inconsistency statistic of >50% to indicate significant heterogeneity. Where significant heterogeneity was present, we carried out sensitivity analysis based on the quality of studies, with particular attention paid to allocation concealment, blinding and loss to follow-up (missing data). In cases where there was inadequate allocation concealment, unclear blinding, high loss to follow-up ( 20% missing data for studies 2 years follow-up and 30% for those with >2 years follow-up) or differential missing data, this was examined in a sensitivity analysis. For the latter, the duration of follow up was also taken into consideration prior to including in a sensitivity analysis. Assessments of potential differences in effect between subgroups were based on the chi-squared tests for heterogeneity statistics between subgroups. If no sensitivity analysis was found to completely resolve statistical heterogeneity then a random effects (DerSimonian and Laird) model was also explored to provide a more conservative estimate of the effect. The means and standard deviations of continuous outcomes were required for meta-analysis. However, in cases where standard deviations were not reported, the standard error was calculated if the p-values or 95% confidence intervals were reported and meta-analysis was undertaken with the mean and standard error using the generic inverse variance method in Cochrane Review Manager (RevMan5) software. Where p values were reported as less than, a conservative approach was undertaken. For example, if the p value was reported as p 0.001, the calculations for standard deviations will be based on a p value of 0.001. If these statistical measures were un available then the methods described in section 16.1.3 of the Cochrane Handbook Missing standard deviations were applied as the last resort.
Update 2011
3.2.6
Appraising the quality of evidence by outcomes

The evidence for outcomes from the included RCT studies were evaluated and presented using an adaptation of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) toolbox developed by the international GRADE working group (http://www.gradeworkinggroup.org/). The software (GRADEpro) developed by the GRADE working group was used to assess the quality of each outcome, taking into account individual study quality and the meta-analysis results. The summary of findings was presentedas an evidence profile, a single table that includes details of the quality assessment as well as pooled outcome data, where appropriate, an absolute measure of intervention effect and the summary of quality of evidence for that outcome. In this table, the columns for intervention and control indicate the sum of the sample size for continuous outcomes. For binary outcomes such as number of patients with an adverse event, the event rates (n/N: number of patients with events divided by sum of number of patients) are shown with percentages. Reporting or publication bias was only taken into consideration in the quality assessment and included in the Clinical Study Characteristics table if it was apparent.
19

2011 Methods
Each outcome was examined separately for the quality elements listed and defined in Table 1 and each graded using the quality levels listed in Table 2: The main criteria considered in the rating of these elements are discussed below (see 3.2.7 Grading of Evidence). Footnotes were used to describe reasons for grading a quality element as having serious or very serious problems. The ratings for each component were summed to obtain an overall assessment for each outcome. GRADE is currently designed only for randomised trials and observational studies. Table 1: Description of quality elements in GRADE for intervention studies.
Description Limitations in the study design and implementation may bias the estimates of the treatment effect. Major limitations in studies decrease the confidence in the estimate of the effect. Inconsistency refers to an unexplained heterogeneity of results. Indirectness refers to differences in study population, intervention, comparator and outcomes between the available evidence and the review question, or recommendation made. Results are imprecise when studies include relatively few patients and few events and thus have wide confidence intervals around the estimate of the effect relative to the clinically important threshold. Publication bias is a systematic underestimate or an overestimate of the underlying beneficial or harmful effect due to the selective publication of studies.
Quality element Limitations
Inconsistency Indirectness
Imprecision
Publication bias
Table 2:
Level None Serious
Levels of quality elements in GRADE

Description There are no serious issues with the evidence The issues are serious enough to downgrade the outcome evidence by one level The issues are serious enough to downgrade the outcome evidence by two levels
Update 2011
Very serious
Table 3:
Level High Moderate Low Very low
Overall quality of outcome evidence in GRADE

Description Further research is very unlikely to change our confidence in the estimate of effect Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate Any estimate of effect is very uncertain
3.2.7
Grading the quality of clinical evidence

After results were pooled, the overall quality of evidence for each outcome was considered. The following procedure was adopted when using GRADE: 1. A quality rating was assigned, based on the study design. RCTs start HIGH and observational studies as LOW. 2. The rating for RCTs was then downgraded for the specified criteria: Study limitations, inconsistency, indirectness, imprecision and reporting bias. These criteria are detailed below. Due to the wide diversity of study design, data reported and data analysis methods of the observational studies that were included in this guideline, it was very difficult to compare studies
20

2011 Methods
for quality and therefore observational studies were not downgraded or upgraded in GRADE, and all remained as LOW quality evidence (please see below, section 3.2.12, for details of quality assessment of prognostic studies).. 3. The downgraded marks were then summed and the overall quality rating was revised. For example, all RCTs started as HIGH and the overall quality became MODERATE, LOW or VERY LOW if 1, 2 or 3 points were deducted respectively. 4. The reasons or criteria used for downgrading were specified in the footnotes. The details of criteria used for each of the main quality element are discussed further in the following sections 3.3.5 to 3.3.8/3.3.9 [if section for publication bias is relevant].
3.2.8
Study limitations
The main limitations for randomised controlled trials are listed in Table 4. Table 4:
Limitation Allocation concealment Lack of blinding Incomplete accounting of patients and outcome events Selective outcome reporting Other limitations
Study limitations of randomised controlled trials

Explanation Those enrolling patients are aware of the group to which the next enrolled patient will be allocated (major problem in pseudo or quasi randomised trials with allocation by day of week, birth date, chart number, etc) Patient, caregivers, those recording outcomes, those adjudicating outcomes, or data analysts are aware of the arm to which patients are allocated Loss to follow-up not accounted and failure to adhere to the intention to treat principle when indicated
Update 2011
Reporting of some outcomes and not others on the basis of the results For example: Stopping early for benefit observed in randomised trials, in particular in the absence of adequate stopping rules Use of unvalidated patient-reported outcomes Carry-over effects in cross-over trials Recruitment bias in cluster randomised trials
3.2.9
Inconsistency
Inconsistency refers to an unexplained heterogeneity of results. When estimates of the treatment effect across studies differ widely (i.e. heterogeneity or variability in results), this suggests true differences in underlying treatment effect. When heterogeneity exists (Chi square p<0.1 or I- squared inconsistency statistic of >50%), but no plausible explanation can be found, the quality of evidence was downgraded by one or two levels, depending on the extent of uncertainty to the results contributed by the inconsistency in the results. If inconsistency could be explained based on pre-specified subgroup analysis, the GDG took this into account and considered whether to make separate recommendations based on the identified explanatory factors, i.e. population and intervention. Where subgroup analysis gave a plausible explanation of heterogeneity, the quality of evidence was not downgraded.
21

2011 Methods
3.2.10
Indirectness
Directness refers to the extent to which the populations, intervention, comparisons and outcome measures are similar to those defined in the inclusion criteria for the reviews. Indirectness is important when these differences are expected to contribute to a difference in effect size, or may affect the balance of harms and benefits considered for an intervention.
3.2.11
Imprecision
The criteria applied for imprecision are based on the confidence intervals for pooled or the best estimate of effect as illustrated in Figure 1 and outlined in Table 5. Table 5: Criteria applied to determine precision
Dichotomous and continuous outcomes The 95% confidence interval (or alternative estimate of precision) around the pooled or best estimate of effect: 1. Does not cross either of the two minimal important difference (MID) thresholds (the threshold lines for appreciable benefit or harm); defined as precise Rating for precision: no serious imprecision Crosses one of the two MID thresholds (appreciable benefit or appreciable harm); defined as imprecise Rating for precision: serious Crosses both of the two MID thresholds ( appreciable benefit and appreciable harm); defined as imprecise Rating for precision: very serious
Update 2011
2.
3.
22

2011 Methods
Figure 1: Illustration of precise and imprecise outcomes based on the confidence interval of outcomes in a forest plot
MID
MID
NO SERIOUS IMPRECISION
SERIOUS IMPRECISION -1 VERY SERIOUS IMPRECISION -2 0.75 1.0 1.25
Appreciable benefit Non-appreciable benefit or harm (AEs and harmful outcomes) / appreciable harm (effectiveness and beneficial outcomes)
Appreciable harm (AEs and harmful outcomes) / appreciable benefit (effectiveness and beneficial outcomes)
MID = minimal important difference determined for each outcome. The MIDs are the threshold for appreciable benefits and harms. The confidence intervals of the top five points of the diagram (within the green sector or within the purple sector) are considered precise because the upper and lower limits of the point estimate (diamond shapes) do not cross the pre-defined MID. Conversely, the bottom three points of the diagram are considered imprecise because the upper and lower limits of the point estimates (diamonds) for each of them cross the pre-defined MID and reduce the certainty of the result. The following are the MID for the outcomes in this guideline (as agreed by the GDG). Table 6:
Outcome Mortality from any cause Stroke (ischaemic or haemorrhagic) Myocardial infarction (MI) (including, where reported, silent MI)
Update 2011
MIDs for the outcomes used in this guidance

Relative risk reduction 10% 10% 10%
23

2011 Methods Outcome Heart failure New onset diabetes Vascular procedures (including both coronary and carotid artery procedures) Angina requiring hospitalisation Health-related quality of life (to use what is reported by trials) Major adverse cardiac and cerebrovascular events (MAACE): fatal and nonfatal MI, fatal and non-fatal stroke, hospitalised angina, hospitalised heart failure, revascularisation (and different composites of this outcome) Study drug withdrawal rates (surrogate for adverse effects of drug treatment and for adherence Angioedema in black people of African and Caribbean descent Blood pressure Relative risk reduction 10% 10% 10% 10% As defined in literature for each specific QoL measure 15%
10% 10% 5 mmHg (mean difference, continuous outcome)
3.2.12
Prognostic studies
All prognostic study designs were included for the prognostic questions. The quality of the prognostic studies was assessed using the quality checklist in the NICE Guidelines Manual April 2009. The main criteria considered in assessing study quality were: The study sample represents the population of interest with regard to key characteristics, sufficient to limit potential bias to the results Loss to follow-up is unrelated to key characteristics (that is, the study data adequately represent the sample), sufficient to limit potential bias The prognostic factor of interest is adequately measured in study participants, sufficient to limit potential bias The outcome of interest is adequately measured in study participants, sufficient to limit bias Important potential confounders are appropriately accounted for, limiting potential bias with respect to the prognostic factor of interest The statistical analysis is appropriate for the design of the study, limiting potential for the presentation of invalid results The methodological flaws of the prognostic studies included in the guideline update, have been summarised in tables within appendix F, in order to give an overview of the quality of each individual study, since GRADE is not currently designed for prognostic studies. Odds ratios, relative risks or hazard ratios, with their 95% confidence intervals, from multivariate analyses were extracted from the papers. Data for selected outcomes has been summarised in tables within the relevant review chapter. Full data for all the outcomes has been reported in the evidence tables (see appendix F) for each individual prognostic study. Taking into consideration the advice on prognostic reviews in the NICE guidelines manual, meta-analysis was not undertaken for prognostic studies.
Update 2011
3.3 Evidence of cost-effectiveness

Evidence on cost-effectiveness related to the key clinical issues being addressed in the guideline was sought. The health economist undertook: A systematic review of the economic literature New cost-effectiveness analysis in priority areas
24

2011 Methods
3.3.1
Literature review
The Health Economist: Identified potentially relevant studies for each review question from the economic search results by reviewing titles and abstracts full papers were then obtained. Reviewed full papers against pre-specified inclusion / exclusion criteria to identify relevant studies (see below for details). Critically appraised relevant studies using the economic evaluations checklist as specified in The Guidelines Manual.430 Extracted key information about the studys methods and results into evidence tables (evidence tables are included in Appendix G: Evidence tables health economic studies. Generated summaries of the evidence in NICE economic evidence profiles (included in the relevant chapter write-ups) see below for details. Inclusion/exclusion Full economic evaluations (studies comparing costs and health consequences of alternative courses of action: costutility, cost-effectiveness, cost-benefit and cost-consequence analyses) and comparative costing studies that addressed the review question in the relevant population were considered potentially applicable as economic evidence. Studies were excluded if they only reported cost per hospital (not per patient), or only reported average cost effectiveness without disaggregated costs and effects. Abstracts, posters, reviews, letters/editorials, foreign language publications and unpublished studies were excluded. Studies judged to have an applicability rating of not applicable were excluded (this included studies that took the perspective of a non-OECD country). Remaining studies were prioritised for inclusion based on their relative applicability to the development of this guideline and the study limitations. For example, if a high quality, directly applicable UK analysis was available other less relevant studies may have been excluded and this is noted in the relevant section. For more details about the assessment of applicability and methodological quality see the economic evaluation checklist (The Guidelines Manual, Appendix H 430 and the health economics research protocol in Appendix E: Review protocols. When no relevant economic analyses were identified in the economic literature review, relevant UK NHS unit costs were presented to the GDG to inform consideration of cost effectiveness. NICE economic evidence profiles The NICE economic evidence profile has been used to summarise cost and cost-effectiveness estimates. The economic evidence profile shows, for each economic study, an assessment of applicability and methodological quality, with footnotes indicating the reasons for the assessment. These assessments were made by the health economist using the economic evaluation checklist from The Guidelines Manual, Appendix H.430 It also shows incremental costs, incremental outcomes (for example, QALYs) and the incremental cost-effectiveness ratio from the primary analysis, as well as information about the assessment of uncertainty in the analysis. See Table 7 for more details. If a non-UK study was included in the profile, the results were converted into pounds sterling using the appropriate purchasing power parity.468
Update 2011
25

2011 Methods
Table 7:
Item Study
Content of NICE economic profile

Description First author name, reference, date of study publication and country perspective. An assessment of methodological quality of the study(a): Minor limitations the study meets all quality criteria, or the study fails to meet one or more quality criteria, but this is unlikely to change the conclusions about cost effectiveness. Potentially serious limitations the study fails to meet one or more quality criteria, and this could change the conclusion about cost effectiveness Very serious limitations the study fails to meet one or more quality criteria and this is very likely to change the conclusions about cost effectiveness. Studies with very serious limitations would usually be excluded from the economic profile table. An assessment of applicability of the study to the clinical guideline, the current NHS situation and NICE decision-making(a): Directly applicable the applicability criteria are met, or one or more criteria are not met but this is not likely to change the conclusions about cost effectiveness. Partially applicable one or more of the applicability criteria are not met, and this might possibly change the conclusions about cost effectiveness. Not applicable one or more of the applicability criteria are not met, and this is likely to change the conclusions about cost effectiveness. Particular issues that should be considered when interpreting the study. The mean cost associated with one strategy minus the mean cost of a comparator strategy. The mean QALYs (or other selected measure of health outcome) associated with one strategy minus the mean QALYs of a comparator strategy. Incremental cost-effectiveness ratio: the incremental cost divided by the respective QALYs gained. A summary of the extent of uncertainty about the ICER reflecting the results of deterministic or probabilistic sensitivity analyses, or stochastic analyses of trial data, as appropriate.
Limitations
Applicability
Other comments Incremental cost Incremental effects ICER Uncertainty
a) Limitations and applicability were assessed using the economic evaluation checklist from The Guidelines Manual, 430 Appendix H
3.3.2
Undertaking new health economic analysis

As well as reviewing the published economic literature for each review question, as described above, new cost-effectiveness analysis was undertaken by the Health Economist in priority areas. Priority areas were agreed by the GDG after formation of the review questions and consideration of the available health economic evidence.
Update 2011
Additional data for the analysis were identified as required through additional literature searches undertaken by the Health Economist, and discussion with the GDG. Model structure, inputs and assumptions were explained to and agreed by the GDG members during meetings, and they commented on subsequent revisions. Results were presented in GDG meetings for discussion and interpretation. The priority area identified for new economic analysis was diagnosis of hypertension see Appendix J: Cost-effectiveness analysis blood pressure monitoring for confirming a diagnosis of hypertension (new 2011) for full methods. The 2006 cost-effectiveness analysis of drug treatment was also updated see Appendix I: Cost-effectiveness analysis pharmacological treatment (updated 2011) for full methods.
26

2011 Methods
3.3.3
Cost-effectiveness criteria
NICEs report Social value judgements: principles for the development of NICE guidance sets out the principles that GDGs should consider when judging whether an intervention offers good value for money.429,430 In general, an intervention was considered to be cost effective if either of the following criteria applied (given that the estimate was considered plausible): a) The intervention dominated other relevant strategies (that is, it was both less costly in terms of resource use and more clinically effective compared with all the other relevant alternative strategies), or b) The intervention cost less than 20,000 per quality-adjusted life-year (QALY) gained compared with the next best strategy. If the GDG recommended an intervention that was estimated to cost more than 20,000 per QALY gained, or did not recommend one that was estimated to cost less than 20,000 per QALY gained, the reasons for this decision are discussed explicitly in the from evidence to recommendations section of the relevant chapter with reference to issues regarding the plausibility of the estimate or to the factors set out in the Social value judgements: principles for the development of NICE guidance.429
3.4 Developing recommendations

Over the course of the guideline development process, the GDG was presented with: Evidence tables of the clinical and economic evidence reviewed from the literature. All evidence tables are in Appendix E: Evidence Tables Clinical studies and Appendix G:Evidence tables health economic studies. Summary of clinical and economic evidence and quality Forest plots and summary ROC curves A description of the methods and results of the cost-effectiveness analysis undertaken for the guideline The main considerations specific to each recommendation are outlined in the link from evidence to recommendation section preceding the recommendation section.
Update 2011
3.4.1
Research recommendations
When areas were identified for which good evidence was lacking, the guideline development group considered making recommendations for future research. Decisions about inclusion were based on factors such as: the importance to patients or the population national priorities potential impact on the NHS and future NICE guidance ethical and technical feasibility
3.4.2
Validation process
The guidance is subject to a four week public consultation and feedback as part of the quality assurance and peer review the document. All comments received from registered stakeholders are responded to in turn and posted on the NICE website when the pre-publication check of the full guideline occurs.
27

2011 Methods
3.4.3
Updating the guideline

Following publication, and in accordance with the NICE guidelines manual, NICE will ask a National Collaborating Centre or the National Clinical Guideline Centre to advise NICEs Guidance executive whether the evidence base has progressed significantly to alter the guideline recommendations and warrant an update.
3.4.4
Disclaimer
Health care providers need to use clinical judgement, knowledge and expertise when deciding whether it is appropriate to apply guidelines. The recommendations cited here are a guide and may not be appropriate for use in all situations. The decision to adopt any of the recommendations cited here must be made by the practitioners in light of individual patient circumstances, the wishes of the patient, clinical expertise and resources. The National Clinical Guideline Centre disclaims any responsibility for damages arising out of the use or non-use of these guidelines and the literature used in support of these guidelines.
Update 2011
3.4.5
Funding
The National Clinical Guideline Centre was commissioned by the National Institute for Health and Clinical Excellence to undertake the work on this guideline.
28

2004 Methods
4 2004 Methods
4.1.1 Review methods
The aim of reviewing was to identify and synthesise relevant published and unpublished evidence to allow recommendations to be evidence-based wherever possible.630 The search was carried out using the electronic databases MEDLINE, EMBASE and CENTRAL, attempting to locate systematic reviews and meta-analyses, and original randomised trials using a combination of subject heading and free text searches. We made extensive use of high quality recent review articles and bibliographies, as well as contact with subject area experts. New searches were concentrated in areas of importance to the guideline development process, for which existing systematic reviews were unable to provide valid or up to date answers. The expert knowledge and experience of group members also backed up the search of the literature. Electronic searches used a sensitive search strategy based on a combination of text and index terms to locate randomised controlled trials of treatments relevant to the guideline. If data necessary for our analyses were not reported, we wrote to authors or sponsoring agencies. We are grateful to investigators and sponsors who provided unpublished information to aid our work. We assessed the quality of relevant studies retrieved and their ability to provide valid answers to the clinical questions addressed by the group. Assessment of study quality concentrated on internal validity (the extent to which the study measured what it intended to measure), external validity (the extent to which study findings could be generalised to other treatment settings) and construct validity (the extent to which measurement corresponded to theoretical understanding of a disease).
139
Table 8:
Quality Criteria for Randomised Controlled Trials
Appropriateness of inclusion and exclusion criteria Concealment of allocation Blinding of patients Blinding of health professionals Blinding of data collectors/outcome assessors Completeness and length of follow up Appropriateness of outcome measures
Once data had been abstracted from individual papers and their quality assessed, the information was synthesised. Individual trials often have an insufficient sample size to identify significant outcomes with confidence81, so where appropriate, the results of randomised studies were combined using meta-analytic techniques 175. Questions were answered using the best evidence available. When considering the effect of an intervention, if this could be addressed by the best study design then weaker designs were not reviewed. Where studies were of poor quality, or contained patient groups considered likely to have different responses, the effects of inclusion or exclusion were examined in sensitivity analyses. No trials that met our inclusion criteria were excluded from the primary analyses. However, where data on relevant outcomes were not available, these studies could not be included, thus leading to the potential for publication bias. Review criteria Scoping work revealed a vast number of trials of pharmaceutical interventions. Recent work suggests that study size is a useful proxy for study quality.189,224 Consequently to achieve the task in the timescale provided we reviewed only those pharmaceutical studies which enrolled 200 or more patients. Since the prime motivation for treatment in hypertension, an asymptomatic condition, is
29

2004 Methods
the prevention of mortality and morbidity, we reviewed those studies with a planned follow-up of at least a year since such studies are likely to have been designed to inform about these endpoints. Few non-pharmacological studies directly address cardiovascular endpoints or feature substantial durations of follow-up. Consequently in these areas we evaluated blood pressure reduction as a proxy endpoint and included trials with a follow-up of 8 weeks follow-up or more, which compared a group receiving a lifestyle intervention with a control group who received no treatment, usual treatment, sham therapy or a placebo. Statistical methods Pharmacological interventions The outcomes analyzed were: all cause mortality, fatal and non-fatal myocardial infarction, fatal and non-fatal stroke. We did not consider the following endpoints: renal disease (rare in non-diabetic patients); heart failure (inconsistently reported in trials); cardiovascular events (a concatenation of myocardial infarction and stroke). For each trial, the risk ratios comparing the risk of each outcome in the active treatment and control groups - or, for head-to-head trials, in the different treatment groups - were calculated. Results of trials were combined in a meta-analysis using the DerSimonian and Laird random effects model175, to estimate an overall pooled risk ratio (RR) and its 95% confidence interval (95%CI). This model assumes that there are different effects of treatment in different populations, which are clustered about a mean effect; the pooled RR gives the best estimate of this mean effect. In the placebo-controlled trials reported in this guideline, a RR less than 1 favours treatment and a RR greater than 1 favours control. If the 95%CI include 1, there is no statistically significant difference between the treatments being compared. Finally, we assessed the tolerability of the interventions by comparing the rate of overall withdrawal (percentage of patients who withdrew each year) in each treatment arm of a trial and calculating the difference in these rates (called the 'incident risk difference'). These incident risk differences were combined in a meta-analysis using the DerSimonian and Laird random effects model175, to estimate an overall pooled incident risk difference and its 95% confidence interval. We assessed heterogeneity between trials using a chi-squared statistic (Q). This assesses whether the trials are sufficiently similar to be validly combined. Although the test for heterogeneity is weak, it is usually assumed that if it gives p-values greater than 0.10, there is no significant heterogeneity and it is valid to discuss the combined findings. We also assessed whether the effect in individual trials was related to the size of the trial; any such trend might indicate publication bias, e.g. where small trials were published only if they showed a positive effect. Again, this test for systematic variation in the magnitude of the estimated effect with the size of the trial is weak, but it is usually assumed that if it gives a p-value greater than 0.10, there is unlikely to be any such bias. Lifestyle interventions None of the studies identified were designed to quantify significant changes in rates of death or cardiovascular events, so we analysed the surrogate endpoint of reduced blood pressure. For each trial, the difference in the final value mean blood pressure in the treatment and control groups - or, for head-to-head trials, in the different treatment groups - was calculated. Change scores from baseline were used where complete data for final values was unavailable. These mean differences were weighted according to the precision of each trial (which depends largely on its size, with larger trials getting more weight) and combined in a meta-analysis using the DerSimonian and Laird random effects model175, to estimate an overall pooled weighted mean difference and its 95% confidence interval. While most of the trials were of parallel design (two or more groups received the various interventions at the same time), some were of crossover design (all participants received both active
30

2004 Methods
treatment and control interventions, but in a random order). Crossover trials have about four times greater precision than parallel trials of the same size, so we used methods have been developed recently to combine the parallel and crossover trials in the same meta-analysis.147,193 Heterogeneity and the potential for publication bias were assessed in the same way as for pharmaceutical trials. The mean percentage achieving a reduction of 10mmHg or more in systolic blood pressure was then estimated from the cumulative normal distribution637 and confidence intervals were estimated using the delta method.51 Finally, we assessed the tolerability of the interventions by comparing the proportion of withdrawals (% of patients who withdrew) in each treatment arm of a trial and calculating the difference in these proportion (called the 'risk difference'). These risk differences were combined in a meta-analysis using the DerSimonian and Laird random effects model,175 to estimate an overall pooled risk difference and its 95% confidence interval.
4.1.2
Group process
The guideline development group was run using the principles of small group work and was led by a trained facilitator. The group underwent initial exercises to set its own rules to determine how it wanted to function and received brief training on reviewing methods, economic analysis and grading methodology. Additional training was provided in the group as the need arose in subsequent meetings. Findings, expressed as narratives, statements of evidence and recommendations, were reached by informal consensus. There was no obligation to force an agreement where none existed after discussion: dissensions were recorded in the guideline narrative.471
4.1.3
Evidence statements and recommendations

The guideline development group process produces summary statements of the evidence concerning available treatments and healthcare and from these makes its recommendations. Evidence statements and recommendations are commonly graded in guidelines reflecting the quality of the study designs on which they are based. An established scheme adapted from the Agency for Health Care Policy and Research (AHCPR) Classification is shown in Table 9 and Table 10.14 Table 9:
Ia: Ib: IIa: IIb: III: IV:
AHCPR derived categories of evidence
Level of evidence evidence from meta-analysis of randomised controlled trials evidence from at least one randomised controlled trial evidence from at least one controlled study without randomisation evidence from at least one other type of quasi-experimental study evidence from non-experimental descriptive studies, such as comparative studies, correlation studies and case-control studies evidence from expert committee reports or opinions and/or clinical experience of respected authorities
Table 10: AHCPR derived strengths of recommendations

Strength of evidence A B C D directly based on category I evidence directly based on category II evidence or extrapolated recommendation from category I evidence directly based on category III evidence or extrapolated recommendation from category I or II evidence directly based on category IV evidence or extrapolated recommendation from category I, II or III evidence
31

2004 Methods
Two grading schemes were used when developing this guideline, the one above and a new scheme called GREG (Guideline Recommendation and Evidence Grading).392 The new scheme seeks to address a number of problems, by extending grading from treatment to include diagnosis, prognosis and cost, and to handle the subtleties of clinical evidence more sensitively (Table 11). Table 11: GREG scheme for assessing evidence and writing recommendations
EVIDENCE Evidence statements provide information about disease, diagnosis and treatment, and are used to support recommendations. Each evidence statement is graded by scoring the study design and applying quality corrections. Design Design scores Treatment Randomised controlled trial Non-randomised controlled study Uncontrolled study Diagnosis Blinded cohort study Unblinded cohort study Other design Prognosis Incidence cohort study Other cohort study Descriptive data Population data Representative sample Convenience sample Quality corrections Flawed design, conduct or analysis Imprecise findings Lack of consistency or independence Inadequate relevance Very strong association Evidence Grade I: High II: Intermediate 2 III: Low Notes Notes i. Blinding refers to independent interpretation of a test and reference standard. ii. An incident cohort is identified and followed in time from a defined point in the progress of disease or care. iii. Important flaws may be judged to occur when adequate standards of research are not followed or are unreported in published findings. Potential examples include failure to analyse by intention-to-treat, over-interpretation of secondary analyses, failure to adjust for potential confounding in non- randomised designs. For diagnostic studies this includes the need for an adequate reference standard and to apply different tests in an adequately short timescale. iv. Sparse data (too few events or patients) are the most common reason for imprecision. A confidence interval including both no effect and a clinically important effect is an example of an imprecise finding. v. Consistency in [1] design: involves methods, patients, outcome measures; and [2] findings: involves homogeneity of summary estimates. Independence refers to the availability of research from at least two independent sources. Evidence of publication bias also denotes lack of consistency. vi. Adequate relevance requires [1] use in studies of a relevant patient-oriented health outcome or a strongly linked surrogate endpoint; and [2] a sufficiently representative and relevant patient group or mix. vii. In comparative designs a very strong association can raise the quality score.
1 2 3
1 2 3
1 2 3 1 2 3
+1 +1 +1 +1 +1 -1
1 2 3
Recommendations Recommendations provide guidance about appropriate care. Ideally, these should be based on clear evidence: a robust understanding of the benefits, tolerability, harms and costs of alternative patterns of care. They also need to be feasible in the healthcare setting addressed. There are three unique categories, and each recommendation may be positive or negative, conditional or unconditional reflecting current evidence and the understanding of the guideline group.
32

2004 Methods EVIDENCE Evidence statements provide information about disease, diagnosis and treatment, and are used to support recommendations. Each evidence statement is graded by scoring the study design and applying quality corrections. A. Recommendation There is robust evidence to recommend a pattern of care. B. C. Provisional recommendation On balance of evidence, a pattern of care is recommended with caution. Consensus Opinion Evidence being inadequate, a pattern of care is recommended by consensus.
Use of the two schemes was evaluated in this and another guideline being developed contemporaneously. Both groups consistently favoured the new scheme and so the guideline is presented using the new grading scheme. The evaluation of the two schemes will be reported separately. The key point of note is that any assessment of evidence quality is ultimately a subjective process. How bad does a trial have to be before it is flawed or how sparse do the findings have to be before we lose confidence in the findings? The purpose of an evidence grading scheme is to characterise the robustness of outcomes from studies, and the random and systematic biases that pertain to them. Similarly recommendation grading must credibly assimilate evidence and health service context to credibly advise lines of care for average patients. Clinicians must use their judgement and awareness of patients' circumstances and values when considering recommendations from guidelines.
4.1.4
Costs and consequences

Approaches to cost-effectiveness have assisted in reaching recommendations in a series of primary care evidence-based guidelines.188,393 This guideline involves a systematic appraisal of effectiveness, compliance, quality-of-life, safety and health service resource use and costs of a medical intervention provided in the British health care setting. Using the most current, pertinent and complete data available, the economic analysis attempts a robust presentation showing the possible bounds of costeffectiveness that may result. The guiding principle behind economic analysis is that it is desirable to use limited healthcare resources to maximise health improvements in the population. Well defined but narrow notions of health improvement may not reflect all aspects of value to patients, carers, clinicians or society. For example, evidence may lead the guideline group to recommend targeting additional resources to certain patient groups when unequal access to care is apparent. The group process allows discussion of what should be included in the definition of 'improved health' and more broadly of other concepts of value to society such as fairness, justice, dignity or minimum standards of care. The range of values used to generate cost-effectiveness estimates reflects the available evidence and the concerns of the guideline development group. Recommendations are graded reflecting the certainty with which the costs and consequences of a medical intervention can be assessed. This practice reflects the desire of group members to have simple, understandable and robust information based on good data. It is not generally helpful to present an additional systematic review of previous economic analyses that have adopted a variety of differing perspectives, analytic techniques and baseline data. However, the economic literature is reviewed to compare guideline findings with representative published economic analyses and to interpret any differences in findings when these occurred. A commentary is included when the group feel this aids understanding.
33

2004 Methods
4.2 2006 methods

4.2.1
4.2.1.1
Clinical evidence
Methodological introduction Study inclusion and reporting criteria A systematic search of the literature was performed on EMBASE and MEDLINE for randomised controlled trials comparing any combination of antihypertensive drugs from among the following five classes of drugs: ACE inhibitors (ACEi) angiotensin-II receptor antagonists (ARB) beta-receptor blockers (BB) calcium-channel blockers (CCB) thiazide-type diuretics (TD). Placebo-controlled studies were not included because the main aim of this rapid partial update was to make recommendations regarding the optimal sequencing of drug treatment for hypertension, for which head-to-head studies are required, and because sufficient placebo-controlled studies of the main drug classes had been considered in the original NICE guideline. However, placebo-controlled studies were sought for isolated systolic hypertension because of a lack of comparator studies. The cut-off date for evidence to be considered in the previous guideline was July 2004, so this update only searched for English-language titles published after that date. Papers published up to and including 19 December 2005 were considered this constitutes the cut-off for evidence for this rapid update. Studies were excluded due to: inadequate or no randomisation inadequate study power, defined as a sample size of less than 200 patients, or having a follow-up period of less than 12 months having an exclusive diabetic or paediatric patient population, unrepresentative of the general UK hypertensive population stroke, myocardial infarction, and mortality outcomes not being reported. The following outcomes were recorded for each study, where available: mortality from any cause stroke (ischaemic or haemorrhagic) myocardial infarction (including, where reported, silent MI) heart failure new-onset diabetes mellitus vascular procedures (including both coronary and carotid artery procedures) incidence of unstable angina (or angina episodes requiring hospitalisation) study drug withdrawal.
34

2004 Methods
Interpretation and analysis of results All outcomes, with the exception of study drug withdrawal, vascular procedures and unstable angina, were entered into a meta-analysis for each drug combination using RevMan 4.2 software (The Nordic Cochrane Centre). The overall effect size was reported as the relative risk (RR) with 95% confidence intervals in each case. A p-value less than 0.05 was considered statistically significant for overall effect. Forest plots for each comparison are included in Appendix A. In recording the outcomes, stroke was considered to be synonymous with 'cerebrovascular event'. Reports of 'cardiovascular events' or other composite outcomes other than those listed above were not considered. Sensitivity analyses were performed based on the inclusion and exclusion of silent myocardial infarction and the inclusion and exclusion of secondary prevention studies. Additional subgroup analyses were performed to identify the source of any significant heterogeneity in study results (defined as an I2 statistic greater than 50%). Where the heterogeneity has I2 greater than 50%, the trials are reported individually in the evidence statements. The following outcomes were not subject to meta-analysis due to potential variability or subjectivity in diagnosis or treatment protocols, and were reported as a narrative only: unstable angina revascularisation procedures study drug withdrawal. Following consultation on the draft guideline, heart failure as an outcome was included in the metaanalysis. Because of inconsistency in definition of heart failure in the trials, this was analysed using a random effects model. Secondary analyses In addition to results in general hypertensive populations, the following subgroups were also considered separately: those patients with isolated systolic hypertension (ISH) black people of African and Caribbean descent younger patients (defined as under 55 years). For ISH, due to the lack of evidence comparing different antihypertensive drugs, the results from placebo-controlled trials were also considered. These results included pre-defined subgroup analyses from trials in general hypertensive populations as well as one trial comprising only ISH patients. The results were entered into a meta-analysis according to the same procedure specified above. The definition of ISH varied slightly between studies: permitting a diastolic blood pressure up to 95 mmHg in one study (SYST-EUR43,124,555) and 90 mmHg in the others (SHEP483,536,537,606, SHEP-P281,484,485). No trials comprising only non-white patients were found, although two pre-defined subgroup analyses from trials in general hypertensive populations were found (ALLHAT589-591, LIFE154,176,222,369,370,507,618,619). Results involving placebo comparisons in non-white populations were not considered. Evidence on younger patients was extremely sparse, and evidence consideration was therefore extended to include papers pre-dating July 2004 and in which blood pressure lowering effect was the main outcome measure.
35

2004 Methods
4.2.2
Cost-effectiveness evidence
The GDG drafted recommendations on the basis of the clinical evidence. A health economic analysis was then conducted to balance the clinical outcomes and to test the cost effectiveness of different initial antihypertensive medications.
Update 2011
See Appendix I: Cost-effectiveness analysis pharmacological treatment (updated 2011) for full methods note that analysis was updated as part of the 2011 update.
36

Guideline summary
5 Guideline summary
5.1 Algorithms
Figure 2: Diagnosis of Hypertension
Update 2011
37

Guideline summary
Figure 3: Treatment of Hypertension
Update 2011
38

Guideline summary
5.2 Key priorities for implementation

From the full set of recommendations, the GDG selected ten key priorities for implementation. The criteria used for selecting these recommendations are listed in detail in The Guidelines Manual.430 The reasons that each of these recommendations was chosen are shown in the table linking the evidence to the recommendation in the relevant chapter. Diagnosing hypertension If the clinic blood pressure is 140/90 mmHg or higher, offer ambulatory blood pressure monitoring (ABPM) to confirm the diagnosis of hypertension. [new 2011] When using ABPM to confirm a diagnosis of hypertension, ensure that at least two measurements per hour are taken during the persons usual waking hours (for example, between 08:00 and 22:00). Use the average value of at least 14 measurements taken during the persons usual waking hours to confirm a diagnosis of hypertension. [new 2011] When using home blood pressure monitoring (HBPM) to confirm a diagnosis of hypertension, ensure that: o for each blood pressure recording, two consecutive measurements are taken, at least 1 minute apart and with the person seated and o blood pressure is recorded twice daily, ideally in the morning and evening and o blood pressure recording continues for at least 4 days, ideally for 7 days. Discard the measurements taken on the first day and use the average value of all the remaining measurements to confirm a diagnosis of hypertension. [new 2011] Initiating treatment Offer antihypertensive drug treatment to people aged under 80 years with stage 1 hypertension who have one or more of the following: o target organ damage o established cardiovascular disease o renal disease o diabetes o a 10-year cardiovascular risk equivalent to 20% or greater. [new 2011] Offer antihypertensive drug treatment to people of any age with stage 2 hypertension. [new 2011] For people aged under 40 years with stage 1 hypertension and no evidence of target organ damage, cardiovascular disease, renal disease or diabetes, consider seeking specialist evaluation of secondary causes of hypertension and a more detailed assessment of potential target organ damage. This is because 10-year cardiovascular risk assessments can underestimate the lifetime risk of cardiovascular events in these people. [new 2011] Monitoring treatment and blood pressure targets For people identified as having a white-coat effect, consider ABPM or HBPM as an adjunct to clinic blood pressure measurements to monitor the response to antihypertensive treatment with lifestyle modification or drugs. [new 2011] Choosing antihypertensive drug treatment Offer people aged 80 years and over the same antihypertensive drug treatment as people aged 5580 years, taking into account any comorbidities. [new 2011]
Updte 2011
39

Guideline summary
Step 1 treatment Offer step 1 antihypertensive treatment with a calcium-channel blocker (CCB) to people aged over 55 years and to black people of African or Caribbean family origin of any age. If a CCB is not suitable, for example because of oedema or intolerance, or if there is evidence of heart failure or a high risk of heart failure, offer a thiazide-like diuretic. [new 2011] If diuretic treatment is to be initiated or changed, offer a thiazide-like diuretic, such as chlortalidone (12.525.0 mg once daily) or indapamide (1.5 mg modified-release or 2.5 mg once daily) in preference to a conventional thiazide diuretic such as bendroflumethiazide or hydrochlorothiazide. [new 2011] For people who are already having treatment with bendroflumethiazide or hydrochlorothiazide and whose blood pressure is stable and well controlled, continue treatment with the bendroflumethiazide or hydrochlorothiazide. [new 2011] Step 4 treatment For treatment of resistant hypertension at step 4: o Consider further diuretic therapy with low-dose spironolactone (25 mg once daily) if the blood potassium level is 4.5 mmol/l or lower. Use particular caution in people with a reduced estimated glomerular filtration rate because they have an increased risk of hyperkalaemia. o Consider higher-dose thiazide-like diuretic treatment if the blood potassium level is higher than 4.5 mmol/l. [new 2011]
5.3 Full list of recommendations

1. Healthcare professionals taking blood pressure measurements need adequate initial training and periodic review of their performance. [2004] 2. Because automated devices may not measure blood pressure accurately if there is pulse irregularity (for example, due to atrial fibrillation), palpate the radial or brachial pulse before measuring blood pressure. If pulse irregularity is present, measure blood pressure manually using direct auscultation over the brachial artery. [new 2011] 3. Healthcare providers must ensure that devices for measuring blood pressure are properly validated, maintained and regularly recalibrated according to manufacturers instructions. [2004] a 4. When measuring blood pressure in the clinic or in the home, standardise the environment and provide a relaxed, temperate setting, with the person quiet and seated, and their arm outstretched and supported. [new 2011] 5. If using an automated blood pressure monitoring device, ensure that the device is validated and an appropriate cuff size for the persons arm is used. [new 2011] 6. In people with symptoms of postural hypotension (falls or postural dizziness): measure blood pressure with the person either supine or seated. measure blood pressure again with the person standing for at least a minute prior to measurement. [2004, amended 2011] 7. If the systolic blood pressure falls by 20 mmHg or more when the person is standing: review medication measure subsequent blood pressures with the person standing
a
Update 2011
A list of validated blood pressure monitoring devices is available on the British Hypertension Societys website (see www.bhsoc.org). The British Hypertension Society is an independent reviewer of published work. This does not imply any endorsement by NICE.
40

Guideline summary
consider referral to specialist care if symptoms of postural hypotension persist. [2004, amended 2011] 8. When considering a diagnosis of hypertension, measure blood pressure in both arms: If the difference in readings between arms is more than 20 mmHg, repeat the measurements. If the difference in readings between arms remains more than 20 mmHg on the second measurement, measure subsequent blood pressure in the arm with the higher reading. [new 2011] 9. If the clinic blood pressure is 140/90 mmHg or higher, offer ambulatory blood pressure monitoring (ABPM) to confirm the diagnosis of hypertension. [new 2011] 10.If a person is unable to tolerate ABPM, home blood pressure monitoring (HBPM) is a suitable alternative to confirm the diagnosis of hypertension. [new 2011] 11.If the person has severe hypertension, consider starting antihypertensive drug treatment immediately, without waiting for the results of ABPM or HBPM. [new 2011] 12.While waiting for confirmation of a diagnosis of hypertension, carry out investigations for target organ damage (such as left ventricular hypertrophy, chronic kidney disease and hypertensive retinopathy) (see 21) and a formal assessment of cardiovascular risk using a cardiovascular risk assessment tool (see 20). [new 2011] 13.If hypertension is not diagnosed but there is evidence of target organ damage such as left ventricular hypertrophy, albuminuria or proteinuria, consider carrying out investigations for alternative causes of the target organ damage. [new 2011]
Update 2011
14.If hypertension is not diagnosed, measure the persons clinic blood pressure at least every 5 years subsequently, and consider measuring it more frequently if the persons clinic blood pressure is close to 140/90 mmHg. [new 2011] 15.When using ABPM to confirm a diagnosis of hypertension, ensure that at least two measurements per hour are taken during the persons usual waking hours (for example, between 08:00 and 22:00). Use the average value of at least 14 measurements taken during the persons usual waking hours to confirm a diagnosis of hypertension. [new 2011] 16.When using HBPM to confirm a diagnosis of hypertension, ensure that: for each blood pressure recording, two consecutive measurements are taken, at least 1 minute apart and with the person seated and blood pressure is recorded twice daily, ideally in the morning and evening and blood pressure recording continues for at least 4 days, ideally for 7 days. Discard the measurements taken on the first day and use the average value of all the remaining measurements to confirm a diagnosis of hypertension. [new 2011] 17.Refer the person to specialist care the same day if they have: accelerated hypertension, that is, blood pressure usually higher than 180/110 mmHg with signs of papilloedema and/or retinal haemorrhage or suspected phaeochromocytoma (labile or postural hypotension, headache, palpitations, pallor and diaphoresis). [2004, amended 2011] 18.Consider the need for specialist investigations in people with signs and symptoms suggesting a secondary cause of hypertension. [2004, amended 2011] For NICE guidance on the early identification and management of chronic kidney disease see Chronic kidney disease (NICE clinical guideline 73, 2008).
41

Guideline summary
19.Use a formal estimation of cardiovascular risk to discuss prognosis and healthcare options with people with hypertension, both for raised blood pressure and other modifiable risk factors. [2004] 20.Estimate cardiovascular risk in line with the recommendations on Identification and assessment of CVD risk in Lipid modification (NICE clinical guideline 67) b. [2008] 21.For all people with hypertension offer to: test for the presence of protein in the urine by sending a urine sample for estimation of the albumin:creatinine ratio and test for haematuria using a reagent strip take a blood sample to measure plasma glucose, electrolytes, creatinine, estimated glomerular filtration rate, serum total cholesterol and HDL cholesterol examine the fundi for the presence of hypertensive retinopathy arrange for a 12-lead electrocardiograph to be performed. [2004, amended 2011] 22.Offer antihypertensive drug treatment to people aged under 80 years with stage 1 hypertension who have one or more of the following: target organ damage established cardiovascular disease renal disease diabetes a 10-year cardiovascular risk equivalent to 20% or greater. [new 2011] 23.Offer antihypertensive drug treatment to people of any age with stage 2 hypertension. [new 2011] 24.For people aged under 40 years with stage 1 hypertension and no evidence of target organ damage, cardiovascular disease, renal disease or diabetes, consider seeking specialist evaluation of secondary causes of hypertension and a more detailed assessment of potential target organ damage. This is because 10-year cardiovascular risk assessments can underestimate the lifetime risk of cardiovascular events in these people. [new 2011] 25.Use clinic blood pressure measurements to monitor the response to antihypertensive treatment with lifestyle modification or drugs. [new 2011] 26.For people identified as having a white-coat effect, consider ABPM or HBPM as an adjunct to clinic blood pressure measurements to monitor the response to antihypertensive treatment with lifestyle modification or drugs. [new 2011] 27.Aim for a target clinic blood pressure below 140/90 mmHg in people aged under 80 years with treated hypertension. [new 2011] 28.Aim for a target clinic blood pressure below 150/90 mmHg in people aged 80 years and over with treated hypertension. [new 2011] 29.When using ABPM or HBPM to monitor the response to treatment (for example, in people identified as having a white-coat effect c and people who choose to monitor their blood pressure at home) aim for a target average blood pressure during the persons usual waking hours of: below 135/85 for people aged under 80 years below 145/85 in people aged over 80 years and over. [new 2011]
b c
Update 2011
Clinic blood pressure measurements must be used in the calculation of cardiovascular risk. A discrepancy of more than 20/10 mmHg between clinic and average daytime ABPM or average HBPM blood pressure measurements at the time of diagnosis.
42

Guideline summary
For NICE guidance on the prevention of obesity and cardiovascular disease see Obesity (NICE clinical guideline 43, 2006) and Prevention of cardiovascular disease at population level (NICE public health guidance 25, 2010). 30.Lifestyle advice should be offered initially and then periodically to people undergoing assessment or treatment for hypertension. [2004] 31.Ascertain peoples diet and exercise patterns because a healthy diet and regular exercise can reduce blood pressure. Offer appropriate guidance and written or audiovisual materials to promote lifestyle changes. [2004] 32.Relaxation therapies can reduce blood pressure and people may wish to pursue these as part of their treatment. However, routine provision by primary care teams is not currently recommended. [2004] 33.Ascertain peoples alcohol consumption and encourage a reduced intake if they drink excessively, because this can reduce blood pressure and has broader health benefits. [2004] 34.Discourage excessive consumption of coffee and other caffeine-rich products. 35.Encourage people to keep their dietary sodium intake low, either by reducing or substituting sodium salt, as this can reduce blood pressure.[2004] 36.Do not offer calcium, magnesium or potassium supplements as a method for reducing blood pressure. [2004] 37.The best current evidence does not show that combinations of potassium, magnesium and calcium supplements reduce blood pressure. [2004] 38.Offer advice and help to smokers to stop smoking. [2004] 39.A common aspect of studies for motivating lifestyle change is the use of group working. Inform people about local initiatives by, for example, healthcare teams or patient organisations that provide support and promote healthy lifestyle change. [2004] 40.Where possible, recommend treatment with drugs taken only once a day. [2004] 41.Prescribe non-proprietary drugs where these are appropriate and minimise cost. [2004] 42.Offer people with isolated systolic hypertension (systolic BP 160 mmHg or more) the same treatment as people with both raised systolic and diastolic blood pressure. [2004] 43.Offer people aged 80 years and over the same antihypertensive drug treatment as people aged 5580 years, taking into account any comorbidities. [new 2011] 44.Offer antihypertensive drug treatment to women of child-bearing potential in line with the recommendations on Management of pregnancy with chronic hypertension and Breastfeeding in Hypertension in pregnancy (NICE clinical guideline 107). [2010] 45.Offer people aged under 55 years step 1 antihypertensive treatment with an angiotensinconverting enzyme (ACE) inhibitor or a low-cost angiotensin-II receptor blocker (ARB). If an ACE inhibitor is prescribed and is not tolerated (for example, because of cough), offer a low-cost ARB. [new 2011] 46.Do not combine an ACE inhibitor with an ARB to treat hypertension. [new 2011] 47.Offer step 1 antihypertensive treatment with a calcium-channel blocker (CCB) to people aged over 55 years and to black people of African or Caribbean family origin of any age. If a CCB is not
Update 2011
43

Guideline summary
suitable, for example because of oedema or intolerance, or if there is evidence of heart failure or a high risk of heart failure, offer a thiazide-like diuretic. [new 2011] 48.If a diuretic is to be initiated or changed, offer a thiazide-like diuretic, such as chlortalidone (12.5 mg25.0 mg once daily) or indapamide (1.5mg slow release or 2.5 mg once daily) in preference to a conventional thiazide diuretic such as bendroflumethiazide or hydrochlorothiazide. [new 2011] 49.For people who are already having treatment with bendroflumethiazide or hydrochlorothiazide and whose blood pressure is stable and well controlled, continue treatment with the bendroflumethiazide or hydrochlorothiazide. [new 2011] 50.Beta-blockers are not a preferred initial therapy for hypertension. However, beta-blockers may be considered in younger people, particularly: those with an intolerance or contraindication to ACE inhibitors and angiotensin-II receptor antagonists or women of child-bearing potential or people with evidence of increased sympathetic drive. [2006] 51.If therapy is initiated with a beta-blocker and a second drug is required, add a calcium-channel blocker rather than a thiazide-like diuretic to reduce the persons risk of developing diabetes. [2006] 52.If blood pressure is not controlled by Step 1 treatment, offer step 2 treatment with a CCB in combination with either an ACE inhibitor or an ARB d. [new 2011] 53.If a CCB is not suitable for step 2 treatment, for example because of oedema or intolerance, or if there is evidence of heart failure or a high risk of heart failure, offer a thiazide-like diuretic. [new 2011] 54.For black people of African or Caribbean family origin, consider an ARB in preference to an ACE inhibitor, in combination with a CCB. [new 2011] 55.Before considering step 3 treatment, review medication to ensure step 2 treatment is at optimal or best tolerated doses. [new 2011] 56.If treatment with three drugs is required, the combination of ACE inhibitor or angiotensin-II receptor blocker, calcium-channel blocker and thiazide-like diuretic should be used. [2006] 57.Regard clinic blood pressure that remains higher than 140/90 mmHg after treatment with the optimal or best tolerated doses of an ACE inhibitor or an ARB plus a CCB plus a diuretic as resistant hypertension, and consider adding a fourth antihypertensive drug and/or seeking expert advice. [new 2011] 58.For treatment of resistant hypertension at step 4: Consider further diuretic therapy with low-dose spironolactone (25 mg once daily) e. If the blood potassium level is 4.5 mmol/l or lower. Use particular caution in people with a reduced estimated glomerular filtration rate because they have an increased risk of hyperkaelemia. Consider higher-dose thiazide-like diuretic treatment if the blood potassium level is higher than 4.5 mmol/l. [new 2011] 59.When using further diuretic therapy for resistant hypertension at step 4, monitor blood sodium and potassium and renal function within 1 month and repeat as required thereafter. [new 2011]
d e
Update 2011
Choose a low-cost ARB. At the time of publication (August 2011), spironolactone did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented.
44

Guideline summary
60.If further diuretic therapy for resistant hypertension at step 4 is not tolerated, or is contraindicated or ineffective, consider an alpha- or beta-blocker. [new 2011] 61.If blood pressure remains uncontrolled with the optimal or maximum tolerated doses of four drugs, seek expert advice if it has not yet been obtained. [new 2011] 62.Provide appropriate guidance and materials about the benefits of drugs and the unwanted side effects sometimes experienced in order to help people make informed choices. [2004] 63.People vary in their attitudes to their hypertension and their experience of treatment. It may be helpful to provide details of patient organisations that provide useful forums to share views and information. [2004] 64.Provide an annual review of care to monitor blood pressure, provide people with support and discuss their lifestyle, symptoms and medication. [2004] 65.Because evidence supporting interventions to increase adherence is inconclusive, only use interventions to overcome practical problems associated with non-adherence if a specific need is identified. Target the intervention to the need. Interventions might include: suggesting that patients record their medicine-taking encouraging patients to monitor their condition simplifying the dosing regimen using alternative packaging for the medicine using a multi-compartment medicines system. (This recommendation is taken from Medicines adherence, NICE 408 clinical guideline 76). [new 2011]
Update 2011
5.4 Key research recommendations

1. Which automated blood pressure monitors are suitable for people with hypertension and atrial fibrillation? 2. In people aged under 40 with hypertension, what is the most accurate method of assessing the lifetime risk of cardiovascular events and the impact of therapeutic intervention on this risk? 3. In people aged under 40 with hypertension, what are the appropriate thresholds for intervention? 4. In adults with primary hypertension, does the use of out-of-office monitoring (HBPM or ABPM) improve response to treatment? 5. In people with treated hypertension, what is the optimal systolic blood pressure? 6. In adults with hypertension, which drug treatment (diuretic therapy versus other step 4 treatments) is the most clinically and cost effective for step 4 treatment?
45

Measuring blood pressure
6 Measuring blood pressure

For many years blood pressure has been measured using a brachial pressure cuff and auscultation of the brachial artery to identify the appearance and disappearance of Korotkoff sounds. Increasingly, automated devices for measuring blood pressure are now used in the clinic, hospitals and by people in their homes. In addition, ambulatory blood pressure measurement devices are available that are programmed to allow blood pressure to be measured repeatedly during the day and night. Blood pressure (BP) can be highly variable and this variability is due to the inherent variability in BP itself and the influence of factors such as posture, room temperature and pain/discomfort or stress. In addition there are factors related to the process of BP measurement itself that can contribute to BP variability such as the appropriateness of the cuff size, the rate of inflation and deflation of the cuff and the accuracy of the process of measurement or the automated BP monitor being used.
6.1 Techniques for measuring blood pressure

6.1.1 Manual blood pressure measurement
The cuff is inflated to block the brachial pulse. The first sound occurring with the return of the brachial pulse is the systolic pressure (the point at which the heart pumping at its hardest overcomes the pressure exerted by the cuff to push blood past the obstruction). Intermediate sounds follow as the cuff pressure drops, with muffling and then the disappearance of sounds indicating the diastolic pressure (the point at which the heart is not pumping outward and the residual arterial pressure is sufficient to overcome the pressure exerted by the cuff). The interpretation of the sounds was later developed by Ettinger.579 Three types of error have been identified for the RRK technique. Failure to accurately identify the Korotkoff sounds can lead to over or under estimation. Digit preference refers to the tendency of clinicians to round readings up or down, often to the nearest zero. Observer prejudice occurs when clinicians alter readings toward their prior expectation, a particular concern when close to a threshold which changes management.64,482 Supervised training and reassessment may help minimise errors. Systolic pressure is estimated by first palpating the brachial pulse with slow deflation of the cuff. The cuff is reinflated before listening for Korotkoff sounds. The first pass is important since sometimes the first sounds disappear as pressure is reduced (the auscultatory gap) leading to an underestimation of systolic pressure by auscultation alone. In a case series, 21% of 168 untreated hypertensive patients demonstrated an auscultatory gap.121 A number of summaries are available highlighting good technique: an adaptation of these is shown in Table 12. Table 12: Estimating blood pressure by manual auscultation
Manual auscultation Standardise the environment as much as possible: Relaxed, temperate setting, with the patient seated and rested Arm out-stretched, in line with mid-sternum and supported Correctly wrap a cuff containing an appropriately sized bladder around the upper arm and connect to a manometer. Cuffs should be marked to indicate the range of permissible arm circumferences; these marks should be easily seen when the cuff is being applied to an arm. Palpate the brachial pulse in the antecubital fossa of that arm. Rapidly inflate the cuff to 20 mmHg above the point where the brachial pulse disappears. Deflate the cuff and note the pressure at which the pulse reappears: the approximate systolic pressure. Re-inflate the cuff to 20 mmHg above the point at which the brachial pulse disappears.
46

Measuring blood pressure Manual auscultation Using one hand, place the stethoscope over the brachial artery ensuring complete skin contact with no clothing in between. Slowly deflate the cuff at 23 mmHg per second listening for the Korotkoff sounds. Phase I: The first appearance of faint repetitive clear tapping sounds gradually increasing in intensity and lasting for at least two consecutive beats: note the systolic pressure. Phase II: A brief period may follow when the sounds soften and or 'swish'. Auscultatory Gap: In some patients the sounds may disappear altogether. Phase III: The return of sharper sounds becoming crisper for a short time. Phase IV: The distinct, abrupt muffling of sounds, becoming soft and blowing in quality. Phase V: The point at which all sounds disappear completely: note the diastolic pressure. When the sounds have disappeared, quickly deflate the cuff completely if repeating the measurement. When possible, take readings at the beginning and end of consultations.
There has been some controversy as to whether phase IV or phase V sounds should be used to record diastolic blood pressure. Commonly, the difference in pressure between phase IV and V is less than 5 mmHg but occasionally can be substantial. Phase V can be absent with sounds audible to zero cuff pressure notably in some children, during pregnancy, with anaemia, aortic insufficiency and with elderly people. Phase V correlates better with direct measurement, is commonly used in clinical trials of antihypertensive therapies, and is more reproducible when assessed by different observers. There is now general consensus that phase V should be taken as the diastolic pressure except when absent.
27,64,99
6.2 Cuffs
Modern cuffs consist of an inflatable cloth-enclosed bladder which encircles the arm and is secured by Velcro or by tucking in the tapering end. The width of the bladder is recommended to be about 40%, and its length 80%, of the arm circumference. Manufacturers are now required to provide markings on the cuff indicating the arm circumference for which it is appropriate (BS EN 1060-1) 21; these marks should be easily seen when the cuff is being applied to an arm. When the bladder is too small (under-cuffing) it is possible to overestimate blood pressure. The existence of over-cuffing and consequent underestimation is contentious although likely to be of smaller magnitude.482,553,636
6.3 Conditions and environment

Blood pressure is maintained by a combination of mechanical, neuronal and endocrine selfregulating systems in the body. These systems can alter blood pressure in response to changes in environment. Individual readings are influenced (for example) by age, ethnicity, disease, the time of day, posture, emotions, exercise, meals, drugs, fullness of bladder, pain, shock, dehydration, acute changes in temperature and changes in altitude. These influences can be substantial, altering systolic readings by as much as 20 mmHg.65 Standardising the environment in which blood pressure measurements are made reduces variation and enhances the interpretation of a series of readings taken over time.27,99 A quiet, comfortable location at normal room temperature is optimal. Ideally, the patient should not need to pass urine, not recently have eaten, smoked or taken caffeine or exercise. Allowing the patient to rest at least five minutes before measurement is also advised.27,65,99
47

Blood pressure readings tend to increase as patients move from the supine to standing position. The change may not be significant, but it is traditional for measurements to be taken whilst seated. Certain patients demonstrate a significant lowering of blood pressure when standing (postural hypotension).27,65,66,99,452 Blood pressure readings also tend to increase as the patient's arm is lowered below the horizontal and decrease when the arm is raised. When blood pressure is measured in the clinic setting, the patients arm should be out-stretched, level with their heart and in line with their mid sternum, and supported by a table or some other means.27,65,66,99,452 Blood pressure is usually measured in the nondominant arm, especially when using home or ambulatory monitoring. Differences in readings may occur between arms. A BP difference of <10mmHg can be considered normal, however, a difference of more than 20mmHg between arms is unusual, occurring in <4% of people and is usually associated with underlying vascular disease. Clinicians are advised to take readings in both of the patient's arms initially, and use the arm with the higher reading for subsequent measurements of blood pressure. . Consistent inter-arm differences of over 20/10 mmHg may suggest pathology warranting specialist referral.27,65,99
6.4 White Coat Hypertension

The observation that clinicians (signified by their white coats) can cause spuriously high blood pressure readings in patients was first described in the 1940s.58 Additionally, sympathetic symptoms such as sweating, tachycardia and palpitation sometimes occur. The effect is short-lived with blood pressure dropping to normality after or near the end of the consultation. Consequently, a patient may present as hypertensive in clinic (in a primary or secondary care setting) but be normotensive otherwise. White Coat Hypertension (WCH) is reported to occur in as many as 15% to 30% of the population,448 although this may be inflated due to inadequate evaluation of patients. It is more common in pregnancy and with increasing age although poorly understood otherwise.569 The size of white coat effect in individuals can vary over time and a small proportion (4%) may demonstrate atypical very high clinic readings.27 Failing to identify WCH makes inappropriate treatment for hypertension in normotensive patients a possibility. Similarly, hypertensive individuals can also exhibit WCH and may receive inappropriate dose titrations or additional antihypertensive agents.490,506,635 Patients have historically been enrolled in trials using clinic BP values, and these trials will almost certainly have included a proportion of patients with WCH. It is unknown whether benefits of treatment differ substantially in those with or without WCH. White Coat Hypertension: A difference between clinic BP and home or ambulatory blood pressure averages is expected. This difference has been reported to average approximately 10/5mmHg but this will vary considerably and is usually greater in people with a higher baseline blood pressure and as people age. White coat hypertension is defined when a patient has a persistently elevated clinic BP and a normal home or ambulatory BP day time average, i.e. <135/85mmHg. White coat Effect in people with hypertension: People with true hypertension, treated or untreated, can also exhibit a White Coat Effect, for example a clinic BP reading that is disproportionately greater than their home or ambulatory BP averages, but their home or ambulatory BP averages are in a hypertensive range. Such patients are at risk of receiving more BP medication than they need and will require out of office measurement to monitor the efficacy of their BP treatment.
Update 2011
48

6.5 Blood pressure measurement devices

There is considerable guidance about the range of appropriate devices for measuring blood pressure.100,171,446 and about their maintenance and periodic recalibration [172 Local medical physics and biomedical/clinical engineering departments can often give further advice.
6.5.1
Mercury sphygmomanometer
The mercury sphygmomanometer has been used for the traditional measurement of blood pressure. It is reliable and provides the reference standard for indirect measurement. However it is bulky, fragile and there are particular safety and economic concerns about the toxic effects of mercury. Mercury is being phased out of clinical use and mercury sphygmomanometers have already been removed from clinical areas in hospitals and primary care. Thus, alternatives to mercury sphygmomanometry are now required for routine clinical use. Non-mercury devices that operate in a similar way to the traditional mercury column devices are available and provide a suitable alternative to mercury devices when manual auscultation is required to measure blood pressure.
6.5.2
Aneroid sphygmomanometers
Aneroid sphygmomanometers measure pressure using a lever and bellows system. They may be less accurate than mercury sphygmomanometers and their alternatives (see above), especially over time. Using the manual auscultation technique they are subject to the same sources of observer error.64
6.5.3
Automated devices
Automated devices are increasingly being used in hospitals and primary care. All sphygmomanometers need regular maintenance. Rubber tubing can crack and leak making cuff deflation hard to control, underestimating systolic and overestimating diastolic readings. Faulty valves can cause similar problems.64
6.6 Ambulatory blood pressure monitors

Ambulatory Blood Pressure monitoring (ABPM) involves a cuff and bladder connected to electronic sensors which detect changes in cuff pressure and allow blood pressure to be measured oscillometrically. The cuff is inflated by a battery powered compressor and sensors within the cuff detect changes in pressure oscillations during cuff deflation. Systolic and diastolic pressure readings are deduced from the shape of these oscillometric pressure changes using an algorithm built into the measuring device. Developed as a research tool in the 1960s, these devices have considerably reduced in size and now can be described properly as ambulatory. Thus a patient's blood pressure can be automatically measured at repeated intervals (commonly every 30 minutes) throughout the day and night, while they continue routine activities. Systolic and diastolic pressure can be plotted over time, with most devices providing average day, night and 24 hour pressures.448 (see Figure 2, page 41) An advantage of ABPM is the removal of observer error with automated reading. However, oscillometric measurement may be difficult in the presence of arrhythmias, particularly rapid atrial fibrillation, and in a subgroup of the general population in whom oscillometric readings are inaccurate for unknown reasons.445,448 A number of ABPM devices are available varying in size, weight, noise level, data manipulation and cost.450,452 Devices should be independently validated to one or both of two internationally accepted standards from the British Hypertension Society and the Association for the Advancement of Medical
Update 2011
49

Instrumentation.41,447,451 See British Hypertension Society website www.bhsoc.org for a list of validated monitors. When using ABPM, patients need some understanding of how the device works and instruction about manual deflation, missed readings, arm position, and machine location: fitting takes 1530 minutes. An appropriately sized cuff is necessary as with non-ambulatory monitoring and if one arm gives a higher reading at baseline then this should be used subsequently. Patients may be asked to make diary records of events that are known to affect blood pressure so that readings can be related to them, for example, periods of sleep. Sleeping times can be recorded or fixed times may be predefined, including preparing for sleep (e.g. 9pm midnight) and waking up (e.g. 6am 9 am).448,450
6.7 Home blood pressure monitors

Home monitoring devices are oscillometric, measuring BP on the upper arm, the wrist or the finger. Home monitoring potentially offers some similar benefits to ABPM. Frequent measurement produces average values that may be more reproducible and reliable that traditional clinic measurement. Potentially, white coat hypertension, systematic error, terminal digit preference and observer prejudice can be removed.104,449,556 Home monitoring allows patients to assess their own response to antihypertensive medication, which may increase compliance with treatment. It has been argued that better evaluation provided by home monitoring may reduce unnecessary treatment, increase compliance and thus deliver cost savings.490,556 Home blood pressure devices are thought by some professionals to cause anxiety or obsessive self interest.449,452,556,569 Potential disadvantages stem from the need for appropriate training to avoid biased measurement. Use of inappropriately sized cuffs, isometric exercise when not resting the arm, measurement after or during exercise and observer prejudice (for non-automated recording) are possible.27 One study found that only 30% of patients using a manual home blood pressure monitor correctly adhered to the protocol. Further, less than 70% of the self-reported measurements were identical to those simultaneously recorded by the machine.303 Observer bias was more apparent in those patients who were more hypertensive or whose readings showed more variation. As with ABPM, home monitoring devices are oscillometric and may have difficulty measuring pressure in cases of arrhythmias, and in certain patients for no apparent reason. See British Hypertension Society website www.bhsoc.org for a list of validated monitors.
Update 2011
6.8 Recommendations
1. Healthcare professionals taking blood pressure measurements need adequate initial training and periodic review of their performance. [2004] 2. Because automated devices may not measure blood pressure accurately if there is pulse irregularity (for example, due to atrial fibrillation), palpate the radial or brachial pulse before measuring blood pressure. If pulse irregularity is present, measure blood pressure manually using direct auscultation over the brachial artery. [new 2011] 3. Healthcare providers must ensure that devices for measuring blood pressure are properly validated, maintained and regularly recalibrated according to manufacturers instructions. [2004] 4. When measuring blood pressure in the clinic or in the home, standardise the environment and provide a relaxed, temperate setting, with the person quiet and seated, and their arm outstretched and supported. [new 2011]
50

5. If using an automated blood pressure monitoring device, ensure that the device is validated f and an appropriate cuff size for the persons arm is used. [new 2011] 6. In people with symptoms of postural hypotension (falls or postural dizziness): measure blood pressure with the person either supine or seated. measure blood pressure again with the person standing for at least a minute prior to measurement. [2004, amended 2011] 7. If the systolic blood pressure falls by 20 mmHg or more when the person is standing: review medication measure subsequent blood pressures with the person standing consider referral to specialist care if symptoms of postural hypotension persist. [2004, amended 2011]
6.9 Research recommendation

1. Which automated blood pressure monitors are suitable for people with hypertension and atrial fibrillation? Atrial fibrillation is common in older people and may prevent accurate blood pressure measurement with automated devices. It would be valuable to know if this can be overcome.
Update 2011
A list of validated blood pressure monitoring devices is available on the British Hypertension Societys website (see www.bhsoc.org). The British Hypertension Society is an independent reviewer of published work. This does not imply an endorsement by NICE.
51

D iagnosis of Hypertension
7 Diagnosis of Hypertension
Hypertension is diagnosed and subsequently treated to reduce the risk of developing stroke, ischaemic heart disease, heart failure, peripheral vascular disease, renal disease, dementia and premature death. A persons risk is not only determined by their blood pressure but also by the presence of target organ damage, established cardiovascular disease and other risk factors for cardiovascular disease such as lifestyle (e.g. diet, smoking, obesity and lack of exercise), diabetes and dyslipidaemia . The assessment of a person when contemplating a clinical diagnosis of hypertension must take account of these additional factors which are discussed in Chapter 8 of the guideline. Blood pressure is highly variable and the 2004 guidance emphasised that hypertension should not be diagnosed nor treatment offered on the basis of a single BP measurement. Consequently, people with suspected hypertension have been required to undergo repeated measurements of their clinic BP on repeated clinic visits to confirm or refute the diagnosis of hypertension. The exception being the rarer occasions when patients present with severe elevations of BP, usually associated with evidence of target organ damage, when treatment is needed more urgently. The emergence of automated BP monitoring, either for home use, or ambulatory BP monitoring devices, has revealed that there can be marked discrepancies between clinic BP measurement and home or ambulatory BP averages , which are known as either white coat hypertension (see 6.4) or masked hypertension (where clinic BP is normal but ABPM and/or HBPM measurements are elevated). The identification of these discrepancies has prompted consideration as to whether the conventional clinic blood pressure measurement method is still the most accurate at predicting the risk of future cardiovascular disease and establishing the diagnosis of hypertension.
Update 2011
7.1 Predicting outcome using clinic, home and ambulatory measurements

Review question: In adults with suspected primary hypertension, what is the best method to measure blood pressure (HBPM versus ABPM versus CBPM) to predict the development of cardiovascular events?
7.1.1
Clinical evidence 2004

If clinic blood pressure measurements are inaccurate this may weaken the relationship between blood pressure and cardiovascular risk. Studies were systematically identified and retrieved that prospectively compared the ability of ambulatory, home and clinic measures of blood pressure to predict fatal or non-fatal cardiovascular events. Studies addressing markers of evolving disease, such as left ventricular mass or hypertrophy, were not included because of their uncertain relationship with patient outcome. Details of six reports relating to four cohorts of patients were abstracted. Studies were conducted in London, England,324 Ohasama, Japan,465,523 Umbria, Italy,526,613-615 and the final cohort was provided by European patients enrolled in a drug trial.557 Two further studies are ongoing.87,385,472 The four cohorts included about 4,500 participants; approximately 50% of participants were male and their mean age was nearly 55 years. Most participants were Caucasian or Japanese reflecting the location of the studies. The mean length of follow-up was five years. The British study investigated ambulatory blood pressure using an intra-arterial cannula, and thus its findings may not generalise to indirect ambulatory measurement. This limitation accepted, 24 hour, day or night direct measurements predicted cardiovascular events whereas clinic measurement did not.
52

The Ohasama study compared self-measured home BP and clinic BP. Neither method demonstrated superior prediction of first stroke, although home measurement appeared to be a better predictor of cardiovascular mortality. In the Italian cohort, ambulatory 24-hour systolic blood pressure was a better predictor than clinic assessment for cardiovascular morbidity and mortality. The analysis suggested that white coat hypertension and nocturnal dipping are independently associated with the risk of cardiovascular disease, the implication being that those not demonstrating a white coat effect or nocturnal dipping are at greater risk. It is plausible that a nocturnal reduction in blood pressure may protect target organs, although the definition of 'non-dippers' currently varies between studies (examples include a mean nocturnal pressure fall of less than 10% or an absolute reduction of less than 10/5 mmHg). Varying definitions, as well as classification of day and night periods, may explain differences in the prevalence of non dippers seen in studies. The SYST-EUR trial enrolled 4,695 patients into a trial comparing calcium-channel blocker initiated blood pressure control and placebo. A sub-study conducted in 46 of the 198 participating centres compared the prognostic value of ambulatory and clinic blood pressure readings. When treatment and placebo groups were taken together, this study provided no evidence that ambulatory values more accurately predicted cardiovascular morbidity or mortality than clinic readings. Combining the evidence from these four cohorts, the difference in prognostic accuracy of home, ambulatory and clinic measures appears small and inconsistent. None of these studies adequately described their approach to analysing their data or the statistical robustness of models produced. A further potential confounder was the adequacy of clinic baseline measurements. It is possible that SYST-EUR, which had better baseline clinic assessment, minimised the 'regression to the mean' phenomenon and obtained more representative values. On the other hand, it is clear from large epidemiological studies that there is a very precise relationship between periodic clinic based blood pressure measurements and risk of cardiovascular disease.361,379
7.1.2
Clinical evidence 2011

Three pooled analyses of prognostic studies210,254,326 and 11 individual prognostic studies77,86,159,178,211,253,284,404,438,564 were found that fulfilled the inclusion criteria and looked at the ability of clinic, home or ambulatory blood pressure measurements to predict outcomes. Outcomes of interest were mortality, stroke, MI, heart failure, diabetes, vascular procedures, hospitalisation for angina, and other major adverse cardiac and cerebrovascular events (MAACE). The three pooled analyses210,254,326 were meta-analyses of individual data from prospective studies. The individual studies included in these pooled analyses were excluded from our review in order to avoid duplication / double counting of data. Two of the pooled analyses254,326 used data from four studies of random populations with longitudinal follow-up of fatal and non-fatal CV outcomes. They both included the same studies, however the people they included in the final analyses were different (one study326 excluded people with no night-time data available, and the other study254 excluded people with no daytime data available). The third pooled analysis210 used data from three studies in the Belgian Ambulatory Blood Pressure Monitoring database (which contains individual data of HT patients from studies performed in Europe and coordinated by the university of Ghent or Leuven). Patients had a history of CV disease. All prognostic studies were observational and were found to be methodologically sound / have a low risk of bias (see quality assessment summary tables in appendix F). Studies that were published before 2003 (the cut-off date of the original guideline, CG18436) were excluded. Studies were categorised into those which compared: Home versus clinic measurements (five studies)86,211,438,534,564 ABPM versus clinic measurements (11 studies)77,159,178,210,253,254,284,326,404
Update 2011
53

ABPM versus home versus clinic measurements (two studies)211,534 Four studies were conducted in people who were known or suspected to have hypertension86,159,178,404 and the rest of the studies were in population samples which would have contained both hypertensive and non-hypertensive people. Mixed population studies are a better representation of how BP monitoring would be used in clinical practice and the prognostic ability of the blood pressure measurement methods to determine clinical outcome. NOTE: The Hansen 2007 study254 only assessesd daytime ABPM measurements; the Dawes 2006 study159 only assessed 24h ABPM measurements; and the Fagard 2005 and Fagard 2008 studies210,211 only assessed daytime and night-time ABPM, and not 24h measurements. All other studies assessed and compared separately all three types of ABPM measurements - 24h, daytime and night-time). The protocol used for measuring blood pressure (for example, the intervals between each ABPM reading and definitions of daytime and night-time periods) varied between studies.
Update 2011
7.1.3
Evidence statements clinical

The table below (Table 13) summarises the overall results of the prognostic studies included for this review. Table 14summarises the numerical results for selected outcomes of the prognostic studies included for this review. The full data for all outcomes can be found in the evidence tables in the appendix. NOTE: The best method was chosen as the method of measuring BP that best predicted (ie. statistically significant predictors and higher HR values) clinical outcomes (after adjustment for covariates in multivariate analyses).
54
Table 13: Summary of included prognostic studies

Study Home vs clinic Bobrie 2004
86
N 4939
438
Follow-up time Mean 3.2 years Mean 6.8 years
Outcome CV events Mortality and CV events
Best method Home Home
Representative of real life home BP measurements? Yes measurements over 4 days Yes measurements over 7 days; but home BP threshold (for HT diagnosis) not given Yes measurements over 3 days; but small study , and home BP threshold (for HT diagnosis) not given
Niiranen 2010
2081
Stergiou 2007
564
665
Mean 8.2 years
CV events
NS difference
Update 2011
ABPM vs clinic Bjorklund 2004

77
872
Mean 6.6 years
CV morbidity
SBP: Office and ABPM (daytime SBP added more) ABPM (daytime) ABPM (especially night-time) ABPM (especially night-time) ABPM ABPM (CV events); but no difference for mortality (total and CV) ABPM (night-time DBP) No difference
n/a
Dawes 2006 Dolan 2005
159
10,129 5292 302 1700 7030
Median 10 years Mean 7.9 years Median 6.8 years Up to 9.5 years Median 9.5 years
Mortality CV mortality Mortality, CV mortality, CV events Mortality and CV mortality CV death, stroke, cardiac events and CHD
n/a n/a n/a n/a n/a
178
Fagard 2008* Hansen 2005
210
253 254
Hansen 2007*
Ingelsson 2006 Kikuya 2007*
284
951 5682
Up to 9.1 years Median 9.5 years
CHF CV death, stroke, cardiac events
n/a n/a
326
Study Mesquita-Bastos 404 2010 Home vs ABPM vs clinic Fagard 2005

211
N 1200
Follow-up time Mean 8.2 years
Outcome and CHD CV events and stroke
Best method ABPM (especially night-time) Home equal to ABPM and better than office No difference
Representative of real life home BP measurements? D iagnosis of Hypertension
n/a
391
Median 10.9 years
Major CV events
No home BP measurement performed y investigator rather than patient. No only measured home BP on 1 day; home BP threshold (for HT diagnosis) not given
Update 2011
USega 2005
534
2051
Mean 10.9 years
Mortality

Table 14: Summary of numerical results for prognostic studies (selected outcomes)
Study Home vs clinic Bobrie 2004
86
Outcome CV events
Best method Home
HR (95% CI) for SBP measurement Home: 1.02 (1.01, 1.02) p=<0.001 Clinic: 1.01 (1.00, 1.01) p=0.09 Per 1mmHg rise in SBP Home: 1.22 (1.09, 1.37) p<0.001 Clinic: 1.01 (0.92, 1.12) p=0.80 per 10mmHg rise in SBP Home: 1.00 (0.99, 1.02) p=0.68 Clinic: 1.01 (0.99, 1.03) p=0.08 Per 1mmHg rise in SBP ABPM (24h): 1.23 (1.07, 1.42) p<0.05 ABPM (daytime): 1.23 (1.07, 1.42) p<0.05 Clinic: 1.21 (1.04, 1.41) p<0.05 per 1SD rise in SBP ABPM (daytime): 1.51 (1.25, 1.83); p<0.001 Clinic: 1.02 (0.84, 1.24); p=0.90 highest quartile of SBP compared to ?lowest
Niiranen 2010
438
CV events
Home
Stergiou 2007
564
CV events
No difference
ABPM vs clinic Bjorklund 2004

77
CV morbidity
SBP: Office and ABPM (daytime SBP added more) ABPM (daytime) ABPM (especially night-time) ABPM (especially night-time) ABPM
Dawes 2006
159
Mortality
Dolan 2005
178
Update 2011
CV mortality
ABPM (24h): 1.19 (1.14, 1.26) p<0.001 ABPM (night-time): 1.21 (1.16, 1.27) p<0.001 Clinic: 1.06 (1.02, 1.10) p<0.01 per 10mmHg rise in SBP ABPM (24h): 1.20 (0.91-1.58) NS ABPM (daytime): 1.03 (0.77-1.36) NS ABPM (night-time): 1.34 (1.06-1.69) p<0.01 Per 1SD rise in SBP ABPM (24h): 1.51 (1.28, 1.77) p<0.0001 ABPM (daytime):1.50 (1.27, 1.76) p<0.0001 Clinic: 1.25 (1.10, 1.42) p<0.001 per 10mmHg rise in SBP Cardiac events ABPM (daytime): 1.13 (1.04, 1.23) p<0.0001 Cardiac events Clinic: 1.06 (0.99, 1.13) p>0.05 CV events ABPM (daytime): 1.17 (1.10, 1.24) p<0.0001 CV events Clinic: 1.05 (1.00, 1.10) p>0.05 per 10mmHg rise in SBP ABPM (24h): 1.13 (0.91, 1.40) p>0.05 ABPM (night-time): 1.21 (0.98, 1.49) p>0.05 Clinic: 1.25 (0.98, 1.59) p>0.05 per 1SD rise in SBP ABPM (24hrs): 1.20 (1.13, 1.27) p<0.0001 ABPM (daytime): 1.16 (1.09, 1.23) p<0.0001 Clinic: 1.09 (1.04, 1.15) p<0.001 per 10mmHg rise in SBP
Fagard 2008*
210
CV events
Hansen 2005
253
CV mortality
Hansen 2007*
254
Cardiac events / CV events
ABPM (CV events); but no difference for mortality (total and CV) ABPM (night-time)
Ingelsson 284 2006
CHF
Kikuya 2007*
326
Cardiac events
No difference
57

D iagnosis of Hypertension Best method ABPM (esp. night-time)
Study Mesquita-Bastos 404 2007
Outcome CV events
HR (95% CI) for SBP measurement ABPM (24h): 1.41 (1.20-1.65) <0.001 ABPM (daytime): 1.33 (1.10-1.60) <0.01 ABPM (night-time): 1.57 (1.32-1.86) p<0.001 Per 1SD rise in SBP Home: 1.32 (1.06, 1.64) p=0.01 ABPM (daytime): 1.33 (1.07, 1.64) p<0.01 ABPM (night-time): 1.42 (1.16, 1.74) p<0.001 Clinic: 1.13 (0.88, 1.45) p=0.34 Per 1mmHg rise in SBP No HRs given, but all entry BP values had a direct exponential relationship with the risk of all-cause death or CV death Goodness of fit of the relationship of BP to risk of death (CV and all-cause) was not less for clinic, compared to home and ambulatory. Coefficient ABPM (24h): 0.0557 0.0008 p<0.0001 ABPM (daytime): 0.0479 0.008 p<0.0001 ABPM (night-time): 0.0559 0.007 p<0.0001 Coefficient the increase in risk per 1mm Hg increase in SBP
Home vs ABPM vs clinic Fagard 2005

211
Major CV events
Home equal to ABPM and better than office No difference
Sega 2005
534
Mortality
Summary Studies showed that for predicting clinical outcomes: ABPM versus CBPM (nine studies): ABPM was superior to CBPM (eight studies) There was no difference between ABPM and CBPM (one study) HBPM versus CBPM (three studies): HBPM was superior to CBPM (two studies) There was no difference between HBPM and CBPM (one study) HBPM versus ABPM versus CBPM (two studies): HBPM was similar to ABPM and both were superior to CBPM (one study) There was no difference between HBPM, ABPM and CBPM (one study)
Update 2011
7.2 Sensitivity and specificity of clinic, home and ambulatory measurements

Review question: In adults with suspected primary hypertension, what is the best method to measure blood pressure (HBPM versus ABPM versus CBPM) to establish the diagnosis of hypertension?
7.2.1
Clinical evidence
One systematic review/meta-analysis275 was found that fulfilled the inclusion criteria and looked at the best method of measuring blood pressure for diagnosing hypertension. Studies were included in
58

the SR/MA if they were: RCTs, adult population (all ages), all settings except hospitalised (the main focus was to be on primary care). Studies were excluded from the SR/MA (unless these groups could be excluded from other data within a paper) if they: did not specify the diagnostic thresholds used, had spectrum bias (no normotensives or hypertensives in one measurement group), patients were pregnant, hospitalised, or were receiving treatment at the time of the comparison. The systematic review/meta-analysis included 20 studies (N=5863) and compared the sensitivity and specificity of CBPM and HBPM measurements (using ABPM as the reference standard as ABPM has been shown to be the best blood pressure method for indicating prognosis). The systematic review/meta-analysis was of good quality, however the quality of the studies it included ranged from poor to good. The population included in the 20 studies consisted of: primary care primary care at risk secondary care the general population general population at risk community volunteers The 20 studies included in the SR/MA differed in terms of: Mean age (range <33 to 60 years) Gender: % male (range 16 to 69%) Sample size (range N=16 to N=2370) Mean baseline BP of population Sensitivity (Home vs ABPM range 0.48 to 0.91; clinic vs ABPM range 0.17 to 1.0) Specificity (Home vs ABPM range 0.34 to 0.92; clinic vs ABPM range 0 to 0.98) Number of measurements for ABPM (range: 24 to 111 in the daytime) Number of measurements for clinic BP (range: 2 to 18) Number of measurements for home BP (range: 18 to 56) Period of ambulatory measurement (range: 6 to 24 hours) BP thresholds used (range: ABPM SBP 91-144 mmHg; clinic SBP 90 to 160 mmHg; home SBP 127 to 140 mmHg)) Quality assessment (QUADAS criteria) of the included studies showed that they: had good reporting of attrition had good selection criteria of participants had reporting bias: all studies had lack of clarity of reporting avoided both partial and differential verification bias (i.e. all patients in the studies received the same comparison measurement tests, regardless of initial results) used validated devices for all strands of monitoring: 11/20 studies limited evidence of blinding to previous BP results from monitoring assessors NOTE: only 10 of the 20 studies were ultimately included in the meta-analysis of data. Only studies with the same reference test threshold and same index test threshold were pooled and included in the meta analysis. Eight studies used a 135/85 mmHg ABPM threshold and a 140/90 mmHg clinic BPM threshold to diagnose hypertension, whilst three studies used a threshold of 135/85 mmHg for both ambulatory and home diagnosis. However, one of the clinic comparison studies used the full 24 hour mean ABPM rather than mean daytime readings and was therefore not comparable to the others and excluded from the analysis.
Update 2011
59

7.2.2

One SR/MA275 found the following sensitivities and specificities for CBPM and HBPM when using ABPM as the reference standard (Table 15): Table 15: CBPM and HBPM for diagnosing Hypertension. The thresholds used in the SR/MA for diagnosis were: ABPM (daytime) 135/85 mmHg; clinic BP 140/90 mmHg; home BP 135/85 mmHg.
Clinic / ABPM 219,461,540,566,567,602,603 (7 studies) 74.62 (60.72, 84.83) 74.61 (47.88, 90.38) Home / ABPM 62,167,567 (3 studies) 85.65 (77.95, 90.97) 62.44 (47.98, 74.98) Statistical significance (pvalue) NS (p-value not reported) NS (p-value not reported)
Parameter / BP test Sensitivity,% % (95% CI) Specificity, % (95% CI)
Clinic versus Home BP (Table 15): o there was NS difference between the BP measurement methods for sensitivity or specificity In a sensitivity analysis for CBPM which included only studies with mean BPs close to or above the diagnostic threshold (ie. a typical general practice screening population with no normotensives): CBPM sensitivity increased to 85.6% (CI 81.0 to 89.2) and specificity decreased to 45.9 (CI 33.0 to 59.3). o NOTE: The home BP studies already used a typical general practice screening population with no control group of normotensives and so the values remained the same.
Update 2011
This made HBPM the same as CBPM for sensitivity but better for specificity Clinic BP thresholds (140/90 mmHg vs 150/90 mmHg);Table 16: sensitivity decreased with increasing BP threshold, however, the change was NS. specificity increased with increasing BP threshold, however, the change was NS. Home BP thresholds (135/85 mmHg vs 140/90 mmHg and 130/80 mmHg);Table 16: Sensitivity significantly decreased with increasing threshold Specificity significantly increased with increasing threshold Summary: Home BP is a better measurement than clinic BP for diagnosing HT (in a typical general practice screening population), but is not as good as ABPM. A higher BP threshold (for clinic BP) resulted in worse sensitivity and better specificity for diagnosing HT (compared to the current standard threshold used for diagnosis: 140/90 mmHg), however the effect was NS. A higher BP threshold (for home BP) resulted in a significantly worse sensitivity and significantly better specificity for diagnosing HT (compared to the current standard threshold used for diagnosis: 135/85 mmHg) A lower BP threshold (for home BP) resulted in significantly better sensitivity and significantly worse specificity for diagnosing HT (compared to the current standard threshold used for diagnosis: 135/85 mmHg)
60

Table 16: CBPM and HBPM sensitivity and specificity of different thresholds for diagnosing Hypertension. The thresholds used in the SR/MA for diagnosis by ABPM (daytime) was 135/85 mmHg.
Test threshold (referm=nces not provided in SR/MA) 140/90 (n=7) 150/90 (n=1) Relative specificity, % (95% CI) 1.00 (reference) 1.15 (0.71 to 1.88), p=0.57 1.42 (1.20 to 1.68), p<.0001 1.00 (reference) 0.73 (0.57 to 0.93), p=0.01
Sensitivity, % (95% CI) 74.73 (61.73 to 84.43) 66.34 (28.28 to 90.79) 52.56 (34.71 to 69.78) 83.15 (76.09 to 88.45) 91.75 (84.37 to 95.82)
Relative sensitivity, % (95% CI) 1.00 (reference) 0.89 (0.51 to 1.55), p=0.68 0.63 (0.45 to 0.88), p=0.01 1.00 (reference) 1.10 (1.03 - 1.18), p=0.01
Specificity, % (95% CI) 74.75 (49.82 to 89.82) 86.16 (24.80 to 99.16) 80.32 (67.88 to 88.74) 56.68 (46.42 to 66.40) 41.35 (30.13 to 53.53)
Clinic BP thresholds
Home BP thresholds 140/90 (n=1) 135/85 (n=3) 130/80 (n=1)
7.3 Cost-effectiveness of clinic, home and ambulatory measurements

7.3.1 Economic evidence literature review
An economic evaluation should ideally compare all relevant alternatives. No studies were identified comparing all of clinic blood pressure monitoring (CBPM), ambulatory blood pressure monitoring (ABPM) and home blood pressure monitoring (HBPM) at diagnosis. One study (Krakoff 2006338) was identified that examined the cost effectiveness of ABPM compared with CBPM in the diagnosis of hypertension. This is summarised in the ABPM versus CBPM economic evidence profile below (Table 17, Table 18). A full evidence table is also provided in Appendix G: Evidence tables health economic studies (2011 update). One study was identified that examined HPBM and CBPM in the diagnosis of hypertension but was excluded as it was judged to have serious methodological limitations.225 Table 17: ABPM versus CBPM (diagnosis) economic study characteristics
Study Krakoff 2006 USA
338
Update 2011
Applicability Partially applicable(a)
Limitations Potentially serious(b)
Other Comments CBPM diagnosed population. CBPM vs CBPM+ABPM at diagnosis. Decision analytic model incorporating prevalence of white coat hypertension, rate of conversion to true hypertension and drop-out rate from treatment. 5-year time horizon. Costs: ABPM (diagnosis and annual follow-up) and hypertension treatment.
a) Does not incorporate all relevant comparators. Does not incorporate health effects (possibly conservative towards ABPM).Some uncertainty about the applicability of USA costs. Discounting not applied. b) Source of prevalence of white coat hypertension unclear but varied in sensitivity analysis (15-20%). Limited sensitivity analysis.
61

Table 18: ABPM versus CBPM (diagnosis) economic summary of findings (mean per person)
Study Krakoff 2006 USA
338
Incremental cost () -80(a)
Incremental effects N/a
ICER N/a
Uncertainty -28 to -132(b)
a) Converted from 2005 US dollars. b) Two way sensitivity analysis varying white coat hypertension rate 15%-20% and the annual conversion rate of white coat hypertension to true hypertension 5%-20%.
7.3.2
Economic evidence - original economic analysis

The GDG considered the clinical evidence reviewed as part of the guideline update to suggest that ambulatory blood pressure monitoring (ABPM) may be more accurate at diagnosing patients with hypertension than clinic blood pressure monitoring (CBPM) or home blood pressure monitoring (HBPM); however it is also the most expensive option in terms of monitor costs. HBPM was found to be more specific than CBPM but was also associated with additional monitor costs. The use of ambulatory or home monitoring instead of clinic monitoring to confirm a diagnosis of hypertension was identified as the highest economic priority by the GDG due to it being a significant change in practice that would require considerable investment in new devices by primary care. As described above, no cost-effectiveness analyses comparing all of ABPM, HBPM and CBPM were identified from the published literature. A protocol for a cost-effectiveness analysis in development was submitted, in response to the call for evidence in this area (see Methods), by a UK research group g who had also undertaken a systematic review and meta analysis of the sensitivity and specificity of CBPM and HBPM compared to ABPM that was included in the guideline as part of the clinical evidence review275. However, the cost-effectiveness analysis would not be completed within the timeframe of the guideline update and so a collaboration was agreed between the GDG and the research group. Below is a summary of the analysis that was undertaken. For full details please see Appendix J: Costeffectiveness analysis).
Update 2011
7.3.2.1
Methods A cost-utility analysis was undertaken to look at different blood pressure monitoring methods for confirming a diagnosis of hypertension. A Markov model was used to estimate lifetime qualityadjusted life years (QALYs) and costs from a current UK NHS and personal social services perspective. Both costs and QALYs were discounted at a rate of 3.5% per annum in line with NICE methodological guidance427. Uncertainty was explored through probabilistic analysis and extensive sensitivity analyses. The population used for the analysis was people with suspected hypertension those with a screening clinic blood pressure measurement equal or above 140/90 mmHg. Analyses were run for ten gender and age (40, 50, 60, 70, 75 years) stratified subgroups. The comparators selected for the model were confirmation of diagnosis with: Clinic blood pressure monitoring (CBPM) Home blood pressure monitoring (HBPM)
g Richard McManus, Professor of Primary Care Cardiovascular Research, University of Birmingham; Sue Jowett, Senior Lecturer in Health Economics, University of Birmingham; Pelham Barton, Reader in Mathematical Modelling, University of Birmingham; James Hodgkinson, Research Fellow, University of Birmingham; Jonathan Mant, Professor of Primary Care Research, University of Cambridge; Una Martin, Reader in Clinical Pharmacology, University of Birmingham; Carl Heneghan, Reader in Evidence-Based Medicine, University of Oxford; Richard Hobbs, Head of Primary Care Clinical Sciences, University of Birmingham.
62

Ambulatory blood pressure monitoring (ABPM) The population entering the model comprised people suspected of having hypertension based on a screening clinic blood pressure reading. This group therefore included both those that were truly hypertensive (true positive following screening) and those that were not (false positive following screening). The diagnosis process aimed to correctly confirm both true hypertensives (in order to reduce their cardiovascular risk via treatment) and true normotensives (in order to reduce unnecessary treatment). The key differences between diagnostic options were their ability to accurately diagnose both these groups. One of the key inputs in the model was therefore the sensitivity and specificity of the different diagnostic options and this was based on the meta analysis275 included as clinical evidence in the guideline. In addition the comparators varied in terms of the time they took to confirm a diagnosis (and so receive treatment and the benefits of treatment in terms of cardiovascular risk reduction). Key model assumptions (these are discussed in more detail in the full write-up in Appendix J: Costeffectiveness analysis blood pressure monitoring for confirmation of diagnosis of hypertension): People with hypertension have a higher risk of cardiovascular events than people without hypertension. Once a diagnosis of hypertension has been made (correctly and incorrectly; that is true positives and false positives) people receive treatment including antihypertensive drugs. Only people who are truly hypertensive (true positives receive benefit in terms of cardiovascular risk reduction from treatment. o People who are truly normotensive but are treated (false positives) do not receive any health benefits. People who are truly normotensive at entry to the model may develop hypertension over time. People diagnosed as not hypertensive (correctly or incorrectly; that is true negatives and false negative) will have a blood pressure check-up with CBPM every 5 years. o At this check-up, it is assumed that they will again screen positive and so be suspected of having hypertension again and their diagnosis is confirmed using the same method as previously (CBPM, HBPM or ABPM) People who have had a cardiovascular event experience reduced quality of life and have an increased risk of death. Diagnosis confirmations using CBPM, HBPM or ABPM are associated with different initial costs. As they also vary in terms of their ability to correctly diagnose people with and without hypertension the downstream costs (including hypertension treatment, CVD costs and checkups in those diagnosed as not hypertensive) and QALYs also vary. Model inputs were based on the clinical effectiveness review undertaken for the guideline, other published data and expert opinion where required. These are described in full in the technical report in Appendix J. All model inputs and assumptions were validated by the GDG and research group. The cost of confirming a diagnosis with CBPM, HBPM and ABPM took into account device costs, maintenance and healthcare professional time. In the base-case analysis the cost per person was 38.00 for CBPM, 39.13 for HBPM and 53.40 for ABPM. This was based on the following assumptions: CBPM was assumed to require at least a further two sets of readings should be taken at monthly intervals. For costing purposes it was assumed in the base case that two sets of readings would be taken; the first with a practice nurse and the second with a GP (as this may involve a treatment consultation). A cost for the CBPM monitor was not included in the costing as GPs will still require clinic monitors even if HBPM or ABPM at diagnosis in instigated and so this cost will not vary dependant on the diagnosis strategy. HBPM was assumed to require measurements over 7 days. For costing purposes it was assumed that two healthcare consultations would be required; an initial appointment with a practice nurse
Update 2011
63

to explain to the patient how to use the monitor and a second once the monitoring was complete with a GP to review the results and provide treatment advice if necessary. ABPM was assumed to take place over a single 24 hour period. For costing purposes it was assumed that two healthcare consultations would be required: an initial appointment with a practice nurse to fit the monitor and a second with a GP to review the results and provide treatment advice if necessary. In addition time for a nurse to download the ABPM data was factored in. HBPM and ABPM device costs per person were calculated based on median published costs to the NHS and assuming a lifetime of 5 years, no resale value, a discount rate of 3.5% and uses per year per machine of 40 and 125 respectively. Alternative diagnosis costs were used in a series of sensitivity analyses. This included scenarios with lower uses per year per machine and ABPM via direct access at hospital. 7.3.2.2 Results This analysis of cost-effectiveness found that, confirming a diagnosis of hypertension with ABPM instead of CBPM or HBPM was the most cost-effective option in all age/gender subgroups (40, 50, 60, 70 and 75 years). In fact, ABPM was cost saving compared to CBPM when long term costs were taken into account. The key driver of cost savings with ABPM compared to CBPM was hypertension treatment costs avoided due to more accurate diagnosis (increased specificity). Results are summarised in Table 19. In most subgroups ABPM was associated with higher QALYs, as well as lower costs, than CBPM and HBPM (that is ABPM was the dominant option). The exception was in the subgroups with starting age 40 years and the female subgroup with staring age 50 years, where ABPM still had lower costs but was associated with a small reduction in QALYs; however, ABPM was still the most cost effective option in these scenarios. Table 19: Basecase analysis results (probabilistic analysis) cost effectiveness (incremental costs and QALYS, and optimal strategy)
Incremental QALYs vs CBPM Subgroup Male, 40 years Male, 50 years Male, 60 years Male, 70 years Male, 75 years Female, 40 years Female, 50 years Female, 60 years Female, 70 years Female, 75 years HBPM -0.001
(CI: -0.006, 0.004)
Update 2011
Incremental costs vs CBPM HBPM -48

(CI: -128, 17)
ABPM -0.004
(CI: -0.009, 0.005)
ABPM -235
(CI: -322, -117)
Most CE strategy ABPM ABPM ABPM ABPM ABPM ABPM ABPM ABPM ABPM ABPM
Probab ility CE 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%
0.001
(CI: -0.009, 0.009)
0.006
(CI: -0.003, 0.017)
-34
(CI: -89, 11)
-156
(CI: -233, -62)
0.003
(CI: -0.010, 0.015)
0.017
(CI: 0.006, 0.029)
-26
(CI: -70, 7)
-112
(CI: -178, -43)
0.005
(CI: -0.009, 0.017)
0.022
(CI: 0.012, 0.035)
-23
(CI: -65, 7)
-89
(CI: -150, -30)
0.004
(CI: -0.007, 0.015)
0.021
(CI: 0.012, 0.030)
-16
(CI: -49, 6)
-56
(CI: -105, -10)
-0.001
(CI: -0.004, 0.001)
-0.006
(CI: -0.008, -0.003)
-68
(CI: -167, 25)
-323
(CI: -389, -222)
-0.001
(CI: -0.006, 0.004)
-0.001
(CI: -0.006, 0.007)
-40
(CI: -106, 15)
-182
(CI: -256, -79)
0.001
(CI: -0.006, 0.008)
0.006
(CI: 0.000, 0.015)
-32
(CI: -83, 11)
-146
(CI: -220, -55)
0.003
(CI: -0.005, 0.011)
0.014
(CI: 0.008, 0.021)
-20
(CI: -59, 8)
-82
(CI: -142, -25)
0.002
(CI: -0.004, 0.007)
0.010
(CI: 0.006, 0.015)
-17
(CI: -52, 11)
-63
(CI: -121, -8)
CE= cost effective at a 20,000 threshold; CI = 95% confidence interval; QALYs = quality-adjusted life years.
64

The conclusion that ABPM is cost-effective compared to CBPM and HBPM was robust to a wide range of sensitivity analyses including those varying the cost of ABPM. As might be expected, the conclusion was sensitive to changes to the accuracy of diagnosis with each method and in some scenarios HBPM became the most cost-effective option. The conclusion was somewhat sensitive to the assumption that check-ups for those diagnosed without hypertension are undertaken every 5 years; in the two lower age subgroups HBPM became cost-effective when check-ups were done annually. The conclusion was also sensitive to the assumption that people who were not hypertensive but were treated did not receive benefits from treatment; when non-hypertensive people also received a risk reduction from treatment CBPM became the most cost-effective option as there was now benefit to misdiagnosing people. 7.3.2.3 Interpretation & limitations This analysis suggests that ABPM is the most cost-effective method of confirming a diagnosis of hypertension in a population suspected of having hypertension based a CBPM screening measurement >140/90 mmHg, compared with further CBPM or HBPM. This conclusion was consistent across a range of age/gender stratified subgroups. Uncertainties in the analysis were explored through extensive sensitive analysis which in most cases did not change conclusions. Where conclusions were impacted this was discussed by the GDG and it was felt that these should not change the overall conclusion. It was noted that the analysis is most probably conservative in terms of ABPM in a number of places. For example, ABPM reduces treatment costs compared to CBPM and HBPM and the cost of these used in the basecase analysis is most likely on low side as it is based on most commonly used generic drug costs and a single clinic visit per year. In addition, the basecase does not incorporate any negative quality of life impacts of being on treatment and when even a 1% reduction in quality of life is incorporated into the analysis QALYs differences between options are considerably more favourable for ABPM. These effects were omitted from the basecase analysis because side effects of antihypertensive drugs are generally fairly mild and rare and patients can often change drugs if they experience side effects but also because no appropriate data was identified to quantify any effects. However, it is not implausible that there may be a small negative impact of being on pharmacological treatment due to side effects. In was noted in GDG discussions that there were potentially some additional benefits of ABPM that were not captured by the model but that would be valued by patients. With ABPM less people are incorrectly diagnosed as having hypertension when they do not. These patients will therefore avoid unnecessarily drug treatment which will mean they wont experience side effects, incur prescription costs or be labelled as having a medical condition, with the potential psychological and practical impacts this can have305. With ABPM patients will also get a definitive diagnosis more quickly that with CBPM. Sensitivity and specificity inputs The relative sensitivity and specificity of CBPM, HBPM and ABPM is the key differentiator between treatments in the model and as such is an important input. However, there were a number of limitations to the estimates of sensitivity and specificity used in the model. A key assumption in the model, and the meta analysis used for sensitivity and specificity estimates, was that ABPM is the reference standard for diagnosing hypertension and so has 100% sensitivity and specificity. This is a potential limitation in that ABPM probably does not have 100% sensitivity and specificity. However, prognostic studies indicated that ABPM was most predictive of prognosis
Update 2011
65

and so this was considered a reasonable assumption for the analysis; without making this assumption it would not be possible to undertake the analysis. Conclusions were however somewhat sensitive to variations in the sensitivity and specificity values, with HBPM becoming cost effective in some scenarios. However, while there is uncertainty around the assumption that ABPM is the gold standard with 100% sensitivity and specificity, the instances when conclusions were changed were generally quite extreme. For example, when the sensitivity and specificity of ABPM were set equal to that of HBPM or when the sensitivity of HBPM was increased to 100%. In addition, while it is known that sensitivity and specificity vary with disease prevalence (and so age) data was not available to allow this to be incorporated into the basecase analysis. However, when examined in exploratory sensitivity analyses it seemed that it would probably not impact conclusions. The GDG carefully considered the uncertainty around the estimates of sensitivity and specificity but given the currently available evidence felt that it should not impact the overall conclusion that ABPM was the preferred option. Treating those who are not hypertensive The basecase conclusion that ABPM was a more cost-effective option for confirming a diagnosis of hypertension than CBPM or HBPM was sensitive to the assumption that only people who were hypertensive received benefits (cardiovascular risk reduction) from treatment. When a risk reduction was also applied to people who were treated but who were not hypertensive (people incorrectly diagnosed as having hypertension), CBPM was the most cost effective option across all subgroups. The basecase assumption was based on the clinical GDG members opinion that there is currently insufficient evidence of benefit for initiating treatment below the currently recommended thresholds. While there is evidence of a continuous relationship between blood pressure and cardiovascular risk361, it is not well established that initiating blood pressure treatment below 140/90 mmHg reduces that risk in people with uncomplicated hypertension. The meta analysis reported by Law and colleagues351 was used to inform the cardiovascular risk reduction in the model for people with and without hypertension as results were stratified by pre-treatment blood pressure; people with hypertension therefore got a greater risk reduction than people without in the analysis. This meta analysis was reviewed as part of the guideline update in relation to the question of what the treatment initiation threshold should be (Chapter 9.1). This analysis asserts that cardiovascular risk reduction is obtained at all levels of pre-treatment blood pressure. However, the GDG noted that the analysis included studies with a range of populations and those that provided information for risk reduction where pre-treatment blood pressure was below 140/90 mmHg were generally in populations with a history of cardiovascular disease or other increased risk that are not necessarily representative of the more general hypertension population. The sensitivity analysis results, with CBPM more cost-effective than ABPM or HBPM, suggests that misdiagnosing people as having hypertension when they do not is a good thing because the health benefits of doing so are worth the additional cost of treatment. This result is therefore more to do with what the diagnostic threshold should be rather than the method that should be used to confirm diagnosis. It should also be noted that potential negative effects of treatment (in terms of reducing people quality of life) were not considered in this sensitivity analysis. The basecase analysis reflects the GDGs interpretation of the clinical data relating to treatment thresholds and as such was considered to reflect the most appropriate analysis for informing which method should be used to confirm a diagnosis of hypertension.
Update 2011
66

Differential treatment initiation threshold In the model it is assumed for practical reasons that all people diagnosed with hypertension (CBPM 140/90 mmHg; HBPM/ABPM 135/85 mmHg) receive pharmacological treatment. However, this guideline recommends a differential treatment initiation threshold whereby people diagnosed with hypertension (by the above definition) generally receive pharmacological treatment if their blood pressure is >160/100 mmHg (HBPM/ABPM >150/95 mmHg), or they have an estimated 10-year cardiovascular risk equivalent to 20% or greater, target organ damage, pre-existing cardiovascular disease, renal disease or diabetes. In those with hypertension but not eligible for pharmacological treatment it is recommended they receive lifestyle advice and an annual check-up. The implications of this simplification are likely to be that the analysis somewhat overestimates the costs of treating hypertension as some people wont need to be treated and somewhat overestimates the benefits of treatment (QALY gain), as some people wont get treated and so wont get the risk reduction from treatment. However, the cost implications will be mitigated by the fact that many people will eventually need drug treatment and that nearly half the cost of hypertension treatment in the model is the annual check-up which will still be required in those that have hypertension but not receiving drug treatment. The treatment costs used in the basecase analysis are also potentially conservative. In addition, the QALYs implications will be mitigated by the fact that the people who do not receive treatment will be at lower risk so the people who remain in the model will have higher risk and benefit more on average and lifestyle advice will provide some risk reduction in some patients at least. In addition to the above considerations, the implication of the differential pharmacological treatment initiation threshold is effectively a reduction in the number of people eligible for treatment. This is therefore somewhat addressed by the sensitivity analysis where the prevalence of true hypertension in the model is varied through a wide range. The conclusion that ABPM was the most cost-effective option was maintained through a prevalence of true hypertension is the suspected hypertension population of 10-80%. Check-up frequency In the basecase analysis it was assumed that people who were diagnosed without hypertension were checked-up every 5 years. In a sensitivity analysis where this was change to an annual check-up, ABPM was no longer cost-effective in younger age groups. The GDG discussed the implications of this finding and felt that, while check-up frequency will vary between patients, on balance this should not impact the overall conclusion that ABPM should be used. It was however noted that in younger patients diagnosed as not hypertensive but in whom frequent follow-up is planned, it might be considered reasonable to use an alternative to ABPM to avoid high diagnosis costs. Model input uncertainty Throughout this report it has been highlighted where there have issues with model input uncertainty this is a limitation of the analysis. In some places there was a lack of data to inform inputs; this included CVD event and post-event costs and the prevalence of true hypertension in a population of people with suspected hypertension. In other places there was variability between settings or patients, such as the cost of ABPM and the frequency of check-ups in those diagnosed without hypertension. The best available or more likely inputs were used for the basecase analysis and these were varied in sensitivity analyses.
Update 2011
7.3.3
Evidence statements economic

One partially applicable study with potentially serious limitations found that ABPM was cost saving compared to CBPM; the treatment costs avoided from not treating patients with WCH were greater than the additional costs of ABPM.
67

New economic analysis from a current UK NHS and PSS perspective comparing CBPM, HBPM and ABPM for confirming a diagnosis of hypertension in a population with suspected hypertension found ABPM to be the most cost effective option across a range of age subgroups in both men and women. In most subgroups ABPM was found to both improve health (increased QALYs) and reduce costs overall. The conclusion was robust to the majority of sensitivity analyses undertaken including those varying the cost of ABPM.
7.4 Measurement protocols for diagnosing hypertension

7.4.1 Ambulatory blood pressure measurement
Review question: In adults with primary hypertension, what protocol should be used when measuring ambulatory blood pressure for treatment and diagnosis? 7.4.1.1 Clinical evidence The literature was searched for all years (as this was not addressed in the previous guidelines)425,436 and all study types were included. Studies were excluded if the population consisted of people who were exclusively diabetic or had CKD. Validation studies of ABPM machines were also excluded. 53 studies77,88,111,151,178,190,200,210,211,237,253,271,272,284,325,326,363,387,405,416,456,491,534,562,563,573,622 46,52,56,114,131,133,150,196,353,386,389,390,420,473,527,530,531,538,541,557,576,595,600,608,609,654 were found that fulfilled the inclusion criteria and assessed what protocol should be used when measuring ambulatory BP for the treatment and diagnosis of adults with primary hypertension. The studies addressing the question were categorised into two different types: 1. Prognostic studies (17studies;17 papers)77,88,131,178,210,211,237,253,284,325,326,363,405,491,534,557,576 those that assess the prognostic significance of ambulatory BP and the optimal schedule for measurement based on outcome data 2. Reliability / reproducibility studies (36 studies; 36 papers)46,52,56,111,114,133,150,151,190,196,200,271,272,353,386,387,389,390,416,420,456,473,527,530,531,538,541,562,563,573,595,600,608,609,62 2,654 - those that assessed any of the following - the optimal ambulatory BP schedule based on: a) b) c) d) e) the reproducibility of ABPM its stability over time (variability of BP over time) the relationship (correlation) between day and night values with mean 24h ABPM values its ability to identify people diagnosed with HT / NT / ICH or dippers and non-dippers changes in BP in response to treatment
Update 2011
Reliability /repeatability studies were deemed to be applicable to the question because they showed which aspects of the ABPM protocol (daytime, night-time, or 24h blood pressure measurements) were the most reliable, and therefore served as an indication of the best / optimal ABP measurements to be taken. Details of all the studies are included in Table 20and Table 26.
68

Table 21summarises the numerical results for selected outcomes of the prognostic studies included for this review. The full data for all outcomes can be found in the evidence tables in the appendix. A summary of the measurement intervals for BP readings used by each of the studies is summarised in Table 20, Table 22 and Table 23. All prognostic studies were found to be methodologically sound / have a low risk of bias (see quality assessment summary tables in appendix F) except for the Li 2008 study363 which was rated as unclear for a number of potential methodological flaws. NOTE: For the prognostic studies, the best method was chosen as the method of measuring BP that best predicted (ie. statistically significant predictors and higher HR values) clinical outcomes (after adjustment for covariates in multivariate analyses). For the reproducibilty/reliability studies the best method was chosen as the the method / protocol of measuring blood pressure that was the most reliable or repeatable.
Update 2011
69
Prognostic studies Table 20: Study details and results for prognostic studies assessing the optimal ABPM protocol
Reference / study type Bjorklund et al., 77 2004 within-group comparison Boggia et al., 2007 Pooled analysis of other study data, within-group comparisons (IDACO) Clement et al., 2003
131 88
N 872
Population General population (HT and NT)
Device AUS
Follow-up time Mean 6.6 years
Time and frequency of measurement every 20 mins
Outcomes CV mortality
Proposed protocol (authors conclusions) best prognostic ability 24h, daytime and night-time are all predictors Use SBP not DBP
7458 analy sed
General population (HT and NT)
OSC or AUS
Median 9.6 years
D range 15-30 mins N range 30-60 mins
Total mortality, CV mortality, nonCV mortality, CV events, stroke, cardiac events
Both daytime and night-time BP (need to record ABPM throughout the whole day). NOTE: 24h BP was not measured.
Update 2011
70
2232
HT
Median 5 years
D 30 mins N <60 mins
Total mortality, CV mortality, CV events, MI, stroke
24h and daytime (are better than nighttime, especially SBP)
Within-group comparison Dolan et al., 2005 within-group comparison Fagard et al., 211 2005 within-group comparison 391 General population in primary care practice (HT and NT) Median 10.9 years D 15 mins N 30 mins
178
5292
HT
OSC
Mean 7.9 years
every 30 mins
All-cause mortality; Cardiac mortality; CV mortality CV events
Night-time (better than daytime or 24h)
Night-time (better than daytime)
Reference / study type Fagard et al., 210 2008 Pooled analysis of other study data ,within-group comparisons Gosse et al., 2001 within-group comparison
237
N 302
Population HT (with history of CV disease)
Device not specifie d
Follow-up time Median 6.8 years
Time and frequency of measurement D range 15-30 mins (10am 6pm) N range 30-60 mins (12am 6am)
Outcomes All-cause mortality; CV mortality; composite of major CV events CV complications
Proposed protocol (authors conclusions) best prognostic ability
Update 2011 Update 2011 Update 2011
Night-time
256
HT
AUS
Mean Mean 84 months
D 15 mins N 15 or 30 mins
24h, daytime, night-time and arising BP are all predictors (24h, daytime and arising slightly stronger predictors) Single BP value on rising in the morning (is as good as mean daytime or mean 24h measurements) Use SBP not DBP Night, day and 24h SBPs and DBPs DBP better than SBP
Hansen et al., 253 2005 within-group comparison Ingelsson et al., 284 2006 within-group comparison Khattar et al., 325 2001 within-group comparison
1700
OSC
Up to 9.5 years
D 15 mins N 30 mins
All-cause mortality; CV mortality
71
951
AUS
Up to 9.1years (mean range 0.1 11.4 years) Mean 9.2 years
D 20 or 30 mins N 20 or 60 mins
CHF
Night-time (better than daytime or 24h)
688
HT
Intraarterial ABPM
Every hour
Non-CV death, coronary death, CeV death, peripheral vascular death, nonfatal MI, nonfatal stroke,
24h, daytime and night-time all predictors SBP and DBP in age <60 Only SBP in age >60
Reference / study type
Population
Device
Follow-up time
Time and frequency of measurement
Outcomes coronary revascularisation. CV events; coronary events; cardiac events; fatal/non-fatal stroke
Kikuya et al., 326 2007 Pooled analysis of other study data, within-group comparisons (IDACO) Li et al., 2008
363
5682
General population (HT and NT); <10% had underlying CV disease
Median 9.5 years
1 study: every 20 mins 1 study: every 30 mins 1 study: 15 mins day, 30 mins night 1 study: 20 mins day, 45 mins night
24h, daytime and night-time (SBP and DBP)
7458
Summary of prospective population studies (case series) Metoki et al., 405 2006 within-group comparison 1542
not specifie d
Median 9.6 years
D interval not specified N interval not specified
CV mortality, nonCV mortality, CV events, stroke, cardiac events
Daytime and night-time (depending on which outcome) Night-time better for mortality outcomes Daytime better for non-CV mortality Both for CV events and stroke Need to record ABPM throughout the whole day Night and early morning 2h SBP (CeV and CV mortality) Elevated daytime 2h SBP (Haem stroke mortality) elevated night-time 2h SBP (cerebral infarction and HD mortality) High BP at different times of day is associated with different subtypes of CeV and CV mortality risk. Daytime for cardiac events, night-time for stroke One summary measure not enough to predict different clinical outcomes
72
OSC
Mean 10.6 years
30 mins over 24 hours Weekday average of 4 SBP = 2hr SBP value at different periods
Mortality risk from CeV and CV events
Pickering et al., 491 2007 Summary of prospective
8945
1 study: general population (HT and NT) 6 studies: HT
OSC or AUS
Mean 5.8 years
15-30 mins over 24 hours
Cardiac events; stroke
Reference / study type population studies (case series) Sega et al., 2005 within-group comparison (PAMELA study) Staessen et al., 557 1999 Within-group comparison: substudy ofSyst-Eur trial Suzuki et al., 576 2000 Within-group comparison
534
Population (NT controls)
Device
Follow-up time
Time and frequency of measurement
Outcomes
2051
OSC
Mean 10.9 years
every 20 mins
All cause mortality; CV mortality
Nighttime better than daytime SBP better than DBP
837
HT (ISH)
OSC
Mean 4.4 years
D - 30 mins N - 30mins
Total mortality, CV mortality, CV events, stroke, cardiac events
Night-time (better than daytime) Excluding the first 2h does not improve accuracy
73
324
HT and NT
OSC
Mean 51.5 months
D 30 mins N 30 mins
CV events
Higher 24-h and nighttime BP (SBP and DBP) are associated with a higher incidence of CV events
NT = normotensives; HT = hypertensives; ISH = isolated sytolic HT; AUS = auscultatory device; OSC = oscillometric device; D = daytime; N = night-time
Table 21: Summary of numerical results for prognostic studies (for selected outcomes)
Study Bjorklund et al., 2004
77
Outcome CV mortality
HR (95% CI) for SBP measurement ABPM (24h): 1.23 (1.07, 1.42) p<0.05 ABPM (daytime): 1.23 (1.07, 1.42) p<0.05 ABPM (night-time): 1.18 (1.03, 1.34) p<0.05 per 1SD rise in SBP ABPM (24h): not given ABPM (daytime): 1.16 (1.07-1.26) p<0.001 ABPM (night-time): 1.21 (1.12-1.30) p<0.001 Per 1SD rise in SBP No HRs given. Relative Risks: ABPM (24h): 1.34 (1.11-1.62) ABPM (daytime): 1.30 (1.08-1.58) ABPM (night-time): 1.27 (1.07-1.51) Per 1SD rise in SBP ABPM (24h): 1.19 (1.14, 1.26) p<0.001 ABPM (daytime): 1.15 (1.10, 1.21) p<0.001 ABPM (night-time): 1.21 (1.16, 1.27) p<0.001 per 10mmHg rise in SBP ABPM (24h): Not given ABPM (daytime): 1.33 (1.07, 1.64) p<0.01 ABPM (night-time): 1.42 (1.16, 1.74) p<0.001 Per 1mmHg rise in SBP ABPM (24h): 1.20 (0.91-1.58) NS ABPM (daytime): 1.03 (0.77-1.36) NS ABPM (night-time): 1.34 (1.06-1.69) p<0.01 Per 1SD rise in SBP No HRs given,only characteristics of people with vs without complications and the statistical difference. ABPM (24h): 133 16 vs. 143 14 (p<0.001) ABPM (daytime): 138 16 vs 149 15 (p<0.01)
Boggia et al., 2007 * CV events
88
Clement et al., 2003
131
CV events
Dolan et al., 2005
178
74
CV mortality
Fagard et al., 2005
211
CV events
Fagard et al., 2008 *
210
Composite of major CV events
Gosse et al., 2001
237
CV complications
Study
Outcome
HR (95% CI) for SBP measurement ABPM (night-time): 121 17 vs 129 14 (p<0.05) SBP mm Hg without vs with complications MeanSD
Hansen et al., 2005
253
CV mortality
ABPM (24h): 1.51 (1.28, 1.77) p<0.0001 ABPM (daytime):1.50 (1.27, 1.76) p<0.0001 ABPM (night-time): 1.41 (1.23, 1.62) p<0.0001 per 10mmHg rise in SBP ABPM (24h): 1.13 (0.91, 1.40) p>0.05 ABPM (day-time): 1.08 (0.85, 1.36) p>0.05 ABPM (night-time): 1.21 (0.98, 1.49) p>0.05 per 1SD rise in SBP <60 yrs ABPM (24h): 1.01 (1.00, 1.02)p=0.04 < 60 yrsABPM (daytime): 1.01 (1.00, 1.02)p=0.04 <60 yrs ABPM (night-time): 1.01 (1.00, 1.02)p=0.04 >60 yrs ABPM (24h): 1.02 (1.00, 1.03) p=0.003 >60 yrsABPM (daytime): 1.02 (1.00, 1.03)p=0.004 >60 yrs ABPM (night-time): 1.02 (1.00, 1.03) p=0.007 No info on the reference rise of SBP, but likely per 1mmHg ABPM (24hrs): 1.24 (1.19, 1.30) p<0.0001 ABPM (daytime): 1.20 (1.15, 1.25) p<0.0001 ABPM (night-time): 1.18 (1.14, 1.23) p<0.0001 ABPM (24hrs): 1.20 (1.13, 1.27) p<0.0001 ABPM (daytime): 1.16 (1.09, 1.23) p<0.0001 ABPM (night-time): 1.16 (1.10, 1.22) p<0.0001 per 10mmHg rise in SBP ABPM (24h): not given ABPM (daytime): 1.16 (1.07-1.26) <0.001 ABPM (night-time): 1.21 (1.12-1.30) <0.0001 per 1SD rise in SBP ABPM (24h): 1.76 (1.39-2.25) p<0.002
Ingelsson et al., 2006
284
CHF
Khattar et al., 2001
325
all cause mortality. (no results for cornonary death)
75
Kikuya et al., 2007
326
CV events defined as CV endpoints in the evidence table (also used cardiac events in red)
Li et al., 2008 * CV events
363
Metoki et al., 2006
405
Mortality risk from
Study
Outcome CeV and CV events
HR (95% CI) for SBP measurement ABPM (daytime): 1.59 (1.25-2.01) p<0.002 ABPM (night-time): 1.78 (1.40-2.27)p<0.002 Per 1SD rise in SBP ABPM (24h): not given ABPM (daytime): HR = 1.29(95% CI: 1.20-1.39); p < 0.0001 ABPM (night-time): HR = 1.22(95% CI: 1.14-1.30); p < 0.0002 per 10mmHg rise in SBP No HRs given, but all entry BP values had a direct exponential relationship with the risk of all-cause death or CV death Goodness of fit of the relationship of BP to risk of death (CV and all-cause) was not less for clinic, compared to home and ambulatory. Coefficients: ABPM (24h): 0.0557 0.0008 p<0.0001 ABPM (daytime): 0.0479 0.008 p<0.0001 ABPM (night-time): 0.0559 0.007 p<0.0001 Coefficient the increase in risk per 1mm Hg increase in SBP ABPM (24h): 1.20 (0.98-1.49) NS ABPM (daytime): 1.17 (0.96-1.44) NS ABPM (night-time): 1.23 (1.03-1.46) 0.05 per 10mmHg rise in SBP ABPM (24h): 1.28 (1.05 to 1.54) p< 0.05 ABPM (daytime): No HR reported ABPM (night-time): 1.34 (1.13 to 1.58)p < 0.01 per 10mmHg rise in SBP
Pickering et al., 2007 *
491
Cardiac events
Sega et al., 2005
534
CV mortality
76
Staessen et al., 1999
557
CV events
Suzuki et al., 2000
576
CV events
Reliability and reproducibility studies Table 22: Study details and results for reliability/reproducibility studies assessing the optimal ABPM protocol
Reference / study type Frequency of measurements N Population Device Follow-up Consecu tive reading s 24h Time of measurement Mathematical method Proposed number of measurements (authors conclusions) Differences in BP measurements (3 measurements) was only significant during waking hours
Antivalle et al., 46 1990 case-series: RCT substudy
22
HT
AUS and OSC
4 weeks (3 measuremen ts: baseline, 2 and 4 weeks)
Daytime Night-time 24h intervals not given
Reproducibility of BP (between the 3 measurements over time)
77
Asagami et al., 52 1996 within-group comparison Asmar et al., 56 2001 RCT 30 HT 64 Borderline HT AUS and OSC 1-2 years on a work day 24h 1 month (2 measuremen ts1 month apart) 24h
Daytime (30 mins) Night-time (1 hr) 24h Daytime (15 mins) Night-time (30 mins) 24h
Long-term reproducibility of BP (between the 2 measurements over time): SD Reproducibility of BP (between the 2 measurements over time, after placebo treatment)
Daytime BPwas better (vs nighttime and 24h)
Placebo administration resulted in SS reductions between baseline and 1 month 24h ABPM (SBP), and daytime SBP/DBP. No treatment resulted in NS differences between baseline and 1 month for 24h, daytime and night-time SBP/ DBP. This suggests a placebo effect on
Frequency of measurements BP.
Calvo et al., 2003 Case-series
111
823
HT
OSC
48 h
48h
D 20 mins (07.0023.00) N 30 mins (23.0007.00) ABPM started on a weekday (Mon, Wed or Fri)
Comparison of day-to-day variations in BP
ABPM for 48 h revealed a statistically significant pressor response (this could largely be due to the novelty of wearing an ABPM device for the first time). The pressor effect remains statistically significant for the first 10 h of monitoring, independent of gender, day of the week of monitoring and number of a-HT drugs used. Nocturnal mean BP was similar between both days of sampling. The effect diminished, but was not eliminated, in extent and duration for successive sessions of ambulatory monitoring. ABPM for just 24 h may be insufficient for a proper diagnosis of HT, evaluation of treatment efficacy and identification of dipping status in relation to targetorgan damage. 24h BP was more reproducible over time than daytime and nighttime BP measurements.
78
Campbell et al., 114 2010 within-group comparison Coats et al., 100 HT 72 HT and NT OSC 2 years (2 measuremen ts 2 years apart) 1 month 24h Daytime (15 mins) Night-time (30 mins) 24h 24h Daytime only (30 mins) Reproducibility of BP (between the 2 measurements over time) Reproducibility
Average daytime ABPM DBP was
Reference / study type 1992

133
Frequency of measurements (2 measuremen ts1 month apart) 208 HT OSC 3 weeks (2 measuremen tswithin 3 weeks) 24h Daytime (15 mins) Night-time (20 mins) 24h 24h D (working day) 15 mins (07.00-23.00) N 20 mins (23.0007.00) of BP (between the 2 measurements over time) Reproducibility of BP (between the 2 measurements over time) more reproducible than a single measuremnt from daytime. There was improved reproducibility with more measurements during the day There was no change in diurnal BP variations. This indicates that the short term reproducibility of diurnal changes in BP in the early phases of untreated essential HT, is overall satisfactory. Classification of ICH based on a single ABPM (using cut-offs suggested in major HT guidelines) has limited short-term reproducibility Repeated ABPM measurements at a short time interval should be used to ensure correct diagnosis of ICH and improve CV risk stratification, allowing a more appropriate treatment strategy Reproducibility of ABP levels and BP varaiblity was fairly good. Reproducibility of BP reductions was fairly good for ABP levels, so a single ABPM before and during treatment is acceptable in a drug intervention trial.
within-group comparison Cuspidi et al., 150 2002 case-series
Cuspidi et al., 151 2007 Case-series
611
ICH
OSC
2 x 24h periods (1-4 weeks apart)
Correlation with clinical diagnosis of ICH Reproducibility of ICH diagnosis (repeated ABPM measurements)
79
Eguchi et al., 190 2010 within-group comparison
43
HT
OSC
Measureme nts twice within a 2week interval between measuremen ts
24h
Every 30 mins
Reproducibility of ABP, BP variability and BP reduction
Frequency of measurements
Enstrom et al., 196 1996 RCT
80
HT and NT
OSC
14 days (2 measuremen ts: 1 work and 1 nonwork day)
24h
Daytime Night-time 24h All: 20 min intervals
Reproducibility on work and non-work days: SD; reproducibility over time (2 measurements, 2 weeks apart)
BP was higher during the work day. Daytime and night-time: there was a SS difference in BP measurement between the 2 readings There was NS difference for nighttime BP between the 2 readings There were no major differences in reporducibility if 1, 2 or 3 recordings / hour were used. Arbritrary dividing lines for day/night or according to patients own statement did not have any major effect on the result. But it may be wise to perform recordings not less than every 30 mins for patients After excluding the first hour, correlations for mean SBP the subsequent 3-, 5- and 7-hour periods demonstrated greatest improvement in correlation when session is increased from 4 to 6 hours. 6-hour ABPM can approximate the overall mean BP obtained from full 24-hour ABPM. Shortened sessions do not characterise the influence of circadian variation over the 24hour mean BP and may
80
Ernst et al., 2008
200
post-hoc analysis (DIDIMA study)
1004 ABPM sessio ns (529 studie s)
Borderline HT, suspected WCH, suspected hypotension, MHT, Tx resistance, aHT treatment
OSC
24h
3 readings /hr (daytim e) 2 readings /hr (nighttime)
D 20 mins (6am 6, 8 or 10pm) N 30 mins (6, 8 or 10pm 6am)
Correlation of shorter ABPM periods with 24h ABPM
Frequency of measurements overestimate 24-hour BP levels.
Hermida et al., 271 2002 Case-series
538
HT
OSC
48 h
48h
D 20 mins (07.0023.00) N 30 mins (23.0007.00) ABPM started on a weekday (Mon, Wed or Fri)
Comparison of variations in BP
BP is significantly increased by the novelty of wearing an ABPM device for the first time (the ABPM effect). Pressor effect remains statistically significant for the first 6-8h of monitoring, independent of gender, day of the week of monitoring and number of a-HT drugs used. Differences between successive days of ABPM are no longer significant when patients were evaluated for second or successive times. ABPM for just 24 h may be insufficient for a proper diagnosis of HT, evaluation of treatment efficacy and identification of dipping status in relation to targetorgan damage. The percentages of patients classified as non-dipper for the first 24 h, the second 24 h and the 48 h average were 47, 50 and 48% respectively. When the first and second 24-h periods were compared, 147 (24%) subjects switched from dipper (D) to non-dipper (ND) or vice-versa.
81
Hernandez-del Rey 272 et al., 2007 Historical caseseries 611 HT OSC 48h 24h / 48h Night and day defined based on patients diary; at least 14 measurements during period of activity and at least 7 during period of rest Reproducibility of BP dipping pattern in 24-h vs 48-h ABPM
Frequency of measurements Recording intervals (minutes between measurements) not given When the first 24-h period was compared to the 48-h average, 66 (11%) subjects switched patterns. The proportions were similar separately for SBP and DBP, and between treated and untreated patients. In subjects with poor ABPM reproducibility, night-to-day ratios were of an intermediate value between those of subjects always classified as Dipper or non-dipper. Categorisation of D or non-dipper based on a single 24-h ABPM is moderately reproducible, since one out of every five patients change profile over the following 24 h. A more reliable classification of the BP circadian profile should be performed by repeating a second ABPM within a short period, but the use of 48-h ABPM in clinical practice should be assessed according to cost-effectiveness criteria. Similarities in BP measurements between 3 FoMs BP was similar in the three FoMs studied at daytime and night-time. There is therefore no strong argument to perform ABPM at high FoM
82
Lede et al., 1997 case-series
353
49
Pregnant women with preeclampsia (DBP90mm
AUS
24h
24h
3 different frequencies of monitoring (FoM) readings/ hour: High FoM = 7/hr
Frequency of measurements Hg and proteinuria >300mg). Low FoM = 1/hr Medium FoM = 2/hr BP measurement at a lower FoM may be better for the patient and reduce equipment deterioration whilst providing equivalent information as supplied by a high FoM Reproducibility of BP (between the 2 measurements over time; and hourly vs mean 24h, SDD) The second ABPM recording was lower but was NS different from the first Reproducibility was lower for hourly rather than 24h average BP. This suggests that ABPM measurement loses its advantages for reproducibility if results are analysed over hourly periods Treatment-induced reduction in BP is smaller for the night-time than daytime average BP The effect of anti-HT treatment is unevenly distributed between day and night Results advocate a more systematic adoption of ABP monitoring in trials assessing CV protection by anti-HT drugs
Mancia et al., 386 1992 case-series
29
HT
AUS
4 weeks (2 measuremen ts4 weeks apart)
24h
Daytime (15 mins) Night-time (20 mins) 24h
83
Mancia et al., 387 2004 SR / MA of 44 trials 6000 HT (treated) AUS or OSC 1 week 36 months Daytime: not given Night-time: not given 24h: not given Change in BP response by different measurement methods Mansoor et al., 389 1994 25 HT AUS and OSC Mean 23 months 24h Daytime Night-time Reproducibility of BP (between 2 repeated studies and
24h and night-time BP had better reproducibility than daytime BP (between studies and between readings over time)
Reference / study type within-group comparison
Frequency of measurements over time): SDD, co-efficient of variance and % of people within 10mm and 5mm SBP and DBP Diagnostic accuracy with varying number of measurements Comparison of weekly variations in BP Increasing the number of measurements led to a reduction in diagnostic error due to random variability of BP. Monday surge in BP was found in the awake and morning BP but not in the asleep BP Morning BP surge on Monday was higher than on the other days of the week except for Tuesday Morning BP surge on a Monday may be in accord with clinical evidence that CV events more frequently occur in the morning on Monday There was high agreement between the 4 readings BP values were lower during the 4th reading (vs 1st) People should not be labelled as HT based on initial readings, since initial ABPM may yield higher
24h All: 15 min intervals
Mar et al., 1998 within-group comparison
390
138
HT (newly diagnosed)
OSC
Not given
24h
Daytime (20 mins) Night-time (1 hr) 24h
Murakami et al., 416 2004 within-group comparison
135
OSC
7 days
Fitted on Thursday between 10am 2pm; D - every 30 mins (0700 to 2200 hours) N - 60 mins (2200 to 0700 hours).
84
Musso et al., 420 1997 case-series
40
NT
OSC
3 months (4 measuremen ts each 28 days apart)
24h
Frequency of measurements values than later monitoring
Octavio et al., 456 2010 within-group comparison
450
Suspected arterial HT
not specifie d
24h
24h
Group
BP readi ng interv al Day (0600 2300) Night (2300 0600) 30 min 20 min 30 min
Reliability of conventional vs time-weighted quantification of 24-h ABP
I II III Palatini et al., 473 1994 case-series 24h 6461 ISH or high DBP OSC 3 months 2 (3 months apart)
15 min 15 min 30 min
Higher number of readings per hour during daytime leads to an overestimation of conventional 24h average BP, particularly in individuals with preserved nocturnal BP dipping. This can be avoided either by scheduling the same number of readings/h throughout 24 h or by performing a time-weighted quantification of 24-h BP The clinical implications of these different approaches deserve further investigation.
85
Daytime (10 mins) Night-time (30 mins)
Reproducibility over time (2 measurements, 3 months apart)
Small but SS decreases in average daytime BP / no change in average nighttime BP occur when ABPM is performed twice 3 months apart. There was a SS increase in SBP when the period between midnight and 5 am was considered in nighttime analysis. ABPM shows better reproducibility than office BP, particularly for 24h BP. Nighttime BP was less reproducible than daytime BP,
Frequency of measurements probably due to sleep disturbance which was reported in 2/3 of patients.
Schillaci et al., 527 1994 case-series
24
HT
OSC
1 week (2 measuremen ts1 week apart)
24h
Daytime (15 mins) Night-time (15 mins session 1, 1hr session 2) 24h
There was NS difference in daytime or night-time systolic or diastolic BP and heartrate between the two sessions A low number of cuff measurements of BP during the night (1 per hour) provides similar results to a high number of measurements in terms of sleep BP, and changes of BP from wake to sleep. Men: had greater BP variation (SBP and DBP) during the inactive period (vs. active period) Women: SBP there was NS difference in BP variation during the inactive period (vs. active period). DBP as for men.
86
Schwartz et al., 530 2000 within-group comparison
143
NT
AUS
1 week
24h
Active period (daytime) Inactive period (nighttime) All: 10 min intervals
Intraindividual BP variability (SDs), during the active (daytime) and inactive (nighttime) periods of the day Intraindividual BP variability (SDs), during active (daytime) and inactive (nighttime) periods of the day
Schwartz et al., 531 2000 within-group comparison
240
NT
AUS
1 week
24h
Active period (daytime) Inactive period (nighttime) All: 10 min intervals
Men and women: there was greater BP variation (SBP) during the inactive period (vs. active period) Women: DBP there was NS difference in BP variation during the inactive period (vs. active period)
Reference / study type Sheps et al., 538 1994 within-group comparison Shinagawa et al., 541 2002 case-series
Frequency of measurements 294 HT and NT AUS 2 months (2 measuremen ts2 months apart) 24h Daytime (7.5 mins) and other time frequencies Reproducibility of BP (between the 2 measurements over time): As few as six hours of monitoring with two to three readings/hour achieved most of the gain in precision obtainable by going from single BP readings toward continuous measurement during an entire awake period The average SBP (daytime) is higher on the first day of monitoring vs the other 6 days. Daytime BP was lowest on Sundays and the day-night ratio was optimal on weekends. Average ABPs are highly reproducible in patients with uncomplicated essential HT of limited duration. Nocturnal dipping pattern also reproduced satisfactorily. White coat effect and variability are greatly attenuated during repeated measurements, and these measures may thus be of less utility in clinical practice. ABP and pulse pressure and of nocturnal fall in BP have the most prognostic relevance and are of great value in clinical practice. Mean 24h (was better than awake or asleep BP)
56
unclear
OSC
7 days
7 days of 24h recordin gs
Daytime (30 mins) Night-time (1 hour) 24h
BP variability on different days of the week
Stenehjem et al., 562 2004 within-group comparison
75
HT
AUS
4 weeks measuremen ts before and after 4 week observation period (2 separate work days)
24h
D 20 mins (0700 2200) N 30 mins (2200 0700)
Reproducibility of BP variability, white coat effect and dipping pattern
87
Stergiou et al., 563 2002
133
HT (untreated)
OSC
2 work days
24h
Every 20 mins
Test-retest variability (correlations
Reference / study type within-group comparison Suarez et al., 573 2003 retrospective diagnostic caseseries
Frequency of measurements and SDD) 261 HT OSC 24h 24h D 20 mins (07002400) N 30 mins (2400 0700) Reference standard: mean 24h ABP (125/80) Index test: mean awake ABP (<135/85) Agreement between ABP daytime average and 24h average for diagnosing HT and assessing effects of antiHT treatments (sensitivity / specificity) In 90% of the records there was agreement between both criteria Daytime and 24 h average BP may carry similar information for diagnosing HT and assessing the effects of anti-HT treatment in clinical practice. ABPM used only during the daytime could be better tolerated and agreed to by patients than 24 h monitoring. 24h and Daytime ABPM was better than night-time BP (all were better than clinic)
88
Thijs et al., 1992
595
102
ISH
OSC
within-group comparison: substudy of SystEur trial
1 month (2 measuremen ts 1 month apart)
24h
Daytime Night-time 24h All intervals not <30 mins
Consistency (median differnce between the 2 recordings); repeatability (2 x SD of the changes between the 2 recordings) Reproducibility of BP (between the 2 measurements over time)
Trazzi et al., 600 1991 case-series
34
HT
AUS
4 weeks (2 measuremnt s 4 weeks apart)
24h
There WAS NS differnce in SBP / DBP measurements 4 weeks apart (24h ABPM) 24h ABPM was more reporducible than office BP due to a larger number of measurements.
Reference / study type Van der Steen et 608 al., 1999
Frequency of measurements 45 HT AUS device may not be truly ABPM OSC 2-3 weeks (2 measuremnt s 2-3 weeks apart) 24h 24h Daytime (15 mins) Night-time (30 mins) 24h 24h Daytime (15 mins) Night-time (20 mins) long fixed sleep period: waking 7am-10pm and sleeping 10pm-7am short fixed sleep period: waking 10am to 11pm and sleeping 1am-7am pts diary sleep period: actual sleep times 24h Reproducibility of BP (between the 2 measurements over time) Differnce in BP using long and short sleep periods vs actual sleep period (pts diary) There was poor reproducibility. 24h and daytime BP were better than night-time measurements.
within-group comparison Van Ittersum et al., 609 1995 retrospective caseseries 20 HT and WCH
A short sleeping period gives accurate measures of blood pressure during sleep. The long sleeping period method should be avoided as it can overestimate BP during sleep.
89
Wallace et al., 622 2005 Retrospective comparative study with historical control 31 HT AUS 2 separate weekdays, 23 days apart SAME group: monitoring began at same time of day 24h
SAME group: first reading 177-1900; OPP group: sessions randomised to begin in morning (0700-0900) or evening (1700-1900). D - 15 5 minutes (0600-2200)
Reproducibility of BP variables: averages, 24-h, day-time, nighttime, crest, trough, trough:crest (Intra-class correlation)
For SBP the ABPM was only reproducible when monitoring began at the same time of day and not when variables were measured at opposite times of day TrBP and average 24-h SBP were significantly higher when the monitoring session began in the morning compared with the evening Reproducibility of DBP was similar
Frequency of measurements OPP group: sessions randomised to begin in morning or evening N - 30-45 5 minutes (2200-0600) between SAME and OPP conditions. Ambulatory BP variables were consistently higher when monitoring session began in the morning Reproducibility over time (2 measurements, 2 weeks apart) There was no difference between the 4 readings (over time) for 1h, 24h daytime or night-time (SBP or DBP)
Zakopoulos et al., 654 2001 case-series
25
HT
OSC
4 months Four times (four(interval s of 1 week each)
24h
Daytime Night-time 24h
Update 2011
All: 15 min intervals and 1 hr intervals

NT = normotensives; HT = hypertensives; ICH = isolated clinic HT; AUS = auscultatory device; OSC = oscillometric device; D = daytime; N = night-time; TrBP = trough BP.
90
Reference / study type DAY and NIGHT and 24h Hansen et al., 2005 Kikuya et al., 2007
253
Table 23: Day and night intervals and results for prognostic studies assessing the optimal ABPM protocol
N 1700 5682 688 324 Follow-up time Up to 9.5 years Median 9.5 years Mean 9.2 years Mean 51.5 months Day protocol (mins) 15 15, 20, 30 60 30 Night protocol (mins) 30 20, 30, 45 60 30 Best: day (D), night (N) or 24h D + N + 24h All intervals are the same. D + N + 24h D + N + 24h N + 24h
326
Khattar et al., 2001 NIGHT and 24h Suzuki et al., 2000 DAY and 24h
325
576
Reference / study type Gosse et al., 2001 DAY and NIGHT Boggia et al., 2007
88 237 131
N 256 2232 7458 analysed

237
Follow-up time Mean 84 months Median 5 years Median 9.6 years Mean 7.4 years Mean 5.8 years Mean 6.6 years Median 9.6 years Mean 10.6 years Median 10.9 years Median 6.8 years Mean 10.9 years Up to 9.1years (mean range 0.1 11.4 years) Mean 4.4 years Mean 7.9 years
Day protocol (mins) 15 30 15-30 15 15-30 20 30 15 15-30 20 20 or 30
Night protocol (mins) 15 or 30 <60 30-60 30 15-30 20 30 30 30-60 20 30 or 60
Best: day (D), night (N) or 24h Morning was as good as D + 24h D + 24h D+N D+N D+N D+N D+N D+N N N N N
Clement et al., 2003
Cipriano and Gosse et al., 2001 Pickering et al., 2007 Li et al., 2008 NIGHT Fagard et al., 2005 Fagard et al., 2008
211 210 363 405 491 77
741 8945 872 7458 1542 391 302 2051
Bjorklund et al., 2004 Metoki et al., 2006
Sega et al., 2005
534 284
91
Ingelsson et al., 2006
951
Staessen et al., 1999 Dolan et al., 2005

178
557
837 5292
30 30
30 30
N N
D = daytime; N = night-time
Table 24: Day and night intervals and results for reliability/reproducibility studies assessing the optimal ABPM protocol
Reference / study type DAY and NIGHT and 24h Zakopoulos et al., 2001 DAY + 24h Van der Steen et al., 1999 Suarez et al., 2003 Thijs et al., 1992 NIGHT + 24h Palatini et al., 1994
473 389 573 595 608 654
N 25 45 261 102 6461 25 22 24 143 240 64 72 133 1004 sessions
Follow-up time 4 months 2-3 weeks 24h 1 month 3 months Mean 23 months 4 weeks 1 week 1 week 1 week 1-2 years 2 years 2 work days 24h 48 h 48 h 2 months
Day protocol (mins) 15 15 20 30 10 15 15 10 10 30 15 20 20 20 20 7.5, 20 or 30
Night protocol (mins) 15 30 30 30 30 15 15 or 60 10 10 60 30 20 30 30 30 -
Best: day, night or 24h D + N + 24h D + 24h D + 24h D + 24h N + 24h N + 24h N + 24h D + N (60minswas fine for night) D D D 24h 24h 6h 24h >24h >24h 20 and 30 mins are almost as good (for D)
Mansoor et al., 1994 Antivalle et al., 1990 DAY + NIGHT Schillaci et al., 1994
46
527
92
DAY Schwartz et al., 2000 Schwartz et al., 2000 Asagami et al., 1996 24h Campbell et al., 2010 Stergiou et al., 2002 Ernst et al., 2008 >24h Hermida et al., 2002 Calvo et al., 2003
111 271 200 114 563 530 531
52
538 823 294
OTHER INTERVALS SPECIFIED Sheps et al., 1994

538
Reference / study type Lede et al., 1997

353 386 456
N 49 29 450
Follow-up time 24h 4 weeks 24h
Day protocol (mins) 7.5, 30 or 60 15 15 or 30
Night protocol (mins) 7.5, 30 or 60 20 20 or 30
Best: day, night or 24h All times are similar 24h was better than hourly D had lower readings,or perform the same number of readings for 24h 20, 30 or 60 mins are fine Increased measurements are better More day measurements are better -
Mancia et al., 1992
Octavio et al., 2010
Enstrom et al., 1996 Mar et al., 1998

390
196
80 138 100
14 days Not given 1 month
20 20 30
20 60 -
Coats et al., 1992 NOT SPECIFIED
133
Trazzi et al., 1991
600 609
34 20 208 611 30 31
562
4 weeks 24h 3 weeks 1-4 weeks 1 month 2-3 days 4 weeks 2 weeks 7 days 7 days 1 week 36 months 3 months 48h
10 15 15 15 15 15 20 30 30 30 15 -
20 20 20 20 30 30-45 30 30 60 60 30 -
Van Ittersum et al., 1995 Cuspidi et al., 2002 Cuspidi et al., 2007 Asmar et al., 2001
150 151
93
56 622
Wallace et al., 2005 Eguchi et al., 2010
Stenehjem et al., 2004

190
75 43 56 135 6000 40 611
Shinagawa et al., 2002 Murakami et al., 2004 Mancia et al., 2004 Musso et al., 1997
387 420
541
416
Hernandez-del Rey et al., 272 2007

+ = or ; D= daytime; N = night-time

7.4.1.2
Health economic evidence No relevant economic studies were identified relating to ABPM measurement protocols.
7.4.1.3
Evidence statements clinical The 17 prognostic studies recommend the following regimens (as the best predictors of CV events) : All day measurements are needed (11 studies): o day and night day and night measurements predict different outcomes (four studies)88,363,405,491 o 24h, day and night were all good predictors of outcome (five studies)77,237,253,325,326 o 24h and day were the best predictors of outcome (one study)131 o 24h and night were the best predictors of outcome (one study)576 Night BP only is sufficient (a good predictor of outcome) (six studies)178,210,211,284,557534 A single BP measurement on rising is sufficient this is as good as using the 24h or daytime mean for predicting outcome (one study)237 Excluding the first two hours does not improve accuracy (one study)557 SBP is sufficeint (a good predictor of outcome) but DBP is not (four studies: one study - SBP in >60 years, DBP<60 years)77,237,325,534 DBP is sufficient (a good predictor of outcome) but SBP is not (two studies: one study - SBP in >60 years, DBP<60 years)253,325
Update 2011
The 36 reliability/reproducibility studies showed the following: 1. The optimum interval between measurements: Repeat ABPM over a short time interval (one study)151 A greater number of readings/hr leads to an overestimation of BP: use the same number readings over 24 hours or use a time-weighted calculation of 24h BP (one study)456) One reading per hour for night-time is equivalent to a 15 min interval for night-time BP (one study)527 A short sleep period (1-7am) is more accurate than using a long sleep (10pm 7am) (one study)609 Daytime BP: taking more measurements is better than just one measurement (one study)133 More measurements taken lead to less diagnostic error (one study)390 Taking 2-3 readings/hr for 6 hours is almost as good as continuous measuring every 7.5 mins for daytime ABPM (one study)538 There is no difference between taking 1, 2 or 3 recordings per hour, but using an interval of <30 mins is probably not so good for the patient (one study)196 There was no differnce between taking one, two or seven recordings per hr. However a lower number of recordings is probably better for the patient and for the longevity of the equipment (one study)353 2. When to begin measurements: SBP take measurements at the same time of day, not at opposite times (one study)622 Mean 24h BP is higher if measurements are started in the morning rather than the evening (one study)622 DBP readings are not affected by the time of day that measurements are taken (one study)622
94

3. The best time of day to take measurements All day measurements are needed (16 studies): o One hour (one study), 24h, day, night (two studies)150,654 o Day and night are best (two studies)387,527 o Day and 24h are best one study showed 24 hour BP was slightly better but using 6 hour BP was sufficient if patients are not able to tolerate / comply with 24 hours of measuring (four studies)473,573,595,608 o Night and 24 hour measurements gave greater reproducibility (two studies)46,389 o Daytime measurements are best (especially for men in one study; three studies)52,530,531 o Mean 24 hour measurements are best (two studies)114,563 o 24h BP is similar to 6 hour BP: but 6 hour BP may overestimate the value as it does not account for 24 hour BP variation (one study)200 4. How often to repeat measurements (over time) Twice - four weeks apart: there was decreased variability and WCH (one study)562; similar measurements were found at both times (one study)600 Twice - two weeks apart (one study)190 Twice (second) or successive times, or 48 hours this accounts for: circadian variation, the ABPM effect (higher BP the first time ABPM is used), the pressor effect (lower BP readings achieved with consecutive measurements) - three studies111,271,272 Four times (four weeks apart): there was high agreement between the measuerments but the fourth measurement gave a lower BP reading therefore dont label someone as being HT on the basis of an initial ABPM (1 study)420 Twice (three months apart): BP was SS lower in the day but not at night or over 24h BP measurement (one study)473 The first day of monitoring gave higher BP readings than measurements of the other six days (one study) 541 5. What day of week to perform ABPM: Monday morning BP surge is greater than on other days (one study)416 The day of the week does not affect the pressor effect ie. lower BP values are obtained with consecutive measurements (two studies)111,271 Daytime BP is lowest on Sunday; the optimal day-night ratio occurs on weekends (one study)541 BP is higher on a work day (one study)196 7.4.1.4 Evidence statements economic No relevant cost-effectiveness evidence was identified.
Update 2011
95

7.4.2
Home blood pressure measurement

Review question: In adults with primary hypertension, what protocol should be used when measuring blood pressureat home for treatment and diagnosis?
7.4.2.1
Clinical evidence The literature was searched for all years and studies published since the original guideline (2003 onwards) were included. All study types were included, if the population did not consist of people who were exclusively diabetic or had CKD. Validation studies of home blood pressure machines were excluded. Eight studies53,191,203,302,315,316,464,565,611,612 were found that fulfilled the inclusion criteria and assessed what protocol should be used when measuring home BP in for the treatment and diagnosis of adults with primary hypertension. Two of the studies (1 study;53,464 one study315,316) were each published as two separate papers reporting different assessment methods or outcomes, so these studies have only been counted once, however results from both papers are reported and referenced here. The studies addressing the question were categorised into two different types: Prognostic studies (two studies; three papers)53,53,565 those that assess the prognostic significance of home blood pressure and the optimal schedule for measurement based on outcome data Reliability / reproducibility studies (seven studies; eight papers)191,203,302,315,316,565,611,612 - those that assess any of the following - the optimal home blood pressure schedule based on: o the reproducibility of home blood pressure o its stability over time o its relationship (correlation) with ABPM values o its ability to identify people diagnosed with Hypertension / Normotension o its ability to identify treatment responders Reliability /repeatability studies were deemed to be applicable to the question because they showed which aspects of the HBPM protocol were the most reliable, and therefore served as an indication of the best / optimal HBP measurements to be taken. All prognostic studies were found to be methodologically sound / have a low risk of bias (see quality assessment summary tables in appendix F). Details of all the studies are included in Table 25 and Table 26. NOTE: all home blood pressure measurements in the studies were taken when the patient was seated. NOTE: For the prognostic studies, the best method was chosen as the method of measuring BP that best predicted (ie. statistically significant predictors and higher HR values) clinical outcomes (after adjustment for covariates in multivariate analyses). For the reproducibility/reliability studies the best method was chosen as the the method / protocol of measuring blood pressure that was the most reliable or repeatable.
Update 2011
7.4.2.2
Economic evidence No relevant economic studies were identified relating to HBPM measurement protocols.
7.4.2.3
Evidence statements clinical The studies showed the following:
96

The optimum number of readings to take (seated) Only one reading is sufficient (two studies)123,283 Two or >two readings are needed: (two studies) 203,302 Three readings are needed: (two studies)191,612 The optimum interval between measurements Take a one minute interval, not every ten seconds (one study)191 Should any readings be discarded? The first and second reading are both fine (one study)565 Discard the first reading (three studies, four papers) 315,316,565,568 Discard day one readings (one study)565 Discard day one readings (two studies) 565,568 Keep day one readings (one study)302 Discard day one and daytwo readings (one study)612 The best time of day to take measurements Morning and evening are best (two studies, three papers)53,464,565 Morning only is sufficient (one study)283 Morning and evening are best (one study) 302 How many days to take measurements Three days (four studies)123,228,283,568 Four or more days (one study)302 Five or more days (two studies)203,612
Update 2011
Seven days (one study, two papers) 315,316
97
Table 25: Study details and overall results for prognostic studies assessing the optimal home blood pressure protocol
Frequency of measurements Reference / study type Stergiou et al., 565 2010 Within-group comparison (DIDIMA STUDY) Ohkubo ey al., 2004 and Asayama et al., 53,464 2006 Within-group comparison (OHASAMA STUDY) 1766 General population (HT and NT) SOD 2 4 weeks N Population Device Consecutive readings 2 Days Time of measurement Outcomes Proposed protocol (authors conclusions) best prognostic ability more readings averaged (from 1-12) increased the prognostic ability. Take the 1st or 2nd readings; morning or evening are equally good; discard 1st day Morning and evening are equally good; there is no threshold (1-14 measurements) but take as many measurements as possible (preferably >14 measurements)
665
HT
AOD
M seated, after 5 mins rest E seated, after 5 mins rest
CV events (fatal / nonfatal)
Update 2011
M seated, within 1hr waking E seated, just before going to bed
98
Stroke
NT = normotensives; HT = hypertensives; AOD = automatic oscillometric device; SOD = semiautomatic oscillometric device; E = evening; M = morning; MS = mercury sphygmomanometer
Reliability / reproducibility studies Table 26: Study details and results for reliability/reproducibility studies assessing the optimal home blood pressure protocol
Frequency of measurements Reference / study type Verberk et al., 2005 SR study 1: 123 Celis et al., 1997 Within-group comparison
611
Population
Device
Consecutive readings
Days
Time of measurement
Mathematical method
Proposed number of measurements (authors conclusions)
MODERATE QUALITY systematic review of 4 within-group comparison observational studies (studies below) 74 Elderly HT MS 1 100 M lying in bed M after 10 mins standing E standing before going to bed E lying in bed for 10 mins M (6 10am) E (5 11am) Variability (SD); t-test Take one reading / day for 3 consecutive days
Update 2011
99
SR study 2: Stergiou et al., 568 1998 Within-group comparison SR study 3: Garcia-Vera et al., 228 1999 Within-group comparison SR study 4: 283 Imai et al., 1993 871 NT and HT SOD 1 28 48 HT SOD 1 8 189 HT AOD 2 3 workdays
Test-retest variability (SD), correlation with ABPM
Take the average of the 2nd and 3rd working day
M E At work
Test-retest variability (SD), Generalisability theory
Take one reading at work and one at home for 3 consecutive days for reliable estimates for 2 months Take one reading/day in the morning for 3 consecutive days
M - <1h after awakening
Variability (SD)
Frequency of measurements Within-group comparison Other studies Stergiou et al., 565 2010 Within-group comparison (DIDIMA STUDY) Kawabe et al., 2005 315,316 and 2008 Within-group comparison 700 General population (HT and NT) SOD 3 7 665 HT AOD 2 3 M seated, after 5 mins rest E seated, after 5 mins rest Variability (SD) More readings averaged reduced variability (from 112); discard the first day (as this gave unstable values)
M seated, within 1hr waking (before breakfast and medication, after urination) E seated, before bed (not within 30 mins bathing) M 10sec or 1 min intervals (randomised to eaither) E - 10sec or 1 min intervals (randomised to either) M 1-2 min intervals E 1-2 min intervals Mean number 27.5
Correlation with clinical diagnosis of HT / NT
Take 7 day measurements for diagnosis (more pronounced using 1st vs. mean 2nd and 3rd measurements or evening BP): this led to a diagnosis of HT more frequently, and NT less frequently Take a 1 min interval of 3 measurements (this gave a better estimate of average daytime ABPM level; 10sec intervals gave higher readings than 1 min) Take duplicate measurements, at least 4 days (evening and morning); dont discard 1st day measurements (there was NS difference in
100
Eguchi et al., 2009 Cohort study
191
57
Known or suspected HT
AOD
8 weeks (4days/ week)
Correlation with ABPM and Office BP
Johansson et al., 302 2010 Cohort study
464
HT
AOD
Correlation with ABPM
Frequency of measurements correlation with ABPM when the 1st day was excluded) Ewald et al., 2006
203
53
HT
AOD
12 weeks
M E
Post-hoc analysis of RCT (OLMETEL STUDY): thus cohort
Identification of treatment responders (sensitivity/ specificity); response to Treatment Correlation with ABPM
Take at least 2 measurements/day (this gives a better response to treatment); take at least 5 readings/week (this was the threshold for correctly predicting response to treatment) Take a minimum of 5 days; 3 consecutive morning and evening measurements; discard 1st two days and 1st reading of each triplicate (for calculating mean values) this is a time consuming protocol, so use it for a decision to start or change treatment, or for special patient groups
Verberk et al., 612 2006
216
HT
AOD
Post-hoc analysis of RCT (HOMERUS STUDY) thus cohort
M seated, after 5 mins rest (1 min interval between measurements) E seated, after 5 mins rest (1 min interval between measurements)
101
NT = normotensives; HT = hypertensives; AOD = automatic oscillometric device; SOD = semiautomatic oscillometric device; E = evening; M = morning; MS = mercury sphygmomanometer
Hypertension (partial update) D iagnosis of Hypertension
7.4.2.4
Evidence statements health economic No relevant cost-effectiveness evidence was identified.
7.5 Link from evidence to recommendations

Clinic blood pressure measurement (CBPM) on repeated clinic visits has long been the standard method for the diagnosis of hypertension and subsequent monitoring blood pressure control on treatment in clinical practice. The increased availability of automated blood pressure measuring devices has led to their increased use in clinical practice and clinical studies. Home blood pressure measurement (HBPM) or ambulatory blood pressure measurement (ABPM) both provide multiple measurements of blood pressure away from the clinic setting in a more usual environment. This raised the question as to whether ABPM and/or HBPM may provide better prognostic information with regard to the relationship between blood pressure and clinical outcomes. The predictive value for clinical outcomes of blood pressure measurement based on clinic blood pressure measurement (CBPM), home blood pressure measurement (HBPM) and ambulatory blood pressure measurement (ABPM) were compared. Three pooled analyses were identified. 210,254,326 The clinical outcomes of interest were mortality, stroke, MI, heart failure, diabetes, vascular procedures, hospitalisation for angina, and other major adverse cardiac and cerebrovascular events (MACCE). All other studies identified were observational and comprised nine prognostic studies77,159,178,210,253,254,284,326,404 that compared CBPM with ABPM, five studies86,211,438,534,564 that compared CBPM with HBPM and two studies211,534 that compared all three methods for blood pressure measurement. The studies included adult patients with normal blood pressure, suspected hypertension and known hypertension across a wide age range (30 to 71 years). All of the studies were deemed to have a low risk of bias. The results of this analysis showed that when CBPM was compared to ABPM in 8 out ofthe 9 studies77,159,178,210,253,254,284,404 ABPM was superior to CBPM at predicting clinical events there was no difference in one study. 326 ABPM can also provide data on the 24 hour average BP, daytime average BP and night-time average BP. The GDG noted that in some studies the daytime ABPM average was the most predictive of clinical outcomes, whereas in others the ABPM night-time average was the most predictive but there was no conclusive evidence suggesting a preference for day versus nighttime averages. The GDG noted that from a practical perspective, when comparing different methods, ABPM daytime averages are preferred because they allow easier comparison with CBPM and HBPM averages which are also usually taken during the daytime. There was less data comparing CBPM with HBPM in only three studies. 86,438,564 HBPM was superior to CBPM at predicting clinical outcomes in two of these studies86,438 and no difference between the methods was noted in one small study. 564 All three blood pressure measurement methods were compared with each other in only two studies in one of which there was no difference in their predictive value and in the other, ABPM and HBPM were similar to each other but superior to CBPM at predicting clinical outcomes. Taken together, the GDG concluded that the analysis of these studies showed that CBPM was never superior to ABPM or HBPM at predicting clinical outcomes. Furthermore, ABPM was never inferior to other methods and was most often the best predictor of clinical outcomes. HBPM also appeared superior to CBPM at predicting clinical outcomes but there was less data with HBPM when compared ABPM. The GDG concluded that multiple blood pressure measurements away from the clinic setting are the best predictor of blood pressure-related clinical outcomes and that to date, studies with ABPM provided the most robust evidence. The GDG considered the reasons for this and noted that this in part, could relate to the fact that ABPM and HBPM are providing more measurements and more representative data of a persons usual blood pressure away from the clinic setting. It could
Update 2011
102
Hypertension (partial update) D iagnosis of Hypertension also relate to the fact that some people diagnosed as hypertensive based on their CBPM in reality have much lower blood pressures according to their ABPM or HBPM averages, i.e. white coat hypertension or a white coat effect, and consequently are at much lower risk of clinical outcomes than their CBPMs suggest. That said, the GDG felt that more prospective data from epidemiological studies and clinical intervention trials, comparing the prognostic value of CBPM versus HBPM versus ABPM should be undertaken to better inform this prognostic relationship and better define treatment thresholds and targets according to daytime versus night-time averages and the optimal protocols for HBPM and ABPM measurement. As well as looking at prognostic studies the GDG reviewed studies that compared the sensitivity and specificity of CBPM, HBPM and ABPM in order to address the important question of which is the best method to measure blood pressure to diagnose hypertension. A recent systematic review and metaanalysis 275 examined the relative effectiveness of CBPM or HBPM versus ABPM for establishing the diagnosis of hypertension. ABPM was used as the reference standard for this analysis on the basis that: i) it is a superior predictor of clinical outcomes (see above); and ii) ABPM is the test resorted to in clinical practice when there is uncertainty about the diagnosis of hypertension, thus, ABPM is the de facto reference standard for confirming the diagnosis of hypertension in clinical practice. Thus, the GDG agreed that it was appropriate to adopt ABPM as the reference standard for the analysis of the three different BP monitoring modalities to establish the diagnosis of hypertension. This systematic review included 20 studies (N=5863). For the purposes of the analysis, an ABPM daytime average of 135/85mmHg was taken as the threshold for the diagnosis of hypertension and the performance of CBPM or HBPM versus this reference standard was compared. The CBPM and HBPM thresholds for diagnosis of hypertension were 140/90mmHg and 135/85mmHg respectively. Nine studies that used these thresholds were meta-analysed. The meta-analysis found that, compared with ABPM, CBPM had a mean sensitivity of 74.6% (95% CI, 60.7 to 84.8) and specificity of 74.6% (47.9 to 90.4) for the diagnosis of hypertension and HBPM had a mean sensitivity of 85.7% (78.0 to 91.0) and specificity of 62.4% (48.0 to 75.0). Neither differences in sensitivity or specificity between HBPM and CBPM were significant. In this context, sensitivity is the number of people who are diagnosed with hypertension according to CBPM or HBPM as a proportion of all those who actually have hypertension as defined by the ABPM reference standard. Specificity is the number who test negative for hypertension according to CBPM or HBPM as a proportion of all those that actually do not have hypertension as defined by ABPM. Thus based on the specificity results from the primary analysis of the meta-analysis CBPM will misdiagnose 25% of people who do not have hypertension as hypertensive; with HBPM this figure is 38%. In addition, based on sensitivity, with CBPM 25% of people with hypertension will mistakenly be diagnosed as not hypertensive; with HBPM that figure is 14%. However, the studies included in the meta-analysis for CBPM were in a range of populations and a sensitivity analysis was also reported which included only studies with a mean BP close to or above the diagnostic threshold. This is relevant because sensitivity and specificity vary with disease prevalence while it is often asserted that sensitivity and specificity are independent of disease prevalence it has been demonstrated that when categorisation is based on a continuous trait, as with hypertension, this is not the case. 98 In this analysis CBPM sensitivity increased to 85.6% (CI 81.0 to 89.2) and specificity decreased to 45.9 (CI 33.0 to 59.3). The HBPM studies were all in this restricted population and so the analysis for HBPM remained the same. With this restricted analysis CBPM and HBPM are virtually identical in terms of sensitivity, but HBPM was now more specific than CBPM. This sensitivity analysis was considered by the GDG to be more relevant to the guideline as screening the general population is outside of its scope. The GDG also considered a sensitivity analysis looking at the impact of the diagnostic threshold on the performance of the different diagnostic methods. Perhaps not surprisingly, the specificity of
Update 2011
103
Hypertension (partial update) D iagnosis of Hypertension CBPM for diagnosing hypertension improved when the CBPM blood pressure threshold for diagnosis is increased, i.e. those defined as hypertensive when their CBPM is higher are more likely to be hypertensive according to ABPM. However, the corollary was also true, i.e. that the accuracy of diagnosis of hypertension when comparing CBPM with the ABPM reference standard is most uncertain in those who blood pressure is close to the CBPM diagnostic threshold of 140/90mmHg. This detailed analysis suggested that the current practice of using CBPM to define hypertension will lead to drug treatment being offered to a substantial number of people who are normotensive according to ABPM. The GDG recognised that these data have profound implications for the diagnosis of hypertension. Firstly, they suggest that some patients randomised and treated in clinical outcome trials on the basis of their CBPM, may not have been hypertensive, potentially diluting and underestimating the true benefits of treatment in those who were hypertensive. Secondly and perhaps more importantly, these findings suggest that the current practice of using a series of CBPM alone for the diagnosis of hypertension can lead to inaccurate diagnosis. Screening for hypertension was outside the scope of this guideline. However, the GDG agreed it is not practical to use ABPM or HBPM as a screening tool, despite them potentially offering greater accuracy than CBPM. The working assumption was that CBPM would still be used for screening patients and that the key decision that remained was how the diagnosis should be confirmed. Taking into account the prognostic data and the meta-analysis of sensitivity and specificity, the GDG agreed that ABPM appeared to provide the best method of confirming a diagnosis of hypertension. The GDG also considered that a change in practice as profound as this required clear evidence that ABPM would not only be a more effective means of diagnosis but also, a more cost-effective means of establishing the diagnosis of hypertension.
Update 2011
The GDG agreed the most practical method to diagnose hypertension would be to use CBPM as a screening tool and that those people with a CBPM 140/90mmHg measured using the recommended standardised conditions, should then be offered ABPM to confirm or refute the diagnosis of hypertension based on a diagnostic threshold of an ABPM daytime average of 135/85mmHg. The GDG reviewed the data regarding the number of measurements required to establish the ABPM daytime average blood pressure. The number of measurements taken during prognostic studies varied from every 15 minutes to every hour during the daytime. The GDG concluded that two measurements per hour should be taken during normal waking hours, e.g. 08.00hrs to 22.00hrs and that a minimum of 14 readings should be used to derive the daytime average blood pressure. This means that patients would not necessarily need to wear the ABPM monitor for a full 24hrs, depending on the time the monitoring session was initiated. For practical reasons and efficiency in use of the monitors, not every monitoring session will begin at 08.00hrs and some patients will start their session in the afternoon. In these patients continuation of monitoring for 24hrs will be required to capture the normal waking hours across a spread of 24hrs. Consideration would also need to be given to shift and night workers whose normal waking hours will differ. When ABPM is poorly tolerated, inconvenient for the patient, or the patient does not want to undergo ABPM, HBPM should be offered to establish the diagnosis of hypertension. HBPM may also be preferred to monitor the control of blood pressure in treated patients with a significant white coat effect, or where this is the patients preference for monitoring their blood pressure control (see section 9.6 monitoring blood pressure control). Regarding use of HBPM, the GDG noted that a range of strategies had been used in studies to establish the HBPM average blood pressure reading. The optimal timing of measurements and the number of measurements required was reviewed. The GDG concluded that a standardised approach was needed and recommended that patients should measure their blood pressure whilst seated and relaxed and that at each measurement session, two blood pressure measurements should be taken, at least one minute apart, in the morning and the evening. The recording should continue for at least 4 days and ideally 7 days. The readings on the
104
Hypertension (partial update) D iagnosis of Hypertension first day should be discarded and the readings for all remaining days should be used to establish the HBPM average. The GDG discussed a number of caveats to recommendations regarding the use of ABPM to establish the diagnosis of hypertension. Some people may have severe hypertension at screening with CBPM (i.e. systolic BP 180mmHg and/or diastolic BP 110mmHg) and in such cases, clincians should not delay treatment whilst awaiting the results of ABPM in these cases, the subsequent ABPM will serve to confirm the diagnosis and severity of the hypertension; ii) some people will have atrial fibrillation or other significant pulse irregularity that might render automated BP monitoring (ABPM and HBPM) inaccurate or impossible, in such cases manual auscultation of blood pressure in the clinic would be the only alternative; and iii) some people may not tolerate ABPM in these people HBPM can be used an alternative on the grounds of better prognostic value and better specificity for hypertension. However, the GDG noted that based on current data, HBPM could not be considered equivalent to ABPM with regard to accuracy of diagnosis and emphasised that that ABPM is the preferred means of confirming or refuting the diagnosis of hypertension. The GDG also discussed whether ABPM was necessary for confirmation of diagnosis in all patients, or whether it could be used more selectively, e.g. only in those close to the diagnostic threshold. The GDG noted that even in people with stages 2, or resistant hypertension, a significant white coat effect can occur, which would be important to document to facilitate decisions about the best strategy for subsequent monitoring of blood pressure control on treatment. The need for ABPM for people with evidence of target organ damage, e.g. LVH or albuminuria was also discussed by the GDG. It was noted that target organ damage may not always be due to hypertension, even when the two appear to co-exist. For example, the presence of ECG LVH in a patient subsequently shown not to be hypertensive on ABPM would prompt consideration of alternative causes for the ECG abnormality. Furthermore, some people have higher blood pressures away from the clinic (so called masked hypertension) and ABPM could reveal much worse blood pressure control levels than apparent in the clinic this would be important to know. Finally, the GDG noted that people with target organ damage are a higher risk group and the best possible assessment of their blood pressure level when initiating treatment seemed appropriate, mindful of the better prognostic value of ABPM when compared to CBPM. Overall, the GDG could not identify a strong evidence-base or clinical argument against the use of ABPM to improve the accuracy of diagnosis of hypertension, which for many people results in exposure to life-long treatment. The residual concern in the GDG deliberations was not whether this was the right thing to do but rather, whether the strategy would be cost-effective (see below) and whether the practical challenges of implementing an ABPM-based strategy for diagnosis could be overcome. The GDG were also mindful of the concerns about the accuracy of automated devices for measuring blood pressure in people with atrial fibrillation and considered this an important area for technology development to see if such problems can be overcome. The GDG noted that In some patients with chronic atrial fibrillation with good rate control, automated devices can function effectively but concluded that until automated devices, validated for routine clinical use are available for people with atrial fibrillation, manual auscultation over the brachial artery is the only practical alternative to measure blood pressure in people with significant cardiac rhythm irregularity. As noted above, evaluation of the effectiveness of different methods for measuring blood pressure to establish the diagnosis of hypertension suggested that ABPM would be the most accurate method, avoiding clinical disease labelling and treatment of people who were not truly hypertensive according to their ABPM average blood pressure. The GDG noted, however, that despite the clear effectiveness of ABPM in improving the specificity and sensitivity of diagnosis for hypertension, ABPM devices are considerably more expensive than simple desk top blood pressure monitors and the GDG recognised the obvious potential cost implications of recommending the more widespread use of ABPM for the routine diagnosis of hypertension. The GDG thus identified modelling of the cost effectiveness of different methods for blood pressure measurement as the highest priority for economic analysis as a
Update 2011
105
Hypertension (partial update) D iagnosis of Hypertension prior literature search had identified no published work addressing this key question in sufficient detail. The cost-effectiveness analysis compared CBPM, HBPM or ABPM for confirming a diagnosis in people with suspected hypertension. The GDG spent considerable time discussing the various factors that would potentially impact on the costs of using ABPM and also HPBM as an alternative to current standard practice of using a series of CBPM readings to confirm the diagnosis of hypertension. These included the number and type of healthcare appointments required to confirm a diagnosis with each method, the failure rate associated with ABPM and HBPM and the number of uses of the devices each year. As well as initial diagnosis costs, the analysis took into account downstream costs including hypertension treatment, checkups and development of cardiovascular disease. Health benefits were quantified in terms of QALYs. A summary of the cost-effectiveness analysis is provided in Section 7.3 with full details available in Appendix J: Cost-effectiveness analysis. Contrary to what might have been expected and mindful of the higher costs of ABPM devices, the cost-effectiveness analysis found ABPM to be the most cost effective option for the diagnosis of hypertension across a range of age groups in both men and women. Remarkably, in most groups ABPM was found to actually improve health (increased QALYs) and reduce costs, suggesting that use of ABPM for the diagnosis of hypertension has the potential to be cost saving for the NHS. The GDG noted that this conclusion was robust to a wide range of sensitivity analyses including those varying the cost of ABPM, the failure rate for ABPM, the level of CVD risk and the prevalence of true hypertension in the population. Unsurprisingly, the conclusion was sensitive to assumptions regarding the accuracy of diagnosis with each method, e.g. when the other methods (CBPM or HBPM) were assumed to be as accurate as ABPM which the effectiveness analysis suggests they are not. The conclusion was also sensitive to the assumption that people who were not hypertensive but were treated did not receive benefits from treatment, which they might. On the other hand, the analysis did not model the impact of unnecessarily treating people who are not hypertensive and the costs, inconvenience, adverse effects of treatment and impact disease labelling may have on individual patients incorrectly diagnosed as hypertensive. The extensive GDG deliberations on the cost effectiveness analysis concluded that the use of ABPM for the routine diagnosis of hypertension, using a daytime average threshold of 135/85mmHg, in people who have previously been identified as potentially hypertensive at a threshold of 140/90mmHg using a CBPM, would be both cost-effective and in almost all cases, cost saving for the NHS, as well as improving the accuracy of diagnosis for patients. The GDG thus recommended that ABPM should be implemented for the routine diagnosis of hypertension in primary care. The GDG also discussed other important aspects when considering the diagnosis of hypertension including: i) whether there might be an underlying secondary cause for the elevated blood pressure that might warrant referral for specialist evaluation; ii) whether the patient might have accelerated hypertension requiring emergency in-patient care; and iii) the need to assess for the presence of target organ damage and formally assess cardiovascular disease risk. The GDG recognised and discussed the considerable challenges for implementation of this recommendation. Sufficient numbers of validated ABPM devices would need to be procured and adequately maintained. Staff would need to be trained in their use and the interpretation of data generated by the ABPM reports. The existing recommendations on use of appropriate cuff size (see section 6.2) and recognition that automated measurements may be unreliable or impossible in people with significant pulse irregularity (e.g. atrial fibrillation) (see section 6.5) still apply. Some people will not tolerate ABPM and in others the procedure will fail. The GDG modelled an anticipated failure rate of 5%, ranging to a more extreme failure rate of 10% in sensitivity analyses in the cost effective analysis and ABPM remained the most cost effective option for the diagnosis of hypertension. In those unable to tolerate or unwilling to undergo ABPM, the GDG recommended HBPM as an alternative means of confirming the diagnosis of hypertension with emphasis that ABPM
Update 2011
106
Hypertension (partial update) D iagnosis of Hypertension is the preferred method. For those with significant pulse irregularity, ABPM and HBPM are likely to be unreliable methods for blood pressure measurement and a series of CBPM readings via manual auscultation (see section 6.1.1) remains the only suitable option. Finally, the GDG discussed the practicalities of implementing this strategy for the diagnosis of hypertension. That implementation of this strategy is a challenge is acknowledged. Presently, some but not all primary care practices have access to ABPM devices, others do not. Some practices access ABPM through referral to secondary care. Few practices presently have sufficient numbers of devices to increase their use as required by this guideline recommendation. The GDG discussed the fact that models of future care cannot just be based on what we do now and considered it likely that alternative models of service provision would emerge, reflecting first and foremost what was best and most convenient for patients and local demand. The GDG considered it inevitable that the costs of ABPM devices will fall as demand for their use increases and that different models of ABPM provision will evolve over time to meet local demand.
7.6 Recommendations
8. When considering a diagnosis of hypertension, measure blood pressure in both arms: If the difference in readings between arms is more than 20 mmHg, repeat the measurements. If the difference in readings between arms remains more than 20 mmHg on the second measurement, measure subsequent blood pressure in the arm with the higher reading. [new 2011] 9. If the clinic blood pressure is 140/90 mmHg or higher, offer ambulatory blood pressure monitoring (ABPM) to confirm the diagnosis of hypertension. [new 2011] 10.If a person is unable to tolerate ABPM, home blood pressure monitoring (HBPM) is a suitable alternative to confirm the diagnosis of hypertension. [new 2011] 11.If the person has severe hypertension, consider starting antihypertensive drug treatment immediately, without waiting for the results of ABPM or HBPM. [new 2011] 12.While waiting for confirmation of a diagnosis of hypertension, carry out investigations for target organ damage (such as left ventricular hypertrophy, chronic kidney disease and hypertensive retinopathy) (see 21) and a formal assessment of cardiovascular risk using a cardiovascular risk assessment tool (see 20). [new 2011] 13.If hypertension is not diagnosed but there is evidence of target organ damage such as left ventricular hypertrophy, albuminuria or proteinuria, consider carrying out investigations for alternative causes of the target organ damage. [new 2011] 14.If hypertension is not diagnosed, measure the persons clinic blood pressure at least every 5 years subsequently, and consider measuring it more frequently if the persons clinic blood pressure is close to 140/90 mmHg. [new 2011] 15.When using ABPM to confirm a diagnosis of hypertension, ensure that at least two measurements per hour are taken during the persons usual waking hours (for example, between 08:00 and 22:00). Use the average value of at least 14 measurements taken during the persons usual waking hours to confirm a diagnosis of hypertension. [new 2011] 16.When using HBPM to confirm a diagnosis of hypertension, ensure that:
Update 2011
107
Hypertension (partial update) D iagnosis of Hypertension for each blood pressure recording, two consecutive measurements are taken, at least 1 minute apart and with the person seated and blood pressure is recorded twice daily, ideally in the morning and evening and blood pressure recording continues for at least 4 days, ideally for 7 days. Discard the measurements taken on the first day and use the average value of all the remaining measurements to confirm a diagnosis of hypertension. [new 2011] 17.Refer the person to specialist care the same day if they have: accelerated hypertension, that is, blood pressure usually higher than 180/110 mmHg with signs of papilloedema and/or retinal haemorrhage or suspected phaeochromocytoma (labile or postural hypotension, headache, palpitations, pallor and diaphoresis). [2004, amended 2011] 18.Consider the need for specialist investigations in people with signs and symptoms suggesting a secondary cause of hypertension. [2004, amended 2011]
Update 2011
108
Hypertension (partial update) Assessing cardiovascular risk, target organ damage and secondary causes of hypertension
8 Assessing cardiovascular risk, target organ damage and secondary causes of hypertension
There are four key objectives in the assessment of a person with suspected hypertension; i) to confirm whether or not blood pressure is elevated (see section xxx); ii) to document the presence or absence of blood pressure related target organ damage damage (e.g. left ventricular hypertrophy, hypertensive retinopathy, increased albumin:creatinine ratio); iii) to evaluate the persons cardiovascular risk either due to established cardiovascular disease or high cardiovascular disease risk states (e.g. diabetes or CKD), or by calculation of their 10 year CVD risk estimate (ref section and NICE guidance), and iv) to consider whether their may be secondary causes for the hypertension. The risk of clinical events associated with hypertension is not only determined by the level of blood pressure but also by; i) the presence of target organ damage; ii) the presence of established cardiovascular disease (iscahemic heart disease or heart failure, cerebrovascular disease, peripheral vascular disease) or concomitant disease associated with high cardiovascular disease risk, e.g. diabetes or CKD; or iii) the calculated cardiovascular risk (estimated from factors such as age, gender, smoking history, etc.). Therefore, routine assessment of simple markers of target organ damage, a clinical history and examination to identify associated cardiovascular disease and when indicated, cardiovascular risk calculation, all form part of the routine assessment of a patient with suspected or confirmed hypertension. This assessment will also help clinicians to decide the appropriate blood pressure threshold at which to consider drug therapy for the treatment of hypertension and whether any additional therapies to reduce cardiovascular disease risk (e.g. statins and antiplatelet therapy) should also be offered to the patient. The clinical history, examination and routine blood and urine tests will also alert the clinician to possible secondary causes of hypertension, some of which are potentially life threatening (e.g. phaeochromocytoma), and others which might be amenable to potentially curative interventions (e.g. Conns adenoma, fibromuscular dysplasia).
8.1.1
Hypertension and cardiovascular disease

An analysis of 61 prospective observational studies, involving nearly one million individuals, explored the relationship between blood pressure level and 12,000 strokes and 34,000 ischaemic heart disease events over an average of 13.2 years follow-up361. Across age bands from 40 to 89, reduction in usual diastolic blood pressure of 20 mmHg systolic or 10 mmHg diastolic blood pressure was associated with reductions in death from stroke and ischemic heart disease of about one half, slightly more in the youngest and slightly less in the oldest. Findings were similar for men and women, for different types of stroke, and consistent across the range of blood pressure (down to 115/75 mmHg). An earlier analysis of nine observational studies, involving 420,000 individuals explored the relationship between blood pressure level and 843 subsequent strokes and 4,856 coronary events over an average of 7 years follow-up379. Reductions in usual diastolic blood pressure of 5, 7.5 and 10 mmHg were associated with reductions in stroke of 34%, 46% and 56% and coronary heart disease of 21%, 29% and 37% respectively. The relationship between blood pressure and disease was constant over a wide range suggesting there is no clear threshold below which further reduction in blood pressure becomes unbeneficial or harmful. The implication of these two studies is that some or all of the predicted benefits, found by comparing individuals with different usual blood pressure levels, could be obtained by one patient maintaining a similar reduction. A systematic review of 14 antihypertensive randomised drug trials (diuretics or beta-blockers compared with placebo) included 37,000 patients135. A mean reduction in diastolic blood pressure of
109
Hypertension (partial update) Assessing cardiovascular risk, target organ damage and secondary causes of hypertension 56 mmHg over 5 years achieved a relative reduction in stroke of 42% (95% CI: 3350%) and CHD of 14% (95%CI: 422%). The authors concluded that virtually all of the epidemiologically observed benefit from reduced stroke and over half of the reduction in coronary heart disease could be achieved by lowering blood pressure.
8.2 Routine clinical investigations

A full cardiovascular assessment should be conducted in patients with persistently raised blood pressure who do not have established cardiovascular disease. There is no firm evidence from which to define the exact composition of assessment and recommendations are consensus-based. Medical history, physical examination, and limited diagnostic testing serve to identify an individual patient's profile of cardiovascular risk factors including age and gender, smoking, hyperlipidaemia, diabetes, and family history of cardiovascular disease. Testing may detect diabetes and identify signs of developing target organ damage such as left ventricular hypertrophy and angina. It may also detect secondary causes of hypertension. The guideline group identified the following tests as necessary to obtain an accurate profile of cardiovascular risk. These tests may help identify diabetes, evidence of hypertensive damage to the heart and kidneys, and secondary causes of hypertension such as kidney disease: Urine strip test for blood and protein Blood electrolytes and creatinine, and eGFR Blood glucose Serum total and HDL cholesterol 12 lead electrocardiogram.
Update 2011
8.2.1
Urine testing for proteinuria

The presence of protein in urine identifies patients with kidney damage, but does not distinguish between patients who have renal disease and secondary hypertension and those in whom kidney damage is due to essential hypertension. The test consists of dipping a test strip, which is impregnated with chemicals which react to protein, into a sample pot of urine. After 3060 seconds (or according to manufacturer's instructions) the strip is read alongside a colour code provided. A more sensitive test for urine protein is available by requesting the local chemical biochemistry laboratory to assay microalbumin in a random specimen of urine. For further information refer to NICE Clinical Guideline 73.
8.2.2
Blood electrolyte, urea, creatinine, glucose and total/HDL cholesterol levels

These are measured in serum or plasma (glucose) using standard clinical biochemistry methods. Sodium and potassium levels are checked to exclude hypertension resulting from adrenal disease. Likewise, urea and creatinine measurements, which reflect kidney function, are measured to exclude kidney disease as a secondary cause of hypertension Glucose levels are tested to evaluate diabetes and cholesterol profiles are used to assess cardiovascular risk. 12 lead electrocardiogram. Refer to NICE guidance on Diabetes (Clinical Guidelines 15 and 87). From an ECG it is possible to determine heart rate, rhythm, conduction abnormalities, left ventricular size and damage to specific regions of the heart muscle. The presence of electrocardiographic left ventricular hypertrophy is a variable used in cardiovascular risk calculators. An echocardiogram might be considered, to confirm or refute the presence of LVH suggested by ECG findings.
110
8.3 Cardiovascular Risk Assessment

Risk models have been developed (as charts, graphs or computer programmes) to allow clinicians to predict the likelihood of patients developing coronary or cardiovascular disease using lifestyle and clinical markers (See NICE Lipids Modification, CG67). Although they vary in detail, risk models may estimate an individual's risk of coronary heart disease and stroke over the next ten years using their gender, age, diabetic status, smoking status, total serum cholesterol (TC), high density lipoprotein cholesterol (HDL-C) and blood pressure. An important aspect of risk models is that they lead the clinician to address a patient's overall profile of risk rather than treat one risk factor in isolation. Risk factors have a cumulative effect, and an individual with a number of modest risk factors may be at greater risk of developing cardiovascular disease than an individual with one high risk factor23. Since several risk factors are potentially modifiable, an important aspect is which of these to address and in what order.
8.4 Secondary Hypertension

An identifiable cause of hypertension is more likely when hypertension occurs in younger patients (less than 40 years of age), worsens suddenly, presents as accelerated hypertension (BP more than 180/110 mmHg with signs of papilloedema and/or retinal haemorrhage) or responds poorly to treatment. [III] An elevated creatinine or reduced eGFR indicates renal disease. Labile or postural hypotension, headache, palpitations, pallor and diaphoresis are potential signs of pheochromocytoma. Hypokalaemia, abdominal or flank bruits, or a significant rise in serum creatinine when starting an ACEi or ARB may indicate renovascular hypertension. Isolated hypokalaemia may be due to hyperaldosteronism. Potential signs of Cushing syndrome include osteoporosis, truncal obesity, moon face, purple striae, muscle weakness, easy bruising, hirsutism, hyperglycemia, hypokalaemia, and hyperlipidaemia. [III] Secondary hypertension refers to high blood pressure from an identifiable underlying cause. It may occur in up to 10% of hypertension cases, the most common cause being chronic renal disease. Other principal identifiable causes are renovascular hypertension, pheochromocytoma, Cushing syndrome, and primary aldosteronism. Signs and symptoms of the main causes of secondary hypertension and available diagnostic tests are summarised below, although many of these techniques are not provided in primary care but accessed through specialist referral. We retrieved no useful diagnostic studies which might establish primary care screening characteristics for secondary causes of hypertension as a basis for referral: current advice is simply to be aware of signs and symptoms and refer on the basis of a high index of suspicion and where the findings are likely to necessitate specialist management.
8.4.1
Renal and renovascular disease

Chronic kidney disease is the most common identifiable cause of hypertension occurring in 2% to 5% of patients182. The British National Formulary advises against routinely using ACEi or ARBs in patients with known or suspected renovascular disease26. Signs and symptoms indicating that hypertension may be associated with renal disease are: young onset of hypertension (before 40 years of age), sudden onset of hypertension or progressive deterioration in middle age, accelerated hypertension (BP more than 180/110 mmHg with signs of papilloedema and/or retinal haemorrhage), oliguria (urine output <250 ml/day) or anuria (<50 ml/day), oedema, acidosis (acidic blood, <pH), abnormal serum urea or reduced eGFR, systolic or diastolic bruit467, drug resistant hypertension or increased creatinine with ACEi or ARB, hypertension onset > 60 years, DBP >110 mmHg, and anaemia (lowered red blood cell count) resulting in insufficient oxygen to tissues and organs. Although renal artery stenosis is suggested by the presence
111
Hypertension (partial update) Assessing cardiovascular risk, target organ damage and secondary causes of hypertension of an abdominal or flank bruit, it is an insensitive test (sensitivity=65%; specificity=90%). When present it is a good marker (positive likelihood ratio=6.5) but when absent does not rule out renal artery stenosis (negative likelihood ratio=0.4)182,505. Renal disease may be diagnosed by elevated serum levels of urea or creatinine (found by a blood test) or reduced eGFR . Specialist investigation includes magnetic resonance angiography for imaging of the kidneys, and duplex ultrasound scanning directly measuring the size of the kidneys467, 35. Test sensitivities have been reported for these investigations182.
8.4.2
Pheochromocytoma
A pheochromocytoma is a tumour which produces and releases large amounts of adrenaline and noradrenaline (hormones) into the blood. It is rare and may occur in between 0.04% and 0.1% of patients; about 10% are malignant. Adrenaline causes an increase in heart rate and contractility, while noradrenaline increases systemic vascular resistance. Patients with signs and symptoms of pheochromocytoma need immediate specialist investigation given the seriousness of the condition and risk to the patient. The definitive treatment of pheochromocytoma is surgical removal of the tumour. Signs and symptoms include a rapid heart rate, headache, high blood glucose levels, elevated basal metabolic rate, facial flushing, nervousness, sweating, decreased gastrointestinal movements and oedema. Diagnostic techniques include plasma or 24 hour urine collections for metadrenaline and normetadrenaline 22,250. Following positive findings two types of imaging study may be used to locate the tumour: metaiodobenzyl-guanidine (MIBG) scintigraphy and computed tomography (CT).
8.4.3
Hyperaldosteronism (primary aldosteronism)

Aldosterone is a hormone that regulates sodium and water balance. Hyperaldosteronism can due to bilateral adrenal hyperplasia or Conns adenoma occurring in 0.01% to 0.03% of patients182,570], although its prevalence is contested and may be much higher [364. Signs and symptoms include sodium retention, and hypokaelaemia leading to heart rhythm irregularities and possibly muscle weakness. The hypokaelaemia may only occur when diureticinduced hypokalaemia is not explained by natural causes467. Measurement of plasma aldosterone levels and plasma renin activity as the aldosterone:renin ratio may be used to detect primary aldosteronism250. As with any laboratory test, standardisation of laboratory assays is important.
8.4.4
Cushing's syndrome
Cushing's syndrome is a syndrome generated by excess glucocorticoids. Cushings Disease specifically refers to over-production of ACTH by the pituitary gland and is the most common form of the syndrome. Over-production of cortisol can also be due to a tumour in the adrenal gland, either benign (an adenoma), or malignant (a carcinoma) and in this variant is not dependent on ACTH. Production of ACTH in an organ or gland other than the pituitary or adrenal gland (e.g. thymus gland, lung, pancreas) is called ectopic corticotrophin-releasing production469. Cushing's syndrome may occur in 0.1% to 0.6% of patients. Signs and symptoms include hypertension, sudden onset of weight gain, central obesity, moon face, weakness, fatigue, backache, headache, glucose intolerance, oligomenorrhoea (infrequent menstruation), amenorrhoea (abnormal discontinuation of periods), increased thirst, increased
112
Hypertension (partial update) Assessing cardiovascular risk, target organ damage and secondary causes of hypertension urination, impotence, muscle atrophy, depression, insomnia, thinning of the skin, cutaneous hyperpigmentation (darkening of the skin), osteoporosis469. Diagnosis of Cushing's syndrome begins with a single dose overnight dexamethasone-suppression test. A differential diagnosis is achieved by measuring plasma ACTH together with either a long dexamethasone suppression test or a corticotrophin-releasing hormone (CRH) stimulation test217,437.
8.5 Other identifiable causes of hypertension

8.5.1 Hypothyroidism
Hypothyroidism is under production of the hormone thyroxine (which controls metabolism) by the thyroid gland. Hypertension in hypothyroid patients may result from altered levels of renin, angiotensin and aldosterone. After thyroid replacement therapy diastolic blood pressure returns to normal in patients with hypothyroidism suggesting a cause-and-effect relationship185,329,509. Signs and symptoms include lethargy, fatigue, weight loss, hair loss, confusion, nausea, bone pain, muscle weakness, slow heart rate. Hypothyroidism is associated with increased diastolic blood pressure75,572. Hypothyroidism is diagnosed by measuring thyroid stimulating hormone levels467.
8.5.2
Hyperthyroidism
Hyperthyroidism is the excessive secretion of thyroxine by the thyroid gland. Signs and symptoms include increased systolic blood pressure, increased metabolic rate, enlargement of the thyroid gland, tachycardia (increased heart rate), exophthalmia (abnormal protrusion of the eyeball in the orbit), oedema, dry hair and skin, weight gain, goitre (enlarged thyroid gland)314. Hyperthyroidism is diagnosed by measuring thyroid stimulating hormone levels467.
8.5.3
Obstructive sleep apnoea

Obstructive sleep apnoea is caused by the upper airway becoming obstructed during sleep. It is more prevalent in men. Signs and symptoms include daytime somnolence (unnatural drowsiness and sleepiness), obesity, snoring, lower extremity oedema, nocturia and morning headaches. The main diagnostic technique is a polysomnograph to monitor normal and abnormal physiological activity during sleep 250,467. Please refer to NICE Technology Appraisal 139 (www. http://guidance.nice.org.uk/TA139/Guidance/pdf/English) for guidance on continuous positive airway pressure (CPAP).
8.5.4
Coarctation of aorta
Coarctation of aorta is a congenital condition where a segment of the aorta is too narrow, reducing oxygenated blood flow around the body. Signs and symptoms include high blood pressure, decreased or delayed femoral pulse, abnormal chest radiograph. Diagnostic techniques: doppler or CT imaging of the aorta467.
8.5.5
Acromegaly
Acromegaly is due to excess production of growth hormone. Signs and symptoms of acromegaly include hypertension, cardiomegaly, enlarged facial features, enlarged jaw, headache and arthralgia, hypertrichosis, excessive sweating, tiredness, weakness, somnolence and impaired glucose tolerance360. Acromegaly is diagnosed by evidence of increased growth hormone secretion360.
113
8.5.6
Drugs
A number of medications are known to cause raised blood pressure. These include decongestant found in inhaled cold remedies, may raise diastolic blood pressure517,547. Oral contraceptive pills containing oestrogen may cause small, and occasionally pronounced, rises in blood pressure. In rare cases accelerated hypertension may occur535. Other drugs that may raise blood pressure include immunosuppressive agents, nonsteroidal anti-inflammatory drugs, COX-2 inhibitors, weight loss agents, stimulants (for example, cocaine), mineralocorticoids, antiparkinsonian agents, monoamine oxidase inhibitors, anabolic steroids, sympathomimetics467.
8.6 Recommendations
For NICE guidance on the early identification and management of chronic kidney disease see Chronic kidney disease (NICE clinical guideline 73, 2008). 19.Use a formal estimation of cardiovascular risk to discuss prognosis and healthcare options with people with hypertension, both for raised blood pressure and other modifiable risk factors. [2004] 20.Estimate cardiovascular risk in line with the recommendations on Identification and assessment of CVD risk in Lipid modification (NICE clinical guideline 67) h. [2008] 21.For all people with hypertension offer to: test for the presence of protein in the urine by sending a urine sample for estimation of the albumin:creatinine ratio and test for haematuria using a reagent strip take a blood sample to measure plasma glucose, electrolytes, creatinine, estimated glomerular filtration rate, serum total cholesterol and HDL cholesterol
Update 2011
examine the fundi for the presence of hypertensive retinopathy arrange for a 12-lead electrocardiograph to be performed. [2004, amended 2011]
8.7 Research recommendations

2. In people aged under 40 with hypertension, what is the most accurate method of assessing the lifetime risk of cardiovascular events and the impact of therapeutic intervention on this risk? Current short-term (over 10 years) risk estimates are likely to substantially underestimate the lifetime cardiovascular risk of younger people (aged under 40) with hypertension, because shortterm risk assessment is powerfully influenced by age. Nevertheless, the lifetime risk associated with untreated stage 1 hypertension in this age group could be substantial. Lifetime risk assessments may be a better way to inform treatment decisions and evaluate the cost effectiveness of earlier intervention with pharmacological therapy.
Clinic blood pressure measurements must be used in the calculation of cardiovascular risk.
114
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets
9 Initiating and monitoring treatment, including blood pressure targets

The diagnostic threshold for defining hypertension has been progressively lowered over the past 50 years as treatment of hypertension has been shown to be beneficial at reducing cardiovascular morbidity and mortality when initiated at progressively lower blood pressure thresholds. During that time, the focus also shifted from hypertension diagnosed purely on the basis of diastolic pressure towards systolic pressure thresholds being the most common indication for treatment this reflects the increased prevalence of hypertension with ageing and the usual progressive rise in systolic pressure with age. In the 2004 guideline, two different grades of hypertension were defined, Grade 1 hypertension (140-159/90-99mmHg) and Grade 2 hypertension (i.e 160/100mmHg). The guideline recommended that patients with Grade 2 hypertension should be offered pharmacological treatment. The guideline was more cautious with regard to pharmacological treatment for uncomplicated Grade 1 hypertension (i.e. in those without evidence of target organ damage, cardiovascular disease, CKD or diabetes or at a calculated 10 year CVD risk <20%). This 2011 guideline partial update reviewed evidence published since the cut point of the last review (2003) to determine whether the existing recommendations for blood pressure thresholds for diagnosis and treatment of hypertension should be revised. Furthermore, in light of the recommendation in this guideline update that an ABPM daytime average blood pressure will hereafter be the preferred method for confirming the diagnosis of hypertension, the thresholds for diagnosis and grades of hypertension also needed to be reviewed with regard to ABPM daytime averages. Once a decision has been made to initiate pharmacological treatment for hypertension, the next key question was how low should blood pressure be lowered? i.e. what is the recommended blood pressure target? The 2004 guideline noted that the evidence base to support a recommendation for an optimal treatment target for hypertensiion was less substantial than it should be. International consensus has specified an optimal treatment target for hypertension of <140/90 mmHg and in some cases even lower targets for people with established cardiovascular or renal disease or diabetes. There has also been concern but little evidence, as to the efficacy, safety and appropriate blood pressure target for the people at advanced age with hypertension (greater than 80 years). Consequently, studies examining optimal treatment targets have been reviewed.
Update 2011
9.1 Blood pressure thresholds for initiating pharmacological treatment

Review question: In adults with primary hypertension, at what blood pressure should treatment be initiated?
9.1.1
Clinical evidence
The literature was searched for studies published since the original guideline (2003 onwards). All study types were included, if the population did not consist of people who were exclusively diabetic or had CKD. Studies were excluded if they did not stratify results into more than one different BP value / threshold. Thirty studies (31 papers)49,50,54,57,60,61,68,89,101,119,136,165,206,208,213,243,244,247,269,285,291,313,331,332,340,351,454,466,521,546,629 were found that fulfilled the inclusion criteria and assessed at what BP should treatment be initiated (appropriate threshold for intervention). One of the studies60,61 was published as two separate papers reporting different assessment outcomes, so this study has only been counted once, however results from both papers are reported and referenced here.
115
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets The studies addressing the question were categorised into three different types: 1. SRs / MAs (three studies)54,206,351 . The SRs/MAs were of high quality however the studies they included were either low quality (observational)54,206 or low to high (RCTs).351. 2. Prognostic studies (27 studies; 28 papers)49,50,57,60,61,68,89,101,119,136,165,208,213,243,244,247,285,291,313,331,332,340,454,466,521,546,629 - those that assess the risk of developing clinical outcomes (over time) at different BP values. Most of the prognostic studies were found to be methodologically sound (see quality assessment summary tables in appendix F) except for the following eight studies which had (or were rated as unclear for) three or more of the six potential methodological flaws (Fagard 2007, Gudmundsson 2005, Obara 2007, Okayama 2006, Sleight 2009, Fagard 2004, Britton 2009, Conen 2007101,136,206,208,243,454,466,546). Prognostic studies were divided into four categories: those that assessed BP measured by either clinic, home, ambulatory or self-reported / unknown methods. 3. Blood pressure equivalence studies (one study)269 those that calculate equivalent blood pressures using different measurement methods (home, ABPM or clinic), in order to set thresholds for the diagnosis and treatment of HT. All these studies were observational and therefore low quality. Data from the included studies was not pooled into a meta-analysis. This was because for many studies only HRs were given rather than the number of patients with events, and data was often stratified differently in the studies (for example, by age, gender, treated/untreated or other population characteristics), making it not possible to pool together. Additionally, it was deemed inappropriate to pool the studies because the studies themselves differed considerably in their design and analysis, particularly regarding the following areas: blood pressure values, groups and thresholds used blood pressure measurement methods used outcome measures (and definitions of outcomes) used follow-up times used covariates taken into account in analyses Details of all the studies are included in Table 27, Table 28 and Table 30. Table 29 summarises the numerical results for selected outcomes of the prognostic studies included for this review. The full data for all outcomes can be found in the evidence tables in the appendix.
Update 2011
116
Systematic reviews/Meta-analyses
Initiating and monitoring treatment, including blood pressure targets
Table 27: Study details and results for SRs/MAs assessing the risk of developing clinical outcomes at different BP thresholds.
BP measureme nt method Clinic Followup Mean 9.5 years BP values at baseline (groups / thresholds); mmHg Optimal: <120/ <80 Normal: 120-129/80-84 High normal: 130-139/85-89 Grade 1 (mild) HT: 140-159/ 90-99 Grade 2 (moderate) HT: 160179/ 100-109 Grade 3 (severe) HT: 180/110 10mm SBP increments from 120 180 mmHg Best BP threshold (authors conclusions) Untreated groups: risk (HR) of first stroke increased linearly with BP. Treated people with optimal BP had higher risk of stroke than untreated people with optimal BP.
Reference Asayama et 54 al., 2009 MA of data from 4 cohort studies
N 4571
Population General population (HT and NT)
Study design Prognostic: Risk (HR) of developing clinical outcomes
Outcomes Stroke; death from stroke
Update 2011
Law et al., 351 2009 SR/MA of 108 RCTs
248,445
HT and NT People of any age, disease status, preTreatment BP and use of other drugs
Clinic
Mean 3.5 years
BP difference trials designed to achieve a difference in BP between randomised groups
CHD events; stroke
BP treatment reduced risk of CVD and stroke, regardless of patients pre-treatment BP (as low as 110 SBP and 70 DBP; mmHg). Lowering BP by 10mmHg SBP or 5mmHg DBP reduced CVD events by around 25%, heart failure (by about 25%) and stroke (by about 33%). Authors concluded that BP lowering drugs should be offered to anyone at high risk (whatever the reason for high risk, e.g. age, cardiovascular disease event) not just to
Reference
Population
BP measureme nt method
Followup
Study design
Outcomes
BP values at baseline (groups / thresholds); mmHg
Best BP threshold (authors conclusions) people with high BP, because a given BP reduction lowers the risk of coronary heart disease and stroke by a constant proportion irrespective of pretreatment BP. NS difference between WCH and NT for incidence of CV events; worse CV events in MH and sustained HT
Fagard et al., 206 2007 SR/MA of 7 studies
11,502
General population, primary care and secondary care (HT and NT)
Clinic and ABPM (to give diagnoses)
Mean 8 years
Risk of developing events in people diagnosed as NT, WCH, MH or sustained HT
CV events
NT: normal BP clinic and ABPM; mean BP 121.8/75.6 and 119.7/72.6 respectively WCH: clinic HT, normal ABPM; mean BP 148.2/86.2 and 125.6/74.9 respectively MH: normal clinic, ABPM HT; mean BP 129.9/78.6 and 141.1/83.2 respectively Sustained HT: clinic HT and ABPM HT; mean BP 157.7/88.5 and 152.4/85.7 HT diagnosis - cut off BP Clinic: 140/90 mmHg ABPM: 135/85 mmHg (except 1 study 135/83mmHg)
Prognostic studies Table 28: Study details and results for prognostic studies assessing the risk of developing clinical outcomes at different BP thresholds
Reference Arima et al., 49 2006 Sub-analysis of RCT (PROGRESS) Arima et al., 50 2009 Cohort (HISAYAMA) Assmann et al., 57 2005 Cohort (PROCAM) Barengo et al., 2009 and 60,61 2009 Cohort 41,895 (study 1) 47,610 (study 2) General population (HT and NT) Median 20 years 5389 General population (HT and NT) 10 years 1621 N 6105 Population HT and NT (Cerebrova scular disease) Follow-up Mean 3.9 years Study design Risk of developing events in people with different baseline BP values Outcomes Stroke, CV events BP values at baseline (groups / thresholds); mmHg SBP values <120 (median 114) 120-139 (median 130) 140-159 (median 149) 160 (median 169) Optimal: <120 /<80 Normal: 120-129 /80-84 High normal: 130-139 /85-89 Grade 1 HT: 140-159 /90-99 Grade 2 HT: 160-179 /100-109 Grade 3 HT: 180 /110 NT: 140 /90 New HT: SBP >159 and/or DBP>94 Adequately treated HT: <160 /95 Inadequately treated HT: 160/95 NT:<160/95 and no Tx HT (160 SBP or 95 DBP or Tx in last 7 days); treated and controlled (<160/95mmHg) HT: Tx and not controlled HT and aware (HT diagnosis or current Tx) but untreated Best BP threshold (authors conclusions) The benefits of treatment were comparable for patients who were or were not HT at baseline, for baseline BP levels extending down to 115/75mmHg. Initiating and monitoring treatment, including blood pressure targets
Clinic BP measurements
32 years
Risk of developing events in people with different baseline BP values (grouped)
Stroke
Age-adjusted incidence of total stroke rose progressively with higher BP in both genders
Risk of developing events in people with different baseline BP values (grouped) Risk of developing events in people with different baseline BP values (grouped)
Major coronary event
In all HT men, including those receiving adequate antihypertensive Tx, the 10-year risk of CHD was at least doubled.
Study 1: Mortality (all cause and CV) Study 2: stroke
In men, all-cause and cardiovascular mortality were significantly higher in all hypertensive groups compared with the normotensive group. In women, the mortality in those whose hypertension was controlled was not significantly
Reference
Population
Follow-up
Study design
Outcomes (fatal or non-fatal)
BP values at baseline (groups / thresholds); mmHg HT but unaware
Best BP threshold (authors conclusions) different from the normotensive group, suggesting that these women benefitted from achieving normal BP, although the uncontrolled, untreated and unaware groups had higher mortality. The risk of stroke was significantly higher in men and women in all hypertensive groups compared with the normotensive group. It may be higher in treated than untreated patients if they have had hypertension longer and it is more severe (also unaware were significantly younger so had lower risk).
Carlsson et al., 119 2009 Cohort study
2280
26 years
Mortality; CV mortality
NT/optimal: <130 / <85 Pre-HT: 130-139 and/or 85- 89 DBP High: 140 - 159 and/or 90-94 DBP Very high: 160 and/or DBP 95 NT/high-NT:<140 /<90 Mild-moderate HT: 140-179 /90109 Severe HT: 180 /110
Risk of Events increased with increasing BP; Very high blood pressure (160/95mmHg) is an independent risk factor for allcause and CV mortality in men and women. Patients treated for HT whose BP is not controlled have a higher risk of mortality than those whose BP is controlled. (Note: Tx target <160/<95mmHg; treatment not
Gudmundsson 243 et al., 2005 Cohort study
3246
Up to 20 years (mean 13.6 for men and 14.4 for women)
Mortality; CV mortality
Reference
Population
Follow-up
Study design
Outcomes
Best BP threshold (authors conclusions) as aggressive as it would be today; number controlled to <140/90mmHg was less than half those labelled controlled in this study.) Risk of stroke higher among HT vs. NT patients, and treated vs. non-treated HT, even when BP controlled to <140/90mmHg Untreated HT might have had a shorter duration of HT (and therefore lower risk of stroke) or have WCH (also lower risk).
Ishikawa et al., 291 2008 Cohort (JMS)
11,103
Mean 10.7 years
Stroke
NT: <140/90, no treatment HT: treated (receiving Tx, irrespective of current BP) C: Controlled (<140/90) U: Uncontrolled (140 and/or DBP 90) HT: untreated (140 /90 without Tx) M: Mild (SBP 140-159 or DBP 9099) MS: Moderate-severe (SBP 160 and/or DBP 100) SBP values NT: <140 Mild HT: 140-159 moderate-severe HT: >160 Optimal: <120 /<80 Normal: 120-129 /80-84 High normal: 130-139 /85-89 Stage 1 HT: 140-159 /90-99 Stage 2/3 HT: 160 /100 Very few people in stage 3 so combined into stage 2 values
Kagiyama et 313 al., 2008 Cohort
639
General population (HT and NT) but elderly (80 years) General population (HT and NT)
4 years
Mortality and CV mortality
No association between total mortality and SBP in the very elderly overall (however increased risk with increase BP), but there was an association in those with CVD or on Tx. Normal and high normal BP were a risk factor for the incidence of stroke and MI in men compared with optimal BP, as well as hypertension stage 1 or more. In women, the risk was seen at hypertension stages but not at normal/high normal BP (although numbers of events
Kokubo et al., 331 2008 Cohort (SUITA)
5494
Mean 11.7
CV events (MI or Stroke)
Reference Kono et al., 332 2005 Case-control Kshirsagar et 340 al., 2006 Cohort (ARIC)
N 708
Population HT (with vs. without CV event)
Follow-up n/a as casecontrol study
Study design Risk of developing events in people with different baseline BP values (grouped) Risk of developing events in people with different baseline BP values (grouped)
Outcomes CV events
BP values at baseline (groups / thresholds); mmHg SBP values NT: <140 Mild HT: 140-159 moderate-severe HT: >160 Optimal: <120 /<80 Normal: 120-129 /80-84 High normal: 130-139 /85-89
Best BP threshold (authors conclusions) were lower in women). Positive relationship between BP status and risk of cardiovascular events
8960
Mean 11.6 years
CVD
Normal BP and high normal BP were associated with a greater risk of incident cardiovascular disease compared with optimal BP. The risk was also higher for black people of African and Caribbean descent, older people (55-64 compared with 45-54), those with diabetes, high BMI, raised LDL cholesterol or renal insufficiency. In a relatively old cohort (mean age 60 years), risk of cardiovascular disease increased in higher BP groups
Obara et al., 454 2007 Post-hoc analysis (cohort)
1798 General population (HT and NT)
10,300 personyears
Onset of or death due to circulatory disease (stroke, angina, MI, cardiac death) Mortality; CV mortality
Optimal: <120 /<80 Normal: 120-129 /80-84 High normal: 130-139 /85-89 Grade 1 HT: 140-159 /90-99 Grade 2 HT: 160-179 /100-109 Grade 3 HT: 180 /110
Okayama et 466 al., 2006 Cohort (NIPPON DATA 80)
4244
19 years
SBP values Group 1: <120 Group 2: 120-139 Group 3: 140-159 Group 4: 160-179 Group 5: >179
Increased BP associated with cardiovascular disease mortality at all ages
Reference
Population
Follow-up
Study design
Outcomes
BP values at baseline (groups / thresholds); mmHg DBP values Group 1: <80 Group 2: 80-84 Group 3: 85-89 Group 4: 90-99 Group 5: >99
Best BP threshold (authors conclusions) Initiating and monitoring treatment, including blood pressure targets
Sairenchi et 521 al., 2005 Cohort
97,153
Mean 8.7 years (men), 8.9 years (women)
Risk of developing events in people with different baseline BP values (grouped) Risk of developing events in people classed into baseline BP quartiles
Mortality
Optimal: <120 /<80 Normal: 120-129 /80-84 High normal: 130-139 /85-89 Stage 1 HT: 140-159 /90-99 Stage 2/3 HT: 160 /100 SBP values (quartiles) 130 mmHg 130-142 mmHg 142-154 mmHg >154 mmHg
Impact of SBP and DBP on cardiovascular disease around 2 times larger among middle-aged than elderly subjects (men and women); generally an increase in risk with increase BP values No relationship found between SBP reduction and risk of MI, congestive heart failure and cardiovascular death. Avoid excessive SBP reduction (below 130mmHg) in older sicker high-risk patients For the primary outcome, there is a J-shaped pattern (nadir 130mmHg) in the relationship between on-treatment SBP (deciles) and adjusted risk of events; this was also true for cardiovascular mortality (nadir 130mmHg) and MI (126mmHg) but not for stroke.
Sleight et al., 546 2009 Post-hoc analysis of RCT (ONTARGET)
25,558
People with atheroscler otic disease or diabetes with end organ damage (High risk)
Mean 56 months
CV events (CV death, MI, Stroke, HF)
Reference Haider et al., 247 2003 Cohort (Framingham heart study subset)
N 2040
Population General population
Follow-up Mean 17.4 years
Study design Risk of developing events in people classed into baseline BP groups
Outcomes Congestive HF
BP values at baseline (groups / thresholds); mmHg SBP values 87-125 mmHg 126-141 mmHg 161 mmHg DBP values 49-74 mmHg 75-82 mmHg 83 mmHg
Best BP threshold (authors conclusions) Both SBP and DBP were associated with CHF, but SBP conferred greater risk than DBP. Increased risk of events with increased BP value. Initiating and monitoring treatment, including blood pressure targets
Benetos et al., 68 2003 Case-control
34,776
NT, HT and HT (Tx)
8-12 years
Risk of developing events in people iwth higher and lower BP values (and in Tx and unTx HT).
CVD, CHD and associated mortality
Treated (mean BP ~151/93 mmHg) Untreated (mean BP ~136/83 mmHg) High BP (140/90 mmHg) Lower BP(<140/90)
Treated HTs had higher SBP (+ 15 mmHg) and higher DBP (+ 9 mmHg), and a higher prevalence of associated risk factors and diseases. Treated HTs vs. untreated HTs presented a twofold increase in the RR for CV mortality and CHD mortality. Adjustment for unmodifiable risk factors only slightly decreased the excess CV risk observed in treated people. After additional adjustment for modifiable associated risk factors, the increased mortality in treated people persisted. Only after additional adjustment for SBP were CV mortality and CHD mortality similar in the two groups of people. Therefore, the increased CV mortality in treated HT vs.
Reference
Population
Follow-up
Study design
Outcomes
Best BP threshold (authors conclusions) untreated HT is mainly due to high SBP levels under treatment.
Weitzman et 629 al., 2006 Cohort
9611
23 years
Risk of developing events in people classed into baseline BP groups
Mortality (stroke, CHD and all-cause)
SBP values 80-119 mmHg 120-129 mmHg 130-136 mmHg 137-149 mmHg 150-260 mmHg DBP values 40-77 mmHg 78-80 mmHg 81-85 mmHg 86-90 mmHg 91-150 mmHg
Borghi et al., 89 2003 Cohort (Brisighella Heart Study)
2939
23 years
Risk of developing events in people classed into baseline BP groups
Mortality, CHD, MI, CeVD
SBP values <120 mmHg 120-139 mmHg 140-159 mmHg >159 mmHg DBP values <70 mmHg 70-79 mmHg 80-89 mmHg >89 mmHg
There is a consistent, strong, graded association between SBP (but not DBP) and cardiovascular events Increase in combined SHD and cerebrovascular disease risk was already evident with highnormal SBP
Fang et al., 213 2006
26,587
Mean 9.5 years
Risk of developing events in people classed into
Stroke
ISH: 140 / <90 mmHg SDH: 140 / 90mmHg IDH: <140 / 90 mmHg (with or
Highest risk of stroke in people with ISH and SDH vs IDH and MHT.
Reference Cohort
Population
Follow-up
Study design baseline BP groups
Outcomes
BP values at baseline (groups / thresholds); mmHg without a-HT Tx) MHT: <140 / <90 (and controlled BP by a-HT Tx) NT: <140 / <90 (without history of HT)
Best BP threshold (authors conclusions) Initiating and monitoring treatment, including blood pressure targets
People with SDH are at the highest risk of stroke and should be treated more aggressively.
Home BP measurements no studies (one included in Fagard meta-analysis) Ambulatory BP measurements Fagard et al., 208 2004 Cohort subanalysis of RCT (Syst-Eur) Inoue et al., 285 2007 Cohort; subanalysis of RCT (OHASAMA) Gustavsen et 244 al., 2003 Cohort 566 General population (NT, HT and WCH) Mean 10.2 years 1,271 HT Mean 11.2 years 295 HT (SBP) Median 7.5 years Risk of developing events in people classed as normal, abnormal or high BP Risk of developing events in people classed as HT (SBPDBP; ISH, IDH) vs. NT CV events Normal ABP: <140mmHg Abnormal ABP: 140-159mmHg High ABP: 160mmHg Baseline ABP predicts cardiovascular events. Increased events with increase in BP
Stroke
NT: <135 / <80 mmHg SDH: 135 / 80 mmHg ISH: 135 / <80 mmHg IDH: <135 / 80 mmHg
ISH determined by ABPM was associated with a high risk of stroke, similar to that found for patients with combined systolicdiastolic HT.
Risk of developing events in people classed as NT, WCH and HT
Death and CV events
NT: <140; mean = 129.1 mmHg HT: SBP >140; mean = 160.3 mmHg WCH: CBP>140, mean = 136.3; ABPM <135/90 mmHg
There is an increased cardiovascular risk in WCH compared to normotensive controls; the level of risk is the same as that seen with EHs (even though WCH had a lower average ABP than NT). Linear relationship between NT SBP (120-129mmHg and 130-
Self-reported / unknown BP measurement method Britton et al., 101 2009 18,876 HT Mean 20.7 years Risk of developing events in people HF SBP values
Reference Cohort
Population
Follow-up
Study design with different baseline BP values
Outcomes
BP values at baseline (groups / thresholds); mmHg NT (not on Tx) <120 mmHg 120-129 mmHg 130-139 mmHg HT (or on Tx) <130 mmHg 130-139 mmHg 140-149 mmHg 150-159 mmHg 160 mmHg
Best BP threshold (authors conclusions) 139mmHg) and risk of heart failure risk, as well as for HT SBP
Conen et al., 136 2007 Cohort (subanalysis of RCT) Deckers, 165 2006 Post-hoc analysis of RCT (EUROPA)
39,322
NT and HT women
Median 10.2 years
Risk of developing events in people with different baseline BP values
CV death, stroke or MI
Optimal: <120/ <75 Normal: 120-129/75-84 High normal: 130-139/85-89 HT: 140 /90
The CV risk of women with high normal BP is higher than those with normal BP; there was a strong and consistent increase in events down to the optimal BP category. Higher baseline BP associated with increased risk.
12,218
HT with CAD
Median 4.1 years
Risk of developing events in people with different baseline BP values
CV death, non-fatal MI
SBP values 130 mmHg >130-160 mmHg >160 mmHg
Table 29: Summary of numerical results for prognostic studies (for selected outcomes)
Study Arima et al., 49 2006 Outcome Stroke HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] SBP values (%, events/ person years) No HR values given 120 (median 114): 6.8% 120-139 (median 130) : 12.2% 140-159 (median 149): 12.5% 160 (median 169): 19.0% Men Optimal: <120 /<80: Reference Men Normal: 120-129 /80-84: 1.64 (0.76-3.56) p>0.05 Men High normal: 130-139 /85-89: 1.52 (0.70-3.31) p>0.05 Men Grade 1 HT: 140-159 /90-99: 3.31 (1.73-6.32)p<0.05 Men Grade 2 HT: 160-179 /100-109: 4.22 (2.16-8.25)p<0.05 Men Grade 3 HT: 180 /110: 5.75 (2.93-11.30)p<0.05 Women Optimal: <120 /<80: Reference Women Normal: 120-129 /80-84: 1.53 (0.60-3.89)p>0.05 Women High normal: 130-139 /85-89: 2.19 (0.93-5.16)p>0.05 Women Grade 1 HT: 140-159 /90-99: 3.92 (1.84-8.35)p<0.05 Women Grade 2 HT: 160-179 /100-109: 4.89 (2.24-10.67)p<0.05 Women Grade 3 HT: 180 /110: 7.51 (3.39-16.64)p<0.05 Assmann et al., 57 2005 Major coronary event NT: 140 /90 New HT: SBP >159 and/or DBP>94 Adequately treated HT: <160 /95 Inadequately treated HT: 160/95 No HR values given NT:<160/95 and no Tx : Reference HT (160 SBP or 95 DBP or Tx in last 7 days): No HR given HT treated and controlled (<160/95mmHg) 2.25 (1.70-2.99) HT: Tx and not controlled 2.41 (2.01-2.89) Initiating and monitoring treatment, including blood pressure targets
Arima et al., 50 2009
Stroke
Barengo et al., 60,61 2009 and 2009
CV mortality (MEN)
Study
Outcome
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] HT and aware (HT diagnosis or current Tx) but untreated 1.92 (1.65-2.23) HT but unaware 1.49 (1.33-1.68) Initiating and monitoring treatment, including blood pressure targets
Benetos et al., 68 2003
CVD, CHD and associated mortality
Treated (mean BP ~151/93 mmHg) Untreated (mean BP ~136/83 mmHg) High BP (140/90 mmHg) Lower BP(<140/90) No HRs given SBP values <120 mmHg Reference 120-139 mmHg 1.48 (1.04-2.10), p=0.0313 140-159 mmHg 1.92 (1.32-2.80), p=0.0006 >159 mmHg 2.38 (1.61-3.50), p<0.0001 Men NT/optimal: <130 / <85 Reference Men Pre-HT: 130-139 and/or 85- 89 DBP 1.07 (0.58-1.97) Men High: 140 - 159 and/or 90-94 DBP 1.17 (0.66-2.09) Men Very high: 160 and/or DBP 95 3.12 (1.84-5.26) Women NT/optimal: <130 / <85 Reference Women Pre-HT: 130-139 and/or 85- 89 DBP 1.89 (0.76-4.68) Women High: 140 - 159 and/or 90-94 DBP 2.34 (1.01-5.45) Women Very high: 160 and/or DBP 95 3.84 (1.62-9.12)
Borghi et al., 89 2003
Mortality
Carlsson et al., 119 2009
CV mortality
Fang et al., 2006
213
Stroke
NT: <140 / <90 (without history of HT) Reference ISH: 140 / <90 mmHg 2.35 (1.91-2.90) SDH: 140 / 90mmHg 2.96 (2.49-3.52) IDH: <140 / 90 mmHg (with or without a-HT Tx) 2.16 (1.69-2.76) MHT: <140 / <90 (and controlled BP by a-HT Tx) 1.33 (0.96-1.84) Men NT/high-NT:<140 /<90 Reference Men Mild-moderate HT: 140-179 /90-109 RR: 1.30 (0.79-2.14)
Gudmundsson et 243 al., 2005
CV mortality
Study
Outcome
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] Men Severe HT: 180 /110 RR: 1.23 (0.72-2.11) Women NT/high-NT:<140 /<90 Reference Women Mild-moderate HT: 140-179 /90-109 RR: 1.56 (0.85-2.86) Women Severe HT: 180 /110 RR: 2.57 (1.36-4.87) Only RRs given for above categories. However, per 1SD rise in SBP (22.4mmHg for men and 22.5 mmHg for women), HRs for Cv mortality are: 1.00 (0.87-1.15) for men and 1.34 (1.16-1.55),p<0.001 for women Initiating and monitoring treatment, including blood pressure targets
Haider et al., 247 2003
Congestive HF
SBP values 87-125 mmHg Reference 126-141 mmHg 1.48 (0.99-2.21), p=0.06 161 mmHg 3.07 (2.10-4.49), p<0.001 Men NT: <140/90, no treatment Reference Men HT: treated (receiving Tx, irrespective of current BP) RR:3.00 (2.00-4.51) Men C: Controlled (<140/90) RR 2.96 (1.66-5.26) Men U: Uncontrolled (140 and/or DBP 90) RR 3.05 (1.92-4.85) Men HT: untreated (140 /90 without Tx) RR 2.56 (1.83-3.57) Men M: Mild (SBP 140-159 or DBP 90-99) RR 2.34 (1.62-3.37) Men MS: Moderate-severe (SBP 160 and/or DBP 100) RR 3.17 (2.02-4.97) Women NT: <140/90, no treatment Reference Women HT: treated (receiving Tx, irrespective of current BP) RR 3.34 (2.29-4.87) Women C: Controlled (<140/90) RR 3.69 (2.20-6.17) Women U: Uncontrolled (140 and/or DBP 90) RR 3.16 (2.06-4.85) Women HT: untreated (140 /90 without Tx) RR 1.93 (1.35-2.76) Women M: Mild (SBP 140-159 or DBP 90-99) RR 1.95 (1.32-2.87)Women MS: Moderate-severe (SBP 160 and/or DBP 100) RR 1.87 (1.08-3.24) Only RRs given for above categories (but unclear). No HRs given
Ishikawa et al., 291 2008
Stroke
Kagiyama et al.,
CV mortality
SBP values
Study 313 2008
Outcome
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] NT: <140: Reference Mild HT: 140-159: RR:1.71 (0.56-5.24) moderate-severe HT: >160: RR: 2.15 (0.51-8.97) Only RRs given for above categories. No HRs given Initiating and monitoring treatment, including blood pressure targets
Kokubo et al., 331 2008
CV events (MI or Stroke)
Men Optimal: <120 /<80 Reference Men Normal: 120-129 /80-84 2.04 (1.19-3.48) Men High normal: 130-139 /85-89 2.46 (1.46-4.14) Men Stage 1 HT: 140-159 /90-99 2.62 (1.59-4.32) Men Stage 2/3 HT: 160 /100 3.95 (2.37-6.58) Women Optimal: <120 /<80 Reference Women Normal: 120-129 /80-84 1.12 (0.59-2.13) Women High normal: 130-139 /85-89 1.54 (0.85-2.78) Women Stage 1 HT: 140-159 /90-99 1.35 (0.75-2.43) Women Stage 2/3 HT: 160 /100 2.86 (1.60-5.12) Overall Optimal: <120 /<80 Reference Overall Normal: 120-129 /80-84 1.62 (1.08-2.43) Overall High normal: 130-139 /85-89 2.08 (1.42-3.05) Overall Stage 1 HT: 140-159 /90-99 2.06 (1.42-2.98) Overall Stage 2/3 HT: 160 /100 3.53 (2.43-5.13)
Kono et al., 332 2005
CV events
SBP values NT: <140 reference Mild HT: 140-159 Adjusted OR: 1.69 (1.10-2.60) moderate-severe HT: >160 Adjusted OR: 2.20 (1.08-4.45) Only adjusted ORs given. No HRs given Optimal: <120 /<80 Reference Normal: 120-129 /80-84 1.69 (1.37-2.09)
Kshirsagar et al., 340 2006
CVD
Study Obara et al., 454 2007
Outcome Onset of or death due to circulatory disease (stroke, angina, MI, cardiac death) CV mortality
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] High normal: 130-139 /85-89 2.33 (1.85-2.92) Optimal: <120 /<80 Normal: 120-129 /80-84 Reference High normal:130-139 /85-89 RR:1.19 (0.89-1.20), p=0.3 Grade 1-3 HT: 140->180 RR: 1.46 (1.00-1.17), p=0.011 Only adjusted RRs given. No HRs given SBP values Group 1: <120 Reference Group 2: 120-139 Age adjusted RR: 2.36 (1.17-4.77) Group 3: 140-159 Age adjusted RR: 3.00 (1.51-5.94) Group 4: 160-179 Age adjusted RR: 3.46 (1.75-6.84) Group 5: >179 Age adjusted RR: 5.13 (2.59-10.16) No HRs given for categories above, but multivariate adjusted HRs for 1SD increase in SBP: 1.31 (1.17-1.47) Men Optimal: <120 /<80 Reference Men Normal: 120-129 /80-84 RR: 1.48 (0.50-4.44) Men High normal: 130-139 /85-89 RR:2.89 (1.07-7.86) Men Stage 1 HT: 140-159 /90-99 RR:3.06 (1.15-8.16) Men Stage 2/3 HT: 160 /100 RR:5.99 (2.13-16.8) Women Optimal: <120 /<80 Reference Women Normal: 120-129 /80-84 RR:0.86 (0.34-2.20) Women High normal: 130-139 /85-89 RR:1.19 (0.50-2.84) Women Stage 1 HT: 140-159 /90-99 RR:2.02 (0.93-4.38) Women Stage 2/3 HT: 160 /100 RR:4.09 (1.70-9.85) Only RRs for men and women aged 40-59 given above. No HRs given Initiating and monitoring treatment, including blood pressure targets
Okayama et al., 466 2006
Sairenchi et al., 521 2005
Mortality
Sleight et al., 546 2009
CV events (CV death, MI, HF,
SBP values (quartiles) CV death
Study
Outcome Stroke)
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] 130 mmHg Reference 130-142 mmHg 0.98 (0.86-1.12) 142-154 mmHg 0.93 (0.81-1.06) >154 mmHg 0.98 (0.86-1.11) MI 130 mmHg Reference 130-142 mmHg 0.87 (0.74-1.01) 142-154 mmHg 0.88 (0.75-1.02) >154 mmHg1.03 (0.88-1.20) CHF 130 mmHg Reference 130-142 mmHg 0.85 (0.71-1.01) 142-154 mmHg 0.87 (0.74-1.04) >154 mmHg0.84 (0.71-0.99) Stroke 130 mmHg Reference 130-142 mmHg 1.11 (0.92-1.33) 142-154 mmHg 1.32 (1.11-1.58) >154 mmHg1.51 (1.28-1.79) Initiating and monitoring treatment, including blood pressure targets
Weitzman et al., 629 2006
Mortality (stroke, CHD and all-cause)
SBP values 80-119 mmHg 120-129 mmHg 130-136 mmHg 137-149 mmHg 150-260 mmHg No HRs given, nor any other RRs or ORs relevant to the categories above.
Study Fagard et al., 208 2004
Outcome CV events
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] Normal ABP: <140mmHg Reference Abnormal ABP: 140-159mmHg RR: 1.27 (0.64-2.52) High ABP: 160mmHg RR: 2.13 (1.09-4.13) No HRs given, but unadjusted RRs above calculated from data in outcome table. NT: <140; mean = 129.1 mmHg Reference HT: SBP >140; mean = 160.3 mmHg HR p<0.001 WCH: CBP>140, mean = 136.3; ABPM <135/90 mmHg HR 6.6 (p<0.001) HR p values given as shown, but no CIs and no HR value for HT were provided. NT: <135 / <80 mmHg Reference SDH: 135 / 80 mmHg 2.39 (1.48-3.87), p=0.0004 ISH: 135 / <80 mmHg 2.24 (1.33-3.76), p=0.0024 IDH: <135 / 80 mmHg excluded from model as number of subjects (n=37) and events (number not stated) were too low SBP values NT (not on Tx) <120 mmHg Reference 120-129 mmHg 1.10 (0.89-1.37) 130-139 mmHg 1.35 (1.09-1.68) HT (or on Tx) <130 mmHg 1.91 (1.36-2.68) 130-139 mmHg 2.61 (2.04-3.34) 140-149 mmHg 2.04 (1.63-2.55) 150-159 mmHg 2.66 (1.99-3.55) 160 mmHg 3.42 (2.33-5.04) CV Optimal: <120/ <75 0.51 (0.40-0.64) Normal: 120-129/75-84 0.61 (0.48-0.76) High normal: 130-139/85-89 Reference HT: 140 /90 1.30 (1.08-1.57) Age adjusted HR used SBP values 130 mmHg Initiating and monitoring treatment, including blood pressure targets
Gustavsen et al., 244 2003
CV events
Inoue et al., 285 2007
Stroke
Britton et al., 101 2009
HF
Conen et al., 136 2007
Major event
Deckers, 2006
165
CV death
Study
Outcome
HR (95% CI) for BP measurement (SBP/DBP) [HRs given unless indicated. Available RRs or ORs have been given if no HRs available] >130-160 mmHg >160 mmHg HRs not provided for above comparisons but multivariate HR for a 1mmHg increase in systolic BP: 1.01 (1.00-1.01) Initiating and monitoring treatment, including blood pressure targets
Equiavlence studies Table 30: Study details and results for equivalence studies determining thresholds for diagnosis and treatment using different blood pressure measurement methods.
Reference Head et al., 269 2010 8575 cross-sectional study NT and HT Immediate ABPM equivalents for clinic BPs N Population Follow-up Study design BP values at baseline (groups / thresholds); mmHg CLINIC MEASUREMENT CATEGORIES: lower limits of grade 3 (severe) HT(180/110 mm Hg) grade 2 (moderate) HT (160/100mmHg) grade 1 (mild) HT (140/90 mm Hg); for target upper limits for HT with associated conditions (130/80 mm Hg) HT with substantial proteinuria (125/75 mm Hg Upper limit of optimal normal (120/80 mm Hg). ABPM predicted from doctor measured seated clinic BP (n=1490) 24h 151/95 138/86 126/78 119/70 116/66 113/70 Night 143/86 128/78 113/69 106/61 102/57 99/61 Day 155/98 142/90 129/81 123/73 120/69 117/70 Clinic and ABPM measurements
Authors conclusions: equivalent thresholds Clinic BP threshold ABPM predicted from staff measured seated clinic BP (n=5327) 24h Grade 3 (severe) HT Grade 2 (moderate) HT Grade 1 (mild) HT Target BP + 1 condition Target BP + proteinuria Normal BP >180/110 >160/100 >140/90 <130/80 <125/75 <120/80 163/101 148/93 133/84 125/76 121/71 117/76 Night 157/93 139/84 121/76 112/67 107/63 102/67 Day 168/105 152/96 136/87 128/78 124/74 120/78
9.1.2
Evidence statements - clinical

Details of all the included studies are summarised in Table 31, Table 32 and Table 33. Most studies showed a continuous relationship between BP and risk of developing clinical outcomes (ie. an increased risk of outcome with increasing BP value) This was true regardless of BP measurement method (office, ABPM, self-reported/ not specified) The MA of Law et al.,351 showed that BP treatment reduced CVD risk regardless of pre-treatment BP The Head 2010 study269 provided equivalent threshold values for ABPM and clinic BP measurements for the diagnosis and treatment of HT.
9.1.3

No relevant cost-effectiveness evidence was identified.
9.2 Treatment of people aged 80 years and greater

Review question: in adults with primary hypertension, which is the most clinically and cost effective first-line anti-hypertensive treatment (drug classes) in elderly people (aged 80 years)?
9.2.1
Clinical evidence
The literature was reviewed from December 2005 onwards (the cut-off date of the previous guideline) for systematic reviews, RCTs and subgroup analyses of RCTs which addressed first-line anthypertensive treatment in elderly people (aged 80 years) with primary hypertension. Comparisons could be anti-hypertensive treatment or placebo. RCTs were included if there was: 12 months follow-up and N200 (in accordance with the 2006 guideline criteria) and the population did not consist of people who were exclusively diabetic or had CKD. Two SR/MAs67,419 were found that fulfilled the inclusion criteria and addressed the question. The first SR/MA (Musini et al 2009)419 was a Cochrane review and included N=8 studies. The second SR/MA (Bejan-Angoulvant 2010)67 was an update of a previous SR/MA and included additional data from the newer HYVET and HYVET-PILOT studies. , also consisted of 8 studies in total, and was an update of the Cochrane SR/MA. The Bejan-Angoulvant SR/MA67 was chosen to be included in this review instead of the Cochrane SR/MA becauseit provided data for more outcome measures than the Cochrane review, which pooled some outcomes together. Data was cross-checked between the two SR/MAs. The Began-Angoulvant SR/MA67 compared the development of clinical outcomes in patients who were 80 years old who had been randomised to treatment with either anti-hypertensive drugs or placebo. Data in the MA came from either sub-group analyses of RCTs (data from only the 80 yearold people in the trial), or from RCTs in which only people 80 years were enrolled. The mean followup time was 3.5 years (range 0 11.6) and the total number of patients included was N=6701. The 8 included studies differed in terms of sample size, mean SBP at baseline, follow-up time and the class of anti-hypertensive medication that patients were randomised to in the active treatment arm (D, CCB or BB). However they were similar in terms of the mean age of the study population (83 to 84 years old). NOTE: The HYVET trial which was included in the MA, recruited people who were less ill than those included in the other studies. Participants in HYVET were generally healthier than those in the
Update 2011
136
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets general population: they had low overall rates of stroke and death from any cause and at basline they were generally free of multiple comorbid conditions (low prevalence of previous cardiovascular disease, coronary artery disease and diabetes mellitus; inclusion criteria also excluded people with heart failure, dementia or those requiring nursing care). The evidence profile below (Table 31) summarises the quality of the evidence and outcome data from the SR/MA included in this review,67 comparing treatment vs placebo in people aged 80 years.
Update 2011
137
Table 31: Evidence profile comparing anti-hypertensive treatment versus placebo in people aged 80 years (systematic review/meta-analysis; BejanAngoulvant, 2010)67 NOTE: there was not enough data given in the study to calculate the HRs for these outcomes, so the RRs reported in the paper have been used in the GRADE profile.
Quality assessment No of studies Other considerations Summary of findings No of patients Imprecision anti-HT treatment Placebo Relative (95% CI) Effect Absolute Quality
Design
Limitations
Inconsistency
Indirectness
Mortality (all cause) (follow-up 0-11.6 years) SR/MA based on 8 RCTs* no serious limitations no serious inconsistency1,2 no serious indirectness 1.06 (0.89, 1.25) not enough data given in study to calculate
serious3
none
data not given in study
MODERATE
Update 2011
Coronary events (follow-up 0-11.6 years) SR/MA based on 6 RCTs* no serious limitations no serious inconsistency no serious indirectness 0.83 (0.56, 1.22) not enough data given in study to calculate
138
1 1 1
very serious
none
LOW
Stroke (follow-up 0-11.6 years) SR/MA based on 7 RCTs* no serious limitations no serious inconsistency no serious indirectness no serious imprecision 0.65 (0.52, 0.83) not enough data given in study to calculate
none
HIGH
CV events (follow-up 0-11.6 years) SR/MA based on 6 RCTs* no serious limitations no serious inconsistency no serious indirectness no serious imprecision 0.73 (0.62, 0.86) not enough data given in study to calculate
none
HIGH
Heart failure (follow-up 0-11.6 years)
SR/MA based on 6 RCTs*
no serious limitations
no serious inconsistency
no serious indirectness
no serious imprecision
none
0.50 (0.33, 0.76)
not enough data given in study to calculate
HIGH
coronary death (follow-up 0-11.6 years) SR/MA based on 7 RCTs* no serious limitations no serious inconsistency no serious indirectness 0.99 (0.69, 1.41) not enough data given in study to calculate
very serious
none
LOW
Stroke death (follow-up 0-11.6 years) SR/MA based on 8 RCTs* no serious limitations no serious inconsistency no serious indirectness 0.80 (0.80, 1.11) not enough data given in study to calculate
serious3
none
MODERATE
CV death (follow-up 0-11.6 years) SR/MA based on 8 RCTs* no serious limitations no serious indirectness 0.98 (0.83, 1.15) not enough data given in study to calculate
serious
very serious
none
VERY LOW
139
1 2
*moderate quality SR/MA based on moderate and high quality RCTs significant heterogeneity NS heterogenity when HYVET trial removed 3 95% confidence interval includes both 1) no effect and 2) the MID (appreciable benefit or appreciable harm); or only just crosses the MID 4 95% confidence interval crosses both 1) no effect and 2) appreciable benefit or harm and non-appreciable benefit or harm
9.2.2
Economic evidence
One study (Szucs 2010580) was identified from the update search that examined the costeffectiveness of antihypertensive drug treatment in people over the age of 80 years. This is summarised in the economic evidence profile below (Table 32, Table 33). A full evidence table is also provided in Appendix G: Evidence tables health economic studies (2011 update). Table 32: Antihypertensive treatment versus no treatment in people aged over 80 years economic study characteristics
Study Szucs 2010 ) Switzerland HYVET study
580
Limitations Potentially serious limitations(b)
Other Comments Model based on HYVET RCT Time horizon: 2 years Health outcomes: life years gained Costs: antihypertensive drugs, acute management and follow-up of MI, stroke and heart failure.
639
a) Some uncertainty about applicability of Swiss unit costs. QALYs not used. Discounting not in line with NICE reference case. b) Based on single RCT analysis and so does not incorporate all available evidence for patients over 80 years. Some methodological issues about how health outcomes and costs are calculated and attributed in model.
Table 33: Antihypertensive treatment versus no treatment in people aged over 80 years economic summary of findings (mean per person)
Study Szucs 2010 ) Switzerland HYVET study
580
Incremental effects 0.0457 life years gained
Update 2011
ICER Treatment dominated no treated (lower costs and improved health outcomes)
Uncertainty One way sensitivity analyses of 20% variation in medication cost, cost of stroke, cost of HF, cost of MI, life expectancy. Medication cost and cost of stroke had the biggest impact. Results varied from treatment dominant to 1097 per life year gained.
a) Converted from 2007 Swiss Francs.
9.2.3
Evidence statements Clinical

Study data has come from one moderate quality systematic review/meta-analysis67 which included eight moderate and high quality RCTs. In people aged 80 years old, anti-hypertensive treatment was significantly better than placebo for: stroke CV events heart failure [high quality evidence] [high quality evidence] [high quality evidence]
There was NS difference between anti-hypertensive treatment and placebo in people aged 80 years old for: total mortality coronary events [moderate quality evidence] [low quality evidence]
140
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets coronary death stroke death CV death [low quality evidence] [moderate quality evidence] [very low quality evidence]
9.2.4
Evidence statements Health economic

One partially applicable study with potentially serious limitations found treating people over 80 years of age with hypertension was cost-effective compared to not treating them.
9.3 Link from evidence to recommendations

Two main sources of evidence informed the GDG discussion about blood pressure thresholds; i) observational data examining the relationship between blood pressure and clinical outcomes from normotensive and hypertensive people according to current threshold definitions, and ii) studies examining the impact of treatment of hypertension on clinical outcomes, taking account of the baseline and achieved blood pressure values in clinical trials. It was not possible to pool data from these studies because they included people across varying age ranges, at different levels of baseline cardiovascular risk and patients were either untreated or treated with a range of medications that could have influenced cardiovascular disease risk and clinical outcomes. Thus, studies were examined individually to determine the strength and consistency of evidence to support recommendations for pharmacological treatment thresholds and optimal blood pressure targets for people with treated hypertension. A number of conclusions can be drawn from this analysis; i) there was a positive and continuous relationship between baseline blood pressure levels and the subsequent risk of clinical outcomes; ii) this relationship was consistent for the risk of stroke, ischaemic heart disease, heart failure and cardiovascular mortality; iii) this increased risk was most strongly related to systolic pressure, reflecting the fact that systolic pressure rises with ageing and most studies are conducted in older rather than younger people; iv) there was a paucity of data and no recent studies of the relationship between blood pressure and clinical events in younger people, i.e. <40 years. The GDG noted that clinical trials invariably recruited older patients at high cardiovascular disease risk and that there were no trials that had been specifically designed to examine the appropriate blood pressure thresholds for initiating pharmacological treatment forhypertension. Nevertheless, the individual pharmacological treatment trials had usually randomised people into studies based on systolic blood pressure thresholds of 140 or 160mmHg and diastolic pressure thresholds of 90 or 100mmHg. The GDG also discussed whether recommending specific blood pressure treatment thresholds was justified. The GDG noted that the results of a meta-analysis and systematic review of 248,445 people in 108 randomised controlled trials (Law et al) had shown that blood pressure lowering reduced the risk of cardiovascular disease and stroke irrespective of the patients pretreatment blood pressure, even when pre-treatment pressures were as low as 110/70mmHg suggesting that blood pressure lowering treatment could be offered to any person at high risk of cardiovascular disease, not just those with hypertension. The GDG concluded that such a hypothesis was consistent with the continuous relationship between blood pressure and clinical outcomes. However, it remainsl a hypothesis that requires prospective testing to properly define the balance between efficacy and safety, especially in people with low baseline blood pressure, as well as the cost-effectiveness of such a strategy. With regard to treatment thresholds, the GDG agreed that the current grading of hypertension, i.e. Stage 1 Hypertension (CBPM 140/90mmHg) or Stage 2 hypertension (CBPM160-100) was useful to help stratify people for treatment and should be retained. Furthermore the GDG could see no point in any further grading of hypertension beyond Stage 2 as it would have no impact of treatment stratification or clinical decision making. In light of the fact that this guideline update recommends
Update 2011
141
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets using the ABPM daytime average BP to confirm the diagnosis of hypertension for initiating treatment, it was necessary to define the ABPM daytime average pressures that are equivalent to the thresholds for stages 1 and 2 hypertension, previously defined according to CBPM readings alone. A large study of 8,575 (Head et al., 2010) 269 examined the equivalent Clinic blood pressure and ABPM day time average pressure for normotensive and hypertensive people. Of interest, the difference between Clinic and ABPM was greatest when measured by doctors in the clinic rather than other clinical staff. Based on the clinic staff data, a mean daytime average ABPM of 136/76mmHg was equivalent to Stage 1 hypertension threshold defined according to a CBPM threshold of 140/90mmHg. The 136/76mmHg value was rounded to derive the threshold for defining stage 1 hypertension, i.e. 135/85mmHg according to the ABPM day time average. This ABPM diagnostic threshold is similar to that used as the reference standard in the systematic review of the specificity and sensitivity of the different blood pressure measurement methods for the diagnosis of hypertension. The GDG concluded that an ABPM day time average of 135/85mmHg should be used to define the threshold for Stage 1 hypertension. In the study of Head et al,269 the current CBPM threshold for the diagnosis of Stage 2 hypertension, i.e. 160/100mmHg, was equivalent to an ABPM daytime average of 152/96mmHg, which the GDG rounded to 150/95mmHg. Thus, the GDG concluded that a daytime ABPM average BP 150/95mmHg should be used to define the threshold for stage 2 hypertension. In reviewing treatment thresholds, the GDG first reflected on the existing recommendation (2004) that pharmacological treatment should be offered for stage 2 hypertension, i.e. when the clinic blood pressure is 160-100mmHg (equivalent to an ABPM day time average of 150/95mmHg). This recommendation was based on the evidence review in 2004 which suggested that this level of blood pressure alone was sufficient to convey sufficient risk to benefit from pharmacological therapy for hypertension.The GDG reviewed this recommendation alongside the current evidence review which reinforced the message of the powerful effect of baseline blood pressure on clinical risk across a wide range of blood pressures and that pharmacologic treatment of blood pressure at or above the stage 2 hypertension threshold was associated with a clinical benefits and a reduction in risk. The GDG concluded that adults should be offered pharmacological treatment of hypertension at stage 2 hypertension (ABPM daytime average blood pressure 150/95mmHg). The GDG then discussed whether pharmacologic treatment should be offered to all adults with Stage 1 hypertension, i.e. CBPM systolic pressure 140-159 and/or diastolic pressure 90-99mmHg, and ABPM daytime averages of 135/85mmHg but <150/95mmHg. The existing guidance from 2004 recognised the uncertainty about whether every adult with stage 1 hypertension should be offered treatment. The GDG noted that the current recommendation is to offer treatment to some but not all people with stage 1 hypertension (2004). The treatment being targeted at those with stage 1 hypertension and higher levels of cardiovascular disease risk as indicated by the presence of one or more of; target organ damage, established cardiovascular disease, the presence of concomitant disease that increases cardiovascular disease risk such as diabetes or CKD, or in those whose 10 year cardiovascular risk is estimated to be 20% or more (ref NICE CVD risk) 428. The GDG discussed the fact that most of the people with stage 1 hypertension who would not be offered treatment according to this guidance will be younger (i.e. <40 years) because of their lower 10 year risk risk and lesser likelihood that they will have developed target organ damage or have established cardiovascular disease. Furthermore, there maybe greater uncertainty about the diagnosis of hypertension when blood pressure is close to the threshold for stage 1 hypertension. The GDG concluded that pharmacological treatment should be offered to people with stage 1 hypertension who also have higher levels of cardiovascular disease risk as indicated by the presence of one or more of; target organ damage, established cardiovascular disease, the presence of concomitant disease that increases cardiovascular disease risk such as diabetes or CKD, or in those whose 10 year cardiovascular risk is estimated to be 20% or more (ref NICE CVD risk)428. Moreover, those with stage 1 hypertension without any of these additional higher cardiovascular factors
Update 2011
142
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets indicators, i.e. uncomplicated stage 1 hypertension, would not usally be offered pharmacological therapy for hypertension but; i) would be recomended to undertake lifestyle modifications (see section x), and ii) should also be re-evaluated annually and pharmacological treatment offered if they develop more severe hypertension, i.e. stage 2 hypertension, or they develop target organ damage, diabetes, CKD, cardiovascular disease, or their estimated 10 year cardiovascular disease risk rises to 20% or more. In reality, this means that most people with stage 1 hypertension will be offered pharmacologic treatment because age is a major determinant of CVD risk and the majority of people with hypertension are older rather than younger. However, the GDG discussed the dilemma created by this recommendation about what to advise for younger people (i.e. <40 years) with uncomplicatedstage 1 hypertension. This dilema is created by the fact that younger people with stage 1 hypertension are less likely to have overt evidence of target organ damage or vascular disease and assessment of their CVD risk over a relatively short duration of 10 years is unlikely to adequately reflect their lifetime risk of CVD. The GDG further discussed that this dilemma is compouned by the fact that when compared with older populations; i) in younger people, the time course over which clinical outcomes develop as a consequence of stage 1 hypertension are likely to be very long and much longer then those encountered in conventional clinical outcome trials and epidemiological studies. Thus, there is very much less epidemiological data linking uncomplicated stage 1 hypertension in younger people with adverse clinical outcomes; ii) younger people have not been included in clinical outcome trials in sufficient numbers to evaluate the impact of the pharmacological treatment of stage 1 hypertension on clinical outcomes and probably never will be as such trials would need to be unfeasibly large of too long a duration to be practical; iii) 10 year CVD risk estimates are strongly age dependent and as such, in younger people will rarely provide an indication for treatment of uncomplicated stage 1 hypertension. The GDG concluded that uncomplicated stage 1 hypertension in younger people is unlikely to be benign, blood pressure will most likely rise over time, and that there is uncertainty surrounding whether delayed pharmacological treatment will necessarily reverse any accumulated target organ or cardiovascular damage. The GDG also discussed the need to develop more accurate estimates of the lifetime risk of younger people with uncomplicated stage 1 hypertension and the cost-effectiveness of treatment. In this regard, the GDG recognised the importance of thorough assessment of target organ damage to exclude its presence before deciding not to offer pharmacological treatment of hypertension for younger people with seemingly uncomplicated stage 1 hypertension the GDG thus recommended that evaluation of the potential benefit of treating uncomplicated stage 1 hypertension in younger people with regard to its impact on target organ structure and function should be a priority for future research. Meantime, the GDG recommended that for younger people (i.e. <40years) with uncomplicated stage 1 hypertension, specialist referral for exclusion of secondary causes of hypertension (see section xx) and detailed evaluation of target organ damage e.g. by echocardiography to exclude LVH and dysfunction, should be considered before concluding not to offer treatment. Moreover, when treatment is not offered, careful annual re-evaluation is necessary because blood pressure is likely to rise over time and target organ damage may develop.
Update 2011
9.4 Recommendations
22.Offer antihypertensive drug treatment to people aged under 80 years with stage 1 hypertension who have one or more of the following: target organ damage established cardiovascular disease renal disease diabetes a 10-year cardiovascular risk equivalent to 20% or greater. [new 2011]
143
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets 23.Offer antihypertensive drug treatment to people of any age with stage 2 hypertension. [new 2011] 24.For people aged under 40 years with stage 1 hypertension and no evidence of target organ damage, cardiovascular disease, renal disease or diabetes, consider seeking specialist evaluation of secondary causes of hypertension and a more detailed assessment of potential target organ damage. This is because 10-year cardiovascular risk assessments can underestimate the lifetime risk of cardiovascular events in these people. [new 2011]
9.5 Recommendations for research

3. In people aged under 40 with hypertension, what are the appropriate thresholds for intervention?
There is genuine uncertainty about how to assess the impact of blood pressure treatment in younger people (aged under 40) with stage 1 hypertension, and no overt target organ damage or CVD. In particular, whether those with untreated hypertension are more likely to develop target organ damage and, if so, whether such damage is reversible. Target organ damage and CVD as surrogate or intermediate disease markers are the only indicators that are likely to be feasible in younger people because traditional clinical outcomes are unlikely to occur in sufficient numbers over the time scale of a typical clinical trial. The data will be important to inform treatment decisions for younger people with stage 1 hypertension who do not have overt target organ damage.
9.6 Monitoring treatment efficacy

Review question: In adults with treated primary hypertension, what is the best method to measure blood pressure (home vs ambulatory vs office) for response to treatment?
Update 2011
9.6.1
Clinical evidence
The literature was searched for all years and studies published since the original guideline (2003 onwards) were included. Two SRs/MAs96,290 and 3 RCTs137,439,554 were found that fulfilled the inclusion criteria and assessed which was the best BP measurement method for monitoring treatment in order to reach target BPs. All studies were of moderate to good quality. The first MA96 compared the effects of home monitoring vs usual care on BP lowering and reaching BP targets. The second MA290 compared BP measurements at end of treatment using office or home measurements. The 4 RCTs all assessed the effects of home monitoring vs office or ABPM monitoring on BP lowering and reaching BP targets. NOTE: all RCTs were underpowered to detect a difference in BP. In order to detect a 5mm difference, a sample size of N500 is needed. The evidence profiles below ( Table 35, Table 36, Table 37, Table 38 and Table 39) summarise the quality of the evidence and outcome data from the studies included in this review.96,137,290,439,554.
144
Table 34: Evidence profile comparing self-monitoring vs. usual care (Bray 2010)96
Summary of findings No of patients Effect Other self usual Relative Inconsistency Indirectness Imprecision Absolute considerations monitoring care (95% CI) Change in clinic systolic blood pressure (mm Hg) (Better indicated by lower values) no serious 3.82 lower (5.61 to 2.03 3 4 5 5 none 0 0 serious serious 6 indirectness lower) Change in clinic diastolic blood pressure (mm Hg) (Better indicated by lower values) no serious no serious no serious 1.45 lower (1.95 to 0.94 8 8 0 none 0 9 inconsistency indirectness imprecision lower) Proportion of patients achieving clinic blood pressure target no serious 0/0 1.09 (1.02 to 3 4 11 serious none 0/0 (0%) serious Not estimable 11 6 indirectness (0%) 1.16) Change in daytime ABPM systolic blood pressure (mm Hg) (Better indicated by lower values) 2.04 lower (4.35 lower to no serious no serious no serious 13 13 0 none 0 14 0.27 higher) inconsistency indirectness imprecision Change in daytime ABPM diastolic blood pressure (mm Hg) (Better indicated by lower values) no serious no serious no serious 0.79 lower (2.35 lower to 13 13 none 0 0 15 inconsistency indirectness imprecision 0.77 higher) Quality assessment
No of studies 1
96
Design randomised 1 trials randomised 7 trials randomised 10 trials randomised 12 trials randomised 12 trials
Limitations very 2 serious very 2 serious very 2 serious very 2 serious very 2 serious
Quality
VERY LOW
96
LOW
96
VERY LOW
Update 2011
45
1 2
96
LOW
96
LOW
Meta-analysis of 20 RCTs Unclear randomisation process; unclear allocation concealment; unclear blinding; unclear ITT analysis; unclear drop-out rates 3 I2 >50% 4 95% CI crosses MID 5 Not stated. Total number of patients was 5,898 6 p = 0.000 7 Meta-analysis of 23 RCTs 8 Not stated. Total number of patients was 6,038 9 p = 0.015 10 Meta-analysis of 12 RCTs 11 Not stated. Total number of patients was 2,260 12 Meta-analysis of 3 RCTs 13 Not stated. Total number of patients was 572 14 p = 0.89 15 p = 0.96
Table 35: Evidence profile comparing reduction in blood pressure using clinic and home measurements (Ishikawa 2008)290
Quality assessment No of studies Design Limitations Inconsistency Indirectness Imprecision Other considerations
Summary of findings No of patients Home blood pressure measurement Clinic blood pressure measurement Relative (95% CI) Effect Absolute Quality
290
randomised trials1 randomised trials1
very serious2 very serious2
serious
1 290
1 2 3
serious3
Change in systolic blood pressure (mm Hg) (Better indicated by lower values) no serious serious4 none 05 indirectness Change in diastolic blood pressure (mm Hg) (Better indicated by lower values) no serious serious4 none 05 indirectness
05
MD 0 higher (0 to 0 VERY LOW higher)6 MD 0 higher (0 to 0 VERY LOW higher)7
05
Meta-analysis of 22 RCTs. Data sets in which the methods of clinic BP measurements were not clearly described were excluded Unclear randomisation process; unclear allocation concealment; unclear blinding; unclear ITT analysis; unclear drop-out rates No details 4 Difference in change not stated 5 Not stated. Total number of patients was 6,322 6 Reductions in clinic and home SBP were: -14.70.04 and -11.80.04 respectively; p<0.001 7 Reductions in clinic and home DBP were: -10.70.03 and -8.10.05 respectively; p<0.001
Update 2011
46
Table 36: Evidence profile comparing reduction in blood pressure using home and ambulatory measurements (Ishikawa 2008)290
Quality assessment No of studies Design Limitations Inconsistency Summary of findings No of patients Quality Effect Relative Other Home blood pressure Ambulatory blood pressure Indirectness Imprecision (95% Absolute considerations measuerement measurememnt CI) Change in daytime systolic blood pressure (mm Hg) (Better indicated by higher values) MD 1.6 higher (1.1 to 2.2 no serious no serious 03 none 03 higher)4 indirectness imprecision Change in daytime diastolic blood pressure (mm Hg) (Better indicated by higher values) no serious no serious MD 0.2 higher (0.4 lower none 03 03 indirectness imprecision to 0.8 higher)5
1 290
no serious inconsistency no serious inconsistency
LOW
1 290
LOW
290
no serious inconsistency no serious inconsistency
1 290
1 2
Change in nighttime systolic blood pressure (mm Hg) (Better indicated by higher values) no serious no serious none 03 03 indirectness imprecision Change in nighttime diastolic blood pressure (mm Hg) (Better indicated by higher values) no serious no serious 03 none 03 indirectness imprecision
MD 3.8 higher (3.3 to 4.4 higher)4 MD 1.2 higher (0.6 to 1.8 higher)4
LOW
LOW
Meta-analysis of 5 RCTs. Unclear randomisation process; unclear allocation concealment; unclear blinding; unclear ITT analysis; unclear drop-out rates 3 Not stated. Total number of patients was 801 4 p<0.001 5 p=0.55
Table 37: Evidence profile comparing treatment targeted to home DBP vs.treatment targeted to ambulatory DBP Niiranen 2006439
Quality assessment No of studies Design Limitations Inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency Summary of findings Effect Absolute MD 2.6 higher (2.3 lower to 7.4 higher)3 MD 2.6 higher (0.1 lower to 5.2 higher)4 MD 0.6 higher (3.0 lower to 4.3 higher)5 MD 1.5 higher (1.0 lower to 3.9 higher)6 MD 1.1 higher (3.7 lower to 5.9 higher)7 MD 1.3 higher (5.0 lower to 2.3 higher)8 Quality No of patients Other Home blood pressure Ambulatory blood Relative Indirectness Imprecision considerations measurement pressure measurement (95% CI) Home systolic blood pressure (mm Hg) (follow-up 24 weeks; Better indicated by lower values) no serious serious2 none 52 46 indirectness Home diastolic blood pressure (mm Hg) (follow-up 24 weeks; Better indicated by lower values) no serious none 52 46 serious2 indirectness 24-h systolic blood pressure (mm Hg) (follow-up 24 weeks; Better indicated by lower values) no serious no serious none 52 46 indirectness imprecision 24-h diastolic blood pressure (mm Hg) (follow-up 24 weeks; Better indicated by lower values) no serious no serious none 52 46 indirectness imprecision Clinic systolic blood pressure (mm Hg) (follow-up 24 weeks; Better indicated by lower values) no serious none 52 46 serious2 indirectness Clinic diastolic blood pressure (mm Hg) (Better indicated by lower values) no serious indirectness no serious indirectness serious2 none 52 46 -
Update 2011
47
1 439
randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1
VERY LOW
1 439
VERY LOW
439
LOW
1 439
LOW
1 439
VERY LOW
1 439
VERY LOW
Number of patients who reached target BP (follow-up 24 weeks) 1

439
very serious9
none
30/52 (57.7%)
20/46 (43.5%)
RR 1.33 (0.89 143 more per 1000 (from 48 to 1.99) fewer to 430 more)
VERY LOW
Number of patients progressing to combination therapy (follow-up 24 weeks) 1

1 2 439
randomised very trials serious1
very serious9
none
34/52 (65.4%)
31/46 (67.4%)
RR 0.97 (0.73 20 fewer per 1000 (from 182 to 1.29) fewer to 195 more)
VERY LOW
Unclear allocation concealment; unclear blinding; no ITT analysis 95% CI crosses MID 3 p = 0.29 4 p = 0.06 5 p = 0.72 6 p = 0.23 7 p = 0.66 8 p = 0.46 9 95% CI crosses both MIDs
Table 38: Evidence profile comparing treatment managed with ambulatory measurements vs.treatment managed with clinic measurements (Conen 2009)137
Quality assessment No of patients Effect Quality
48
No of studies 1 137
Design
Limitations Inconsistency Indirectness no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency
randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1 randomised very trials serious1
1 137
1 137
Other Ambulatopry blood Clinic blood pressure Relative considerations pressure measurement measurement (95% CI) Change in 24-h systolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious none 70 66 serious2 indirectness Change in 24-h diastolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious no serious none 70 66 indirectness imprecision Change in clinic systolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious none 70 66 serious2 indirectness Imprecision
Absolute
Update 2011
mean 3.6 lower (7.0 to 0.3 lower)3 MD 0.9 lower (3.0 lower to 1.1 higher)4 MD 4.4 lower (10 lower to 1.1 higher)5
VERY LOW
LOW
VERY LOW
137
1 137
1 137
Change in clinic diastolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious no serious MD 0.4 lower (3.6 lower to 2.8 none 70 66 indirectness imprecision higher)6 Mean number of antihypertensive drugs used (follow-up 1 years; Better indicated by lower values) no serious mean 0.19 lower (0.53 lower very serious7 none 70 66 indirectness to 0.15 higher)8 Patients with controlled 24-h blood pressure (follow-up 1 years) RR 1.41 (1.01 174 more per 1000 (from 4 no serious none 42/70 (60%) 28/66 (42.4%) serious2 to 1.99)9 indirectness more to 420 more)
LOW
VERY LOW
VERY LOW
1
1 2
137
randomised very trials serious1
Patients with controlled office blood pressure (follow-up 1 years) no serious very serious7 none 29/70 (41.4%) 23/66 (34.8%) indirectness
RR 1.19 (0.77 to 1.83)10
66 more per 1000 (from 80 fewer to 289 more)
VERY LOW
No details on allocation concealment; open label; no ITT analysis 95% CI crosses MID 3 p = 0.03 4 p = 0.37 5 p = 0.12 6 p = 0.81 7 95% CI crosses both MIDs 8 p for difference = 0.49
9 10
p = 0.04 p = 0.4
Table 39: Evidence profile comparing treatment managed with home measurements vs.treatment managed with clinic measurements (Staessen 2004)554
Quality assessment No of studies 1 554 No of patients Effect
49
Design randomised serious1 trials randomised serious1 trials randomised serious1 trials randomised serious1 trials randomised serious1 trials 1 554 1 554 1 554 1
554
Update 2011
Quality Limitations Inconsistency Indirectness no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious inconsistency no serious Other Home blood pressure Clinic blood pressure Relative considerations measurement measurement (95% CI) Patients able to permenantly stop antihypertensive drug treatment (follow-up 1 years) no serious no serious RR 2.29 (1.45 none 52/203 (25.6%) 22/197 (11.2%) indirectness imprecision to 3.63)2 Clinic systolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious serious3 none 203 197 indirectness Clinic diastolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious serious3 none 203 197 indirectness Home systolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious serious3 none 203 197 indirectness Imprecision Home diastolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious no serious none 203 197 indirectness imprecision 24-h systolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious serious3 none 203 197 Absolute 144 more per 1000 (from 50 MODERATE more to 294 more) MD 6.8 higher (3.6 to 9.9 higher)4 MD 3.5 higher (1.9 to 5.1 higher)4 MD 4.9 higher (2.5 to 7.4 higher)4 MD 2.9 higher (1.5 to 4.3 higher)4 MD 4.9 higher (2.5 to 7.4 LOW
LOW
LOW
MODERATE LOW
1 554 randomised serious1
trials 1 554
inconsistency
no serious randomised serious1 inconsistency trials 1 Unclear allocation concealment 2 log-rank p<0.001 3 95% CI crosses MID 4 p <0.001
indirectness 24-h diastolic blood pressure (mm Hg) (follow-up 1 years; Better indicated by lower values) no serious no serious none 203 197 indirectness imprecision
higher)4
Update 2011
MD 2.9 higher (1.4 to 4.4 higher)4
MODERATE
50
9.6.2
Economic evidence
An economic evaluation should ideally compare all relevant alternatives. No studies were identified in the update search comparing all of clinic blood pressure monitoring (CBPM), ambulatory blood pressure monitoring (ABPM) and home blood pressure monitoring (HBPM) for assessing blood pressure (BP) control in treated patients. Two studies comparing CBPM and ABPM in treated patients were identified but were excluded as were judged to have serious methodological limitations374,512. One study (Staessen 2004554) was identified that examined the examined the cost effectiveness of HBPM compared with CBPM. This is summarised in the HBPM versus CBPM economic evidence profile below (Table 40, Table 41). A full evidence table is also provided in Appendix G: Evidence tables health economic studies (2011 update). One other study of this comparison was also identified but was excluded in line with the review protocol as the HBPM included a telemonitoring component476. The Staessen 2004 study554 was also included in the clinical review above. Note that this study is in a population diagnosed with CBPM and this may impact the applicability to a population diagnosed by another method. This is because if you are diagnosed by CBPM and then monitored by ABPM to some extent the result will be about the people who were incorrectly diagnosed in the first place not just differences in follow-up monitoring. No cost-effectiveness studies were included in Clinical Guideline 18 relating to this topic.
Update 2011
Table 40: HBPM versus CBPM (assessing response to treatment) economic study characteristics
Study Staessen 2004 Belgium
554
Limitations Potentially serious(b)
Other Comments CBPM diagnosed population who are treated or not treated. CPBM vs HBPM to assess BP control with treatment intensified if DBP >89mmHg, reduced if DBP <80mmHg. Within-RCT analysis. Costs: Antihypertensive drugs, physician visits, HBPM.
a) Some uncertainty about applicability of Belgian resource use and unit costs. Some uncertainty about applicability to a population not diagnosed with CBPM. QALYs not used (cost consequence analysis). b) Given that blood pressure was significantly different, other clinical events and costs of these may be relevant and time horizon may be insufficient. Within trial analysis and so does not incorporate all available evidence on differences between options and results of this study inconsistent with meta analysis included in clinical review; clinical study considered to have methodological limitations.No analysis of uncertainty.
Table 41: HBPM versus CBPM (assessing response to treatment) economic summary of findings (mean per person)
Study Staessen 2004 Belgium
554
Incremental effects BP increased; medication discontinuation increased; no significant difference in left ventricular mass or symptoms
ICER Lower costs with HBPM but worse BP control
Uncertainty NR
a) Converted from 2002 Belgium 2002 using purchasing power parities
468
151
9.6.3

One well-conducted meta-analysis96 found that: Self-monitoring was significantly better than usual care for: o reducing clinic SBP and DBP (SBP: 20 RCTs, N=5898; DBP: 23 RCTs, N=6038) [very low and low quality evidence] o proportion of patients achieving target clinic blood pressure (12 RCTs, N=2260) [very low quality evidence] o There was NS difference between self-monitoring and usual care for reduction in mean daytime SBP and DBP ABPM (3 RCTs, N=572). [low quality evidence] When self-monitoring was accompanied by an additional co-intervention, participants were more likely to meet target blood pressure compared to when there was none. One meta-analysis290 found that: with anti-hypertensive treatment (regardless of drug class used for treatment): o clinic SBP and DBP fell significantly more than home blood pressure [very low quality evidence] home blood pressure fell approximately 20% less than clinic blood pressure changes in clinic blood pressure were linearly related to those of home blood pressure the difference between clinic blood pressure and homeblood pressure was attributable to the difference in baseline blood pressure levels o home blood pressure fell significantly more than daytime ambulatory SBP and night-time ambulatory SBP and DBP [low quality evidence] daytime ambulatory SBP fell 15% less and night-time ambulatory SBP fell 30% less than home blood pressure o the reduction in daytime ambulatory DBP was NS different than the reduction in home blood pressure [low quality evidence] o changes in home SBP were intermediate between clinic and ambulatory SBPs (for 24h, daytime and night-time measurements) One RCT*439 found that there was NS difference between treatment targeted to home DBP vs. targeted to ABPM DBP for: Home SBP and DBP blood pressure measurements (end of trial) Clinic SBP and DBP blood pressure measurements (end of trial) number of patients who reached target blood pressure [very low quality evidence] [very low quality evidence] [very low quality evidence] 24h ABPM SBP and DBP blood pressure measurements (end of trial) [low quality evidence]
Update 2011
intensity of anti-hypertensive treatments (number of patients progressing to combination therapy) [very low quality evidence] One RCT137 found that: treatment managed with ABPM measurements was significantly better than treatment managed with CBPM for: o reductions in mean 24h ABPM SBP [very low quality evidence] o number of patients with controlled 24-hour blood pressure [very low quality evidence] there was NS difference between treatment managed with CBPM measurements versus measured with ABPM for:
152
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets o reductions in mean clinic SBP and DBP [low and very low quality evidence] o reductions in mean 24h ABPM DBP [low quality evidence] o number of patients with controlled clinic blood pressure measurements [very low quality evidence] o number of antihypertensive drugs used [very low quality evidence] One RCT*554 found that: treatment managed with home blood pressure was significantly better than treatment managed with clinic blood pressure measurements for: o number of patients who could permanently stop a-HT treatment [moderate quality evidence] treatment managed with clinic blood pressure was significantly better than treatment managed with home blood pressure measurements for : o reduction in clinic SBP and DBP blood pressure [low quality evidence] o reduction in home SBP and DBP blood pressure [low and moderate quality evidence] o reduction in 24h ABPM SBP and DBP ABPM blood pressure [low and moderate quality evidence] *NOTE: Both groups were given the same target BP for treatment, despite being measured by the two different methods, which would lead to a systematic under-treatment in one of the groups
9.6.4
Evidence statements health economic Update 2011

No cost-effectiveness analyses were identified incorporating all of CBPM, ABPM and HBPM in the assessment of response to treatment. One partially applicable study with potentially serious limitations found that in a population diagnosed with hypertension using CBPM, monitoring response to treatment and adjusting treatment using HBPM was cost saving compared to CBPM; blood pressure control was however worse.
9.6.5
Link from evidence to recommendations

All clinical outcome trials have used CBPM to monitor treatment efficacy. Some of these trials have embedded substudies using HBPM or ABPM to monitor treatment effects but for the primary outcome measures, the blood pressure control was invariably monitored using CBPM. A metaanalysis by Bray et al., 2010 96showed that patients self monitoring their own blood pressure was associated with lower achieved CBPM and a greater liklihood of achieving the clinic blood pressure target. Interestingly another analysis (Ishikawa aet al., 2008)290 also found that HBPM averages fell approximately 20% less than the corresponding CBPM but that the relationship between the two measures was linear. Two studies (Niiranen et al., 2006 and Conen et al., 2009)137,439 examined whether monitoring blood pressure control with CBPM versus ABPM or HBPM impacted on blood pressure control and the number of treatements used to achieve the blood pressure targets and found no differences between blood pressure monitoring methods. The GDG noted that there was inadequate data comparing the use of HBPM or ABPM to monitor blood pressure control and whether they offer any important advantages over CBPM. Routine monitoring with HBPM or ABPM would also require considerable investment in additional monitors beyond that required for diagnosis of hypertension. The GDG recognised that patients may wish to monitor their own blood pressure using HBPM and the possibility that engaging patients in their own blood pressure monitoring process using HBPM could lead to better blood pressure control (NICE Medicines Adherence Guideline, CG76)426. The GDG noted, however, that further data on self-monitoring and
153
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets self management of blood pressurewas required before this could be recommended as the preferred modality for monitoring blood pressure control in people with treated hypertension. The GDG recommended that for people receiving antihypertensive medications, clinic blood pressure readings should usually be used to monitor their response to treatment. The GDG discussed how to monitor blood pressure in people with significant discrepancies between their clinic blood pressure readings, recognising that CBPM may not provide an accurate representation of their blood pressure control. In people identified as having a white coat effect (people who are hypertensive according to their ABPM daytime average blood pressure but with a CBPM at diagnosis that exceeded their ABPM by 20 mmHg systolic, or 10mmHg diastolic) the GDG recommended that HBPM should be considered as an adjunct to CBPM to monitor the response to antihypertensive treatment and/or lifestyle modification.
9.6.6
Recommendations
25.Use clinic blood pressure measurements to monitor the response to antihypertensive treatment with lifestyle modification or drugs. [new 2011] 26.For people identified as having a white-coat effect, consider ABPM or HBPM as an adjunct to clinic blood pressure measurements to monitor the response to antihypertensive treatment with lifestyle modification or drugs. [new 2011]
9.6.7
Research recommendations Update 2011

4. In adults with primary hypertension, does the use of out-of-office monitoring (HBPM or ABPM) improve response to treatment? There is likely to be increasing use of home and ambulatory blood pressure monitoring for the diagnosis of hypertension as a consequence of this guideline update. There are, however, very little data regarding the utility of HBPM or ABPM as means of monitoring blood pressure control or as indicators of clinical outcome in treated hypertension, compared with clinic blood pressure monitoring. Studies should incorporate HBPM and/or ABPM to monitor blood pressure responses to treatment and their usefulness as indicators of clinical outcomes.
9.7 Blood pressure targets for treatment

Review question: in adults with primary hypertension, what is the optimum BP that should be reached for once treatment has been initiated/ targeted for treatment?
9.7.1
Clinical evidence
The literature was searched for studies published since the original guideline (2003 onwards). All study types were included, if the population did not consist of people who were exclusively diabetic or had CKD. Studies were excluded if they did not stratify results into more than 1 different BP value / target. Fifteen studies29,49,82,134,168,209,280,282,298,462,463,539,549,616,623,655 were found that fulfilled the inclusion criteria and assessed what the optimum target blood pressure should be for treating people with primary hypertension. One of the studies (29,298) was published as two separate papers reporting different assessment methods or outcomes, so this study has only been counted once, however results from both papers are reported and referenced here.
154
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets The studies addressing the question were categorised into three different types: 1. More vs less intense treatment studies - (eight studies; eight papers)29,82,280,282,298,463,549,616 those that assess people who were randomised to more intense (strict or intense) BP lowering vs. less intense (mild or standard) BP lowering 2. Within-treatment BP studies (eight studies)49,134,168,209,462,539,623,655 - those that assess withintreatment / achieved BP values and the associated risk of developing clinical outcomes. 3. Target BP studies(one study)462 - those that target people to different specific blood pressure values (for example, according to age groups) Details of all the included studies are summarised in Table 42 and Table 43 and Table 44. NOTE: Data from the more vs less intense treatment studies was not pooled into meta-analysis because the studies varied widely in the following factors: treatment targets, interventions used to reach the target (type of anti-hypertensive drug), follow-up times, BP measurement method and outcome definitions. Therefore GRADE was performed on each individual RCT to give a quality rating for each outcome measure used in the study (see
Update 2011
155
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets Table 45).
156
More versus less intense treatment studies Table 42: Study details and results for optimal blood pressure targets (trials comparing more vs. less intense blood pressure lowering treatment regimens were used to assess this)
Baseline mean BP (SBP/DB P mmHg) 165/104 (<65 years) 173/104 (65 years) Target BP for Treatmen t (SBP / DBP, mmHg) Not specified (just more vs. less intense) Final mean BP (SBP/DBP mmHg) and number people reaching target not reported
Reference / study type BPLTTC, 2008 SR/MA

82
N 190,60 6 31 RCTs
Populatio n HT not clear if underlyin g diabetes / CKD
BP measurem ent method Clinic
Followup Minimu m of 1000 patient years in each trial
Outcomes CV events
Best Target BP (authors conclusions) NS difference between more vs. less intense BP lowering regimens; extent of risk reduction was directly related to the degree of BP lowering
QUALITY LOW and VERY LOW (age <65 and >65 respectively); based on moderate quality SR/MA which included low to high quality RCTs) MODERATE AND LOW
Update 2011
157
Hosohata et al., 280 2007 RCT (HOMEDBP) 971 JATOS study group 2005 and 4320
HT
Home
152/90 (more and less)
12 months
More intense <125/80 Less intense 125134/8084
BP changes/ac hievement of target BP
More: 132/80; 25% Less: 133/79; 45%
NS difference between more vs. less intense BP lowering regimens for change in BP; More people in less intense reached target BP.
HT
Clinic
172/89 (strict
12 months
Strict control
BP changes/ac
12 months: Strict: 139/76;
Strict treatment group was SS better for:
MODERATE
Reference / study type 29,298 2008 RCT (JATOS)
Populatio n
BP measurem ent method
Baseline mean BP (SBP/DB P mmHg) and mild)
Followup and 2 years
Target BP for Treatmen t (SBP / DBP, mmHg) <140 SBP Mild control 140-160 SBP
Outcomes hievement of target BP; morbidity (CVD and renal failure) and mortality
Final mean BP (SBP/DBP mmHg) and number people reaching target 60% Mild: 147/79; 67% 2 years: Strict: 136/75 Mild: 146/78
Best Target BP (authors conclusions) lower final BP value (1 and 2 years) But was SS worse for number of people achieving target BP (1 year) There was NS difference for morbidity and mortality at 2 years
QUALITY
158
Solomon et al., 549 2010 RCT (EXCEED)
228
HT
Clinic
161/90 (intensiv e) 162/94 (standar d)
24 weeks
Intensive treatmen t <130 SBP Standard treatmen t <140 SBP
BP changes/ac hievement of target BP
Intensive: 131/75 Standard: 137/80 Intensive: 46% <130; 82% <140 Standard: 60% <140
More intense treatment was SS better for: lower final BP value More intense treatment increased chance of achieving SBP <140 mmHg
MODERATE AND LOW
Verdecchia et 616 al., 2009 RCT (Cardio-Sis)
1111
HT
Clinic
163/90 (tight and usual control)
2 years
Tight control <130 SBP Usual
BP changes/ac hievement of target BP;
Tight: 132/77 Usual: 136/79 Achieved <140:
Tight control group was SS better for: reduction in CV events percentage achieving SBP (<130 and <140)
MODERATE
Populatio n
BP measurem ent method
Baseline mean BP (SBP/DB P mmHg)
Followup
Target BP for Treatmen t (SBP / DBP, mmHg) control <140 SBP
Outcomes CV endpoint
Final mean BP (SBP/DBP mmHg) and number people reaching target Tight 79% Usual 67% Achieved <130: Tight 72% Usual 27%
Best Target BP (authors conclusions) reduction in BP value
QUALITY
Ichihara et al., 282 2003 RCT
140
HT
Clinic (pulse pressure analyser)
177/101 (mean)
12 months
Intense control <130/85 Moderate control <140/90
BP changes
Intense: 129/78 Moderate: 152/87
Intense control group was SS better for: reduction in BP value
LOW
159
Ogihara et al., 463 2003 RCT (VALISH) 3260 ISH Clinic 169/81 (mean) 3.07 years (median)
NT = normotensives; HT = hypertensives; ISH = isolated systolic hypertensives
Strict control <140 Moderate control 140 to <150 mmHg
BP changes/ac hievement of target BP; CV endpoint
Strict: 137/75 Moderate: 142/77 78% and 48% achieved target (strict and moderate groups respectively)
Strict control group was SS better for: percentage achieving target BPs (<140 and 140 to <150) reduction in BP value There was NS difference between the groups for:: reduction in CV events
MODERATE AND LOW
Within-treatment blood pressure studies Table 43: Study details and results for within-treatment / achieved blood pressure studies assessing the optimal blood pressure target for treatment
BP measur ement method Clinic Baseline mean BP (SBP/DB P mmHg) young: 154/100 old: 174/83 very old: 176/78
Reference / study type Wang et al., 2005623 SR/MA
N 12903 young (3049 years 160/95m mHg) 3 trials; 14323 old (60-79 years 160mmHg / <95mmHg) 5 trials; 1209 very old patients (80 years 160mmHg / <95mmHg) a) low-risk patients (n=13 trials); b) elderly patients (n=11
Population HT
Followup Median young: 5 years; old: 3.9 years; very old: 3.8 years
Outcomes CV events; CV mortality
In-treatment / achieved BPs young: 160 / 95 old and very old: 160 / <95 (ISH)
Best Target BP (authors conclusions) Anti-hypertensive treatment improves outcomes mainly by lowering SBP; Patients with >median SBP reduction risk of outcome decreased regardless of decrease in DBP or achieved DBP. Active treatment tended to reduce the risk of any outcome to a similar extent (i.e. DBP did not lead to differences in cardiovascular outcome as long as SBP substantially decreased.
QUALITY MODERATE quality SR/MA based on low quality observational studies
160
Zanchetti et al., 2009655 SR of different studies
HT (diabetic studies assessed by subgroup analysis)
Clinic
n/a
n/a
Total mortality; CV events; CV mortality
Risk groups (High, medium, low)
Achieved level of risk does not appear to correlate closely with the SBP values achieved. In high risk patients there is a ceiling effect for treatment benefits. Delaying
MODERATE quality SR/MA based on low quality observational studies
N trials); c) diabetic patients (n=11 trials; these would be outside our inclusion criteria); d) high-risk patients (n=18 trials) 6105
Population
BP measur ement method
Followup
Outcomes
In-treatment / achieved BPs
Best Target BP (authors conclusions) therapeutic correction of CV risk factors until a high level of risk is achieved,blunts the full benefits of interventions.
QUALITY
161
Arima et al., 200649 RCT (PROGRESS) Treated as observational study as not using randomised groups Coca et al., 2008134 Treated as observational study as not
Cerebrovasc ular disease (not necessarily HT)
Clinic
Stratifie d into: <120; 120-139; 140-159; 160
Median 3.9 years
Risk of Stroke
Stratified into: <120; 120-139; 140159; 160
Patients with cerebrovascular disease would have lowest risk of recurrence of stroke with BP lowered to approximately 115/75mmHg
LOW
22,576
HT
Clinic
Stratifie d into: SBP <140 vs. 140
61,836 patient years
Fatal/nonfatal stroke; Achieving target BP
SBP Stratified into: <140 vs. 140 DBP Stratified into: <90 vs. 90
Patients who achieved follow up SBP <140mmHg had lower risk of stroke than those with SBP 140mmHg; DBP <90mmHg
LOW
Reference / study type using randomised groups RCT (INVEST) Fagard et al., 2007209 Post-hoc analysis of RCT (Syst-Eur)
Population
Baseline mean BP (SBP/DB P mmHg) DBP: <90 vs. 90
Followup
Outcomes <140/90
Best Target BP (authors conclusions) had lower risk than 90mmHg.
QUALITY
4583
HT (systolic)
Clinic
Mean 174/86
median 2 years; further 4 years+ followup
Treated as observational study as not using randomised groups Shimamoto et al., 2008539 Within-group comparison study (JHEALTH) 26,512 HT Clinic Mean 166/95 Mean 3 years
Cerebrova scular events; CHD events; mortality; CV events; CV mortality
DBP Stratified into: 95; <9585; <85-75; <75-65; <65-55; <55
Antihypertensive treatment can be intensified to prevent cardiovascular events when systolic BP is not under control in older patients with systolic hypertension, at least until diastolic BP reaches 55mmHg, except in patients with coronary heart disease (MI/angina), in whom diastolic should not be lowered to <70mmHg. Clear relationship between BP control and cardiovascular events; incidence of events increased in patients with SBP 140/85mmHg (140/90mmHg in very elderly) and in diabetic patients with BP 130/85mmHg during treatment. Results suggest
LOW
162
Composit e of CV events
SBP Stratified into: <130; 130-139; 140149; 150-159; 160 DBP Stratified into: <75; 75-79; 80-84; 8590; 90
LOW
Population
Followup
Outcomes
Best Target BP (authors conclusions) that BP should be below 140/90 for reducing the risk of CV events. BP was controlled below 140.90 mmHg in the very elderly patients (85 years) and they also had a lower risk of CV events. J-shaped relationship (among each age-group) with on-treatment SBP and DBP and clinical end-points / events. SBP at HR nadir increased with increasing age highest for teh very old (140 mmHg). DBP at HR nadir was only slightly loer for the very old (70 mmHg). Therefore optimal management may involve a higher target SBP and lower target DBP for very old people (80 years) vs other age-groups.
QUALITY
Denardo et al., 2010168 A-priori subanalysis of RCT (INVEST) Treated as observational study as not using randomised groups
22,576
HT
Clinic
Overall mean: 149.5/86 .3
24 months
Mortality, MI stroke
Stratified into agegroups and SBP / DBP nadirs.*
LOW
163
Age <60 60<70 70<80 80
BP nadirs SBP 110 115 135 140 DBP 75 75 75 70
NT = normotensives; HT = hypertensives;
Target BP studies Table 44: Study details and results for target blood pressure studies assessing the optimal blood pressure target for treatment
Baseline mean blood pressure (SBP/DB P mmHg) Overall: 163/92
Reference / study type Ogihara et al., 462 2009 Sub-analysis of RCT (randomised to ARB vs ACEi) treated as observational study as not using randomised groups 4703
Populatio n HT
BP measure ment method Office
Followup Mean 3.2 years
Outcomes CV events
In-treatment / achieved blood pressure All people: 136/78
Best Target blood pressure (authors conclusions) Higher achieved blood pressure was associated with increased risk of CV events.
QUALITY LOW
164
Table 45: GRADE profile for more versus less intense treatment studies
Quality assessment No of studies Other considerations Summary of findings No of patients Imprecision more intense BP lowering less intense BP lowering Relative (95% CI) Effect Absolute Quality
Design
Limitations
Inconsistency
Indirectness
CV events (aged <65 years): SR/MA - BPLTTC (follow-up 1000 patient-years) randomised trials no serious inconsistency no serious indirectness 212/5024 (4.2%) 365/9360 (3.9%) RR 0.88 (0.75 to 1.04) 5 fewer per 1000 (from 10 fewer to 2 more)
serious
serious
none
LOW
CV events (aged >65 years): SR/MA - BPLTTC (follow-up 1000 patient-years) randomised trials no serious inconsistency no serious indirectness 156/2251 (6.9%) 260/4198 (6.2%) RR 1.03 (0.85 to 1.24) 2 more per 1000 (from 9 fewer to 15 more)
serious
very serious
none
VERY LOW
Final home SBP 12 months (Hosohata 2007 study) (follow-up 12 months; measured with: mmHg; Better indicated by lower values)
165
1 1 1
randomised trials
serious4
serious5
none
817
870
MD 1 lower (2.2 lower to 0.2 higher)6
LOW
% reaching BP target (Hosohata 2007 study) (follow-up 12 months) RR 0.44 (0.38 to 0.52)8 252 fewer per 1000 (from 216 fewer to 279 fewer)
randomised trials
serious4
no serious imprecision7
none
163/817 (20%)
392/870 (45.1%)
MODERATE
% reaching BP target (JATOS study group) (follow-up 1 years) RR 0.88 (0.84 to 0.92)8 81 fewer per 1000 (from 54 fewer to 108 fewer)
randomised trials
serious9
none
1288/2165 (59.5%)
1453/2155 (67.4%)
MODERATE
Change in SBP (JATOS study group) (follow-up 1 years; measured with: mmHg; Better indicated by lower values)
randomised trials
serious
none
2165
2155
MD 7.20 lower (8.05 to 6.35 lower)10
MODERATE
Mortality (JATOS study group) . (follow-up 2 years) randomised trials no serious inconsistency no serious indirectness no serious imprecision7 9/2165 (0.4%) 8/2155 (0.4%) RR 1.12 (0.43 to 2.9)11 0 more per 1000 (from 2 fewer to 7 more)
serious9
none
MODERATE
Morbidity (JATOS study group) (follow-up 2 years) randomised trials no serious inconsistency no serious indirectness no serious imprecision7 RR 1.0 (0.74 to 1.33)11 0 fewer per 1000 (from 10 fewer to 13 more)11
serious
none
86/2165 (4%)
86/2155 (4%)
MODERATE
Change in SBP (Solomon 2010) (follow-up 2 years; measured with: mmHg12; Better indicated by lower values) 1 randomised trials serious13 no serious inconsistency no serious indirectness serious5 none 114 114 MD 5.30 lower (0 to 0 higher)
LOW
166
1 randomised trials serious13 no serious inconsistency 1 randomised trials serious15 no serious inconsistency 1 randomised trials serious
15
% reaching target (Solomon 2010) (follow-up 2 years) RR 1.38 (1.16 to 1.64)14 227 more per 1000 (from 95 more to 382 more)
none
94/114 (82.5%)
68/114 (59.6%)
MODERATE
% reaching target (Verdecchia 2009) (follow-up 2 years) RR 1.18 (1.09 to 1.27)10 120 more per 1000 (from 60 more to 181 more)
none
399/507 (78.7%)
334/499 (66.9%)
MODERATE
CV events (Verdecchia 2009) (follow-up 2 years) HR 0.50 (0.31 to 0.79)16 51 fewer per 1000 (from 21 fewer to 71 fewer)
none
27/507 (5.3%)
52/499 (10.4%)
MODERATE
Change in SBP (Verdecchia 2009) (follow-up 2 years) randomised trials no serious inconsistency no serious indirectness no serious imprecision7 399/507 (78.7%) 334/499 (66.9%) RR 1.18 (1.09 to 1.27)17 120 more per 1000 (from 60 more to 181 more) MD 23 lower (0 to 0 higher)19 MD 5.40 lower (6.31 to 4.49 lower)10
serious15
none
MODERATE
Final SBP (Ichihara 2003) (follow-up 2 years; measured with: mmHg; Better indicated by lower values) 1 randomised trials very serious18 no serious inconsistency no serious indirectness no serious imprecision none 71 71 -
LOW
Change in SBP (Ogihara 2010) (follow-up 2 years; measured with: mmHg; Better indicated by lower values) 1 randomised trials serious15 no serious inconsistency no serious indirectness serious5 none 1545 1534 -
LOW
% reaching target (Ogihara 2010) (follow-up 2 years) 1 randomised trials serious15 no serious inconsistency no serious indirectness no serious imprecision7 none 0/1545 (0%) 0/1534 (0%) RR 1.41 (1.33 to 1.5)10 HR 0.89 (0.6 to 1.31)11 4 fewer per 1000 (from 13 fewer to 10 more) MODERATE
167
1 randomised trials serious
15
CV events (Ogihara 2010) (follow-up 2 years) no serious inconsistency no serious indirectness no serious imprecision7 none 47/1545 (3%) 52/1534 (3.4%)
MODERATE
1 2 3
RCTs included were of low to high quality; the SR/MA itself was of moderate quality 95% CI crosses both no effect and the lower MID (appreciable benefit/harm) 95% CI crosses both MIDs (appreciable benefit and appreciable harm) 4 randomised, ITT, but underpowered and attrition bias 5 95% CI crosses the lower MID 6 NS difference between groups 7 95% CI does not cross either MID 8 Favours less intense (p<0.00001) 9 Unclear allocation concealment 10 Favours Intense (p<0.00001) 11 p>0.05 (NS) 12 Favours intense (p=0.03) 13 open label, not true ITT 14 Favours intense (p=0.0002) 15 Inadequate allocation concealment and blinding 16 Favours intense (p=0.03) 17 Favours intense (p<0.001) 18 single blind, inadequate allocation concealment, ITT unclear 19 Favours intense (p<0.05)
9.7.2
Health economic evidence

One study (Jonsson 2003308) was identified from the update search that compared different blood pressure targets. This is summarised in the economic evidence profile below (Table 46, Table 47). A full evidence table is also provided in Appendix G: Evidence tables health economic studies (2011 update). No cost-effectiveness studies were included in Clinical Guideline 18 relating to this topic. Table 46: Treatment targets economic study characteristics
Study Jonsson 2003 Sweden HOT study Comparators Target DBP <90mmHg Target DBP <85mmHg Target DBP <80mmHg Applicability Partially applicable(a) Limitations Potentially serious(b) Other Comments Within RCT analysis (HOT ). Population: Hypertension and DBP110115mmHg Follow-up: mean 3.8year. Costs: antihypertensive drugs, healthcare visits, side effects, cardiovascular hospitalisations.
260
Update 2011
a) Some uncertainty about applicability of international resource use and Swedish unit costs. QALYs not used (clinical outcomes reported as not significantly different). Discounting not applied. b) Within RCT analysis and so does not incorporate all available evidence on differences between targets; issues raised with interpretation of clinical trial as achieved BPs very similar despite different targets.
Table 47: Treatment targets economic summary of findings (mean per person)
Study Jonsson 2003 Sweden HOT study Comparators Target DBP <90mmHg Target DBP <85mmHg Target DBP <80mmHg Incremental cost () Reference Incremental effects Clinical outcomes were reported as not significantly different between groups see clinical evidence review for 260 details . ICER N/a Uncertainty Differences in cost were statistically significant (p<0.01). A sensitivity analysis including non-CV hospitalisations increased total costs but differences between groups were similar.
82(a)
181 (a)
a) Converted from 1995 Swedish Kroner.
Update 2011
9.7.3

More vs. less intense treatment studies (moderate and low quality evidence) showed: NS difference for: o CV events (2 studies)82,463 RRR was related to degree of blood pressure lowering o Change in blood pressure (1 study)280 o Morbidity and mortality (1 study)29,298 Less intense was better for: o More people reaching target (2 studies)29,280,298 More intense was better for: o Lower final blood pressure value (5 studies)29,282,298,463,549,616 o Reduction in CV events (1 study)616 o Percentage reaching target SBP <130 (1 study)616 o Percentage reaching target SBP <140 (3 studies)463,549,616)
168
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets In-treatment / achieved blood pressure studies showed that: Higher achieved blood pressure was associated with increased risk CV events (2 studies and 1 SR/MA)168,539,623 Achieved SBP did not correlate with risk CV events (1 SR/MA)655 Blood pressure <140/90 had a lower risk of CV events (2 studies)134,539 Lowest risk of stroke was at blood pressure 115/75 mmHg (1 study)49 DBP did not lead to risk differences as long as SBP substantially decreased (1 SR/MA)655 DBP <90 had a lower risk of stroke (1 study)134 Up to DBP 55 (had lower risk of stroke) when SBP was controlled; except for MI/angina patients where DBP should not be <70 (1 study)209 Optimal management may involve a higher target SBP and lower target DBP for very old people (80 years) vs other age-groups (1 study)168) Target blood pressure studies showed that: Higher achieved blood pressure was associated with increased risk CV events (1 study)462
9.7.4

One partially applicable within RCT analysis (HOT) with potentially serious limitations found that lower blood pressure targets were associated with higher costs and no significant difference in clinical outcomes.
9.7.5
Link from evidence to recommendations: blood pressure treatment targets.

The GDG assessed a series of studies to define optimal treatment targets for people receiving antihypertensive therapy. The studies addressing this question were categorised into three different types; i) meta-analyses/systematic reviews of trials that had examined more versus less blood pressure lowering on treatment, i.e. people randomised to more intense versus less intense blood pressure lowering; ii) analyses of the relationship between achieved blood pressure on treatment versus clinical outcomes; iii) studies targeting patients to specific blood pressure values. The more versus less studies studies provided more robust evidence for treatment targets because they are randomised controlled trials whereas the studies using post-hoc stratifaction of ontreatment achieved blood pressures versus outcomes are not randomised and are potentially confounded by the fact that the blood pressure response to treatment may reflect underlying vascular damage, i.e. those responding less well to treatment may have more underlying vascular damage and by inference a higher risk of clinical outcomes. Moreover, such studies did not usually adjust the results according to baseline blood pressure, age and other key variables. The results of the more versus less treatment studies failed to show a consistent benefit of the lower blood pressure target on clinical outcomes82,463 but the relative risk reduction did appear to be related to the extent of blood pressure lowering across the range. One study 29,298 did show a benefit of more intensive lowering on cardiovascular morbidity and mortality. More intensive blood pressure lowering, not surprisingly, was associated with more patients reaching a lower final blood pressue value. One smaller study (Verdechia etal., 2009)616 did show better regression of LVH with more intensive BP lowering and also as a secondary analysis, a reduction in a composite of cardiovascular outcomes. In studies randomising patients to less intensive blood pressure lowering, more patients achieved the less intensive blood pressure target29,280,298 reflecting the fact that lower blood pressure targets are more diifuclt to achieve and generally required more medications. In two studies (one a systematic review) examining the impact of achieved blood pressure on treatment versus clinical oucomes, a higher achieved blood pressure was associated with a higher risk of cardiovascular events 168,539,623 and a blood pressure on treatment of <140/90mmHg
Update 2011
169
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets associated with a lower risk of cardiovascular events in two studies134,539. Similarly, in one study, a higher achieved blood pressure was associated with a increased risk cardiovascular events 462. In constrast, in one systematic review, the achieved systolic blood pressure did not correlate with the risk of cardiovascular events (1 SR/MA)655. The risk of stroke appeared particularly sensitive to achieved blood pressure on treatment with the lowest risk in those with the lowest on-treatment blood pressure, down to a value of 115/75 mmHg 49. Similar findings were observed for on-treatment stroke risk in the analysis of Sleight et al (2009). This latter study also stratified on treatment outcomes according to baseline blood pressure and showed that those in patients with a baseline systolic blood pressure <130mmHg, further blood pressure lowering appeared to be associated with an increased risk of cardiovascular events. This latter finding from a large clinical trial of patients at high cardiovascular risk does not support the uncritical adoption of lowering blood pressure in all patients at high risk of cardiovascular disease, irrespective of their baseline blood pressure. A Cochrane analysis of prospective studies of more versus less blood pressure treatment identified only studies randomised on the basis of lowering diastolic pressure and showed no evidence of more versus less blood pressure lowering on clinical outcomes (add ref we did discuss). The same analysis noted an absence of any studies designed to prospectively examine the optinal systolic treatment target. A formal cost effectiveness analysis of more versus less blood pressure lowering was not prioritised as there was no clear evidence of effectivenss. From this perspective, one potentially applicable study was identified (HOT study)260 with potentially serious limitations. This study found that lower blood pressure targets were associated with higher costs, due to the requirement for more treatment and no significant difference in clinical outcomes. Based on these analyses, the GDG concluded that most clinical trials had adopted a treatment target of <140/90 mmHg and that there was no convincing evidence supporting a lower treatment target for the pharmacological treatment of hypertension. That said, the evidence specifically examining optimal treatment targets for hypertension is inadequate and consequently the optimal treatment target could not be clearly defined with certainty. The GDG recommended that the target blood pressure for people treated for hypertension should be <140/90 mmHg (consistent with the usual target bloodpressure in clinical outcome trials), based on clinic blood pressure readings. For those with a white coat effect and thus requiring HBPM to monitor their blood pressure control, or those patients preferring to use HBPM to monitor their blood pressure control, the recommended target should be a HBPM average of <135/85mmHg (based on the equivalent values for CBPM versus HBPM used for diagnosis of hypertension). The GDG also noted the need for further studies prospectively randomising people to more versus less systolic blood pressue lowering to determine the optimal systolic pressure treatment target for people with treated hypertension. Blood pressure thresholds and targets for people over the age of 80 years: Previous guidelines in 2004 and 2006 noted the considerable uncertainty surrounding the balance of benefits and risk when considering initiating blood pressure lowering treatment for people over the age of 80 years. The uncertainty reflected tha fact that people over the age of 80 years had largely been excluded from recruitment into blood pressure treatment trials and thus, the evidence of benefit of treatment in this age group had not been established. Whilst it seemed likely that these people would accrue benefits from blood pressure lowering, it was also conceivable that treatment coud lead to more adverse effects such as syncope and falls, that might have offset any benefits of treatment. The GDG considered one systematic review (Bejan-Angoulvant, 2010)67 which compared the development of clinical outcomes in people aged 80 years who had been randomised to antihypertensive treatment versus placebo. This meta-analysis included data from 8 studies, including subgroups aged 80 years who had been randomized into treatment trials as well as one large study (HYVET study) (Beckett, et al 2009)63 which included only hypertensive people aged
Update 2011
170
Hypertension (partial update) Initiating and monitoring treatment, including blood pressure targets 80years. The total sample size was 6,701 and the mean follow-up was 3.5 years. The baseline blood pressure and initial therapy differed between studies. The results of the analysis showed that in hypertensive people 80 years, pharmacological treatment was significantly better than placebo for reducing the risk of stroke, cardiovascular events and heart failure. The HYVET study provided the most robust and highest quality evidence and had randomised people at a clinic systolic blood pressure threshold of 160mmHg and treated blood pressure to a clinic blood pressure target of <150/90mmHg. The GDG noted that the population randomised into the HYVET study were generally healthier, with lower comorbidity than typically seen in this age group. The GDG recommended that people aged 80 years, should be offered pharmacological treatment for hypertension when they have stage 2 hypertension, i.e. when their ABPM daytime average blood pressure is 150/95mmHg and should be treated to a clinic blood pressure target of <150/90mmHg. If HBPM is being used to monitor blood pressure control in people over the age of 80 years, then the blood pressure target equivakent to the recommended CBPM target of <150/90mmHg, using a HBPM average would be ~140/85mmHg. This recommendation regarding the treatment of people over the age of 80 years applies to people who have stage 2 hypertension but are not currently treated when they reach the age of 80 years. It does not mean that people reaching this age who have been previously treated at lower levels of blood pressure and/or to a lower treatment target of <140/90mmHg should have their treatment back-titrated. There is an important distinction between continuing long-term and well-tolerated treatment in people over the age of 80 years and the initiation of blood pressure lowering therapy at that age. For the latter, the evidence supports initiation of treatment at stage 2 hypertension, treating to a CBPM target of <150/90mmHg. It is conceivable lower thresholds and targets for this age group might be appropriate, however, the balance if safety and efficacy for a more aggressive treatment strategy has not been established. Indeed, before the emergence of the recent evidence (see above), there was genuine uncertainty about the balance of efficacy versus harm with regard to initiating blood pressure treatment in people aged 80 years or over. In this regard, the GDG also noted that the key studies supporting this recommendation generally included older people who were fit and active and had low levels of comorbidities. The GDG recommended that treatment decisions in those aged 80 years should be based on the realistic expectations of clinical benefit from treatment in the context of other comorbidities which might limit life expectancy. Furthermore, the GDG recommended that for older patients who are already receiving antihypertensive treatment when they reach the age of 80 years, the aforementioned evidence supports continuation of treatment.
Update 2011
9.8 Recommendations
27.Aim for a target clinic blood pressure below 140/90 mmHg in people aged under 80 years with treated hypertension. [new 2011] 28.Aim for a target clinic blood pressure below 150/90 mmHg in people aged 80 years and over with treated hypertension. [new 2011] 29.When using ABPM or HBPM to monitor the response to treatment (for example, in people identified as having a white-coat effect and people who choose to monitor their blood pressure at home) aim for a target average blood pressure during the persons usual waking hours of: below 135/85 for people aged under 80 years below 145/85 in people aged over 80 years and over. [new 2011]
171
9.9 Research Recommendation

5. In people with treated hypertension, what is the optimal systolic blood pressure? Data on optimal blood pressure treatment targets, particularly for systolic blood pressure, are inadequate. Current guidance is largely based on the blood pressure targets adopted in clinical trials but there have been no large trials that have randomised people with hypertension to different systolic blood pressure targets and that have had sufficient power to examine clinical outcomes.
Update 2011
9.10 Frequency of review

Antihypertensive medications are used extensively to manage hypertension; dose titrations, symptoms and blood pressure need to be managed and monitored. The guideline development group affirms the importance of fully involving patients in prescribing decisions and supporting them when starting, increasing, reducing or ceasing medicine to promote safety, a good health outcome and patient satisfaction. Periodic review of medicines, lifestyle and patient values and circumstances is thus an important aspect of good patient care. Although there is no evidence for the optimal period, the guideline development group felt that face-to-face medication review should occur once a year as a minimum to provide advice, review symptoms and revise medication when appropriate.
172
Hypertension (partial update) In tegrating the assessment of blood pressure, target organ damage and cardiovascular risk assessment and clinical decision making regarding treatment initiation, treatment and targets
10 Integrating the assessment of blood pressure, target organ damage and cardiovascular risk assessment and clinical decision making regarding treatment initiation, treatment and targets
The algorithms found in Section 5.1 illustrate the recommended schema for the assessment of blood pressure, clinical decision making regarding initation of treatment and review. Clinic blood pressure is usually measured at scheduled reviews in primary care or on occasions opportunistically during health screening. When clinic blood pressure is <140/90mmHg, further investigation is not usually indicated and clinic blood pressure should be re-measured at least every five years. More frequent review should be considered in people whose clinic blood pressure is close to the 140/90mmHg threshold or in those in whom there is evidence of cardiovascular disease or when their estimated 10 year cardiovascular disease risk is close to, or exceeds 20%. People with a clinic blood pressure 140/90mmHg should be offered ABPM to determine whether their daytime ABPM average is 135/95mmHg. If a persons ABPM daytime average is <135/85mmHg they should be offered annual review. If the ABPM daytime average is 135/85mmHg (i.e. stage 1 hypertension), they should be offered lifestyle advice and considered for pharmacological treatment. If their ABPM day time average is 150/95mmHg (i.e. stage 2 hypertension), they should be offered lifestyle advice and pharmacological treatment. All people considered hypertensive should undergo routine clinical evaluation to determine the presence of target organ damage, cardiovascular disease, diabetes or CKD and have their 10 year cardiovascular disease risk estimated. A review of lifestyle factors that may contribute to the development of hypertension and/or increase a patients cardiovascular disease risk should also be undertaken. If the initial clinical evaluation suggests the possibility of secondary hypertension, the patient should be referred for specialist review. If the patient has stage 1 hypertension and evidence of TOD, cardiovascular disease, diabetes, CKD, or their estimated 10 year CVD risk is 20%, they should be offered treatment. If not, they should be offered lifestyle advice and annual review as their blood pressure and cardiovascular disease risk will increase over time. For younger people i.e. aged <40 years, special consideration should be given to the possibility of secondary hypertension and the exclusion of target organ damage before deciding not to initatite therapy for stage 1 hypertension and specialist review should be considered. If not offered pharmacological treatment, they should be offered lifestyle advice and annual review. If the initial clinic blood pressure is 180/110mmHg and there is evidence of target organ damage and/or cardiovascular disease, the initiation of pharmacological therapy should not be delayed whilst awaiting the results of ABPM. If the initial evaluation suggests the possibility of accelerated hypertension or phaechromocytoma, the patient should be referred immediately (same day) for specialist care. When pharmacological treatment is considered, all patients should be offered lifestyle advice (see section 11). People at higher risk, i.e. with target organ damage, established CV disease, diabetes, CKD or an estimated 10 year CVD risk 20%, should be considered for additional therapy to reduce their cardiovascular disease risk (e.g. statins and antiplatelet therapy) if not already initiated (see NICE guidance on CVD risk, statins and antiplatelet therapy).
Update 2011
173
Hypertension (partial update) In tegrating the assessment of blood pressure, target organ damage and cardiovascular risk assessment and clinical decision making regarding treatment initiation, treatment and targets When pharmacological treatment is offered, clinic blood pressure should usually be used to monitor the response to treatment and the target blood pressure is <140/90mmHg in people aged <80 years and <150/90mmHg in people aged 80 years. For people with white coat hypertension (see section 6.4), home blood pressure monitoring (section 9.6) should be considered to monitor the response to treatment - the target blood pressure for optimal treatment is a HPBM average of <135/85mmHg.
Update 2011
174
Hypertension (partial update) Lifestyle interventions
11 Lifestyle interventions
11.1 Overview
A vast epidemiological literature describes an apparent relationship between raised blood pressure and lifestyle choices and habits. For example, observational studies have shown that people with raised blood pressure tend also to have low dietary calcium627. Does inadequate intake of dietary calcium promote raised blood pressure or is the relationship a spurious one, arising from inadequate adjustment for other hard-to-measure influences (a common problem in observational studies). There is similar controversy about the role of diet, exercise, alcohol, caffeine, potassium and magnesium supplements, sodium (table) salt and relaxation therapies. Cause and effect can only be established by repeated and methodologically sound randomized controlled trials, supported by evidence of a plausible biological mechanism, particularly when the potential benefit is small. Randomized controlled trials, enrolling patients who had raised average blood pressure defined as systolic blood pressure 140 mmHg or diastolic blood pressure 85 mmHg, analysing either blood pressure or major cardiovascular endpoints on an intention-to-treat basis, of eight weeks or more follow-up, are included in this review. However, none of the studies identified were designed to quantify significant changes in rates of death or cardiovascular events due to lifestyle interventions: instead they relied on the surrogate endpoint of reduced blood pressure with its epidemiological link to reduced rates of disease. Thus the evidence is less direct than for drug interventions which show reductions in morbidity directly. The requirement that trials have a follow-up of at least eight weeks is arbitrary but it reflects the belief that shorter time frames cannot usefully inform us about enduring changes in blood pressure. We searched electronic databases (Medline, Embase, CENTRAL) from 1998 to July 2003 for reports of relevant randomised controlled trials; articles published before 1998 were identified from hypertension guidelines, systematic reviews and meta-analyses31,118,187,192,214,293,366,388, 37,117,153,204,205,238,239,248,251,268,279,299,300,319-323,444,489,632-634 152,241,350,407 , . Though there were a number of trials informing most of the areas of interest, the trials were commonly small and the intervention of short duration (several months) relative to the progression of raised blood pressure and cardiovascular disease. The quality of reporting of studies was commonly poor (Table 48) and this may reflect poor methodological conduct, further weakening the strength of evidence and consequent recommendations for clinical care.
175
Lifestyle interventions
Table 48: Summary characteristics of trials of lifestyle interventions

Type of intervention Diet Exercise Relaxation Multiple intervention Alcohol reduction Coffee Calcium Magnesium Potassium Sodium Combined salts Number of studies 14 17 23 6 4 0 11 11 5 5 2 Number of participants 1,474 1,357 1,481 413 865 0 414 504 410 420 240 Quality markers: Randomisation description 3 (21%) 1 (6%) 6 (26%) 2 (33%) 1 (33%) 2 (18%) 1 (9%) 3 (60%) 0 (0%) 1 (50%) Concealment of allocation 2 (14%) 0 (0%) 1 (4%) 0 (0%) 0 (0%) 1 (9%) 0 (0%) 2 (40%) 0 (0%) 0 (0%) Baseline comparability a 12 (86%) 13 (76%) 5 (65%) 5 (83%) 2 (67%) 4 (36%) 6 (55%) 2 (40%) 2 (40%) 2 (100%) Blinding of: Participant b 9 (82%) 9 (82%) 3 (60%) 0 (0%) 2 (100%) Treatment provider 9 (82%) 10 (91%) 3 (60%) 0 (0%) 2 (100%) Outcome assessor 4 (29%) 2 (12%) 10 (43%) 4 (67%) 2 (67%) 1 (9%) 0 (0%) 3 (60%) 0 (0%) 0 (0%)
176
a Confirmation of baseline comparability for parallel trials or of no carryover effect for crossover trials. b Neither participant nor treatment provider could be blinded to behavioural interventions.
In overview, 98 trials including 7,993 participants were combined to provide principal findings on lifestyle interventions (see Figure 4) although these were augmented with a number of other trials and reviews. Statistically significant reductions in blood pressure were found, in the short term for improved diet and exercise, relaxation therapies, and sodium and alcohol reduction. For example, our best estimate is that a multiple intervention addressing diet and exercise can reduce systolic and diastolic blood pressure in a cohort of patients, on average, by about 5 mmHg. However this estimate is based on a limited number of patients and is uncertain. The 95% confidence interval shows that (19 times out of 20) the true average reduction may be anywhere between about 2 and 9 mmHg. Individual patients may achieve a greater or lesser reduction than the average and for a combined diet and exercise intervention the best guess is that about one quarter of patients will achieve a reduction in systolic blood pressure of at least 10 mmHg.
Figure 4: Overview of lifestyle interventions: effect on systolic and diastolic blood pressure in randomised trials of patients with raised blood pressure (140/85mmHg)
Most areas featured considerable heterogeneity (i.e. study findings were inconsistent, some positive and some negative) over and above the variation expected by the normal play of chance. This heterogeneity tends to limit the strength of recommendation that can be made about any course of action.
11.1.1
Managing changes in lifestyle

Our systolic (and to a lesser extent our diastolic) blood pressure tends to increase as we grow older. It is unhelpful to think of a single threshold above which we suddenly have problematically high blood pressure, although such thresholds can be useful to spur us into action. A review of our lifestyle helps us to identify changes we can make which may reduce our blood pressure and thus delay, reduce or remove the need for long term drug therapy as well as leading to a healthier life. The cumulative trial evidence suggests that individuals who develop improved habits of regular exercise, sensible diet and relaxation can reduce their blood pressure. Forming these habits will take determination and support. Health care professionals can provide advice, encouragement and materials but ultimately may have limited scope to influence poor dietary habits and inadequate exercise which result in part from the busy and stressful pace of life and in part from personal choice. Much of the research evidence for lifestyle change uses regular time spent together in groups for support and encouragement. Patient and healthcare organisations may be able to help provide patients with, or point them to local groups which encourage lifestyle change, particularly those promoting healthy eating and regular exercise.
11.1.2
Diet
Fourteen randomised controlled trials, including 1,474 participants, met the review inclusion criteria. 18,45,84,138,144,235,262,295,310,406,508,520,545,577,617 380,495,499,502 , . Studies most commonly compared low calorie diets, aimed at overweight patients, with either the patients' usual diet or with a prescribed 'usual care' diet. In addition, one study compared fish oil capsules with olive oil capsules (as a control); one study compared diets supplemented with fibre from oats and wheat; one study compared soy milk with skimmed cows' milk; these studies are discussed separately498, 158, 510. The mean age of study participants was 48 years and 62% were male. Only four studies reported ethnicity and in these about 45% of the participants were white. The median duration of both treatment and follow-up was 26 weeks, ranging from eight weeks to one year.
177
Hypertension (partial update) Lifestyle interventions Randomisation could be confirmed as adequate in only three studies (21%) and concealment of allocation as adequate in only one (7%). Blinding was confirmed as adequate in six studies (43%). Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in 12 studies (86%). Studies varied in their methods and in definitions of diets prescribed. Some focussed primarily on low saturated fat, others primarily on weight reduction but in practice there was considerable overlap of content. Patients were sometimes given advice on other aspects of lifestyle, such as exercise. Dieticians, nurses or counsellors generally delivered interventions although in two studies doctors were primarily involved. Two of the studies provided meals for the participants406,520. Contact between participants and the treatment providers varied considerably from several times weekly through to occasionally. Crucially, we could identify no clear system for sub-grouping diet studies: there were too many confounding influences. There was generally little change in the weight of people in the control groups, whereas average study losses in dietary intervention groups were between two and nine kilograms. Average changes in blood pressure, when comparing treatment and control groups, are shown in Figure 5. Overall, with dietary intervention there was a significant reduction in both systolic (6.0 mmHg, 95% CI: 3.4 to 8.6) and diastolic (4.8 mmHg, 95%CI: 2.7 to 6.9) blood pressure. There was no evidence of reporting bias, but significant heterogeneity existed between studies. Forty percent (95%CI: 33% to 47%) of patients put on diets were likely to show at least a 10 mmHg reduction in systolic blood pressure. There was no overall difference in withdrawal when comparing diet and control arms of studies (treatment vs. control, risk difference 3.6%, 95%CI: 0.1% to 7.2%), although studies varied. Figure 5: Effect of diet on systolic and diastolic blood pressure in randomised trials of patients with raised blood pressure
Omission of a study which enrolled abnormally hypertensive patients (mean baseline BP: 170/110 mmHg)508 resulted in a more modest estimate of reduced blood pressure due to diet: systolic 5.0 mmHg (95% CI: 3.1 to 7.0) and diastolic 3.7 mmHg (95%CI: 2.4 to 5.1).
178
Hypertension (partial update) Lifestyle interventions While soy milk appeared to lower blood pressure when compared to skimmed cows' milk510 and fish oil appeared to lower blood pressure when compared to olive oil135, these findings were from single small short-term studies and require substantiation by other independent studies. In one small study, supplementing the diet with oats did not appear to lower blood pressure when compared to wheat158. The Cochrane Collaboration415 carried out a review which had different inclusion criteria (it included simple interventions reported up to June 1998, had no restriction on length of follow up and also used weight loss as an end point) leaving only four studies common to both reviews. Nevertheless, its conclusions were similar. The recent Canadian guideline reviewed studies between 1966 and 1996355. Although without a formal meta-analysis, it likewise concluded that overweight hypertensive patients should be advised to reduce their weight.
11.1.3
Exercise
Seventeen randomised controlled trials of parallel design84,85,162,184,235,246,249,261,341, 18,45,231,391,513,559,575,583,585 including 1,357 participants, met the review inclusion criteria. Studies most commonly enrolled overweight patients and compared no intervention with a weekly schedule of three to five sessions of aerobic exercise. One study249 offered advice to participants whereas all others provided facilities. Three further studies could not be included because of missing data274,327,604. The mean age of study participants was 53 years and 58% were male. Only five studies reported ethnicity and in these about 80% of the participants were white. The median duration of both intervention and follow-up was 17 weeks, ranging from eight weeks to one year. Randomisation could be confirmed as adequate in only one study (6%), and concealment of allocation as adequate in none (0%). Blinding was confirmed as adequate in one study (6%). Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in 13 studies (76%). Overall, patients receiving exercise-promoting interventions achieved a modest reduction in both systolic (3.1 mmHg, 95%CI: 0.7 to 5.5) and diastolic (1.8 mmHg, 95% CI: 0.2 to 3.5) blood pressure compared to those in control groups (see Figure 6). There was no evidence of reporting bias. Significant heterogeneity existed between studies, although there was no obvious underlying cause for this. There were not enough studies to explore the relative merits of weight training compared to aerobics or differences between low and medium intensity aerobics. Thirty-one percent (95% CI: 23% to 38%) of patients receiving exercise interventions were likely to show at least 10 mmHg reduction in systolic blood pressure. People in the exercise arms were more likely to withdraw from the studies than those in the control arms (treatment vs. control, risk difference: 5.9%, 95%CI: 0.1% to 11.1%), although studies varied.
179
Figure 6: Effect of exercise on systolic and diastolic blood pressure in randomised trials of patients with raised blood pressure
A recent systematic review of studies of the effect of exercise on blood pressure187 included seven studies between 1966 and 1995, all with at least 26 weeks follow-up, and including normotensive and hypertensive participants. The review found exercise had a small and statistically non-significant effect on blood pressure (0.7/0.3 mmHg in 4 studies with hypertensive participants), but noted the poor quality of studies. The recent Canadian guideline reviewed studies between 1966 and 1997132. Although without a formal meta-analysis, it reported short term reductions in blood pressure of 5 to 10 mmHg and recommended 5060 minutes of moderate intensity exercise three or four times per week.
11.1.4
Relaxation therapies
Twenty-three randomised controlled trials of parallel design, including 1,481 participants, met the review inclusion criteria. RCTs of relaxation interventions32,33, 31,34,69,95,115,120,142,221,265,276,277,289,304,367,397,477-479,525,533,610,661 . Twelve further trials could not be included because of missing data128,232,245,345,398,586, 36,80,92,288,418. The mean age of study participants was 49 years and 62% were male. Only six studies reported ethnicity and in these about 84% of the participants were white. The median duration of intervention was 8 weeks, ranging from four weeks to six months; the median duration of follow-up 17 weeks, ranging from eight weeks to four years, reflecting that studies often assessed the longer term impact of interventions well after formal therapy had ceased.
180
Hypertension (partial update) Lifestyle interventions Randomisation could be confirmed as adequate in only seven studies (30%), and concealment of allocation as adequate in only one (4%). Blinding was confirmed as adequate in seven studies (30%). Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in 16 studies (70%). The common component in studies was a strategy to promote relaxation although this could be oriented through education, physical techniques (such as breathing or progressive muscle relaxation), talk therapies, stress management or some combination. Additionally some studies used biofeedback, where the participant received auditory or visual information about their heart rate, peripheral temperature or some other physical marker. There was variation in content, with individual studies incorporating (for example) forms of cognitive training, breathing management, meditation, yoga, behavioural contracts, assertiveness training and anger control techniques. Similarly, delivery varied, being provided by a range of health professionals, most commonly to groups but in a few studies to individuals. Most treatment sessions were about an hour in length (varying from 30 to 90 minutes) and were usually conducted once a week. Control groups received care varying from no intervention to sham group therapy excluding components that investigators believed to be the effective aspects of therapy. Some studies included both types of control groups. Overall relaxation interventions were associated with statistically significant reductions in systolic (3.7 mmHg, 95%CI: 1.3 to 6.0) and diastolic (3.5 mmHg, 95%CI: 1.9 to 5.1) blood pressure (see Figure 7). There was no evidence of reporting bias. However, significant heterogeneity existed between studies. Analysis of the additional value of biofeedback as a component of the intervention was inconclusive when comparing studies that did or didn't include it, or when comparing alternative interventions within trials. Thirty-three percent (95%CI: 25% to 40%) of patients receiving relaxation therapies were likely to show at least a 10 mmHg reduction in systolic blood pressure in the short term. Based on 12 of the studies, there was no significant difference in withdrawal when comparing treatment or control arms of studies (treatment vs. control, risk difference: 3.4%, 95%CI: 0.0% to 6.8%), although studies varied.
181
Figure 7: Impact of relaxation interventions on blood pressure: findings from randomised controlled trials
A recent systematic review of studies of the effect of stress reduction on blood pressure187 included seven studies between 1966 and 1995, all with at least 26 weeks follow-up, and including hypertensive participants. Although the inclusion criteria differed from ours, the review found a small and statistically non-significant effect on blood pressure (1.0/1.1 mmHg) consistent with longer follow-up studies reported here. The review similarly found considerable heterogeneity between studies. The recent Canadian guideline reviewed studies between 1966 and 1997550. It concluded that multifaceted interventions to reduce stress were more likely to be effective than single component therapies and favoured the use of cognitive behavioural therapy, based on the findings of three meta-analyses192,293,366. For hypertensive patients in whom stress appears to be an important issue, they recommended that stress management including individualized cognitive behavioural therapy may be appropriate.
11.1.5
Multiple lifestyle interventions

Six randomised controlled trials, including 413 participants, met the review inclusion criteria. RCTs of multifaceted interventions45,47,84,294,337,337,408,599. Three of the studies essentially provided a therapeutic intervention combining group exercise and diet strategies similar to the lifestyle interventions found in the previous sections45,47,84,337, 599; one study also included relaxation and restriction of intake of common salt337; one study combined a weight loss diet, relaxation and salt restriction294; and one study combined a weight loss diet, exercise and salt restriction408. A further trial, which delivered a health education package to a British population with angina, did not meet
182
Hypertension (partial update) Lifestyle interventions our inclusion criteria for blood pressure and so was excluded from the meta-analysis and is considered separately146. Three further trials could not be included because of missing data274,309,334. The mean age of participants was 52 years, 66% were male and the median follow-up of studies was six months. Five studies reported ethnicity and in these about 75% of the participants were white. Randomisation was confirmed as adequate in only two studies (33%). Concealment of allocation was inadequate or unclear in all six studies. Blinding was confirmed as adequate in four studies (67%). Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in five studies (83%). Overall, multifaceted interventions caused a modest reduction in both systolic (5.5, 95%CI: 2.3 to 8.8) and diastolic (4.5 mmHg, 95% CI: 2.0 to 6.9) blood pressure (see Figure 8). However heterogeneity existed between studies: the study of Jacob (1985) did not demonstrate a reduction in blood pressure. Twenty-six percent (95%CI: 2% to 49%) of patients receiving combined interventions were likely to show at least a 10 mmHg reduction in systolic blood pressure. Data from five studies found no statistically significant difference in withdrawal from treatment and control groups (treatment versus control, risk difference: 4.9%, 95%CI: 2.6% to 12.4%). Figure 8: Impact of combined lifestyle interventions on blood pressure: findings from randomised controlled trials
It was not possible to assess from the available data whether the effects of diet and exercise were additive or whether the combination was no better than either diet or exercise on its own. The large British health promotion study, of 688 participants, lasted longer (two years) and was of older people (mean age 63 years) than the therapeutic studies. It did not show any reduction in blood pressure in response to health advice, but nevertheless reported fewer deaths among those receiving advice (29 in control group and 13 in treatment group), providing a relative reduction in mortality of 55%, an absolute reduction in mortality of 4.6% (95%CI: 1.0% to 8.4%) or a Number Needed to Treat of 22 to prevent a death during two years of follow-up. Patients in this trial, suffering from angina, were at higher risk than most other patients enrolled in lifestyle trials, leading to greater levels of morbidity and mortality. However, the benefit of health promotion in this trial does not appear mediated by reduced blood pressure or any other obvious prognostic marker (smoking, cholesterol or body mass index), and thus needs confirmation from further research. A recent systematic review of studies of multiple interventions for preventing coronary heart disease; included nine studies of normotensive and hypertensive participants, published between 1966 and 1995, and with at least 26 weeks follow-up186. The review found an overall reduction of
183
Hypertension (partial update) Lifestyle interventions 4.2/2.7mmHg, but no significant reductions in morbidity and mortality in studies not including drug interventions.
11.1.6
Alcohol
The epidemiological link between alcohol consumption, blood pressure, cardiovascular disease and all-cause mortality has been studied extensively181,263,497,596. While moderate consumption may do no harm, the literature consistently finds that the move from moderate to excessive drinking (men: more than 21 units/week; women: more than 14 units/week) is associated both with raised blood pressure and a poorer prognosis. (Approximately: one half-pint of beer, glass of wine or a single measure of spirits equals one unit of alcohol or one standard drink and contains 8g or 10ml of alcohol287). Three randomised controlled trials, including 397 participants, met the review inclusion criteria and examined the effect of changes in alcohol consumption on blood pressure148,382,502. Interventions varied in their content but commonly featured a number of visits to a health care practitioner for advice on reducing intake of alcohol. At baseline, patients typically reported drinking 300 to 600 ml of alcohol, or 3060 standard drinks, per week. Although alcoholism was not formally defined, very heavy drinkers were commonly excluded. A further cluster randomized trial with 93 participants was identified and included in a secondary analysis348. The mean age of study participants was 53 years; in the two studies that provided the details all participants were male and three quarters were white. The PATHS study148, with 6 months treatment duration, two year follow-up and 59% of patients, differed in scale from the two other shorter and smaller trials. Randomisation could be confirmed as adequate only in the PATHS study, and concealment of allocation as adequate in none. Blinding was confirmed as adequate in two studies. Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in all three studies, with the possible exception of PATHS which did not report the proportions of men and women in the treatment and control groups. No studies were designed to assess the impact of alcohol reduction on cardiovascular endpoints. Overall, interventions to reduce alcohol consumption caused small but statistically significant reductions in both systolic (3.4 mmHg, 95%CI: 0.9 to 6.0) and diastolic (3.4 mmHg, 95%CI: 1.5 to 5.4) blood pressure. Thirty percent (95%CI: 21% to 39%) of patients receiving a structured intervention to reduce alcohol consumption were likely to achieve a reduction of at least 10 mmHg in systolic blood pressure. No harmful effects of intervention were reported in these trials; withdrawal rates were reported in only one small trial. Inclusion of the single cluster randomized study did not alter qualitatively the summary reduction in systolic (3.7 mmHg, 95% CI: 1.3 to 6.1) or diastolic (3.2 mmHg, 95%CI: 1.4 to 5.0) blood pressure, (see Figure 9). Figure 9: Impact of alcohol reduction on blood pressure: findings from randomised controlled trials
184
The recent Canadian guideline reviewed studies between 1966 and 1996113. Although without a formal meta-analysis, it recommended that alcohol consumption be limited in patients with hypertension to two or fewer standard drinks per day, with consumption not exceeding 14 standard drinks per week for men and nine standard drinks per week for women. For recommendations on preventing the development of hazardous and harmful drinking, see NICE Public Health guidance 24 (http://guidance.nice.org.uk/PH24).
11.1.7
Coffee
Although coffee is a complex beverage containing many chemicals, only the effect of caffeine has been studied extensively516. According to personal taste and type of coffee, the amount of caffeine varies, but typically coffee contains 60 to 120 mg per 150ml cup. This can be compared with tea (20 to 40 mg per 150ml cup) and cola drinks (30 to 50 mg per 330ml can)444, 130. Caffeine consumption has long being associated with raised blood pressure and can demonstrate a dose-related increase of 515 mmHg systolic and 510 mmHg diastolic for several hours following consumption. The most likely mode of action of caffeine is as an adenosine receptor antagonist, which results in vasoconstriction and raises blood pressure. The half life of caffeine in the body is typically about five hours297. We identified no randomised controlled trials examining the impact of coffee or caffeine intake on patients with hypertension, which provided at least eight weeks follow-up. A published systematic review included normotensive as well as hypertensive participants, and shorter durations of followup299. Eleven trials with a total of 522 participants and a median duration of eight weeks (range 2 to 11 weeks) were included. Control groups drank a median of five caffeinated cups of coffee a day, with treatment groups receiving no, or decaffeinated, coffee. The reported overall effect of coffee was an increase in systolic (2.4 mmHg, 95%CI: 1.0 to 3.7) and diastolic (1.2 mmHg, 95%CI: 0.4 to 2.1) blood pressure. Identifying the influence of coffee upon blood pressure, or identifying groups at particular risk, is problematic in the presence of confounding factors such as age, lifestyle, and cardiovascular disease. The small sample sizes and durations of existing trials do not provide an adequate evidence base to infer the long term effects of routine caffeine consumption.
11.1.8
Reducing sodium (salt) intake

Practical steps to reduce sodium intake include choosing low-salt foods (e.g. choosing fresh fruits and vegetables and avoiding processed foods) and reducing or substituting its use in cooking and seasoning. Much dietary salt comes from processed foods whose content should be labelled helping to monitor intake. Five randomised controlled trials (four of parallel design125,212,311,544, one of crossover design10,11), examining the effect of sodium reduction on blood pressure, met the review inclusion criteria and included 420 patients. The findings of one Italian trial in young adults are considered separately141. A further trial could not be included because of missing data395. The mean age of study participants was 52 years and 81% were male. The ethnicity of participants was not reported in any of the studies. The median duration of both intervention and follow-up was 12 weeks. One trial (17%) was double-blinded; blinding could not be confirmed in any of the other studies. Randomisation and concealment of allocation could not be confirmed to be adequate in any of the
185
Hypertension (partial update) Lifestyle interventions studies. Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in 2 studies of parallel design (40%); the crossover study did not report on carryover effects. The studies advised participants to change their diet so as to restrict their sodium intake to below 70100 mmol/day (4.2 6.0g of salt). The Scientific Advisory Committee on Nutrition target for all adults is 6 grams/day532 and NICE public health guidance on the prevention of cardiovascular diseases recommends people aim for a maximum intake of 6 grams per day per adult by 2015 and 3 grams by 2025. Average changes in blood pressure, when comparing treatment and control groups, are shown in Figure 10. Sodium reduction was associated with a statistically significant reductions in systolic (3.4 mmHg, 95%CI: 2.3 to 4.5) and diastolic (2.2 mmHg, 95%CI: 1.5 to 3.0) blood pressure. Twenty-three percent (95%CI: 17% to 30%) of patients who reduced their salt intake were likely to show at least a 10 mmHg reduction in systolic blood pressure. Based on two studies, there was no difference in withdrawal when comparing treatment and control arms of studies (treatment versus control, risk difference: 0.6%, 95%CI: 6.5% to 5.4%). Figure 10: Impact of sodium reduction on blood pressure: findings from randomised controlled trials
One Italian trial enrolled young, borderline hypertensive participants, aged 1631 years. This trial found a dramatic reduction in systolic (18.4 mmHg, 95%CI: 10.1 to 26.7) blood pressure. The trial was poorly described and it is unclear whether the reduction in systolic blood pressure is due solely to the intervention. The authors note that the benefit was found mostly in participants less than 20 years of age. The inclusion of the trial in the meta-analysis increased the average benefit of salt reduction on systolic blood pressure (7.1 mmHg, 95%CI: 2.9 to 11.3), but introduced considerable statistical heterogeneity (Q: p=0.007). Two recent systematic reviews have evaluated advice to reduce salt intake in normotensive and hypertensive adults, in trials with at least 6 months follow-up187,279. The inclusion criteria used in these reviews differ from ours, notably they included studies where the dose of antihypertensive drugs was allowed to vary. Regardless, both reviews found statistically significant reductions in blood pressure in studies with hypertensive participants, of 2.5/1.2 (up to one year follow-up) and 1.1/0.6 (one to six years follow-up)279 and 2.9/2.1 mmHg187, suggesting that reductions in blood pressure tend to diminish over time. The recent Canadian guideline220, citing a previous systematic review, concluded that sodium restriction in adults over 44 years of age resulted in a reduction in blood pressure of 6.3/2.2 mmHg per 100 mmol/day reduction in sodium. Recommendations were made for clinicians to determine
186
Hypertension (partial update) Lifestyle interventions salt intake by interview; aim for a target range of 90130 mmol per day (37 grams per day); provide advice on choosing low-salt foods (e.g. choosing fresh fruits and vegetables and avoiding preprepared foods) and reduce usage in cooking and seasoning.
11.1.9
Calcium supplements
Eleven randomised controlled trials (three of parallel design242,378,442, eight of crossover design227,318,396,571,581,584,627,660), examining the effect of calcium supplementation on blood pressure, met the review inclusion criteria and included 414 patients. Another trial, carried out in patients who were undergoing dialysis, was excluded after consideration of their unusual calcium metabolism but its details are tabulated487. A further trial could not be included because of missing data414. The mean age of study participants was 45 years and 68% were male. Only four studies reported ethnicity and in these 46% of the participants were white. The median duration of both intervention and follow-up was eight weeks. Randomisation could be confirmed as adequate in only two studies (18%) and concealment of allocation as adequate in only one (9%); nine studies (82%) studies were double-blinded treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in one study (33%) of parallel design; three studies (37%) of crossover design confirmed no carryover effect. The intervention was provided as a simple oral supplement taken several times a day. Average changes in blood pressure, when comparing treatment and control groups, are shown in Figure 11. Calcium supplementation was associated with a small reduction in systolic blood pressure 2.3 mmHg, 95%CI: 0.3 to 4.4) which was statistically significant but not robust to minor changes in the reported blood pressure of the participants, and no difference in diastolic blood pressure ( 0.8 mmHg, 95%CI: 2.1 to 0.6). No harmful effects of intervention were reported in these trials; withdrawal rates were on average around 10% in both treatment and control groups. The trials were unable to identify sub-groups of patients that might benefit from calcium. Figure 11: Impact of calcium supplementation on blood pressure: findings from randomised controlled trials
187
11.1.10
Magnesium supplements
Eleven randomised controlled trials (nine of parallel design215,270,365, 91,443,475,621,646,659] 2 of crossover design [317,645), examining the effect of magnesium supplementation on blood pressure, met the review inclusion criteria and included 504 patients. The mean age of study participants was 55 years and 44% were male. Only two studies reported ethnicity and in these 11% of the participants were white. The median duration of both intervention and follow-up was 12 weeks. Ten studies (91%) studies were single or double blinded. Randomisation and concealment of allocation were confirmed to be adequate in one study (9%) and no studies respectively. Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in six studies (67%) of parallel design; neither of the studies of crossover design reported on carryover effects. The intervention was provided as a simple oral supplement taken several times a day. Average changes in blood pressure, when comparing treatment and control groups, are shown in Figure 12. Magnesium supplementation was associated with little change in systolic ( 1.0 mmHg, 95%CI: 4.1 to 2.1) but a statistically significant reduction in diastolic (2.1 mmHg, 95%CI: 3.5 to 0.7) blood pressure. No harmful effects of intervention were reported in these trials; withdrawal rates were reported in only eight studies, where these were on average around 7% in both treatment and control groups. The trials were unable to identify sub-groups of patients that might benefit from magnesium. Figure 12: Impact of magnesium supplementation on blood pressure: findings from randomised controlled trials
11.1.11
Potassium supplementation
Five randomised controlled trials (four of parallel design107,543,543, 578, one of crossover design470), examining the effect of potassium supplementation on blood pressure, met the review inclusion
188
Hypertension (partial update) Lifestyle interventions criteria and included 410 patients. The findings of one African trial are considered separately455. A further trial could not be included because of missing data149. The mean age of study participants was 51 years and 76% were male. Only one study reported ethnicity and in this 86% of the participants were white. The median duration of both intervention and follow-up was 12 weeks. Two studies were triple blinded, two were assessment blinded and one was unclear. Randomisation and concealment of allocation were confirmed to be adequate in one (20%) and two (40%) studies respectively. Treatment and control groups were confirmed as comparable at baseline, with regard to age, sex and initial blood pressure in two studies (50%) of parallel design; the crossover study did not report on carryover effects. The intervention was provided as a simple oral supplement taken several times a day in all but one trial, where dietary advice was provided to increase intake of foods rich in potassium125. Average changes in blood pressure, when comparing treatment and control groups, are shown in Figure 13. Potassium supplementation was not associated with any significant change in systolic (3.5 mmHg, 95%CI: 7.9 to 0.9) or diastolic (0.7 mmHg, 95%CI: 4.9 to 3.6) blood pressure. The findings of the studies were heterogeneous and there are no obvious reasons for this that can be deduced from the limited available evidence. No harmful effects of intervention were reported in these trials; average withdrawal rates of 68% were similar in both treatment and control groups. Figure 13: Impact of potassium supplementation on blood pressure: findings from randomised controlled trials
One trial, which enrolled treatment nave and hypertensive Kenyan participants (DBP 90109 mmHg and SBP>160 mmHg) reported an average reduction of 39/17 mmHg. Although the effect of various salts upon certain ethnic groups is known to vary, a reduction of this magnitude exceeds our understanding and requires confirmation from further independent research. A meta-analysis by Whelton and colleagues found that oral potassium supplementation was associated with a significant reduction in both systolic blood pressure and diastolic blood pressure633, based on 12 trials in normotensive people and 21 in hypertensive people, with a duration ranging from four days to three years (median five weeks). The review found that the blood pressure lowering effect was greater in hypertensive than normotensive people, although the statistical significance of findings in the hypertensive subgroup is not reported. The review also found that the effect was more pronounced in people eating a diet high in sodium chloride (common salt) and therefore recommended potassium supplementation for both prevention and treatment of hypertension, especially in people unable to reduce their intake of sodium.
189
Hypertension (partial update) Lifestyle interventions In contrast, our restriction to trials of at least 8 weeks duration, enrolling only hypertensive patients, resulted in inclusion of only 5 trials with a median duration of 12 weeks and found that the blood pressure lowering effect of oral potassium supplementation was not statistically significant. The group concluded that there is not sufficient relevant evidence to recommend oral potassium supplementation for hypertension.
11.1.12
Combined salt supplements

Two randomised controlled trials studied combinations of the potassium, magnesium, sodium and calcium salts considered individually in previous sections. One study used paired supplements comparing two of calcium, potassium and magnesium with placebo519. None of the combined supplements reduced blood pressure when compared with placebo (see Figure 14). This was consistent with the findings for the individual supplements. Figure 14: Impact of combined supplements on blood pressure: findings from randomised controlled trials
A second study compared a mineral (reduced sodium) salt containing sodium, potassium and magnesium with common sodium table salt. The mineral salt was used in prepared food as well as for seasoning229. The reduction of blood pressure by about 5/4 mmHg consistent with that found with strategies to reduce sodium salt intake. The recent Canadian guideline reviewed studies between 1966 and 1996108. Although without a formal meta-analysis, it recommended against supplementing calcium, magnesium or potassium intake amongst hypertensive participants above the recommended normal daily levels.
11.1.13
Drug therapy versus lifestyle change

Five small randomised controlled trials enrolling 233 patients directly compared the effects of lifestyle interventions and drugs for the treatment of mild to moderate hypertension. Goldstein et al 232 , Murugesan et al 418, Kostis et al 337, MacMahon et al 380, 381, Koopman et al 333. An additional quasirandomised trial, which allocated participants to treatments on the basis of their birth date rather than at random, was also considered (Berglund et al72). All trials were small (between 38 and 66 participants), of short duration (between eight and 52 weeks) and were not designed to assess cardiovascular endpoints. Randomisation and concealment of allocation were either inadequate or not clearly reported in all trials. The outcome assessor was blinded to the treatment status of the participants in three trials333,337,380; blinding was not reported in two trials232,418, and there was no blinding in one trial72. One trial was poorly reported and did not state the total number of participants418. In two trials the confidence intervals on the effects of
190
Hypertension (partial update) Lifestyle interventions treatment could not be estimated, as either the numbers in each treatment group418 or the standard error of the treatment effects were not reported232. The populations studied in the trials differed in: (i) age participants in one trial333 were older, which probably accounted for their higher baseline blood pressure compared to participants in the other trials; (ii) treatment status at the point of recruitment participants were currently untreated or treatment nave in four trials72,232,333,380, currently treated in one trial337, or treatment status was not reported418. The trials compared different drugs with different lifestyle interventions. Typically either a diuretic or a beta-blocker was the class of drug used, although one trial allowed a choice of drugs. Four trials used a low calorie diet: one used diet alone; one combined a low calorie intake with a low sodium and high potassium diet; one used a multiple intervention combining weight loss, a low calorie and low sodium diet, exercise, and relaxation and one combined weight reduction with restricted sodium and alcohol intake. Two trials had relaxation interventions: one considered two separate relaxation interventions (biofeedback and muscular relaxation/breathing exercises); the other used yoga. Five trials reported comparable blood pressure at baseline in both treatment groups and for one trial this was unclear. Within each study, findings for systolic and diastolic blood pressure were similar. Trials comparing diet with drugs provided conflicting evidence (see Figure 15). In the trial of older participants333 who had not received treatment before and had a high baseline blood pressure, drug treatment appears more effective than diet in lowering blood pressure, whereas in a trial of younger participants381 who were currently untreated and had a lower initial blood pressure, diet appears significantly more effective than drug treatment in lowering blood pressure. The one trial337 comparing multiple lifestyle interventions with drugs found both treatments had similar effects on lowering blood pressure. Two trials found drugs to be more effective than relaxation although the confidence intervals on the treatment effects could not be determined418. Figure 15: Comparison of lifestyle and drug interventions: findings from randomised controlled trials
Participants receiving dietary interventions improved their total cholesterol profiles in all four trials compared to participants receiving drugs. Cholesterol levels were not reported in either relaxation trial. Although it was a post hoc exercise, we combined cholesterol reductions found in the dietary trials by imputing missing standard deviations. Using a random effects model, the average reduction in cholesterol was 0.52 mmol/l (95% CI 0.34 to 0.7). Withdrawals were reported in five trials: rates of withdrawal were similar for lifestyle and drug treatments. The current evidence cannot determine whether a lifestyle intervention is generally better than drug treatment for reducing blood pressure. Although cholesterol levels were not a prespecified outcome,
191
Hypertension (partial update) Lifestyle interventions it was observed that, in all four trials with diet interventions, diets were better than antihypertensive drugs at reducing cholesterol. As reduced cholesterol levels are likely to lower the risk of cardiovascular morbidity or mortality irrespective of any change in blood pressure643, a healthier diet may reduce, delay or remove the need for long-term drug therapy in some patients. Thus it seems important that patients are encouraged to try lifestyle changes before proceeding to or increasing drug therapy.
11.1.14
Smoking cessation
A review of the health consequences of smoking and benefit of smoking cessation is not included in this guideline, since there is no direct link to raised blood pressure. However smoking reduces life expectancy and is associated with poor cardiovascular and pulmonary outcomes179,180,357,410,488,648. The NHS website www.smokefree.nhs.uk has facts and information about giving up smoking. Refer to NICEs public health guidance on smoking cessation services in primary care, pharmacies, local authorities and workplaces, particularly for manual working groups, pregnant women and hard to reach communities for more information (www.guidance.nice.org.uk/PH10).
11.1.15
Recommendations
For NICE guidance on the prevention of obesity and cardiovascular disease see Obesity (NICE clinical guideline 43, 2006) and Prevention of cardiovascular disease at population level (NICE public health guidance 25, 2010). 30.Lifestyle advice should be offered initially and then periodically to people undergoing assessment or treatment for hypertension. [2004] 31.Ascertain peoples diet and exercise patterns because a healthy diet and regular exercise can reduce blood pressure. Offer appropriate guidance and written or audiovisual materials to promote lifestyle changes. [2004] 32.Relaxation therapies can reduce blood pressure and people may wish to pursue these as part of their treatment. However, routine provision by primary care teams is not currently recommended. [2004] 33.Ascertain peoples alcohol consumption and encourage a reduced intake if they drink excessively, because this can reduce blood pressure and has broader health benefits. [2004] 34.Discourage excessive consumption of coffee and other caffeine-rich products. 35.Encourage people to keep their dietary sodium intake low, either by reducing or substituting sodium salt, as this can reduce blood pressure.[2004] 36.Do not offer calcium, magnesium or potassium supplements as a method for reducing blood pressure. [2004] 37.The best current evidence does not show that combinations of potassium, magnesium and calcium supplements reduce blood pressure. [2004] 38.Offer advice and help to smokers to stop smoking. [2004] 39.A common aspect of studies for motivating lifestyle change is the use of group working. Inform people about local initiatives by, for example, healthcare teams or patient organisations that provide support and promote healthy lifestyle change. [2004]
192
Hypertension (partial update) Pharmacological interventions
12 Pharmacological interventions
In most hypertensive patients, pharmacological intervention becomes necessary if blood pressure lowering is to be substantial and sustainable. Published epidemiological studies and trials together conclusively demonstrate that a sustained reduction in blood pressure by drugs reduces the incidence of stroke, coronary heart disease, heart failure and mortality. The size of benefit in any period (for example the next 10 years) generally depends on an individual's overall cardiovascular risk. 135,379 For an individual at any age, the greater the cardiovascular risk the greater the potential to benefit from treatment. The Department of Heath National Service Framework for Coronary Heart Disease standards 3 and 4 relate to patients at risk of cardiovascular disease. 'General practitioners and primary care teams should identify all people with established cardiovascular disease and offer them comprehensive advice and appropriate treatment to reduce their risks'. 'General practitioners and primary health care teams should identify all people at significant risk of cardiovascular disease but who have not developed symptoms and offer them appropriate advice and treatment to reduce their risks.' Similarly, the Welsh National Service Framework for Coronary Heart Disease states, 'Everyone at high risk of developing coronary heart disease ... should have access to a multifactorial risk assessment and be offered an appropriate treatment plan'. Based on the findings of trials, a range of drugs (some blood pressure lowering) are offered to patients with existing coronary heart disease. These patients are the subject of a previously published national guideline. 440 The recommendations include the use of aspirin, beta-blockers, statins and ACEi. Once patients are optimally treated to prevent further disease, persistent hypertension should be managed adapting the recommendations from this document.
Update 2011
Trials treating raised blood pressure, and described in this guideline, include patients both with and without cardiovascular disease and thus are relevant to the management of raised blood pressure in all of these patients after any disease specific care has been delivered. Drugs for raised blood pressure are prescribed alone or in combination, and aim to control blood pressure while minimising side effects or toxicity. How the drugs work is not always fully understood. A brief summary of drugs used for essential hypertension is provided in Table 49; further information can be found in the British National Formulary. 306 Drugs for hypertension rarely have serious sideeffects when appropriately initiated and adequately monitored. Table 49: Outline of drugs used for essential hypertension
Commonly used Classes of Antihypertensive Drug Therapies in the United Kingdom (This is intended as a guide and reference to the product label and British National Formulary is recommended for detailed prescribing information) Class Thiazide diuretics Common generic names bendroflumethiazide, hydrochlorthiazide Mode of action Vasodilation and moderate diuresis (increased excretion of sodium, potassium and water). Vasodilation and moderate diuresis (increased excretion of sodium, Duration of action Commonly once daily morning use Usage notes Can cause gout and hypokalaemia and rarely hyponatraemia. Can increase the risk of developing type 2 diabetes Can cause gout and hypokalaemia and rarely hyponatraemia. Can increase the risk of
Thiazide like diuretics
Chlortalidone, indapamide
Commonly once daily morning use
193

Commonly used Classes of Antihypertensive Drug Therapies in the United Kingdom (This is intended as a guide and reference to the product label and British National Formulary is recommended for detailed prescribing information) potassium and developing type 2 water). diabetes Potassiumsparing diuretics Spironolactone amiloride Vasodilation and moderate diuresis (increased excretion of sodium, potassium and water). Once or twice daily Used for resistant hypertension. Spironolactone can cause gynaecomastia in males. Not to be used with potassium supplements. Can cause hyperkalaemia, especially in patients with impaired renal function. Should be avoided in primary care patients with a baseline potassium >4.5mmol/L and used with caution in people with renal impairment. Careful monitoring of potassium and renal function is required.. Not recommended as a preferred therapy for hypertension. Can be considered for resistant hypertension or as an initial therapy for women of child bearing potential. Also used for patients with angina, post myocardial infarction and chronic heart failure. Contraindicated with asthma, heart-block or in combination with a ratelimiting calcium-channel blocker. Reported side-effects include lethargy, depression and sleep disturbance. Increased risk of type 2 diabetes, especially when combined with thiazide or thiazide-like diuretics. Reported side-effects include initial headaches, palpitations, facial flushing and ankle swelling.
Beta-blockers
atenolol, bisoprolol, metoprolol, propranolol, sotalol
Suppress plasma renin production. Negative inotropic and chrontropic effects on the heart. Betablockers with alpha receptor activity also produce vasodilatation
Vary by drug from once to several times daily
Calciumchannel blockers
'dihydropyridines' amlodipine, felodipine, lacidipine nifedipine.
Vasodilatation and natiuresis vasculature.
Vary by drug from once to twice daily. Note only modified release formulation of nifedipine should be
194

Commonly used Classes of Antihypertensive Drug Therapies in the United Kingdom (This is intended as a guide and reference to the product label and British National Formulary is recommended for detailed prescribing information) used to treat hypertension 'rate-limiting CCBs' diltiazem, verapamil Heart rate slowing, vasodilatation and natiuresis Once or twice daily for longer acting forms Caution against use in heart failure or use with a beta-blocker. Reported side-effects similar to dihydropyridines but also include constipation (verapamil) and skin rashes (diltiazem) Contraindicated in pregnancy. .Careful monitoring of potassium levels and renal function required in people with renal impairment. Adverse effects include a persistent dry cough, rash and loss of taste. Rarely angioedema which is more common in black people of African or Caribean origin Contraindicated in pregnancy. Careful monitoring of potassium levels and renal function required in people with renal impairment. Generally well tolerated and unlike ACEi, do not cause cough Consider for the treatment of resistant hypertension. Beneficial side-effect on blood lipid profile. May also be considered for men with symptoms of prostatic outflow obstruction. Caution in women in whom they may cause or worsen symptoms of stress incontinence. Contraindications, cautions and side-effects vary by drug. Most common sideeffects: initial dizziness, postural hypotension,
Angiotensin converting enzyme (ACEi) inhibitors
captopril, enalapril, lisinopril, perindopril, ramipril, trandolapril
Inhibition of angiotensin coverting enzyme and reduced angiotensin II production.
Angiotensin receptor blockers (ARBs)
candesartan, irbesartan, losartan, olmersartan, valsartan, telmisartan
Selective inhibition of the angiotensin AT-1 receptor.
Once daily
Alpha receptor blockers
doxazosin, prazosin, terazosin
Antagonists of the Alpha 1 receptor.
195

Commonly used Classes of Antihypertensive Drug Therapies in the United Kingdom (This is intended as a guide and reference to the product label and British National Formulary is recommended for detailed prescribing information) headache, flushing, nasal congestion, fluid retention, ankle swelling and tachycardia.
12.1 2004 guidance: pharmacological interventions

12.1.1 Placebo controlled trials
An overview of key design characteristics of the 20 placebo controlled trials identified is shown in Table 50 (22 trials are tabulated since two trials had additional treatment arms). Seldom was the method of randomisation or steps to conceal allocation from investigators or patients adequately described, although this reflects contemporary standards of reporting. Patients, clinicians and assessors were commonly blind to the treatment received although individual trials varied. Table 50: Summary of characteristics of placebo controlled trials
Thiazides (High Dose) Number of studies Quality markers: Randomisation description Concealment of allocation Blinding: Participant Treatment provider Outcome assessor Baseline comparability 6 (86%) 4 (57%) 5 (71%) 5 (71%) 5 (100%) 4 (80%) 4 (80%) 5 (100%) 6 (86%) 4 (57%) 6 (86%) 6 (86%) 1 (100%) 1 (100%) 1 (100%) 1 (100%) 1 (100%) 1 (100%) 0 (0%) 1 (100%) 1 (100%) 1 (100%) 1 (100%) 1 (100%) 2 (29%) 0 (0%) 0 (0%) 3 (60%) 3 (43%) 0 (0%) 1 (100%) 0 (0%) 1 (100%) 0 (0%) 1 (100%) 0 (0%) 7 Thiazides (Low Dose) 5 Beta Blockers 7 Ca Channel Blockers 1 Angiotensin Receptor Blockers 1
ACEi 1
Many trials used stepped care regimes aiming to reduce blood pressure to a specified target by adding other drugs to first line therapy: most of these trials provided matching placebo stepped care to the control group (ANBPS, VA-NHLBI, EWPHE, SHEP, SHEP-P, SYST-EUR), but some provided no stepped care in the control group (MRC, MRC-O) and some provided the same active antihypertensive drugs as stepped care to both the active treatment and the control groups (IPPPSH, SCOPE).
196
Hypertension (partial update) Pharmacological interventions 12.1.1.1 Thiazide-type diuretics Thiazide-type diuretics (thiazides for short) include drugs classified by the British National Formulary (BNF) as a thiazide or thiazide like diuretic. Twelve trials were identified that met the review inclusion criteria, see Table 51. Seven trials, with 19,933 participants, starting from as early as 1964, studied high dose thiazides which are no longer used because of the risk of complications due to changed plasma potassium, uric acid, glucose, and lipids, with little additional blood pressure lowering effect compared to low dose thiazides. 26 The mean age of participants was 51, 59% were male and the mean duration of follow-up was 4.0 years. Five trials with 15,086 participants, starting between 1975 and 1989, studied low dose thiazides. Patients had a mean age of 67 years, 53% were male and the mean duration of follow-up was 4.0 years. Only two studies reported ethnicity and in these 86% of participants were Caucasian. 'Low dose' is taken pragmatically to mean the doses used in 'low dose' trials and now normally recommended by the BNF. Although the dichotomisation of low and high dose used in this guideline for placebo and head-to-head trials is the one commonly used by reviewers, individual thiazides may sometimes be used at even lower doses. The underlying risk of disease in patients was proxied by the mortality rate in the control groups of the trials. HSCSG and PATS enrolled patients following a stroke, but it is interesting to note the apparent role of age. The underlying risk in PATS is similar to three other low dose thiazide trials in which patients are, on average, ten years older. It is unclear why the underlying risk in the EWPHE trial is so high, but this may be due to inclusion of patients with coronary heart disease. Two trials, SHEP and SHEP-P exclusively enrolled patients with isolated systolic hypertension (SBP 160219 mmHg and DBP less than 90 mmHg).
197
Pharmacological interventions
Table 51: Description of individual placebo controlled trials of thiazide-type diuretics

Trial Thiazide1 Dose category high3 high high high high high high low3 low low low low Dose, mg 500 1000 10 10 50 1000 100 50 100 2550 2550 2.5 2550 12.5 25 Country Followup, yrs 4.0 2.1 4.9 5.5 >7 3.2 1.5 4.7 5.8 2.0 2.8 4.5 Start year 1973 1966 1977 1972 1965 1964 1978 1975 1982 1989 1981 1985 Age in years Range 3069 <75 3564 4049 <55 2150 60+ 6574 60+ 60+ Mean 50 59 52 45 44 51 38 72 70 60 72 72 Baseline BP, mmHg 157/101 167/100 161/98 156/97 147/99 164/104 183/101 185/91 154/93 172/75 170/77 Number enrolled 3,931 452 12,951 785 422 380 1,012 840 3,294 5,665 551 4,736 Baseline Risk2 5 53 7 4 3 39 0 77 24 28 23 23
ANBPS HSCSG MRC Oslo VAII
Chlorothiazide Methychlothiazide Bendroflumethiazide Chlorothiazide Chlorothiazide Chlorothiazide

3
Australia US UK Norway US US US Europe UK China US US
402
356 548
USPHS
1
VA-NHLBI EWPHE
Chlorthalidone Hydrochlorothiazide Hydrochlorothiazide Indapamide Chlorthalidone Chlorthalidone
6,42,453
198
MRC-O PATS
20
15
SHEP-P SHEP
281,484,485
13,483,536,606
All trials featured co-treatment or stepped care except PATS: see the trial table for details. Control Group death rate per 1000 patients per year. High doses studies were defined as those using starting drugs and doses greater than or equal to chlorthalidone 50mg, hydrochlorothiazide 50mg, chlorothiazide 501 500mg, bendroflumethiazide 5mg, methychlothiazide 5mg .
Hypertension (partial update) Pharmacological interventions A graphical presentation of pooled summary findings is shown in Figure 16 for all cause mortality, fatal or non-fatal myocardial infarction (MI) and fatal or non-fatal stroke. The high dose thiazide trials are of historical interest and, although the findings are more varied, the overall summary for each endpoint is consistent with the findings from the low-dose thiazide trials. The low dose trials show statistically significant reductions in mortality of 9%, in myocardial infarction of 22% and in stroke of 31%: a statistically consistent finding across the range of underlying risk. Figure 16: Meta-analysis of placebo-controlled randomised controlled trials of high and low dose thiazide diuretics
Patients receiving placebo withdrew from treatment at an average rate of 10.7% per year. Overall, withdrawal from active therapy was lower (Incident Risk Difference per year 1.2%, 95%CI: 1.9% to 0.6%) although there was variation between studies (Q, p<0.001). Individual studies varied from a 4% reduction in withdrawal per year to no difference. While rates of overall withdrawal are the most objective estimate of tolerability, they can conceal different problems: lack of efficacy, perceived side-effects, adverse events or disease progression. As the body of evidence increases in favour of new treatments some patients may be withdrawn from placebo-controlled trials because of symptoms or signs indicating the need for active therapy. 12.1.1.2 Beta-blockers Seven trials with 27,433 participants were identified that met the review inclusion criteria (see Table 52). Trials started between 1977 and 1988; enrolled patients had a mean age of 57 years, 49% were male and the mean duration of follow-up was 4.3 years. It is unclear what proportion of participants was from ethnic minorities.
199
Table 52: Description of individual placebo controlled trials of beta-blockers

Trial Coope
19 140
Beta-blocker1 Atenolol Atenolol Oxprenolol Propranolol Atenolol Atenolol

156
Dose, mg 100 50 160320 240 50100 50
Country UK Netherlands International UK UK Sweden Sweden
Follow-up, yrs 4.4 2.7 3.4 4.9 5.8 2.1 2.3
Start year 1978 1986 1977 1977 1982 1985 1988
Age in years Mean 69 52 52 70 76 70 Range 6079 4064 3564 6574 7084 40+
Baseline BP, mmHg 196/99 158/91 173/108 161/98 185/91 195/102 161/89
Number enrolled 884 1,473 6,357 13,057 3,315 1,627 720
Baseline Risk2 34 29 11 6 24 37 75
DUTCH-TIA IPPPSH MRC MRC-O STOP-H TEST

197 7
402 15
Beta-blocker or Diuretic3
All trials featured stepped care, with additional drugs added if necessary Control Group death rate per 1000 patients per year Atenolol (50) or Metoprolol (100) or Pindodol (5)
200
Hypertension (partial update) Pharmacological interventions A graphical presentation of pooled summary findings is shown in Figure 17 for all cause mortality, fatal or non-fatal myocardial infarction (MI) and fatal or non-fatal stroke. Overall, patients on betablockers had a statistically significant reduction in risk of stroke of 19%, and non-significant reductions in risk of death of 6% and of myocardial infarction of 8%. Figure 17: Meta-analysis of placebo-controlled randomised controlled trials of beta-blockers
Patients receiving placebo withdrew from treatment at an average rate of 10.6% per year. Withdrawal per year from active therapy and placebo was similar (Incident Risk Difference per year 0.4%, 95%CI: 1.6% to 0.8%) although there was variation between studies (Q, p<0.001). Individual studies varied from a 5% reduction in withdrawal per year to a 2% increase. 12.1.1.3 ACE inhibitors (ACEi) One trial, with 6,105 participants and a mean follow-up of 3.9 years was identified that met the review inclusion criteria (Table 53). The PROGRESS trial randomised patients following stroke to perindopril with the addition of a diuretic (indapamide) if necessary or placebo. Seventy percent of participants were male and 61% were Caucasian; 58% of patients assigned to the ACEi also received the diuretic. Table 53: Description of individual placebo controlled trials of ACEi
Trial ACEi 1 Dose , mg Country Follow -up, yrs 3.9 Star t year 199 5 Age in years Rang e 26 91 Mea n 64 Baselin e BP, mmHg 147/86 Numbe r enrolle d 6,105 Baselin e Risk2
PROGRES 500 S
Perindopr il
Internation al
27
The PROGRESS trial allowed physicians to add a diuretic if they deemed it appropriate Control Group death rate per 1000 patients per year
PROGRESS did not show an overall reduction in mortality (RR 0.96, 95%CI: 0.83 to 1.12), but statistically significant reductions in coronary events (RR 0.76, 95%CI: 0.60 to 0.96) and stroke (RR 0.73, 95%CI: 0.64 to 0.84). Patients receiving placebo withdrew from treatment during the PROGRESS trial at an average rate of 8% per year. Withdrawal per year from active therapy was similar (Incident Risk Difference per year 0.6%, 95%CI: 0.2% to 1.3%).
201
Hypertension (partial update) Pharmacological interventions The recent HOPE25,652 study randomised patients with two or more cardiovascular risk factors to a fixed dose of ramipril or placebo. The trial was designed similarly to trials of secondary cardiovascular prevention rather than treatment of hypertension; the trial population were not hypertensive and the study is not included in this review. 12.1.1.4 Angiotensin receptor blockers One trial, with 4,964 patients and a mean follow up of 3.7 years, was identified that met the review inclusion criteria (see Table 54). The SCOPE trial randomised elderly patients with mild to moderate hypertension and without cardiovascular disease in the preceding 6 months to candesartan or placebo; approximately one third were male and ethnicity was not reported. Table 54: Description of individual placebo controlled trials of angiotensin receptor blockers
Trial ARB1 Dose , mg Country Follow -up, yrs 3.7 Start year Age in years Rang e 70 89 Mea n 76 Baselin e BP, mmHg 166/90 Numbe r enrolle d 4,964 Baselin e Risk2
SCOPE
371
Candesarta n
816
Europe and N. Americ a
199 7
29
Physicians could add a diuretic and other antihypertensive agents to patients in treatment or control groups if they deemed it appropriate. Control Group death rate per 1000 patients per year.
SCOPE did not show an overall reduction in mortality (RR 0.97, 95%CI: 0.83 to 1.14) or coronary events (RR 1.10, 95%CI: 0.79 to 1.55), but a borderline statistically significant reduction in stroke (RR 0.77, 95%CI: 0.59 to 1.01), primarily due to reduced non-fatal stroke. Patients receiving placebo withdrew from treatment during the SCOPE trial at an average rate of 8% per year. Withdrawal per year from active therapy was similar (Incident Risk Difference per year 0.6%, 95%CI: 1.4% to 0.2%). Two further placebo-controlled trials were identified (IDNT362 and RENAAL97), but not considered adequately relevant to inform this guideline as both enrolled diabetic patients with mild renal impairment. 12.1.1.5 Calcium-channel blockers One trial, with 4,695 participants and median follow-up of two years, was identified that met the review inclusion criteria (see Table 55). The SYST-EUR trial enrolled patients with isolated systolic hypertension, one third of whom were male; ethnicity was not reported.
202
Hypertension (partial update) Pharmacological interventions Table 55: Description of individual placebo controlled trials of calcium-channel blaockers
Trial CCB1 Dos e, mg 10 40 Count ry Follo wup, yrs 23 Sta rt yea r 198 9 Age in years Ran ge 60+ Me an 70 Baseli ne BP, mmH g 174/8 6 Numb er enroll ed 4,695 Baseli ne Risk2 27
SYST-EUR
43,124,207,555,558
Nitrendip ine
Europ e
SYST-EUR featured stepped care, with additional drugs added if necessary. Control Group death rate per 1000 patients per year. Median follow-up.
SYST-EUR demonstrated no overall reduction in mortality (RR 1.06, 95%CI: 0.84 to 1.35), some indication of a possible reduction in coronary events (RR 0.71, 95%CI: 0.45 to 1.10) and a statistically significant reduction in stroke (RR 0.59, 95%CI: 0.41 to 0.84). Patients receiving placebo withdrew from treatment at an average rate of 14% per year. Withdrawal from active therapy per year was greater (Incident Risk Difference per year 2.3%, 95%CI: 0.8% to 3.9%). Two further placebo-controlled trials were excluded because of uncertainty about the validity of randomisation: SYST CHINA16,17,373,624] and STONE [233. 12.1.1.6 Alpha blockers No placebo-controlled trials of alpha blockers in this patient group were identified that met the review criteria.
12.2 2006 rapid pharmacological update: head to head trials

Most studies reported comparisons involving two or more drug classes in each treatment arm administered according to a stepped administration protocol. In such cases, an initial antihypertensive drug would be administered, followed by either: an increase in the dosage of the first drug, and/or the addition of a second drug if blood pressure targets were not reached using the first drug alone. All results should therefore be interpreted as demonstrating the efficacy and tolerability of each drug only when used as the initial step in a wider antihypertensive drug treatment regimen. Many studies permitted a third drug to be added in patients unresponsive to both primary and secondary antihypertensive drugs. Such drugs typically included alpha-blocking drugs such as doxazosin or centrally acting antihypertensive drugs such as clonidine. The update search found no new studies comparing ACEi or angiotensin-II receptor antagonists with beta-blockers, or comparing ACEi with ARBs. Three studies (CONVINCE78,79, NORDIL257,594 and CAPPP256,259,592) included in the original guideline were excluded due to the confounded use of either beta-blocker or thiazide diuretic as first-line antihypertensive therapy within the same treatment arm. A fourth study (MAPHY)640 was a post-hoc follow-up of a subgroup of patients already included in the HAPPHY study641, and so was excluded from the update.
203
Hypertension (partial update) Pharmacological interventions One new study (MOSES)528 identified by the update search was excluded as it reported the primary end-point as a composite of all-cause mortality, cardiovascular, and cerebrovascular events, including all recurrent events, rather than as the first event only.
12.2.1
Clinical evidence statements: head-to-head drug comparisons

ACE inhibitors versus calcium-channel blockers A meta-analysis of three studies (ALLHAT , JMIC-B , STOP-H2 ) comparing ACE inhibitors with calcium-channel blockers (CCBs) showed that ACE inhibitors were associated with a higher incidence of stroke (RR 1.14, 95% CI 1.02 to 1.28) but a lower incidence of new-onset diabetes (RR 0.85, 95% CI 0.75 to 0.98) and heart failure (RR 0.85, 95% CI 0.78 to 0.93). No significant difference was found for mortality. For MI there was substantial heterogeneity among the studies (I2 = 69%). Two studies (ALLHAT 591 650,651 , JMIC-B ) found no significant difference between study drugs in terms of MI incidence, 155,255,258,368 while a third study (STOP-H2 ) found that ACE inhibitors were associated with a reduced incidence of MI (RR 0.77, 95% CI 0.62 to 0.96). Of the two studies (ALLHAT , JMIC-B ) reporting the outcomes of unstable angina and revascularisation procedures, neither found any significant difference. The two studies (ALLHAT , STOP-H2 ) that reported the frequency of study drug withdrawals each found ACE inhibitors to be associated with more withdrawals than CCBs (respectively: RR 1.17, 95% CI 1.12 to 1.23; RR 1.14, 95% CI 1.06 to 1.24). ARBs versus calcium-channel blockers One study (VALUE) was found comparing ARBs with CCBs when used as first-line antihypertensive therapy. ARBs were associated with a higher incidence of MI compared to CCBs (RR 1.17, 95% CI 1.01 to 1.36). There was no significant difference in stroke reduction, mortality or incidence of heart failure. The study also reported frequencies of adverse events for each drug class and showed several differences, but overall these did not particularly favour either drug. Pre-specified adverse events for ARBs versus CCBs included peripheral oedema (14.9% versus 32.9%, p<0.0001), dizziness (16.5% versus 14.3%, p<0.0001) and headache (14.7% versus 12.5%, p<0.0001). Additional adverse events identified included diarrhoea (8.8% versus 6.8%, p<0.0001), serious cases of angina (4.4% versus 3.1%, p<0.0001) and syncope (1.7% versus 1.0 %, p<0.0001). ACE inhibitors versus thiazide-type diuretics A meta-analysis of three studies (ANBP2 , ALLHAT , PHYLLIS ) comparing ACE inhibitors with thiazide-type diuretics showed that ACE inhibitors are associated with a higher incidence of stroke than thiazide-type diuretics (RR 1.13, 95% CI 1.02 to 1.25). However, no difference was found for mortality. For MI, the studies are heterogeneous (I2 = 66.5%). One study based in a relatively elderly and 644 predominantly white population (ANBP2) reported a lower incidence of MI for ACE inhibitors (RR 589-591 657 0.71, 95% CI 0.51 to 0.98), but the remaining studies (ALLHAT , PHYLLIS ) found no significant difference. For heart failure, a meta-analysis of two studies (ALLHAT , ANBP2 ) also demonstrated 589-591 heterogeneity (I2 = 67.1%). ALLHAT reported a higher incidence with ACE inhibitors than 644 thiazide-type diuretics (RR 1.19, 95% CI 1.08 to 1.31), but in ANBP2 there was no significant difference. One study (ALLHAT) reported no significant difference in unstable angina but a higher incidence of revascularisation procedures (RR 1.10, 95% CI 1.00 to 1.21) with ACE inhibitors.
589-591 589-591 644 644 589-591 657 312 589-591 155,255,258,368 589-591 650,651 589589-591 650,651 155,255,258,368
II
II
II
204

Both studies (ALLHAT and ANBP2 ) found ACE inhibitors to be associated with a higher incidence of withdrawal compared to thiazide-type diuretics (RR 1.12, 95% CI 1.08 to 1.17; RR 1.10, 95% CI 1.04 to 1.17). One study (ALLHAT) reported new-onset diabetes as an outcome, and found that the incidence of diabetes after four years of follow-up was significantly higher for thiazide-type diuretics compared to ACE inhibitors (p<0.001). Calcium-channel blockers versus thiazide-type diuretics A meta-analysis of five studies (ALLHAT , INSIGHT , MIDAS , NICS-EH , VHAS ) comparing calcium-channel blockers with thiazide-type diuretics found no significant differences for mortality, MI or stroke. There was a statistically significantly higher incidence of heart failure with CCBs (RR 1.38, 95% CI 1.25 to 1.53). Conversely, based on the results of three studies (ALLHAT , INSIGHT , NICS-EH ), CCBs are associated with a reduced incidence of new-onset diabetes (RR 0.78, 95% CI 0.64 to 0.96). Only the ALLHAT study reported unstable angina as an outcome and found no significant 589-591 difference between the drug classes. For revascularisation procedures, neither ALLHAT nor 90 MIDAS found a significant difference. In terms of study drug withdrawal, one study (INSIGHT) found thiazide-type diuretics to be associated with more withdrawals than CCBs (RR 1.20, 95% CI 1.13 to 1.28), although the other 589-591 90 514,658 studies (ALLHAT , MIDAS , VHAS ) did not find a significant difference between the two drug classes. Outcomes in those with isolated systolic hypertension (ISH) A meta-analysis of three randomised controlled trials (SHEP , SHEP-P, SYST43,122,555 EUR ) compared active antihypertensive drug therapy using either thiazide-based diuretics or a calcium-channel blocker with placebo in patients with isolated systolic hypertension. Antihypertensive drug therapy was associated with a reduced incidence of stroke (OR 0.62, 95% CI 0.51 to 0.77) and myocardial infarction (OR 0.74, 95% CI 0.61 to 0.91), although there was no statistically significant difference in mortality rate. Based on the results of a subgroup analysis from one randomised controlled trial (INSIGHT) initial antihypertensive therapy with the CCB nifedipine was comparable to the thiazide-type diuretic hydrochlorothiazide plus amiloride in terms of mortality.
105,106 483,536,537,606 281,484,485 105,106 589-591 589-591 105,106 343 589-591 105,106 90 343 514,658 589-591 589-591 644
II
II
Based on the results of another subgroup analysis of patients with ISH from a randomised328 controlled trial involving patients with hypertensive LVH (LIFE) , initial therapy with an ARB is associated with a reduced incidence of stroke (RR 0.60, 95% CI 0.38 to 0.92) and a lower mortality rate (RR 0.54, 95% CI 0.34 to 0.87) compared to initial antihypertensive therapy with a beta-blocker. The two drugs were comparable in terms of the incidence of myocardial infarction. Beta-blockers versus thiazide-type diuretics Three studies (HAPPHY , MRC , MRC-0 ) were found comparing the efficacy of beta-blockers and thiazide-type diuretics. One study (HAPPHY) included only male patients. A meta-analysis of these three studies showed no significant difference between the two drug classes in terms of mortality. Heterogeneity in the study results (I2 >75%) suggested that a meta-analysis would be inappropriate for the outcomes of myocardial infarction and stroke. Sensitivity analyses were performed for 15 variation between the studies in terms of age (by including/excluding MRC-0 , in which the average 641 age of participants was 70) and gender (by including/excluding HAPPHY) , but these were unable to account for the observed heterogeneity. One study (MRC-0) found beta-blockers to be associated with a higher incidence of myocardial infarction compared to thiazide-type diuretics (RR 1.63, 95% CI 1.15 to 2.32). No association was
15 641 402 15
Level I
II
205

found in the other two studies
402 402,641
, which considered younger patients.
One study (MRC) in a relatively young population (average age 52 years) found beta-blockers to be associated with a higher incidence of stroke compared to thiazide-type diuretics (RR 2.31, 95% CI 15,641 1.33 to 4.00). However, no association was found in the other two studies . In terms of the frequency of withdrawal of the study drug, two studies (MRC , MRC-0 ) found beta-blockers to be associated with more withdrawals (RR 1.06, 95% CI 1.01 to 1.11; RR 1.29, 95% CI 641 1.22 to 1.37) while the remaining study reported a non-significant result. Angiotensin-II receptor antagonists versus beta-blockers One study (LIFE) was found comparing the angiotensin-II receptor antagonist (ARB) losartan with the beta-blocker atenolol as first-line antihypertensive therapy. The study found no significant difference between the two treatments in terms of myocardial infarction, revascularisation procedures, heart failure or angina. However, the study did find ARBs to be associated with a reduced incidence of stroke (RR 0.75, 95% CI 0.63 to 0.88), new-onset diabetes (RR 0.75, 95% CI 0.64 to 0.88) and fewer study drug withdrawals (RR 0.86, 95% CI 0.82 to 0.91). Although mortality was lower in the ARB treatment group, this result was not statistically significant. Calcium-channel blockers versus beta-blockers A meta-analysis of three studies (ASCOT , ELSA , INVEST ) compared calcium-channel blockers (CCBs) with beta-blockers. There was no statistically significant difference in mortality or myocardial 157 infarction. Based on the results of the two studies reporting stroke as an outcome (ASCOT , 656 ELSA ), CCBs were associated with a reduced incidence of stroke (RR 0.77, 95% CI 0.67 to 0.88). For heart failure, a meta-analysis of two studies (ASCOT , INVEST ) showed substantial heterogeneity (I2 = 67.4%), but neither study alone found a statistically significant difference between CCBs and beta-blockers. Based on the results of one study (ASCOT) , CCBs are associated with a reduced incidence of newonset diabetes (RR 0.71, 95% CI 0.64 to 0.78). ASCOT also found CCBs to be associated with a lower incidence of unstable angina (HR 0.68, 95% CI 0.51 to 0.92) and fewer revascularisation procedures (HR 0.86, 95% CI 0.77 to 0.96) than BBs, but 481 the INVEST study found the association between both classes of drugs to be non-significant for these outcomes. Study withdrawal was reported in two studies. In ASCOT there were fewer withdrawals 481 associated with CCBs (RR 0.64, 95% CI 0.52 to 0.77), but in INVEST there was no significant difference.
157 157 157 157 481 157 656 481 176,222,507,618,619 402 15
II
12.2.2
Meta-analysis results summary

Table 56 summarises the results from the meta-analysis comparing different drug classes in general antihypertensive populations. Included are comparisons and outcomes in which inter-study heterogeneity was considered too great to include the pooled effect size in the evidence statements above and hence these should be treated with caution. Table 56: Summary of effect sizes for each comparison included in the meta-analysis
Comparison 01 Beta-blockers versus thiazides Studies Total n Effect size RR [95% CI] I2 (%)
206

Comparison 01 Mortality 02 Myocardial infarction 03 Stroke 03 ARBs versus beta-blockers 01 Mortality 02 Myocardial infarction 03 Stroke 04 Heart failure 05 Diabetes 01 Mortality 02 Myocardial infarction (inc. silent MI) 03 Myocardial infarction (exc. silent MI) 04 Stroke 05 Heart failure 06 Diabetes 01 Mortality 02 Myocardial infarction 03 Stroke 04 Heart failure 05 Diabetes 02 ARBs versus calcium-channel blockers 01 Mortality 02 Myocardial infarction 02 Stroke 03 Heart failure 05 ACE inhibitors versus thiazides 01 Mortality 02 Myocardial infarction 03 Stroke 04 Heart failure 07 Calcium-channel blockers versus thiazides 01 Mortality 02 Myocardial infarction 03 Stroke 04 Heart failure 05 Diabetes 01 Mortality 02 Myocardial infarction 03 Stroke 5 5 5 5 3 3 3 3 32,195 32,195 32,195 32,195 20,885 9,745 9,745 9,745 0.97 [0.93, 1.02] 1.02 [0.96, 1.08] 0.93 [0.84, 1.04] 1.38 [1.25, 1.53] 0.82 [0.75, 0.90] 0.88 [0.77, 1.01] 0.75 [0.62, 0.91] 0.64 [0.52, 0.78] 0 0 0 0.2 43.8 0 0 0 2 3 3 2 29,697 30,204 30,204 29,697 1.00 [0.94, 1.06] 0.87 [0.60, 1.24] 1.13 [1.02, 1.25] 1.07 [0.81, 1.41] 0% 66.5 0 67.1 1 1 1 1 15,313 15,313 15,313 15,313 1.02 [0.93, 1.12] 1.17 [1.01, 1.36] 1.14 [0.97, 1.33] 0.88 [0.76, 1.01] N/A N/A N/A N/A 1 1 1 1 1 3 3 3 2 2 1 3 3 3 3 2 9,103 9,103 9,103 9,103 7,998 44,075 44,075 44,075 21,499 41,833 14,112 23,625 23,619 23,619 23,619 15,501 0.89 [0.78, 1.01] 1.05 [0.86, 1.28] 0.75 [0.63, 0.88] 0.95 [0.76, 1.18] 0.75 [0.64, 0.88] 0.94 [0.88, 1.00] 0.93 [0.83, 1.03] 0.91 [0.81, 1.02] 0.77 [0.67, 0.88] 0.96 [0.74, 1.26] 0.71 [0.64, 0.78] 1.04 [0.98, 1.11] 0.94 [0.74, 1.19] 1.15 [1.03, 1.27] 0.85 [0.78, 0.93] 0.85 [0.76, 0.94] N/A N/A N/A N/A N/A 5.7 0 0 0 67.4 N/A 0 69.3 5.2 0 15.2 Studies 3 3 3 Total n 15,765 15,765 15,765 Effect size RR [95% CI] 1.04 [0.91, 1.20] 1.15 [0.82, 1.60] 1.27 [0.73, 2.23] I2 (%) 44.1 76.8 77.6
06 Calcium-channel blockers versus beta-blockers
04 ACE inhibitors versus calcium-channel blockers
08 Antihypertensive therapy versus placebo (ISH population)
207
12.3 2011 update: Pharmacological therapy for hypertension

Following the rapid pharmacological update of the guideline in 2006 the use of an algorithm-based approach to treatment was recommended, based on an A,C,D, where A represented an ACEi (or ARB when an ACEi was not tolerated), C respresented a CCB, and D represented a thiazide-type diuretic. The guideline also recommended that initial therapy for primary hypertension (step 1) should be stratified according to age and ethnicity. Specifically, the guideline recommended that for older people aged 55years, treatment should be initiated with a CCB (C) or thiazide-type diuretic (D). For people under the age of 55 years, an ACEi (or ARB id ACEi was not tolerated)(A) was recommended for initial (step 1) therapy. In the absence of clinical outcomes data in younger people, this recommendation was based on data suggesting that an ACEi (or ARB) was likely to produce the most effective blood pressure lowering as initial therapy in younger patients. However, due a lack of headto-head comparison trials, it was unclear in 2006 whether an ARB could be considered equivalent to an ACEi as intial therapy for younger people. The evidence review in 2006 had also suggested that for black people of African and Caribbean descent at any age, a CCB or thiazide type diuretic was the preferred initial therapy at any age. Since 2006, important new data has become available in a number of areas; i) comparison of ACEi with ARB to determine if treatment with an ARB is equivalent at preventing clinical outcomes when compared to treatment with an ACEi; ii) for step 2 therapy, comparison between a a combination of A+C versus A+D on clinical outcomes this is important because if one of these combinations is preferred then it would impact on the preferred step 1 therapy for people aged 55 years, or black people of African and Caribbean descent at any age; iii) new data showing differential effects of antihypertensive treatments on blood pressure variability, suggesting that blood pressure variability per se is an independent predictor of clinical outcomes; iv) a review of diuretic therapy, specifically addressing whether the predominant use of low dose bendroflumethiazide as the preferred diuretic for the treatment of hypertension in the UK is justified when the majority of clinical trials have used different thiazide-type diuretics; and v) new data on antihypertensive therapy options for resistant hypertension (step 4 treatment). Finally, since 2006, the cost of antihypertensive therapies has decreased significantly, some more than others (e.g. CCBs and ARBs) due to generics becoming available. Consequently, this update of hypertension guideline dealing with pharmacological treatment for primary hypertension reviewed recommendations with regard to; i) the equivalence of ACEi versus ARBs on clinical outcomes; ii) the appropriate choice of diuretic therapy for the treatment of hypertension and their place in the hierarchy of treatment; iii) the preferred combination of therapies for step 2 and step 3 treatment; and iv) the treatment of resistant hypertension, i.e. step 4 treatment. This review of pharmacological treatment strategies was supported by an updated cost-effectiveness analysis comparing different treatments with updated costings.
Update 2011
12.3.1
Angiotensin-converting enzyme inhibitors (ACEi) versus Angiotensin Receptor Blockers (ARB)

Forest plots found in Appendix H: Forest plots.
12.3.1.1
Clinical evidence The literature was reviewed from December 2005 onwards (this was the cut-off date of the previous NICE guidance on pharmacological treatment of hypertension, CG34) for systematic reviews and RCTs comparing ACEi vs ARB for first-line treatment in adults with primary hypertension. RCTs were included if there was: 12 months follow-up, N200 and the population did not consist of people who were exclusively diabetic or had CKD.
208
Hypertension (partial update) Pharmacological interventions Three RCTs552,587,653 were found which fulfilled the inclusion criteria and addressed the question and were included in the review. The first RCT653 (the ONTARGET trial) compared treatment with the ACEi ramipril (5 mg/day) vs. the ARB telmisartan (50 mg/day) and vs. a combination of the two (ACEi+ARB) in N=25,620 people with hypertension, and had a median follow-up time of 56 months. Treatment followed a stepped add-on therapy protocol (stepped up to double or triple therapy) for non-responders in each arm. The second RCT587 compared treatment with the ACEi enalapril (20 mg/day) vs. the ARB losartan (50 mg/day) in N=560 people with hypertension, and had a follow-up time of 24 months. Treatment followed a one-step dose adjustment protocol for the ACEi arm. The third RCT552 (CORD IB trial) compared treatment with the ACEi ramipril (5 mg/day) vs. the ARB losartan (50 mg/day) in N=3860 people with hypertension, and had a follow-up time of 12 months. Treatment followed a stepped dose adjustment and add-on therapy protocol (increased dose then if needed added on additional antihypertensive) for non-responders in each arm. NOTE: no quality of life data was found, or data assessing the effects of ACEi vs ARB in people aged 80+ or black people of African and Caribbean descent. NOTE: we additionally looked for outcomes relating to sexual dysfuntion in men, for ACE vs ARB (as this is thought to be an important ussue particulary for erectile dysfunction sufferers). However,no outcomes relating to this were reported in any of the studies. 12.3.1.2 Evidence statements - clinical The evidence profile below (Table 57) summarises the quality of the evidence and outcome data from the three RCTs552,587,653 included in this review, comparing ACEi versus ARB. ARB was significantly better than ACEi for: less study drug withdrawals* mortality (all cause) MI (fatal and non-fatal) stroke (fatal and non-fatal) angina requiring hospitalisation coronary revascularisation new onset diabetes heart failure [moderate quality evidence] [high quality evidence] [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [high quality evidence] [moderate quality evidence] [moderate quality evidence] There was NS difference between ACEi and ARB for:
Update 2011
*There was significant heterogeneity for this outcome when the data from the three trials were pooled together. Heterogeneity could be explained by the fact that both low and high quality trials had been pooled together (details of sensitivity analysis by methodological quality can be found in the forest plot for this outcome). Low quality trials were defined as those which had no blinding or allocation concealment. Data included in GRADE for this outcome was therefore based on the high quality trial alone. However the overall quality rating given by GRADE for this outcome was moderate due to imprecision (reasons outlined in the evidence profile).
209
Table 57: Evidence profile comparing ACEi versus ARBs

Quality assessment No of studies Other considerations Summary of findings No of patients Imprecision ARB ACEi Relative (95% CI) Effect Absolute Quality
Design
Limitations
Inconsistency
Indirectness
Mortality (all cause) (follow-up 12 - median 56 months) 2 CORDIB55

2
ONTARG ET653 2 CORDIB55

2
randomised trials
no serious limitations1
none
995/10443 (9.5%)
1018/10535 (9.7%)
HR 0.98 (0.9 to 1.07)
2 fewer per 1000 (from 9 fewer to 6 more)
HIGH
MI (fatal and non-fatal) (follow-up 12-56 months) randomised trials no serious limitations1 no serious inconsistency no serious indirectness 443/10443 (4.2%) 417/10535 (4%) HR 1.07 (0.94 to 1.22) 3 more per 1000 (from 2 fewer to 8 more)
serious
none
ONTARG ET653 2 CORDIB55

2
MODERATE
210
Stroke (fatal and non-fatal) (follow-up 12 - median 56 months) randomised trials no serious limitations1 no serious inconsistency no serious indirectness 378/10443 (3.6%) 413/10535 (3.9%) HR 0.92 (0.8 to 1.06) 3 fewer per 1000 (from 8 fewer to 2 more)
serious2
none
ONTARG ET653
MODERATE
Hospitalisation for angina (follow-up median 56 months) 1 ONTARG ET653 randomised trials no serious limitations3 no serious inconsistency no serious indirectness 954/8542 (11.2%) 925/8576 (10.8%) HR 1.04 (0.95 to 1.14) 4 more per 1000 (from 5 fewer to 14 more)
serious
none
MODERATE
Coronary revascularisation (follow-up median 56 months) 1 ONTARG ET653 randomised trials no serious limitations3 no serious inconsistency no serious indirectness no serious imprecision 1290/8542 (15.1%) 1269/8576 (14.8%) HR 1.02 (0.95 to 1.1) 3 more per 1000 (from 7 fewer to 14 more)
none
HIGH
New onset diabetes (follow-up 12-56 months)
2 CORDIB55
2
ONTARG ET653
randomised trials
serious2
none
404/7195 (5.6%)
372/7386 (5%)
HR 1.12 (0.97 to 1.29)
MODERATE
Heart failure (follow-up median 56 months) 1 ONTARG ET653 randomised trials no serious limitations3 no serious inconsistency no serious indirectness 537/8542 (6.3%) 514/8576 (6%) HR 1.05 (0.93 to 1.19) 3 more per 1000 (from 4 fewer to 11 more)
serious2
none
MODERATE
Study drug withdrawal (follow-up 12 - median 56 months) 1 ONTARG ET653 HR 0.87 (0.81 to 0.92)7 23 fewer per 1000 (from 14 fewer to 34 fewer)
randomised trials
serious3,4
no serious inconsistency5
no serious indirectness3
serious6
none
1812/10572 (17.1%)
2067/10665 (19.4%)
LOW
1 2
1/2 studies (CORD IB): no blinding, no allocation concealment; but this trial was small compared to the other included one (ONTARGET) so overall weighted as no serious limitations. 95% confidence interval includes both 1) no effect and 2) appreciable benefit or appreciable harm 3 Random, double blind, allocation concealment, powered, ITT analysis. However unclear final dropouts (but treatment withdrawal was <30% for median 56 months follow-up) so acceptable. 4 Patients who entered the trial had already been 'filtered' at run-in to exclude those with poor compliance or who did not perform well. 5 3 studies originally included and pooled but there was significant heterogeneity (p<0.1 and I2 >50%). Low quality trials removed based on sensitivity analysis, and result reported here is from the high quality trial data. 6 95% confidence interval crosses both 1) no effect and 2) appreciable benefit or harm and non-appreciable benefit or harm 7 p<0.0001; favours ARB
211
12.3.1.3
Economic evidence Three studies were identified in the update search that included ACEi and ARB in the comparators but all were excluded due to being judged to have serious methodological limitations202,529,560. In the absence of a published cost effectiveness analysis, current UK drugs costs were presented to the GDG to inform decision making. It was noted that losartan has recently come off patent and other ARBs are also due to come off patent over the next few years.
12.3.1.4
Evidence statements Clinical ARB was significantly better than ACEi for: less study drug withdrawals* [low quality evidence]
There was a non-significant difference between ACEi and ARB for: mortality (all cause) MI (fatal and non-fatal) stroke (fatal and non-fatal) angina requiring hospitalisation coronary revascularisation new onset diabetes heart failure [high quality evidence] [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [high quality evidence] [moderate quality evidence] [moderate quality evidence]
Update 2011
*There was significant heterogeneity for this outcome when the data from the three trials were pooled together. Heterogeneity could be explained by the fact that both low and high quality trials had been pooled together (details of sensitivity analysis by methodological quality can be found in the forest plot for this outcome). Low quality trials were defined as those which had no blinding or allocation concealment. Data included in GRADE for this outcome was therefore based on the high quality trial alone. However the overall quality rating given by GRADE for this outcome was still low for reasons outlined in the evidence profile. 12.3.1.5 Evidence statements Health economics No relevant evidence of cost-effectiveness was available. In terms of drug acquisition costs alone, in December 2010 based on BNF 60 the lowest cost ARB was 25.94 per year (losartan [100mg used for costing]) and the lowest cost ACEi was 20.73 per year (ramipril [10mg used for costing]).
12.3.2
Diuretics
In adults with primary hypertension, which is the most clinically and cost effective thiazide type diuretic (bendrofluazide/bendroflumethiazide, chlorthalidone, indapamide, hydrochlorothiazide) for first line treatment, and does this vary with age and ethnicity?
12.3.2.1
Clinical evidence Thiazide-type diuretics versus placebo or other antihypertensive drug class The literature was searched for all years (as this was not addressed in the previous guidelines)425,436. SRs/MAs and RCTs were included that compared the following TDs
212
Hypertension (partial update) Pharmacological interventions (bendrofluazide/bendroflumethiazide, chlorthalidone, indapamide, hydrochlorothiazide) with either placebo or other classess of a-HT drugs for 1st-line therapy. Studies were excluded if they had sample sizes of N<200, follow-up of <1 year or populations which were exclusively diabetic or had chronic kidney disease. Pre-specified outcomes of interest were only clinical outcomes (e.g. stroke, MI etc.) and not BP measurements. NOTE: in the previous NICE hypertension guidelines 425,436 a lot of the evidence for diuretics was on Chlorthiazide, which is no longer used in the UK and is why many of the studies have not been included in this review. 14 RCTs (21 papers) were identified which fulfilled the inclusion criteria and addressed the question, and were included in the review {1995 6420 /id;Sareli, 2001 489 /id;1978 6415 /id;Beckett, 2008 387 /id;The ALLHAT Officers and Co-ordinators for the ALLHAT Collaborative Research Group, 2000 6139 /id;Weir, 2003 2500 /id;The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT), 2002 752 /id;Wing, 2003 6558 /id;Borhani, 1996 6140 /id;1985 1144 /id;Zanchetti, 2004 80 /id;Zanchetti, 1998 785 /id;Rosei, 1997 786 /id;Perry, 2000 417 /id;SHEP Cooperative Research Group, 1991 470 /id;SHEP Cooperative Research Group, 1988 471 /id;Kostis, 1997 654 /id;Vaccarino, 2001 545 /id;Perry, 1986 418 /id;Hulley, 1985 6137 /id;Perry, 1989 6142 /id;Malacco, 2003 16093 /id;Tresukosol, 2005 1971 /id}. NOTE: several of the studies were published as multiple papers (SHEP: three papers;335,483,606 SHEP-P: three papers;281,484,485 VHAS: two papers;514,658 and ALLHAT: three papers589,591,628) reporting different outcomes, so these studies have only been counted once, however results from all the papers are reported and referenced here483. The table below (Table 58) summarises the studies included in the review. {1995 6420 /id;Sareli, 2001 489 /id;1978 6415 /id;Beckett, 2008 387 /id;The ALLHAT Officers and Co-ordinators for the ALLHAT Collaborative Research Group, 2000 6139 /id;Weir, 2003 2500 /id;The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT), 2002 752 /id;Wing, 2003 6558 /id;Borhani, 1996 6140 /id;1985 1144 /id;Zanchetti, 2004 80 /id;Zanchetti, 1998 785 /id;Rosei, 1997 786 /id;Perry, 2000 417 /id;SHEP Cooperative Research Group, 1991 470 /id;SHEP Cooperative Research Group, 1988 471 /id;Vaccarino, 2001 545 /id;Perry, 1986 418 /id;Hulley, 1985 6137 /id;Perry, 1989 6142 /id;Malacco, 2003 16093 /id;Tresukosol, 2005 1971 /id}. Table 59 summarises the diuretics used in each trial and their doses. Data was categorised into those diuretics that were classed as: thiazide diuretics (TDs): bendrofluazide / bendroflumethiazide (BDZ) and hydrochlorothiazide (HCTZ) thiazide-like diuretics (TDLs): chlorthalidone (CTD) and indapamide (IND) Table 58: Summary of included studies
Study TDs BDZ MRC
8
Update 2011
N 17,354
Intervention BDZ (10mg/day)
Comparison Propanolol (240mg/day) or placebo
Results NS difference in overall mortality, CHD events or cardiovascular events between BDZ and propanolol. BDZ better than propanolol for reduced cerebrovascular events. NS difference in overall mortality or CHD events between BDZ and placebo. BDZ better than placebo for reduced
213

Study N Intervention Comparison Follow-up Results cardiovascular, and cerebro-vascular events
TDs HCTZ THAI elderly{Tresukos ol, 2005 1971 /id} MIDAS

90
200
HCTZ (25-50 mg/day) HCTZ (25 50 mg/day)
CCB (amlodipine) (5-10 mg/day) CCB (isradipine) (2.5- 5mg/daily)
18 months
No difference between HCTZ and CCB for mortality NS differences between HCTZ and isradipine for overall mortality, CHD events, cardiovascular, and cerebro-vascular events NS differences between groups
883
36 months
Sareli et al. 2001

524
409
HCTZ (12.5 mg/day)
CCB (nifedipine SR) (30 mg/day) or CCB (verapamil hydrochloride SR) (240 mg/day) or ACEi (enalapril maleate) (10 mg/day) ACEi (fosinopril) (25mg qid) pravastatin in 50% of patients.
13 months in total but 2 months for monothera py data
PHYLLIS
657
508
HCTZ (25 mg qid) pravastatin in 50% of patients. CTD (50 mg/day initially) CTD (12.5-25 mg/day)
Mean 2.6 years
NS differences in CHD events, cerebrovascular events or cardiovascular events
TDLs CTD VA-NHLBI

3
1012
Placebo
2 years
NS differences between groups CTD better than placebo for reduced CHD events, reduced stroke and reduced cardiovascular events. NS difference for HF (fatal and non-fatal). NS differences between groups NS differences in overall mortality, CHD events, or cerebrovascular No difference between CTD and CCB for mortality, stroke, MI and HF
SHEP
06
335,483,536,537,6
4736
Placebo
4.5 years
SHEP- P
281,484,485
441
CTD (25-50 mg/day) CTD (25mg/day) CTD (12.5-25 mg/day)
Placebo
34 months
VHAS
514,658
1414
CCB (verapamil) (240mg/day) CCB (lacidipine) (4-6 mg/day)
2 years
SHELL
384
1882
Median 32 months
214

Study ALLHAT
589,591,628
N 42,418
Intervention CTD (12.525mg/day)
Comparison CCB (amlodipine) (2.5- 10mg/day) or ACEi (Iisinopril) (10-40mg/day)
Results NS difference between CTD and ACEi I for overall mortality and CHD events. CTD better for cardiovascular and cerebro-vascular events NS difference between CTD vs. CCB for all cause mortality and CHD events, cardiovascular events, and cerebrovascular events CTD worse than enalapril for CHD events. NS difference for overall mortality, cardiovascular and cerebro-vascular events
ANBP2
644
6083
CTD (GPs choice of dose)
ACEi (enalapril) (GPs choice of dose)
Mean 4.1 years
TDLs IND PATS

20
5665
IND (2.5 mg/day)
Placebo
Mean 2 years
IND better for reduced stroke (fatal and nonfatal), total mortality, CV deaths and coronary deaths IND better for reduced MI (fatal and non-fatal), HF (fatal and non-fatal) and mortality. NS difference between groups for stroke
HYVET
63
3845
IND SR (1.5 mg/day)
Placebo
Mean 2.1 years
Table 59: Diuretic and dosage used in trial

Diuretic used TDs HCTZ 5 524 Sareli 644 ANBP2 657 PHYLLIS 90 MIDAS THAI elderly{Tresukosol, 2005 1971 /id} 1 8 MRC 2 20 PATS 63 HYVET 6 591,628 ALLHAT 335,483,536,537 SHEP 384 SHELL 12.5mg/day At GPs discretion 25mg qid 25-50mg/day 25-50 mg/day 10mg/day Number of trials Doses used
BDZ TDLs IND
2.5mg/day 1.5mg/day (SR) 12.5 25mg/day 12.5 25mg/day 12.5-25 mg/day
CTD
215

Diuretic used Number of trials VHAS 484,485 SHEP-P 3 VA-NHLBI
514,658
Doses used 25mg/day 25-50mg/day 50-100mg/day
The evidence profiles below (Table 60 to
216
Hypertension (partial update) Pharmacological interventions Table 67) summarise the evidence and outcome data from the 14 RCTs{1995 6420 /id;Sareli, 2001 489 /id;1978 6415 /id;Beckett, 2008 387 /id;The ALLHAT Officers and Co-ordinators for the ALLHAT Collaborative Research Group, 2000 6139 /id;Weir, 2003 2500 /id;The Antihypertensive and LipidLowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT), 2002 752 /id;Wing, 2003 6558 /id;Borhani, 1996 6140 /id;1985 1144 /id;Zanchetti, 2004 80 /id;Zanchetti, 1998 785 /id;Rosei, 1997 786 /id;Perry, 2000 417 /id;SHEP Cooperative Research Group, 1991 470 /id;SHEP Cooperative Research Group, 1988 471 /id;Kostis, 1997 654 /id;Vaccarino, 2001 545 /id;Perry, 1986 418 /id;Hulley, 1985 6137 /id;Perry, 1989 6142 /id;Malacco, 2003 16093 /id;Tresukosol, 2005 1971 /id} included in this review comparing diureticsvs. placebo or other a-HT drug classes. Data are presented for each diuretic. NOTE: cerebrovascular events in some trials was cited and was synonymous with stroke.
217
Table 60: Bendroflumethazide versus placebo

Quality assessment No of studies 1 MRC8 Design Limitations Inconsistency Indirectness Summary of findings No of patients Other Bendroflumethiazide Imprecision considerations versus placebo Overall mortality (follow-up mean 4.9 years) serious
2
Effect control Relative (95% CI) Absolute 0 fewer per 1000 (from 7 fewer to 9 more) 0 fewer per 1000 (from 7 fewer to 8 more) 9 fewer per 1000 (from 6 fewer to 11 fewer) 11 fewer per 1000 (from 3 fewer to 17 fewer) Quality
randomised trials
serious
none
128/3519 (3.6%)
253/6941 (3.6%)
HR 1 (0.81 to 1.24)
LOW
CHD event (follow-up mean 4.9 years) 1 MRC8 randomised trials serious1 no serious inconsistency no serious indirectness serious2 none 119/3519 (3.4%) 234/6941 (3.4%) HR 1 (0.8 to 1.25)
LOW
Stroke (follow-up mean 4.9 years) 1 MRC8 randomised trials serious

1
serious
none
18/3519 (0.5%)
109/6941 (1.6%)
HR 0.44 (0.30 to 0.63)
LOW
Cardiovascular event (follow-up mean 4.9 years) 1 MRC8

1 2
randomised trials
serious1
serious3
none
140/3519 (4%)
352/6941 (5.1%)
HR 0.78 (0.65 to 0.94)
LOW
Allocation concealment unclear and attrition high 95% CI includes no effect and appreciable benefit or appreciable harm 3 95%CI does not include no effect but crosses both appreciable benefit or harm and non-appreciable benefit or harm
Table 61: Indapamide versus placebo

Quality assessment No of studies 2 PATS20 HYVET
63
Design
Limitations
Inconsistency
No of patients Other Indapamide Indirectness Imprecision control considerations versus placebo Overall mortality (follow-up mean 2.05 years) no serious indirectness serious2 393/4736 (8.3%) 8.90%
Summary of findings Effect Relative Absolute (95% CI) 12 fewer per 1000 (from 1 fewer to 21 fewer) 13 fewer per 1000 (from 1 fewer to 22
Quality
randomised trials
none
342/4774 (7.2%)
HR 0.85 (0.74 to 0.99)
MODERATE
fewer) CHD event (follow-up mean 2.05 years) 2 PATS20 HYVET63 78/4736 (1.6%) none 50/4774 (1%) 1.90% Stroke (follow-up mean 2.05 years) 2 PATS20 HYVET63 286/4736 (6%) none 210/4774 (4.4%) 5.70% Cardiovascular event (follow-up mean 2.05 years) 2 PATS20 HYVET63 259/4736 (5.5%) none 203/4774 (4.3%) 4.70% Quality of life - no limitations in daily activities (follow-up mean 2 years) 1 PATS20
1 2
randomised trials
serious3
serious2
HR 0.53 (0.36 to 0.77)
8 fewer per 1000 (from 4 fewer to 11 fewer) 9 fewer per 1000 (from 4 fewer to 12 fewer) 17 fewer per 1000 (from 8 fewer to 23 fewer) 16 fewer per 1000 (from 7 fewer to 22 fewer) 12 fewer per 1000 (from 4 fewer to 19 fewer) 11 fewer per 1000 (from 3 fewer to 17 fewer) 30 more per 1000 (from 11 more to 52 more)
LOW
randomised trials
serious2
HR 0.72 (0.61 to 0.87)
MODERATE
randomised trials
serious2
HR 0.77 (0.64 to 0.93)
MODERATE
randomised trials
no serious limitations
serious2
none
2125/2841 (74.8%)
2019/2824 (71.5%)
HR 1.09 (1.03 to 1.16)
MODERATE
Both had allocation concealment; attrition was >20% in one trial and no data provided in the other trial 95%CI does not cross the line of no effect but crosses both appreciable benefit or harm and non-appreciable benefit or harm 3 Heterogeneity was 77%. This could be due to different populations. One trial recruited adults aged 80 years+ and the other trial recruited patients with a recent TIA or stroke.
Table 62: Chlorthalidone versus placebo

Quality assessment No of studies 3 SHEP335,483,536,537 SHEP-P484,485 VA-NHLBI 3
3
Design
Limitations
Inconsistency
Indirectness
Other considerations Overall mortality (follow-up mean 2 years) Imprecision serious2
No of patients Chlorthalidone control versus placebo
Summary of findings Effect Relative Absolute (95% CI)
Quality
randomised trials
serious1
none
8/508 (1.6%)
5/504 (1%)
HR 0.87 (0.73 to 1.04)
1 fewer per 1000 (from 3 fewer to 0 more)
LOW
CHD events (follow-up mean 2 years) SHEP SHEP-P484,485 VA-NHLBI3 Stroke 24 fewer per 1000 (from 13 fewer to 33 fewer) 24 fewer per 1000 (from 13 fewer to 34 fewer) 0 more per 1000 (from 0 fewer to 0 more)
335,483,536,537
randomised trials
serious1
serious3
serious4
none
16/508 (3.1%)
8/504 (1.6%)
HR 2.0 (0.86 to 4.67)
16 more per 1000 (from 2 fewer to 56 more) 16 more per 1000 (from 2 fewer to 57 more)
VERY LOW
2 SHEP335,483,536,537 SHEP-P484,485
randomised trials
serious5
none
114/2808 (4.1%)
165/2479 (6.7%)
HR 0.63 (0.49 to 0.80)
MODERATE
Cardiovascular event (follow-up mean 2 years) SHEP335,483,536,537 2 randomised trials serious1,6 no serious inconsistency no serious indirectness no serious imprecision none 2/508 (0.4%) 0/504 (0%) HR 4.31 (0.27 to 68.84)
VA-NHLBI3 1 No ITT analysis conducted on data in one study, attrition >20% in two studies 2 95%CI crosses both no effect and appreciable harm or benefit 3 Heterogeneity 59% 4 95%CI does not cross no effect but includes both appreciable benefit or harm and non-appreciable benefit or harm 5 Attrition >20% 6 ITT analysis not conducted in one study and attrition > 20% in the other study
MODERATE
Table 63: Chlorthalidone versus calcium channel blocker.

Quality assessment No of studies 3 ALLHAT591,628 SHELL384 VHAS514,658 CHD events (follow-up 2 to 4.9 years) 2 ALLHAT591,628 VHAS514,658 1474/9497 (15.5%) 8.90% Stroke (follow-up 2 to 4.9 years) 3 ALLHAT591,628 SHELL384 VHAS
514,658
Design
Limitations
Inconsistency
Indirectness
No of patients Other Chlorthalidone Imprecision control considerations versus CCB Overall mortality (follow-up 2 to 4.9 years)
Summary of findings Effect Relative Absolute (95% CI) 4 more per 1000 (from 4 fewer to 12 more) 2 more per 1000 (from 2 fewer to 7 more) 1 more per 1000 (from 7 fewer to 11 more) 1 more per 1000 (from 4 fewer to 7 more)
Quality
randomised trials
serious1
none
2329/16483 (14.1%)
1406/10439 (13.5%) 7.50%
HR 1.03 (0.97 to 1.10)
MODERATE
randomised trials
serious1
none
2460/15543 (15.8%)
HR 0.94 (0.88 to 1.0)
MODERATE
randomised trials
serious1
serious2
none
717/16483 (4.3%)
419/10439 (4%)
HR 0.94 (0.83 to 1.06)
LOW
Cardiovascular events (follow-up mean 4.9 years) 1 ALLHAT591,628 1 SHELL384 1 SHELL384

1 2
randomised trials
serious3
none
3941/14836 (26.6%)
2432/8790 (27.7%)
HR 0.96 (0.91 to 1.01) HR 0.83 (0.46 to 1.62) HR 1.17 (0.54 to 2.53)
12 more per 1000 (from 0 more to 23 more) 4 fewer per 1000 (from 13 fewer to 15 more) 2 more per 1000 (from 6 fewer to 19 more)
MODERATE
Heart failure (follow-up mean 32 months) randomised trials serious4 no serious inconsistency no serious indirectness very serious2,5 none 19/940 (2%) 23/942 (2.4%) VERY LOW
MI (follow-up mean 32 months) randomised trials serious4 no serious inconsistency no serious indirectness very serious2,5 none 14/940 (1.5%) 12/942 (1.3%) VERY LOW
Attrition was >20% in both trials. There was inadequate explanantion of allocation concealment in one trial 95%CI includes both no effect and appreciable benefit or harm 3 Attirtion >20% 4 Unclear allocation concealment and open blind 5 95%CI includes both no effect and both appreciable benefit and appreciable harm
Table 64: Chlorthalidone versus ACEi Inhibitor

Quality assessment No of studies Design Limitations Inconsistency No of patients Other Chlorthalidone Indirectness Imprecision control considerations versus ACEi Overall mortality (follow-up 4.1 to 4.9 years) Summary of findings Effect Relative Absolute (95% CI) 2 more per 1000 (from 6 fewer to 9 more) 2 more per 1000 (from 5 fewer to 8 more) 40 more per 1000 (from 6 more to 81 more) 29 more per 1000 (from 5 more to 60 more) 4 fewer per 1000 (from 1 fewer to 8 fewer) 5 fewer per 1000 (from 1 fewer to 9 fewer) 11 fewer per 1000 (from 5 fewer to 17 fewer) 17 fewer per 1000 (from 7 fewer to 26 fewer) Quality
2 ALLHAT591,628 ANBP2644
randomised trials
serious1
none
2413/17873 (13.5%)
1509/11822 (12.8%) 10.70%
HR 1.00 (0.94 to 1.07)
MODERATE
CHD events (follow-up 4.1 to 4.9 years) 2 ALLHAT591,628 ANBP2644 1563/11822 (13.2%) 9.50% Stroke (follow-up 4.1 to 4.9 years) 2 ALLHAT591,628 ANBP2644 112/3044 (3.7%) none 107/3037 (3.5%) 4.40% Cardiovascular events (follow-up 4.1 to 4.9 years) 2 ALLHAT591,628 ANBP2644 394/3044 (12.9%) none 429/3037 (14.1%) 20.80% HR 0.91 (0.86 to 0.96) HR 0.88 (0.79 to 0.98) HR 0.97 (0.91 to 1.03)
randomised trials
serious1
none
2533/17873 (14.2%)
MODERATE
randomised trials
serious1
serious2
LOW
randomised trials
serious1
LOW
Table 65: Hydrochlorthiazide versus calcium channel blockers

Quality assessment No of studies Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients HCTZ versus control CCB Summary of findings Effect Relative Absolute (95% CI)
Quality
Overall mortality (follow-up 2 to 36 months) 3 Sareli, MIDAS, THAI{Sareli, 2001 489 /id;Borhani, 1996 6140 /id;Tresukosol, 2005 1971 /id} randomised trials serious1 no serious inconsistency no serious indirectness very serious2 none 10/599 (1.7%) 10/833 (1.2%) HR 1.18 (0.48 to 2.90) 2 more per 1000 (from 6 fewer to 22 more) VERY LOW
CHD events (follow-up 2 to 36 months) 2 Sareli, MIDAS

90,524
randomised trials
serious1
very serious2
none
13/499 (2.6%)
19/733 (2.6%) 2.30%
HR 0.77 (0.37 to 1.57)
12 more per 1000 (from 7 fewer to 51 more) 11 more per 1000 (from 6 fewer to 46 more) 13 more per 1000 (from 7 fewer to 90 more) 14 more per 1000 (from 7 fewer to 92 more) 27 more per 1000 (from 2 fewer to 81 more) 23 more per 1000 (from 2 fewer to 69 more)
VERY LOW
Stroke (follow-up mean 36 months) 1 MIDAS90 6/442 (1.4%) 1.40% Cardiovascular events (follow-up 2 to 36 months) 2 Sareli, MIDAS90,524 26/733 (3.5%) 3% HR 1.8 (0.94 to 3.44) HR 1.99 (0.5 to 7.97)
randomised trials
serious3
very serious2
none
3/441 (0.7%)
VERY LOW
randomised trials
serious1
serious4
none
14/499 (2.8%)
LOW
1 2
None of the trials provide adequate information on allocation concealment. One of the trials had attrition >20% and ITT analysis was not conducted on the data in the other trial 95%CI includes no effect and appreciable benefit and appreciable harm 3 Trial did not provide adequate information on allocation concealment and attrition > 20% 4 95% CI includes both no effect and appreciable benefit or appreciable harm
Table 66: Hydrochlorthiazide versus ACEi Inhibitor

Quality assessment No of studies 1 Sareli
524
Design
Limitations
Inconsistency
Indirectness
Imprecision
Other considerations
No of patients HCTZ versus control ACEi
Summary of findings Effect Relative (95% CI) Absolute
Quality
Overall mortality (follow-up mean 2 months) randomised trials serious1 no serious inconsistency no serious indirectness very serious2 none 1/58 (1.7%) 0/60 (0%) HR 4.06 (0.08 to 204.37) 0 more per 1000 (from 0 fewer to 0 more) VERY LOW
CHD events (follow-up mean 2.6 years) 1 PHYLLIS 1 PHYLLIS657

657
randomised trials
serious3
very serious2
none
3/253 (1.2%)
1/254 (0.4%)
HR 3.02 (0.31 to 29.07)
8 more per 1000 (from 3 fewer to 104 more) 11 more per 1000 (from 4 fewer to 535 more) 12 more per 1000 (from 4 fewer to 541 more) 11 more per 1000 (from 4 fewer to 535 more) 12 more per 1000 (from 4 fewer to 541 more)
VERY LOW
Stroke (follow-up mean 2.6 years) randomised trials serious

3
very serious2
none
0/253 (0%)
1/254 (0.4%)
HR 3.90 (0.08 to 196.36)
VERY LOW
Cardiovascular event (follow-up mean 2.6 years) 1 PHYLLIS657

1 2 3
randomised trials
serious
very serious2
none
0/253 (0%)
1/254 (0.4%)
HR 3.90 (0.08 to 196.36)
VERY LOW
No information on allocation concealment and attrition >20% 95%CI includes both no effect and appreciable benefit and appreciable harm No information on allocation concealment and unclear on attrition
Table 67: Bendroflumethiazide versus Beta blocker

Quality assessment No of studies 1 MRC8 Design Limitations Inconsistency Indirectness Summary of findings No of patients Other Bendroflumethiazide Imprecision considerations versus Beta blocker Overall mortality (follow-up mean 4.9 years) very serious2 none 128/3519 (3.6%) Effect control Relative (95% CI) HR 1.08 (0.84 to 1.39) Absolute Quality
randomised trials
serious1
120/3558 (3.4%)
VERY LOW
CHD events (follow-up mean 4.9 years) 1 MRC8 randomised trials serious1 no serious inconsistency no serious indirectness serious4 none 119/3519 (3.4%) 103/3558 (2.9%) HR 1.17 (0.9 to 1.52) 5 more per 1000 (from 3 fewer to 15 more)
LOW
Stroke (follow-up mean 4.9 years) 1 MRC8 randomised trials serious1 no serious inconsistency no serious indirectness serious3 none 18/3519 (0.5%) 42/3558 (1.2%) HR 0.43 (0.25 to 0.75) 7 fewer per 1000 (from 3 fewer to 9 fewer)
LOW
Cardiovascular events (follow-up mean 4.9 years) 1 MRC

1 2 3 8
randomised trials
serious1
very serious2
none
140/3519 (4%)
146/3558 (4.1%)
HR 1.03 (0.82 to 1.3)
VERY LOW
Allocation concealment unclear and attrition > 20% 95%CI includes both no effect and appreciable benefit and appreciable harm 95%CI does not include no effect but does cross appreciable and non-appreciable benefit and harm 4 95%CI includes no effect and appreciable benefit or appreciable harm
Head to head comparisons The literature was searched for all years (as this was not addressed in the previous guidelines)425,436. SRs/MAs and RCTs were included that compared the fllowing TDs with each other: bendrofluazide/bendroflumethiazide, chlorthalidone, indapamide, hydrochlorothiazide for 1st-line therapy. There was no restriction placed on sample size or follow-up time. Populations which were exclusively diabetic or had chronic kidney disease were excluded. Outcomes of interest were only BP measurements. All studies included in this review measured BP in the office. However two studies94,199 used both office and ABPM or just ABPM measurements. A total of 15 RCTs were found that fulfilled the inclusion criteria. The different comparisons are detailed in the table (Table 1) below. Six RCTs 94,194,339,493,494,551 Emeriau, 2001195 were found which compared Indapamide (IND) vs. Hydrochlorothiazide (HCTZ). Two RCTs 39,76 were found which compared Indapamide (IND) vs. bendrofluazide/bendroflumethiazide (BDZ). Two RCTs
266,503
were found which compared Indapamide (IND) vs. chlorthalidone (CTD). were found which compared Chlorthalidone (CTD) vs. hydrochlorothiazide
Three RCTs (HCTZ).
93 198 216
One RCT5 was found which compared Hydrochlorthiazide (HCTZ) vs. bendroflumethiazide (BDZ). NOTE: several studies194,195,503 assessed additional arms treating people with other classes of a-HT drugs. These were not included because they did not answer this part of the question (TDs vs. TDs) and were not included in the first part of the question (TDs vs. placebo / other a-HT classes) because they did not meet inclusion criteria (ie. were N<200 and/or had <1 year follow-up time). NOTE: all RCTs were underpowered to detect a difference in BP. In order to detect a 5mm difference, a sample size of N500 is needed. NOTE: five studies were cross-over trials: Bowlus 1964, Ernst 2006, Elliott 1991, Hatt 1975, Kreeft 198493,194,198,266,339 The table below (Table 1) summarises the studies included in this review and the results5,39,76,93,94,194,195,198,216,266,339,493,494,503,551 Data was categorised into those diuretics that were classed as: thiazide diuretics (TDs): bendrofluazide / bendroflumethiazide (BDZ) and hydrochlorothiazide (HCTZ) thiazide-like diuretics (TDLs): chlorthalidone (CTD) and indapamide (IND) Table 68: Summary of included studies
Study TDL vs TD Bowlus 93 1964 Ernst, 198 2006 29 CTD (50mg/day) CTD (12.5mg/day) force titrated to 25mg/day HCTZ (100 mg/day HCTZ (25mg/day) force titrated to 50mg/day 6 weeks treatment, 2 weeks washout 8 weeks treatment, 4 weeks washout, 8 weeks treatment NS difference in BP between groups. NS difference (office BP and 24hr ABPM) between groups. N Intervention Control Follow-up Results
Update 2011
30
226

Study Finnerty, 216 1976 N 54 Intervention CTD (50mg/day plus placebo) Control HCTZ (100mg/day) Follow-up 2 weeks no treatment, followed by 4 weeks of treatment in either arm. 2 months placebo run-in, 12 weeks TD drug, 2 months placebo washout, 12 weeks alternate TD drug. 48 weeks Results NS difference in BP between groups.
Kreeft, 339 1984
17
IND (2.5mg/day)
HCTZ (50mg/day)
NS difference in BP between groups.
Plante, 493 1988
47
IND (2.5mg/day)
HCTZ (50 mg/day)
IND better for reduced BP (no P value reported) and was less likely to be associated with hypokalaemia. IND better for reduction in DBP in the recumbent position
Plante, 494 1983
24
IND (2.5mg/day)
HCTZ (50 mg/day)
4-6 washout placebo period, followed by 12 weeks active therapy. 6 months 12 weeks Previously untreated patients. Addition of ACEi at 6 weeks if target BP not met. 4 week washout placebo period; 12 weeks treatment 28 days
Spence, 551 2000 Brandao, 94 2010
39 94
IND (2.5mg/day) IND (1.5 mg/day)
HCTZ (25 mg/day) HCTZ (25 mg/day)
NS difference in BP between groups NS difference in BP (office or ABPM) between groups
Emeriau, 195 2001
524
IND (SR) (1.5 mg/day)
HCTZ (25 mg/day) Amlodipine (5 mg/day)
Similar reduction in BP between groups (equivalence test)
Elliot, 194 1991
11
IND (2.5mg/day) or HCTZ (25 mg/day) IND (2.5mg/day)
Placebo (lactose)
NS difference in BP between groups.
Alem, 39 2008
26
BDZ (2.5 mg/day)
28 days
Both IND and BDZ reduced BP to a significant degree. Equivalent fall in BP in both groups
Bing, 76 1981
20
IND
BDZ
22 weeks
227

Study TDL vs TDL Raki, 503 2002 80 IND (2.5mg/day) CTD (25mg/day) NIC (60mg/day) PPL (120mg/day) CTD (100mg/day) 6 months Significant decreases in BP in all groups N Intervention (2.5mg/day) Control (5 mg/day) Follow-up Results
Hatt, 266 1975
36
IND (5mg/day)
10 days washout, followed by 90 day crossover 12 months
IND better % reduction in DBP.
TD vs TD Anonymou 5 s, 1984 44 HCTZ (12.5mg/day) BDZ (12.5mg/day) NS difference in BP between groups.
Table 69: Thiazide drug and dosages used in trials

TD name CTD Number of trials 5 93 Bowlus, 1964 198 Ernst, 2006 216 Finnerty, 1976 266 Hatt, 1975 503 Raki, 2002 11 5 Anonymous, 1984 194 Elliot, 1991 93 Bowlus, 1964 198 Ernst, 2006 216 Finnerty, 1976 339 Kreeft, 1984 493 Plante, 1988 494 Plante, 1983 551 Spence, 2000 94 Brandao, 2010 195 Emeriau, 2001 11 Doses used 50mg/day 12.5mg/day force titrated to 25mg/day 50mg/day plus placebo 100mg/day 25mg/day 12.5mg/day 25 mg/day 100mg/day 25mg/day force titrated to 50mg/day 100mg/day 50mg/day 50mg/day 50mg/day 25 mg/day 25 mg/day 25 mg/day NOTE: ALL (except one) OF THESE TRIALS STATED THAT THE PREPARATION WAS SR. ALL JUST STATED INDAPMIDE AND THE DOSE. 1.5 mg/day 1.5 mg/day (SR) 2.5mg/day 2.5mg/day 2.5mg/day
HTCZ
Indapamide
Brandao, 2010 195 Emeriau, 2001 39 Alem, 2008 76 Bing, 1981 194 Elliot, 1991 266 Hatt, 1975
94
228

TD name Number of trials Kreeft, 1984 493 Plante, 1988 494 Plante, 1983 503 Raki, 2002 551 Spence, 2000 BDZ 3 39 Alem, 2008 76 Bing, 1981 5 Anonymous, 1984
339
Doses used 5mg/day 2.5mg/day 2.5mg/day 2.5mg/day 2.5mg/day 2.5mg/day 2.5 mg/day 5 mg/day 12.5mg/day
229
Table 70 to Table 75 below summarise the quality of the evidence and outcome data from the studies included in the review 39,76,93,94,194,195,198,216,266,339,493,503,551 Figure 1: TDL vs TD (CTD vs HCTZ)
Table 70: Thiazide-like diuretics versus thiazide diuretics (chlorthalidone versus hydrochlorthiazide)
No of studies Summary of findings Effect Relative Other Limitations Inconsistency Indirectness Imprecision Chlorthalidone HCTZ Absolute (95% considerations CI) SBP seated (change from baseline) BOWLUS (follow-up 6 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 7 lower ( to serious none 29 29 inconsistency indirectness imprecision lower)1 DBP seated (change from baseline) BOWLUS (follow-up 6 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 2.1 lower ( to serious2 none 29 29 inconsistency indirectness imprecision lower)1 SBP seated (change from baseline) ERNST (follow-up 8 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 6.3 higher ( to serious3 none 30 30 inconsistency indirectness imprecision lower)1 DBP seated (change from baseline) ERNST (follow-up 8 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 1.2 lower ( to 3 serious none 30 30 inconsistency indirectness imprecision lower)1 SBP: 24h ABPM (change from baseline) ERNST (follow-up 8 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 5 lower ( to serious3 none 30 30 inconsistency indirectness imprecision lower)1 SBP unknown method (change from baseline) FINNERTY (follow-up 4 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 4 higher ( to serious3 none 26 28 inconsistency indirectness imprecision lower)1 DBP unknown method (change from baseline) FINNERTY (follow-up 4 weeks; measured with: mmHg; Better indicated by lower values) no serious no serious no serious MD 1.3 higher ( to serious3 none 26 28 inconsistency indirectness imprecision lower)1 Quality assessment No of patients Quality
Design
193 193 1198 1198
randomised trials randomised trials randomised trials randomised trials randomised trials randomised trials randomised trials
MODERATE
MODERATE
MODERATE
MODERATE
1198
MODERATE
1216
MODERATE
1216
1 2
MODERATE
NS differnce between groups High dropout rates; no ITT analysis 3 unclear allocation concealment
Table 71: Thiazide-like diuretics versus thiazide-like diuretics (indapimide versus chlorthalidone)
Quality assessment No of studies Design Limitations Inconsistency Indirectness Imprecision Other considerations No of patients Indapamide versus Chlorthalidone Summary of findings Effect Relative control Absolute (95% CI) MD 0 higher (10.14 lower to 10.14 higher) MD 4 lower (9.94 lower to 1.94 higher) MD 3.10 higher (3.08 lower to 9.28 higher)4 MD 3.50 higher (0.22 lower to 7.22 higher)4
Quality
SBP supine (end of follow-up) HATT (Better indicated by lower values) 1266 randomised trials randomised trials randomised trials randomised trials very serious1 no serious inconsistency no serious indirectness very serious2 none 38 38 VERY LOW
1266
very serious1 no serious limitations no serious limitations
1503
1503
1 2
DBP supine (end of follow-up) HATT (Better indicated by lower values) no serious no serious very none 38 38 inconsistency indirectness serious3 SBP supine (end of follow-up) RAKIC (follow-up 6 months; measured with: mmHg; Better indicated by lower values) no serious no serious serious3 none 20 20 inconsistency indirectness DBP supine (end of follow-up) RAKIC (follow-up 6 months; measured with: mmHg; Better indicated by lower values) no serious no serious serious3 none 20 20 inconsistency indirectness
VERY LOW
MODERATE
MODERATE
Although the trial was single blinded, randomisation and allocation concealment was not described and there was no ITT analysis 95%CI includes no effect and both appreciable benefit and appreciable harm 3 95%CI include no effect and appreciable benefit or harm 4 NS difference between groups
Table 72: Thiazide-like diuretics vs Thiazide diuretics (Indapamide versus hydrochlorthiazide)

Quality assessment No of studies Design Limitations Summary of findings Effect Relative Other Indapamide Inconsistency Indirectness Imprecision control Absolute (95% considerations versus HCTZ CI) SBP supine (end of follow-up) (follow-up 28 days to 48 weeks; Better indicated by lower values) No of patients very serious2 no serious indirectness no serious indirectness no serious indirectness no serious imprecision no serious imprecision no serious imprecision none 77 74 MD 8.36 lower (10.92 to 5.8 lower) MD 4.2 lower (5.48 to 2.92 lower) MD 8.74 lower (11.75 to 5.73 lower) Quality
5194,339,493,494,551
randomised trials randomised trials randomised trials
serious1
VERY LOW
DBP supine (end of follow-up) (follow-up 28 days to 48 weeks; Better indicated by lower values) 5
194,339,493,494,551
very serious
serious3
none
77
74
VERY LOW
SBP upright (end of follow-up) (follow-up 28 days to 48 weeks; Better indicated by lower values) 4194,339,494,551 no serious limitations very serious4 none 54 55 LOW
DBP upright (end of follow-up) (follow-up 28 days to 48 weeks; Better indicated by lower values) 4194,339,494,551 randomised trials randomised trials no serious limitations very serious5 no serious indirectness no serious indirectness no serious imprecision no serious imprecision none 54 55 MD 3.85 lower (5.41 to 2.28 lower) MD 3.95 lower (7.03 to 0.87 lower) MD 0.76 lower (2.5 lower to 0.98 higher) MD 12.55 lower (17.11 to 7.99 lower) MD 2.07 lower (7.2 lower to 3.06 higher) MD 5.5 higher (0 to 0 higher)9 MD 5.9 higher (0 to 0 higher)9 MD 7.5 higher (0 to 0 higher)9 MD 2.0 higher (0 to 0 higher)9 LOW
SBP supine (change from baseline) (follow-up 3-6 months; measured with: mmHg; Better indicated by lower values) 2
195,551
serious
none
196
192
MODERATE
DBP supine (change from baseline) (follow-up mean 3-6 months; measured with: mmHg; Better indicated by lower values) 2195,551 randomised trials serious6 no serious inconsistency no serious indirectness no serious imprecision none 196 192 MODERATE
SBP upright (change from baseline) (follow-up mean 6 months; Better indicated by lower values) 1551 randomised trials no serious limitations no serious inconsistency no serious indirectness no serious imprecision none 18 21 HIGH
DBP upright (change from baseline) (follow-up mean 6 months; Better indicated by lower values) 1551 randomised trials no serious limitations no serious inconsistency no serious indirectness serious7 none 18 21 MODERATE
SBP seated (change from baseline) (follow-up 12 weeks; Better indicated by lower values) 194 randomised trials randomised trials randomised trials randomised trials serious8 no serious inconsistency no serious inconsistency no serious inconsistency no serious indirectness no serious indirectness no serious indirectness no serious imprecision no serious imprecision no serious imprecision none 32 33 MODERATE
DBP seated (change from baseline) (follow-up 12 weeks; Better indicated by lower values) 1
94
serious
none
32
33
MODERATE
SBP: 24h ABPM (change from baseline) (follow-up 12 weeks; Better indicated by lower values) 194 serious8 none 32 33 MODERATE
194
1 2
serious8
DBP: 24h ABPM (change from baseline) (follow-up 12 weeks; Better indicated by lower values) no serious no serious no serious none 32 33 inconsistency indirectness imprecision
MODERATE
There were inadequate methodological information in two of the three trials Heterogeneity was 78% 3 Heterogeneity was 76% 4 Heterogeneity was 72% 5 Heterogeneity 68% 6 1/2 studies unclear for allocation concealment 7 95%CI includes no effect and appreciable harm or benefit
8 9
unclear allocation concealment There was NS differnce between groups
Table 73: Thiazode-like diuretic versus thiazide diuretic (Indapamide vs benroflumethiazide)

Quality assessment No of studies Design Limitations Inconsistency Indirectness Imprecision Other considerations Summary of findings No of patients Indapamide versus Bendrofluazide/Bendroflumethiazide control Relative (95% CI) Effect Absolute Quality
SBP supine (end of follow-up) (follow-up mean 22 weeks; Better indicated by lower values) 176 randomised trials very serious no serious inconsistency no serious indirectness serious none 10 10 MD 32 lower (72.34 lower to 8.34 higher) MD 2 lower (32.58 lower to 28.58 higher) MD 5 lower (18.85 lower to 8.85 higher) MD 0 higher (30.97 lower to 30.97 higher) MD 5.6 higher (8.35 lower to 19.55 higher) MD 3.2 higher (1.85 lower to 8.25 higher) VERY LOW
SBP upright (end of follow-up) (follow-up mean 22 weeks; Better indicated by lower values) 176 randomised trials very serious1 no serious inconsistency no serious indirectness no serious imprecision none 10 10 LOW
DBP supine (end of follow-up) (follow-up mean 22 weeks; Better indicated by lower values) 176 randomised trials very serious1 no serious inconsistency no serious indirectness very serious2 none 10 10 VERY LOW
DBP Upright (end of follow-up) (follow-up mean 22 weeks; Better indicated by lower values) 176 randomised trials very serious1 no serious inconsistency no serious indirectness no serious imprecision none 10 10 LOW
SBP (absolute change) (follow-up mean 22 weeks; Better indicated by lower values) 139 randomised trials very serious1 no serious inconsistency no serious indirectness serious3 none 13 10 VERY LOW
DBP (absolute change) (follow-up mean 22 weeks; Better indicated by lower values) 139
1 2
randomised trials
very serious1
serious3
none
13
10
VERY LOW
Lacked most methodological information 95%CI includes no effect and appreciable benefit and appreciable harm 3 95%CI includes no effect and appreciable and non-appreciable harm or benefit
Table 74: Thiazide diuretic vs thiazide diuretic (hydrochlorthiazide vs bendroflumethiazide)

Quality assessment No of studies Other considerations Summary of findings No of patients Imprecision HCTZ BDZ Relative (95% CI) Effect Absolute Quality
Design
Limitations
Inconsistency
Indirectness
SBP supine (change from baseline) (follow-up 12 months; measured with: mmHg; Better indicated by lower values) 15 randomised trials serious1 no serious inconsistency no serious indirectness no serious imprecision none 21 15 MD 1 lower (0 to 0 higher)2 MD 3 higher (0 to 0 higher)2 MD 1 higher (0 to 0 higher)2 MD 4 higher (0 to 0 higher)2 MODERATE
DBP supine (change from baseline) (follow-up 12 months; measured with: mmHg; Better indicated by lower values) 15 randomised trials serious1 no serious inconsistency no serious indirectness no serious imprecision none 21 15 MODERATE
SBP upright (change from baseline) (follow-up 12 months; measured with: mmHg; Better indicated by lower values) 15 randomised trials serious1 no serious inconsistency no serious indirectness no serious imprecision none 21 15 MODERATE
DBP upright (change from baseline) (follow-up 12 months; measured with: mmHg; Better indicated by lower values) 15 randomised trials serious1 no serious inconsistency no serious indirectness no serious imprecision none 21 15 MODERATE
12.3.2.2
Economic evidence No relevant economic studies were included that compared different types of diuretic. Economic studies were considered relevant to the question if they compared one diuretic with another or examine the impact of cost and effectiveness differences between different diuretics on the overall decision about which drug to treat people with. Economic studies that included only one type of diuretic were not considered helpful to decision making and were excluded. In the absence of a published cost effectiveness analysis, current UK drugs costs were presented to the GDG to help inform decision making.
12.3.2.3
Evidence statements - Clinical Diuretics versus placebo or other anti-hypertensive drugs Table 75: Results of studies / meta-analysis
Class of diuretic Diuretic name Outcome measure and statistical significance (arm favoured) MI CV event SS (BDZ) Stroke Mortality CHD event NS HF ADL Studies / references
Diuretics versus placebo TDs BDZ SS (BDZ) NS MRC
Update 2011
TDLs
CTD
SS (CTD)
SS (CTD)
NS
SS (CTD)
SHEP, SHEP-P, VA-NHLBI SS (IND) HYVET, PATS MRC
IND
SS (IND)
SS (IND)
SS (IND)
SS (IND)
Diuretics versus other anti-hypertensive classes TDs BDZ vs BB NS SS (BDZ) NS NS
HCTZ vs ACEi
NS
NS
NS
NS
PHYLIIS, Sareli
HCTZ vs CCB
NS
NS
NS
NS
Sareli, MIDAS, THAI elderly ALLHAT, ANBP2
TDLs
CTD vs ACEi
SS (CTD)
SS (CTD)
NS
SS (CTD)
CTD vs CCB
NS
NS
NS
NS
NS
NS
ALLHAT, SHELL, VHAS
235
Hypertension (partial update) Pharmacological interventions Head to head comparisons NOTE: The results of the meta-analyses comparing IND vs HCTZ for SBP and DBP (supine and upright) should be interpreted with extreme caution due to the observed significant heterogeneity. This appears to be attributed to one of the RCTs494 which reports an effect size in the opposite direction to the other studies and because it has much smaller SDs than the other trials, it has therefore been weighted more highly. If this trial is removed from the MA then heterogeneity is reduced to more acceptable levels of 0% and the effect becomes NS. Removing the two lower quality trials (Plante, 1988 and Kreeft, 1984)339,493 from the analysis did not result in removing the observed heterogeneity. If a random effects model is applied to the pooled estimate, then the effect size also becomes NS.
Update 2011 Update 2011
NOTE: Some data were not provided in a usable format for inclusion in meta-analysis or were unable to be pooled; data from each of these studies has been summarised individually in Table 68 (and in the evidence profiles), along with pooled data where meta-analysis was possible.5,93,94,198,216,503 NOTE: all data given are for between-group differences
236
Table 76: Results of studies / meta-analysis

Diuretic name (interventi on) Diuretic name (comparis on) Outcome measure and statistical significance (arm favoured) Change from baseline Supine SBP DBP Upright SBP DBP Seated SBP NS DBP NS 24h ABPM SBP NS DBP End of follow-up Supine SBP DBP Upright SBP SBP Absolute change unclear method SBP SBP
93,198,216
Studies / references
Thiazide-like diuretic vs Thiazide diuretic CTD HCTZ NS (unclear BP method) SS (IND) NS SS (IND) NS
IND IND
HCTZ BDZ
NS
NS
NS
NS
SS* (IND) NS
SS* (IND) NS
SS* (IND) NS
SS* (IND) NS NS NS
94,194,195,339,49 3,494,551 39,76
Update 2011
Thiazide-like diuretic vs thiazide-like diuretic IND TD vs TD HCTZ BDZ NS NS NS NS

5
237
CTD
NS
NS
NS
NS
266,503
*significant heterogeneity. Hetereogenity is removed if the Plante 2003 trial494 is excluded from the analysis, and the overall effect becomes NS. If a random effects model is applied to the pooled estimate, then the effect size also becomes NS. NOTE: there were no studies found that compared: CTD vs BDZ IND vs BDZ
Update 2011
12.3.2.4
Evidence statements Health economic No evidence comparing the cost-effectiveness of different diuretics was identified. In terms of drug acquisition costs alone, in December 2010 based on BNF 60: bendroflumethiazide (2.5mg) cost 11.86 per year; chlortalidone (50mg i) cost 19.81 per year; indapamide (2.5mg nonproprietary) cost 16.03 per year.
12.4 Cost-effectiveness analysis

This model was developed as part of the 2006 pharmacological update (CG34) to balance clinical outcomes and to test the cost effectiveness of different classes of initial antihypertensive medications. As part of the 2011 update this analysis was rerun with updated costs. The relative risks for ARBs were also updated based on new ACEi vs ARB data. A summary of the analysis methods and results are provided below. Full methods and results including an overiew of the overall impact of the update compared to the previous analysis is available in Appendix I: Cost-effectiveness analysis pharmacological treatment.
Update 2011
12.4.1
12.4.1.1
Methodological introduction
Economic question The aim of the model was to estimate the cost effectiveness of the various blood pressure-lowering drug classes for the management of hypertension in primary care.
12.4.1.2
Population and subgroups The model considered patients with essential hypertension seen in primary care, excluding those with pre-existing cardiovascular disease (CVD), heart failure (HF) or diabetes. It was designed to be run separately for different cohorts, defined by age (55, 65, 75 and 85) and sex. In addition, the model classified these cohorts by baseline CVD risk (0.5%5% per year), by heart failure risk (05% per year) and by diabetes risk (05% per year). A base case analysis was performed for 65-year-old men and women with 2% CVD risk, 1% HF risk and 1.1% diabetes risk, and a sensitivity analysis considered the effect of varying these risk levels. The trial evidence that the model is based on included relatively few younger (under 55) or black people of African and Caribbean descent, so the results may not be reliable for these groups. However, we did conduct sensitivity analyses to explore how different assumptions about treatment effects might impact on the cost-effectiveness results for younger (45) and black people of African and Caribbean descent.
12.4.1.3
Interventions compared The analysis assessed the costs and effects of the various classes of blood pressure-lowering drugs alongside a 'do nothing' comparator. Inclusion of no treatment as an option is important for economic evaluations as it allows us to identify low-risk groups for whom treatment is not likely to be cost effective. The interventions compared were thus: no intervention (NI)
i Note that 25mg was considered the optimal dose but only 50mg tablets were listed in the BNF.
238
Hypertension (partial update) Pharmacological interventions thiazide-type diuretics (D) calcium-channel blockers (C) beta-blockers (B) ACEi/angiotensin-II receptor antagonists (ARBs) (A). At basecase, it was assumed that 80% of patients starting on ACEi would continue with these, but that 20% would switch to ARBs due to an inability to tolerate ACEi (expert opinion). ACEi/ARBs were combined as a strategy as they were considered to have equivalent effectiveness. The costs and effects of the drugs were weighted to take account of this. For simplicity only first-line drugs were considered. However, it should be noted that the relative treatment effects from the meta-analysis include additional benefits from various second and third line treatments offered in the trials. 12.4.1.4 Outcomes The treatment effects were measured in terms of prevention of CVD events (non-fatal unstable angina, MI, heart failure and stroke) and CVD-related deaths. The only adverse effects modelled were onset of HF and diabetes, although we did examine the possible impact of other adverse reactions to the drugs in sensitivity analyses. It should also be noted that the model does not explicitly include cost impacts of withdrawals, nonconcordance or transfers between treatments. The impact of such changes on effectiveness is implicitly included through the use of intention-to-treat trial data. Health outcomes for the cost-effectiveness analysis are summarised in the form of quality adjusted life-years (QALYs), where one QALY represents one year of healthy life. 12.4.1.5 Cost effectiveness The results of cost-effectiveness analysis are usually presented as incremental cost-effectiveness ratios (ICERs), which determine the additional cost of using one drug (X) per additional QALY gained, compared with no intervention or another drug (Y):
Where more than two interventions are being compared, the ICERs are calculated using the following process. The drugs are ranked in terms of cost, from the cheapest to the most expensive (cheapest indicated by LC (lowest cost) in the results table below). If a drug is more expensive and less effective than the previous one, then it is said to be ruled out by 'simple dominated' and is excluded from further analysis (indicated by - in the results table below). ICERs are then calculated for each drug compared with the next most expensive non-dominated option. If the ICER for a drug is higher than that of the next most effective strategy, then it is ruled out by 'extended dominance' (indicated by - in the results table below). ICERs are recalculated excluding any drugs subject to extended dominance (these ICERs are given in the results table below). It is important to bear in mind that comparison between the crude cost-effectiveness ratios for two drugs each compared with 'no intervention' can be highly misleading. To illustrate, the incremental cost of starting antihypertensive therapy with the cheapest drug is relatively low, while the
239
Hypertension (partial update) Pharmacological interventions incremental benefit is high, and thus the ICER is small. A more expensive but more effective drug may also appear to have a relatively small cost-effectiveness ratio when compared with 'no treatment'. However, the more expensive drug may have a larger ICER when it is compared with the cheaper drug the incremental cost of switching from the cheaper drug to the more expensive one may be quite large in relation to the incremental health gain. Nevertheless, the more expensive drug may still be a cost-effective alternative to the cheaper drug if its ICER is less than the maximum amount that we are prepared to pay for a QALY, which is considered to be around 20,000 to 30,000 for NICE decisions. In this situation the most cost-effective option is the more expensive drug, despite its larger ICER. However, if the ICER for the more expensive drug were to exceed the threshold of 20,000 to 30,000 per QALY, then it would not be cost effective and the cheaper option should be preferred.
12.4.2
12.4.2.1
Results of the health economic model

Base case results The base case results are presented in Table 3 for 65-year-old men and women with an annual CVD risk of 2%, HF risk of 1% and diabetes risk of 1.1%. This analysis suggests that antihypertensive treatment is cost effective for this population and that the most cost-effective initial drug in this group is calcium-channel blockers (C). The ICER of C compared with thiazide-type diuretics (D) is 1,520 to 1,960 per QALY gained, which is below the level usually considered to be affordable in the NHS (about 20,000 to 30,000 per QALY). Table 12.77:
Men Cost () D A C B NI Women Cost () D C A B NI 4,310 4,390 4,400 5,050 5,230 Effect (QALYs) 10.65 10.71 10.63 10.29 9.96 ICER (/QALY) LC 1,520 3,910 4,010 4,030 4,550 4,690 Effect (QALYs) 10.22 10.21 10.28 9.89 9.57 ICER (/QALY) LC 1,960 -
Base case results (65-year-old, 2% risk, 1.1% diabetes risk, 1% HF risk)
Update 2011
Beta-blockers (B) are ruled out by simple dominance, since D, A and C are estimated to be cheaper and more effective. This can be seen in Figure 1, since B lies to the northwest of D, A and C. The ACEi/ARB option (A) is also ruled out by extended dominance, since treating some patients with D and the remainder with C would be cheaper and more effective than A; in Figure 18, A lies to the northwest of a straight line joining points D and C. However, it should be noted that the absolute differences between A, C and D are small.
240
Figure 18: Base case results (65-year-old, 2% cardiovascular risk, 1.1% diabetes risk, 1% HF risk)
Men Women
5,400 5,200 5,000 4,400 4,800
Mean cost (2009 UK per person, discounted)
4,800 4,600
Update 2011
4,200 4,000 3,800 9.40
4,600
1,960
9.60 9.80 10.00 10.20 10.40
4,400 4,200 9.80
1,520
10.00 10.20 10.40 10.60 10.80
Mean effect (QALYs per person, discounted)

No Intervention Calcium-channel Blockers ACE Inhibitors/Angiotensin II receptor antagonists
Mean effect (QALYs per person, discounted)

Thiazide-type Diuretics Beta-blockers
QALYs = quality-adjusted life years
The results of this analysis are set out in more detail, together with the sensitivity analyses, in Appendix I: Cost-effectiveness analysis pharmacological treatment (updated 2011).
12.4.3
Conclusions
This analysis found that treating hypertension is highly cost-effective. Treatment resulted in improved health outcomes (higher QALYs) with all of the drug classes in the model and actually resulted in overall cost savings compared to no treatment as the reduction in cardiovascular events led to savings that offset the relatively low cost of antihypertensive medication; although it should be noted that this is based on low cost generic drugs. In most people CCBs were found to be the most cost-effective treatment option for initial treatment of essential hypertension. In terms of how the analysis has changed in 2011 since 2006, the most significant change in the model inputs in the 2011 update was the reduction in drugs costs; in particular the cost of CCBs, ACEs and ARBs. CCBs remained the most cost effective option, meaning no change from 2006 in the interpretation of the base-case result in terms of overall cost effectiveness. The ICER for CCBs did however reduce considerably (from 12,250 to 1,960) making CCBs more cost effective than they were in 2006. CCBs are also no longer the most expensive option, both B and NI being more expensive, meaning that CCBs are now cost saving compared to NI; this was not the case in the 2006 guideline. Another key difference is that the absolute difference between ACEs/ARBs, CCBs and TDs is now much smaller than it was in 2006 with BBs even less cost effective. The results of the subgroup analysis remain largely unchanged apart from that in both men and women, CCBs are cost effective a greater percentage of the time compared with TDs in higher CVD risk and older age groups; however this difference is not very large. Both old and new analyses show similar trends of cost effectiveness but the new analysis has ACE/ARB cost effective in fewer scenarios than before with the heart failure risk where this is the case moving to intermediate/high risk. The considerations that were highlighted in the 2006 guideline are still relevant and are described below. The trials on which the cost-effectiveness calculations are based did not, in general, show large differences in clinical outcomes between drug classes. Some of the outcomes have point estimates of effect that are not statistically significant. In these situations the point estimate is used as the best
Update 2011
241
Hypertension (partial update) Pharmacological interventions estimate of effect and so effects that are not statistically significant have a bearing on the relative cost effectiveness. Where the outcomes have a large effect on quality of life or cost (for example, stroke or death) the effect on overall cost effectiveness may be relatively important. The GDG considered the effect of this uncertainty about important outcomes in reaching their conclusions. The relative cost effectiveness of the agents also depends on the propensity of patients treated with them to develop new-onset diabetes or heart failure. The GDG were aware that both of these adverse outcomes should be treated with some caution in this context. It is not clear that an elevated blood glucose developing as a consequence of drug treatment has the same long-term health impact as in other circumstances, and the same applies to heart failure diagnoses, particularly since the definition of this outcome in some studies would not satisfy currently accepted criteria. The applicability of the model to people under the age of 55 is uncertain, since it is based on trial data from mostly older people. However, sensitivity analysis showed that the drugs that affect the renin-angiotensin system are likely to be the most cost-effective option in this group if they are even slightly more effective in the young than is suggested from the overall trial data. These results are sensitive to the cost of CCBs. The more expensive brands are not likely to be cost effective for use in the NHS. For example, the model estimates that for 65-year-olds at 2% annual CVD risk, 1.1% diabetes risk and 1% heart failure risk CCBs are only cost effective if they cost less than 94 per patient per year. Finally, it should be emphasised that there is still considerable uncertainty about the size of some treatment effects, which translates into uncertainty about the relative cost-effectiveness of the drugs. The evidence base is also difficult to interpret because of the complex nature of some of the treatment protocols and also because of differences in some of the trial populations.
12.5 Step two therapy

12.5.1.1 Clinical evidence The literature was reviewed from December 2005 onwards for systematic reviews and RCTs comparing A+C versus A+D for second-line treatment in adults with primary hypertension. RCTs were included if there was: 12 months follow-up, N200 and the population did not consist of people who were exclusively diabetic or had CKD.
Update 2011
One RCT296 was found that fulfilled the inclusion criteria and addressed the question, and was included in the review. The RCT296 (the ACCOMPLISH trial) compared treatment with the ACEi benazepril (20 then 40mg/day) + the CCB amlodipine (5 mg/day) vs. the ACEi benazepril (20 then 40mg/day) + the diuretic hydrochlorothiazide (12.5 mg/day) in N=11,506 people with hypertension, and had a follow-up time of 24 months. Treatment followed a dose-adjustment protocol for non-responders in each arm. NOTE: no quality of life data was found, or data assessing the effects of ACEi vs ARB in people aged 80+ or black people of African and Caribbean descent. The evidence profile below (Table 78) summarises the quality of the evidence and outcome data from the one RCT296 included in this review, comparing ACEi + CCB vs. ACE + D.
242
Table 78: ACEi + CCB versus ACEi +Diuretic for second line therapy quality assessment
Quality assessment No of studies Design Limitations Inconsistency Indirectness Imprecision No of patients A+D Other A+C considerations Mortality (all cause): ACCOMPLISH trial (follow-up mean 36 months) no serious indirectness serious
2
Summary of findings Effect Relative Absolute (95% CI) HR 0.90 (0.76 to 1.07) 4 fewer per 1000 (from 11 fewer to 3 more) 6 fewer per 1000 (from 0 fewer to 10 fewer) 4 fewer per 1000 (from 8 fewer to 2 more) 3 fewer per 1000 (from 5 fewer to 1 more) 9 fewer per 1000 (from 17 fewer to 0 more) 18 fewer per 1000 (from 5 fewer to 31 fewer)
Quality
randomised trials
none
236/5744 (4.1%)
262/5762 (4.5%)
MODERATE
MI (fatal and non-fatal): ACCOMPLISH trial (follow-up mean 36 months) 1 randomised trials no serious limitations1 no serious inconsistency no serious indirectness serious
3
none
125/5744 (2.2%)
159/5762 (2.8%)
HR 0.78 (0.62 to 0.99)4
MODERATE
Stroke (fatal and non-fatal): ACCOMPLISH trial (follow-up mean 36 months) 1 randomised trials no serious limitations1 no serious inconsistency no serious indirectness serious
2
none
112/5744 (1.9%)
133/5762 (2.3%)
HR 0.84 (0.65 to 1.08)
MODERATE
243
1 randomised trials no serious limitations1 1 randomised trials no serious limitations1 1 randomised trials no serious limitations
1 2
Hospitalisation for unstable angina: ACCOMPLISH trial (follow-up mean 36 months) no serious inconsistency no serious indirectness serious
2
none
44/5744 (0.8%)
59/5762 (1%)
HR 0.75 (0.5 to 1.1)
MODERATE
Coronary revascularisation: ACCOMPLISH trial (follow-up mean 36 months) no serious inconsistency no serious indirectness serious2 none 334/5744 (5.8%) 386/5762 (6.7%) HR 0.86 (0.74 to 1) MODERATE
Study drug withdrawal: ACCOMPLISH trial (follow-up mean 36 months) no serious inconsistency no serious indirectness serious
3
none
1684/5744 (29.3%)
1756/5762 (30.5%)
HR 0.93 (0.88 to 0.98)5
MODERATE
Random, double blind, allocation concealment, powered, ITT analysis. However no washout / run-in and <20% drop-outs (but Tx withdrawal was >30% for median 36 months follow-up). 95% confidence interval includes both 1) no effect and 2) appreciable benefit or appreciable harm 3 95% confidence interval includes both 1) appreciable benefit or harm and 2) non-appreciable benefit or harm 4 p=0.04; favours A+C 5 p=0.01; favours A+C
12.5.2.1
Economic evidence One study was identified in the update search that included A+C and A+D as comparators but was excluded due to being judged to have serious methodological limitations522.
12.5.2.2
Evidence statements - clinical ACEi + CCB was significantly better than ACEi + D for: MI (fatal and non-fatal) less study drug withdrawals mortality (all cause) stroke (fatal and non-fatal) hospitalisation for unstable angina coronary revascularisation new onset diabetes [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [moderate quality evidence]
Update 2011
There was NS difference between A+C and A+D for:
12.5.2.3
Evidence statements health economic No relevant cost-effectiveness evidence was identified.
244
12.6 Resistant hypertension

The GDG agreed to define the term resistant hypertension in the guideline as someone whose blood pressure is not controlled to <140/90mmHg, despite optimal or best tolerated doses of third line treatment. 12.6.1.1 Clinical evidence The literature was searched for all years (as this was not addressed in the previous guidelines)(Newcastle Guideline Development and Research Unit;National Collaborating Centre for Chronic Conditions) and all study types were included. Studies were included that compared 4th-line antihypertensive drugs with placebo,head to head comparisons or gave before-and after data, in people with resistant hypertension (defined as: people whose blood pressure remains uncontrolled, despite taking optimal doses of 3 anti-hypertensive drugs). Populations which were exclusively diabetic or had chronic kidney disease were excluded. Six cohort studies 126,163,226,347,383,511 were found which fulfilled the inclusion criteria and addressed the question, and were included in the review. The first cohort study 163 identifed and categorised people with resistant hypertension receiving treatment with spironolactone (true resistant hypertension), from people with controlled (white coat reisistant hypertension). For those with true resistant hypertension the study then compared data from before to after the introduction of spironolactone. The study had a total of N=236 participants and had a median follow-up time of 15 months. Treatment began with an initial dose of 25mg, and was titrated to 50-100mg/d as required. The second cohort study 347 assessed N=133 participants with resistant hypertension and measured their blood pressure before and after spironolactone 25-50mg/d, with a 3-month and 6-month follow up period. The third cohort study 383 compared two groups of people with hypertension (total of N=69 participants). Group A were untreated hypertensives and Group B were drawn from a hypertension clinic with treatment resistant hypertension. Group A was randomised to receive either spironolactone 50 mg/d or bendroflumethiazide 2.5 mg/d in a crossover design. All people in group B received 50mg/d of spironolactone. Group A received four weeks treatment, four weeks washout, four weeks treatment, and group B had a mean follow up time of 3.7 months. The fourth cohort study 226 assessed N=12 people with resistant hypertension before and after receiving spironolactone (25mg/d and force-titrated to 50mg/d at 4 weeks), and had a follow up time of eight weeks treatment. Other anti-hypertensive treatment was discontinued, if necessary for a low blood pressure. The fifth cohort study 126 reviewed participants with uncontrolled hypertension in the ASCOTBPLA open-label RCT. All participants N=1411 received an anti-hypertensive regimen based on either Atenolol or Amlodopine. The comparison was between those who were prescribed additional spironolactone vs. those who were not prescribed spironolactone. The median follow up time was 5.5 years. The sixth cohort study 511 compared Spironolactone with Doxazosin in N = 198 patients with resistant hypertension. There was no mean follow-up time reported. Participants were followed up until treatment was changed with the addition of a new drug/change in dosage to control blood pressure or when blood pressure was controlled within a pre-specified target.
Update 2011
No evidence profile was generated as GRADE was not performed in this guideline on observational studies. However GRADE automatically assigns a quality rating of low to observational studies.
245
Hypertension (partial update) Pharmacological interventions The table below (Table 79) summarises the quality of the evidence and the outcome data from the six cohort studies 126,163,226,347,383,511 included in this review of the effectiveness of 4th line antihypertensive treatment in resistant hypertension in adults. Table 79: Summary table of studies examining the role of fourth line antihypertensives in resistant hypertension
Study Rodilla et al. 2009{Rodill a, 2009 16014 /id} Mahmud et al. 2005{Mah mud, 2005 15968 /id} Chapman et al. 2007{Chap man, 2007 373 /id} Intervention Spironolactone Comparison Doxazosin Follow-up Until change of treatment/ target blood pressure maintained 3-4 months Results Spironolactone best (decreased home or ambulatory SBP and DBP) Spironolactone effective in reducing BP when used as a 4th line drug Addition of spironolactone effective in reducing BP Evidence Quality Low
Previously untreatedspironolactone/bendro flumethiazide
4th line Spironolacton e
Low
ASCOT trial patients an a-HT regimen based on either Atenolol or Amlodopine Plus addition of Spironolactone Spironolactone
ASCOT trial patients on aHT regimen based on either Atenolol or Amlodopine Before vs. after Spironolacton e
Median 5.5 years
Low
Update 2011
De Souza et al. 2010{de Souza F., 2010 15965 /id} Lane et al. 2007{Lane, 2007 802 /id} Gaddam et al. 2010{Gadd am, 2010 15967 /id}
12 months (Median 15 months, IQR 13-20 months)
Spironolactone effective in reducing office and ambulatory blood pressure. Spironolactone effective in reducing SBP and DBP Addition of spironolactone effective in reducing SBP and DBP
Low
Spironolactone
Before vs. after Spironolacton e Before vs. after Spironolacton e
6 months
Low
Spironolactone
8 weeks
Low
12.6.1.2
Economic evidence No relevant economic studies were identified that examined drugs in patients with resistant hypertension. In the absence of a published cost effectiveness analysis, current UK drugs costs for agents that might be considered for use in resistant hypertension were presented to the GDG to help inform decision making.
12.6.1.3
Evidence statements clinical Six studies found that blood pressure was reduced in people with resistant hypertension who were treated with 4th-line spironolactone.
246
Hypertension (partial update) Pharmacological interventions One study 511 found that 4th line therapy with spironolactone was better than doxazosin for reduction in SBP and DBP [low quality] Three studies163,347 226 found that SBP and DBP was reduced after 4th line spironolactone treatment (vs. before treatment). [low quality]. One study 383 found BP reduced in those treated with spironolactone compared with those previously untreated and reported drop out rates of 10% due to adverse effects [low quality]. One study 126 found the addition of spironolactone (as 4th line therapy) was effective in reducing BP, and an adverse event rate of 13% was reported [low quality].Evidence statements health economic 12.6.1.4 Evidence statements economic No relevant cost-effectiveness evidence was identified. In terms of drug acquisition costs alone, in December 2010 based on BNF 60: spironolactone (25mg) cost 23.73 per year.
Update 2011Update 2011
247
12.7 Special groups for consideration

12.7.1 People aged over 80 years
See section 9 on page 115.
12.7.2
Younger people
Outcomes in younger patients The literature search found no evidence for the clinical outcomes summarised above, therefore 164,177,394 blood pressure response to drug therapy was used as a surrogate. Three studies and an age55 stratified analysis from a fourth study compared blood pressure response across various drug classes and identified ACE inhibitors and beta-blockers as more effective at lowering blood pressure in younger people, when compared to calcium channel-blockers or thiazide-type diuretics. In older people, initial treatment with calcium channel-blockers or thiazide-type diuretics has been shown to be more effective at blood pressure lowering than ACE inhibitors, angiotensin-II receptor 157,312,589-591 antagonists or beta-blockers .
12.7.3
Ethnicity
There are ethnic differences in the prevalence of high blood pressure. In African American patients, the prevalence of hypertension and mortality arising from complications such as cardiovascular, cerebrovascular and renal disease is higher than other ethnic groups. 40,110,127,145,542 Mortality data from England and Wales (198892) shows similar trends, with mortality due to hypertensive complications 3.5 times higher than the national average in the African-Caribbean population. 504 British Asians also exhibit hypertension associated mortality rates 1.5 times higher than the national average. 504 The Whitehall II Study investigated a cohort of London-based civil servants aged 3556 years, between 1985 and 1988.638 A 73% response rate provided a cohort including 8,973 white participants, 577 of South Asian origin and 360 of African-Caribbean origin. Participants were considered hypertensive if they had blood pressure above 160/95 mmHg or were receiving antihypertensive drugs. African-Caribbean (odds ratio: 4.0; 95%CI: 2.8 to 5.7) and South Asian (odds ratio: 2.3; 95%CI: 1.6 to 3.3) participants had a greater prevalence of hypertension than white participants, after findings were adjusted for age, service grade, sex and body mass index. Similarly, diabetes was more common in African-Caribbean (unadjusted odds ratio: 2.8; 95%CI: 1.7 to 4.6) and South Asian (unadjusted odds ratio: 4.2; 95%CI: 3.0 to 5.8) participants. Although both ethnic groups had lower total cholesterol scores that white participants, South Asian people tended to have a poorer lipid profile while African-Caribbean people tended to have a more favourable one. A study conducted in nine practices in South London interviewed men and women aged 4059 years of white, African and South Asian origin.116 Random samples of each group were invited: 64% took some part in the study, although only about one half of these contributed blood pressure data. As with the Whitehall study, individuals were considered hypertensive if they had blood pressure above 160/95 mmHg or were receiving antihypertensive drugs. Age and sex adjusted prevalence ratios for hypertension were 2.6 (95% CI: 2.1 to 3.2) in people of African descent and 1.8 (95% CI: 1.4 to 2.3) in those of South Asian descent. Diabetes prevalence ratios were 2.7 (95% CI: 1.4 to 2.3) and 3.8 (95% CI: 2.6 to 5.6) for those of African and South Asian descent respectively. Differences in ethnic groups (West African vs. Caribbean and Hindu vs. Muslim) were not statistically significant. Similarly to the Whitehall study, people from these ethnic minority groups had lower total cholesterol scores than white participants although a lipid profile was not attempted.
248
Hypertension (partial update) Pharmacological interventions A number of other studies of local populations have explored the relationship between ethnicity and cardiovascular risk factors. These studies raise methodological issues and do not provide a useful picture of hypertension because they did not seek to adjust for treatment. They demonstrate that varying patterns of risk factors may occur in different groups, although these may only be well understood with more definitive epidemiological research. A study comparing South Asian and European participants in Newcastle upon Tyne found that Bangladeshi participants had the poorest lipid profile while Indians had the best, similar to a European profile.74,286 The age-adjusted prevalence of diabetes varied between Bangladeshi (23%), Pakistani (23%), Indian (13%) and European (4%) participants. A London based study drawing from factory worker and general practice populations confirmed the findings of the Whitehall II study, showing similar trends in lipid profile comparing European, South Asian and African-Caribbean participants. 400 Similarly a raised ageadjusted prevalence of diabetes was seen in Sikh (20%), Punjabi Hindu (19%), Gujarati Hindu (20%) and Muslim (19%) groups compared to white participants (5%). A survey of Bangladeshi participants in East London found a poor lipid profile and raised prevalence of diabetes compared to a non-Asian population.399 The evidence thus shows that hypertension and diabetes are more common among certain ethnic groups in the UK. This greater prevalence of hypertension may lead to higher rates of cardiovascular disease and target organ damage.145,230,236,252,409,542 Reasons for this greater prevalence may be environmental as well as physiological. A trend towards increased blood pressure and weight was observed with increasing urbanisation of rural black Africans496, and with the migration of Punjabi participants from India to England.73 12.7.3.1 Clinical evidence The literature was reviewed from December 2005 onwards (the cut-off date of the previous guideline, CG34,425 where this was covered previously) for systematic reviews, RCTs, sub-group analyses of RCTs and cohort studies looking at first-line anti-hypertensive treatment of black people of African or Caribbean descent who have primary hypertension. Studies were included if there was: N1000 and the population did not consist of people who were exclusively diabetic or had CKD. Two subgroup analyses354,492 of an RCT (ALLHAT) were found which fulfilled the inclusion criteria and addressed the question, and were included in the review. The ALLHAT study was originally included in the previous NICE guidelines.425,441 ALLHAT compared ACEi vs TD vs. CCB vs. alpha-blocker and 1/3 of the population were black people (NOTE: the term black was that used in the ALLHAT trial). However, the studies included in the previous guidelines did not give data for the ACEi vs. CCB arms in black people and did not give the incidences of angioedema, which these newer subgroup analyses have looked at. Both the subgroup analyses were planned a-priori as part of the design of the ALLHAT trial. The first subgroup analysis of the ALLHAT RCT492 assessed the incidence of angioedema in people treated within each arm of trial (ACEi vs. TD vs. CCB vs. alpha-blocker) and the incidence of the outcome in different subgroups of people (including different ethnic groups: black people vs. nonblack people). The study follow-up time was mean 4.9 years and the number of people who developed angioedema was N=53 out of the total study group of N=42,418. Because the data we are interested in is the incidence of agioedema in black people vs. non-black people (ie. has come from the subgroup analysis), this study data has been classed as observational (see section below entitled evidence profile). The second sub-group analysis of the ALLHAT RCT354 assessed the incidence of clinical endpoints that occurred in subgroups of patients, including black people vs. non-black people who were randomised to the ACEi and CCB arms of the ALLHAT trial. The study follow-up time was mean 4.9 years and the number of people who developed angioedema was N=53 out of the total study group of N=42,418. This study has been classified as observational because it is a subgroup analysis of an RCT.
Update 2011
249
Hypertension (partial update) Pharmacological interventions The evidence profiles below (Figure 1 and Figure 2) summarises the quality of the evidence and outcome data from the two RCT (ALLHAT) subgroup analyses354,492 included in this review, comparing outcomes in black people and non-black people. Where data was unable to be put into GRADE, it has been written up narratively in the evidence statements.
Update 2011
250
Table 80: Evidence profile comparing ACEi versus other antihypertensive classes (TD, CCB or alpha) in black people and non-black people (data from Piller et al., 2006)492
Quality assessment Summary of findings No of patients Other considerations other a-HT classes (TD, CCB or alpha) Relative (95% CI) Effect Quality
No of studies
Design
Limitations
Inconsistency
Indirectness
Imprecision
ACEi
Absolute
Angioedema (black people) out of total randomised (follow-up mean 4.9 years) randomised trials no serious limitations no serious inconsistency no serious indirectness no serious imprecision 23/3210 (0.7%) 6/10196 (0.1%) RR 12.18 (4.96 to 29.88) 7 more per 1000 (from 2 more to 17 more)
none
HIGH
Angioedema (non-black people) out of total randomised (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness serious2 23/3210 (0.7%) 6/10196 (0.1%) RR 0 (2.47 to 0)3 1 fewer per 1000 (from 1 more to 1 fewer)
251
1 2
none
MODERATE
Angioedema (black people) out of those who developed angioedema (follow-up mean 4.9 years) randomised trials no serious inconsistency no serious indirectness no serious imprecision 23/37 (62.2%) inappropriate to calculate (loss of randomisation) 375 fewer per 1000 (from 375 fewer to 375 fewer)
serious
none
6/16 (37.5%)
MODERATE
Angioedema (non-black people) out of those who developed angioedema (follow-up mean 4.9 years) randomised trials no serious inconsistency no serious indirectness no serious imprecision 14/37 (37.8%) inappropriate to calculate (loss of randomisation) 625 fewer per 1000 (from 625 fewer to 625 fewer)
serious4
none
10/16 (62.5%)
MODERATE
Subgroup analysis of RCT: but pre-specified and the trial deliberately recruited a specific number of black people to be able to do this analysis 95% confidence interval excludes no effect, but the CI includes appreciable benefit and non-appreciable benefit or appreciable harm and non-appreciable harm 3 SS - favours other a-HT classes (p<0.0001) 4 Loss of randomisation in groups (incidence of angioedema in black people and non-black people, out of those who developed angioedema in the trial, rather than all participants randomised in the trial)
Table 81: Evidence profile comparing ACEi vs CCB in black people and non-black people (data from Leenan et al., 2006)354 NOTE: there was not enough data given in the study to calculate the HRs for these outcomes, so the RRs reported in the paper have been used in the GRADE profile
Quality assessment No of studies Other considerations Summary of findings No of patients Imprecision ACEi CCB Relative (95% CI) Effect Absolute Quality
Design
Limitations
Inconsistency
Indirectness
CHD (black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness no serious imprecision data not given in study 1.09 (0.92, 1.03) not enough data given in study to calculate
none
HIGH
CHD (non-black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness data not given in study 0.97 (0.86, 1.10) not enough data given in study to calculate
serious
252
none
MODERATE
Stroke (black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness data not given in study 1.51 (1.22, 1.86) 5 not enough data given in study to calculate
serious
none
MODERATE
Stroke (non-black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness data not given in study 1.07 (0.89, 1.28) not enough data given in study to calculate
very serious
none
LOW
Combined CVD (black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness data not given in study 1.13 (1.02, 1.24)5 not enough data given in study to calculate
serious3
none
MODERATE
Combined CVD (non-black people) (follow-up mean 4.9 years)
randomised trials
serious
none
1.03 (0.96, 1.10)
not enough data given in study to calculate
MODERATE
Heart Failure (black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness data not given in study 0.89 (0.75, 1.06) not enough data given in study to calculate
serious2
none
MODERATE
Heart Failure (non-black people) (follow-up mean 4.9 years) randomised trials no serious limitations1 no serious inconsistency no serious indirectness data not given in study 0.85 (0.75, 0.97)6 not enough data given in study to calculate
serious
none
MODERATE
1 2
Subgroup analysis of RCT: but pre-spcified and the trial deliberately recruited a specific number of black people to be able to do this anlysis 95% confidence interval includes both 1) no effect and 2) appreciable benefit or appreciable harm 3 95% confidence interval excludes no effect, but the CI includes appreciable benefit and non-appreciable benefit or appreciable harm and non-appreciable harm 4 95% confidence interval crosses both 1) no effect and 2) appreciable benefit or harm and non-appreciable benefit or harm 5 SS - favours CCB (p-value not given) 6 SS - favours ACEi (p-value not given)
253
12.7.3.2
Economic evidence No relevant economic studies were identified.
12.7.3.3
Evidence statements One RCT (subgroup analysis)492 found that: Over half (55%) of people who developed angioedema were black people The incidence of angioedema (out of all the people who developed angioedema in the trial) was: o in black people: higher in the ACEi group versus other a-HT classes (TD, CCB or alpha) combined (62% vs. 38%) o in non-black people: lower in the ACEi group versus other a-HT classes (TD, CCB or alpha) combined (38% vs. 63%) [moderate quality evidence] The risk of angioedema in both black people and non-black people was: significantly higher in the ACEi group vs. other a-HT classes (TD, CCB or alpha) combined (as a proportion of the total randomised, see the forest plot in section H.1.4 )
Update 2011
[high and moderate quality evidence] One RCT (subgroup analysis)354 found that: In black people: CCB was significantly better than ACEi for risk of: Combined CVD Stroke CHD HF In non-black people: ACEi was significantly better than CCB for risk of: HF CHD Combined CVD Stroke [moderate quality evidence] [moderate quality evidence] [moderate quality evidence] [low quality evidence] There was NS difference between ACEi and CCB for risk of: [moderate quality evidence] [moderate quality evidence] [high quality evidence] [moderate quality evidence]
There was NS difference between ACEi and CCB for risk of:
No relevant cost-effectiveness evidence was identified.
12.7.4
Chronic kidney disease

For guidance pertaining to people with hypertension and chronic kidney disease refer to NICE Clinical Guideline 73.
254
12.7.5
Type 1 and Type 2 diabetes

For guidance pertaining to people with hypertension and Type 1 diabetes refer to NICE Clinical Guideline 15. For guidance pertaining to people with hypertension and Type 2 diabetes refer to NICE Clinical Guideline 66.
12.7.6
Women who are pregnant or breast-feeding

For guidance on women who are pregnant or breast-feeding, refer to NICE Clinical Guideline 107 http://guidance.nice.org.uk/CG107.
12.8 Stopping treatment

If a patient's blood pressure has been reduced to normal levels by antihypertensive drugs, both patient and doctor may want to know if medication can safely be stopped. Unnecessary drug treatment may put the patient at risk of adverse side effects and is a cost to society. Some patients may be at risk of serious cardiovascular events if they stop taking antihypertensive drugs. It would be useful to be able to identify patients who are likely to be able to stop medication without serious consequences. In studies which have reported on withdrawal of antihypertensive medication240,349,411,561,631, 421, , between 10%433 and 60%349 of patients remained normotensive for at least a year, although studies reporting better success rates were often of highly selected patient populations. Further, the definition of normotension varied between studies, from blood pressure less than 140/85mmHg38 to diastolic blood pressure less than 105mmHg411 and the characteristics of the patients varied, e.g. mean age ranged from 519,411 to 67 years631, baseline blood pressure ranged from 126/80 mmHg240,349 to 152/101mmHg359, number of drugs ranged from one9,201,561,631 to three or more349.
9,38,201,359,413,433,435,582,597
There is consistent evidence, from a systematic review of 5,479 patients who stopped taking antihypertensive medication and who were followed up for at least a year434, and from a subsequent study of 503 patients who were also followed up for a year435, that patients are more likely to remain normotensive if they are younger, have lower blood pressure and have been treated with only one drug. Two studies, of 1,478 patients aged 6084 years, found that on-treatment systolic blood pressure was the best measure of blood pressure to use in predicting success201,435. We identified three randomised controlled trials of interventions - weight loss and restriction of salt and alcohol - which might help patients to successfully stop taking anti-hypertensive medication 349,561,631 . The TONE631 and DISH349 studies were similar: they both evaluated the effects of a weight loss diet and restriction of salt; both randomised obese and non-obese patients independently; both had weekly group counselling sessions during the initial intensive phase of the intervention, followed by less frequent group sessions and individualised counselling during the later maintenance phase; patients in both studies had good blood pressure control (mean baseline blood pressure 129/72 mmHg in TONE and 127/80 mmHg in DISH). The TONE study enrolled patients who had been taking only one antihypertensive drug or a combination of a diuretic and a non-diuretic for a mean duration of 11.7 years. The DISH study enrolled patients who had been on treatment for at least 5 years and included some who were taking three or more antihypertensive drugs. The definitions of normotension - less than 150/90 mmHg in TONE and diastolic blood pressure less than 95 mmHg in DISH - might now be considered high. Meta-analysis of the results of these trials showed that obese patients who were put on a diet to lose weight were more likely to be successful in stopping medication than those who were not (RR = 1.6, 95%CI: 1.4 2.0). Likewise, patients who were encouraged to restrict their salt intake were more likely to remain normotensive (RR=1.4, 95%CI: 1.2
255
Hypertension (partial update) Pharmacological interventions 1.7), with little difference between obese and non-obese patients (see Figure 19). The smaller study by Stamler et al. compared the effects of a multiple intervention, which encouraged loss of weight and restriction of salt and alcohol, with no intervention to support drug withdrawal; it defined normotension as diastolic blood pressure less than 90 mmHg561. This study was combined in a metaanalysis with a similar comparison of two arms of the TONE study of obese patients: a comparison of the combination of weight loss and salt restriction with no intervention. Patients who received a multi-factorial intervention were more likely to successfully stop medication than those who were not (RR = 2.8, 95%CI: 1.9 4.0) and these interventions appeared to be more successful than those which addressed only diet or only salt restriction (see Figure 31). Combining all groups in these three studies349,561,631, 42% of patients who received interventions remained normotensive for at least a year, compared to only 25% in the control groups. This is consistent with the evidence (see Lifestyle interventions) that a healthy diet and reduced salt intake can lower blood pressure. Figure 19: Meta-analysis of RCTs of lifestyle interventions to support withdrawal of antihypertensive drugs
We found little evidence about whether patients became more likely to suffer severe cardiovascular events if antihypertensive medication was withdrawn. One study monitored cardiovascular events for 1232 (average 24) months after withdrawal of medication from 975 patients who had a mean blood pressure of 129/72 mmHg while on one antihypertensive medication336. It found no difference between the rate of cardiovascular events before and after withdrawal of medication, though the statistical power to detect a difference was low, largely because of the short period of monitoring while on medication. The best evidence on the possible effects of drug withdrawal is the epidemiological evidence from over a million adults, that any increase in blood pressure is associated with an increased risk of death from cardiovascular disease361. If patients become hypertensive after stopping drugs, this is most likely to happen in the first six months, although it can happen later434. To avoid this, patients should be carefully followed up and drugs should be withdrawn gradually following manufacturers' guidance.
256
12.9 Link from evidence to recommendations- Pharmacological treatment of hypertension

The pharmacological update of this guideline in 2006 recommended a stepped care approach to treatment. The recommendation for initial treatment (step 1) was stratified by age and ethnicity reflecting data from clinical trials showing differential effects of the different classes of blood pressure lowering drugs on blood pressure lowering and clinical outcomes in younger (<55years) versus older people and in black people of African and Caribbean descent. Antihypertensive therapies were designated A drugs (ACEi or ARBs), C drugs (calcium channel blockers) and D drugs (thiazide-type diuretics). The recommendation for step 1 treatment for younger people was an A drug. At that timethe GDG felt that the benefit from ACEi and ARBs were closely correlated (although lacked head to head evidence) and that they should be treated as equal in terms of efficacy; however, due to cost differences, felt ACE inhibitors should be initiated first and an ARB considered an alternative for when an ACEi was poorly tolerated, usually due to an ACE-inhibitorinduced cough. ACE-inhibitors versus ARBs for step 1 treatment: For this update, the GDG considered evidence from 3 RCTS published since December 2005 comparing ACEi versus ARB for step 1 treatment for adults with primary hypertension. The first RCT653 (the ONTARGET trial) compared treatment with the ACEi ramipril (10 mg/day) versus the ARB telmisartan (80 mg/day) and versus a combination of the two (ACEi+ARB) in 25,620 people considered to be at high cardiovascular disease risk. Many (approximately 70%), but not all of these patients had treated hypertension. The study had a median follow-up time of 56 months. A second RCT587 compared treatment with the ACEi enalapril (20 mg/day) versus the ARB losartan (50 mg/day) in N=560 people with hypertension, for a follow-up time of 24 months. The third study552 (CORD IB trial) compared treatment with the ACEi ramipril (5 mg/day) versus the ARB losartan (50 mg/day) in N=3860 people with hypertension, and had a follow-up time of 12 months. The evidence showed no significant differences between ACEi and ARBs on major clinical outcomes including death, cardiovascular events, stroke and diabetes. There was no consistent trend favouring one drug class over the other. Study drug withdrawal was significantly lower with ARB compared with ACEi. The GDG considered that this most likely reflected better tolerability of the ARB as ACEis are known to cause cough in some patients whereas ARBs do not. There was heterogeneity in the analysis for this latter finding but the lower withdrawal from ARB therapy was a robust finding in the largest trial (ONTARGET). Moreover, the GDG noted that there was an eight week run-in to ONTARGET when patients were prescribed the ACEi to see if they could tolerate the drug, thus, pre-selecting a group with short-term tolerability of the drugs. The results are therefore likely to underestimate the true withdrawal rate from ACEi. The GDG noted that side-effects of a drug are an important consideration in making treatment decisions for the management of a symptomless condition. The ONTARGET study also compared the combination of ACEi + ARB versus ACEi alone and found that there was no advantage of the ACEi + ARB combination on clinical outcomes and a more adverse effects associated with the combination of ACEi + ARB. The GDG concluded that there was no evidence to support the use of ACEi + ARB for the treatment of hypertension and that this combination should not be used for the treatment of primary hypertension. The largest study in the analysis comparing ACEi versus ARB was ONTARGET and the GDG discussed the fact that this study was not a trial designed to specifically examine the treatment of hypertension with initial therapy, but rather looked at the use of an ACEi or ARB for prevention of cardiovascular events. In this regard, the participants in ONTARGET were selected to be at high cardiovascular risk, although 70% of patients in ONTARGET had a history of hypertension and were receiving antihypertensive therapy/s or had discontinued their treatment prior to randomisation to the study drugs. The GDG debated whether ACEi and ARBs could be considered equivalent, based on data
Update 2011 Update 2011
257
Hypertension (partial update) Pharmacological interventions primarily from one large study that was not specifically a hypertensive population. It was noted that ONTARGET was designed to test non-inferiority of the ARB versus the most commonly used ACEi (Ramipril) with regard to clinical outcomes and that further large trials addressing the same question are unlikely to happen - this may, therefore, be the best evidence ever available for a hypertensive population. It was reassuring that the other studies in the analysis, albeit much smaller but studying a more typical hypertensive population, were consistent with the findings of ONTARGET. No relevant cost effectiveness analyses comparing ACEi versus ARBs were identified. However, the difference between the lowest cost ARB and the lowest cost ACEi has reduced considerably due to the recent availability of generic losartan; generic losartan (100mg) is now only about 5 more per year than generic ramipril (10mg). Patent expiry is imminent for many other ARBs too and the GDG considered it likely that the cost of ACEi and ARBs are likely to become similar over the lifetime of this guideline update. The ethnicity of participants was not reported for all of the trials but the GDG did not consider this prevented extrapolation of the findings to a UK population. Finally, the GDG could not identify any quality of life data comparing ACEi versus ARBs. The GDG concluded that the drug classes ACE iand ARBs should be considered equivalent with regard to their effect on clinical outcomes and recommended that people aged <55 years should be offered step one treatment with an ACEi or a low cost ARB. For patients intolerant of ACEi, an ARB should be offered. The GDG also recommended that an ACEi and an ARB should not be combined for the treatment of hypertension. The GDG noted that in women aged <55years and of child bearing potential, the use of ACEi or ARB has been reported to increase the risk of foetal malformation if taken during pregnancy. Women taking these medications should be advised that if they become pregnant, they should discontinue treatment and inform their doctor. In women planning conception, ACEi and ARBs should be avoided during this time and alternative treatments considered if required see clinical Clinical Guideline 97 on Hypertension in Pregnancy. Choice of thiazide-type diuretic therapy for hypertension: The 2006 pharmacological update recommended thiazide-type diuretics as a step 1 treatment option for people aged 55 years or black people of African and Caribbean descent of any age the other step 1 option for this group of people being a CCB. There are many different drugs labelled as thiazide-type diuretics. The predominant thiazide-type diuretic used in the UK for the treatment of hypertension is low dose (2.5mg o.d.) bendroflumethiazide (BFZ). This is somewhat unusual because this thiazide-type diuretic is rarely used anywhere else in the world as the preferred diuretic for the treatment of hypertension. This may be unimportant if the clinical outcomes data with low dose BFZ is equivalent to that with the other, more commonly used thiazide-type diuretics elsewhere in the world. This issue of comparability of different thiazide-type diuretics has been brought into sharper focus by recognition of the fact that, although often grouped together as thiazide-type diuretics, from a pharmacological perspective, there are two broad groups; i) classical thiazide diuretics (e.g. BFZ and hydrochlorthiazide; HCTZ) i.e. the name ends in thiazide, and ii) thiazide-like diuretics (e.g. chlorthalidone; CTD and indapamide; IND). The thiazide-like diuretics retain the main action of thiazide diuretics, i.e. inhibition of the sodium chloride co-transporter in the distal nephrons of the kidney. However, the thiazide and thiazide-like drugs have differential effects on other enzyme effects in the kidney, e.g. carbonic anhydrase inhibition, which can differ by up to 10,000-fold. Differential effects on platelet aggregation and regulation of angiogenesis have also been reported. The relevance of these actions beyond the characteristic thiazide action of inhibition of the sodium chloride cotransporter with regard to blood pressure control and the prevention of clinical outcomes is unknown. Nevertheless, the GDG considered it important to examine the evidence base supporting the use of classical thiazides (BFZ or HCTZ) when compared to the thiazide-like diuretics such as CTD and IND.
Update 2011
258
Hypertension (partial update) Pharmacological interventions Another important element of the data review for thiazide-type diuretic therapy was to examine the doses of diuretics used in the various clinical outcome trials. The trials evaluating clinical outcomes with thiazide-type diuretics have usually been evaluated by grouping all of these various drugs used at various doses altogether. The early diuretic trials used much higher doses than commonly used today. The reduction in dose to what is now known as low dose diuretic therapy resulted from concern about the development of electrolyte disturbances (usually hypokalaemia) and metabolic disturbances (hyperglycaemia) with higher dose diuretic therapy. Consequently, the GDG reviewed the important question as to what is the most clinically and cost effective thiazide-type diuretic for the treatment of adults with primary hypertension? The analysis examined data for the four most commonly used thiazide-type diuretics; i) classical thiazide diuretics (e.g. Bendroflumethiazide (BDZ) and hydrochlorthiazide(HCTZ), and ii) thiazide-like diuretics (e.g. chlorthalidone (CTD) and indapamide (IND). The analysis was complex and the GDG noted that there were no direct comparisons between the different diuretics with regard to clinical outcomes. Where head-to-head comparisons had been undertaken, they were usually based on blood pressure changes as the main outcome. These studies were often of short duration and too small to provide robust data. The GDG considered all of them to be underpowered to detect a significant blood pressure difference between diuretic treatments. There was also considerable variation in the doses of diuretics used in the various studies some early studies using four times the doses used routinely in todays clinical practice making it impossible to pool data for analysis. Consequently, the GDG found it difficult to reach firm conclusions regarding the comparative efficacy of different thiazide-type diuretics with regard to blood pressure lowering. The GDG then reviewed the clinical outcome studies with thiazide-type diuretics and found no direct comparator studies between different diuretics. Furthermore, interpretation of data from head-tohead trials comparing diuretics with placebo or other antihypertensive drugs was complicated by the markedly different diuretic doses used across studies. The GDG noted that the data demonstrating benefits of BFZ on clinical outcomes came from older studies (MRC) in which the dose of BFZ (10mg o.d.) was four times the usual dose of BFZ i.e. 2.5mg o.d., used in clinical practice today. The GDG also noted that there was no study evaluating and confirming the benefit of low dose BFZ on clinical outcomes the only data coming from older studies with much higher doses of BFZ, i.e. 10mg od. This concerned the GDG, mindful of the fact that low dose BFZ (2.5mg o.d.) has been the preferred thiazide-type diuretic for the treatment of hypertension in the UK. The GDG also noted that there was limited evidence confirming benefit of initial therapy on clinical outcomes with low doses of hydrochlorthiazide (12.5-25mg o.d.), the other commonly used thiazide-type diuretic world-wide. The GDG next discussed the evidence for the thiazide-like diuretics, i.e. IND or CTD and noted that the there was evidence showing benefits of low dose IND or low dose CTD on a range of clinical outcomes. The GDG noted that the evidence for IND and CTD was derived from more contemporary studies that had more consistently used lower doses across studies, typically; IND 1.5mg SR or 2.5mg o.d., or CTD 12.5mg or 25mg o.d. Some of the IND studies used an SR formulation, others did not. The GDG concluded that the consistency of the data suggested that the SR formulation was unlikely to have influenced the clinical outcomes in studies with IND. No relevant cost-effectiveness studies were found that compared different types of diuretic. Current UK drugs costs were considered by GDG and it was noted that the aforementioned thiazide-type diuretics were all available as generics. Considering all of the data cited above, the GDG were concerned that there was no evidence confirming a beneficial effect of low dose bendroflumethiazide, i.e. 2.5mg o.d., on clinical outcomes in people with hypertension. This observation is important because bendroflumethiazide 2.5mg od. is the most commonly used thiazide-type diuretic for the treatment of hypertension in the U.K. This does not mean that bendroflumethiazide 2.5mg o.d. is ineffective but it does make it difficult to assess whether it is as effective at preventing clinical outcomes as other thiazide-like diuretics, e.g.
Update 2011
259
Hypertension (partial update) Pharmacological interventions chlortalidone and indapamide for which evidence confirming benefits on clinical outcomes does exist. Having undertaken this analysis it was difficult for the GDG to recommend treatment with low dose thiazide-type diuretics, e.g. bendroflumethiazide or hydrochlorthoazide for which there was no evidence of a benefit on clinical outcomes. Consequently, the GDG recommended that when thiazide-type diuretics are used for the treatment for primary hypertension, thiazide-like diuretics, e.g. chlortalidone (12.5mg -25mg od) or indapamide (1.5mg SR or 2.5mg o.d.) should be preferred to conventional thiazide diuretics, e.g. bendroflumethiazide or hydrochlorthiazide. The GDG did not consider it necessary to recommend that those people already treated with low dose BFZ and in whom blood pressure is controlled, should be switched to CTD or IND. However, when new diuretic therapy was to be initiated, then CTD or IND should be preferred. The cost-effectiveness of pharmacological treatment of hypertension: As part of the 2006 pharmacological update of this guideline (CG34), the cost effectiveness of different classes of antihypertensive medications as initial therapy for hypertension was evaluated. The analysis assessed the costs and effects of the major antihypertensive drug classes; (A), i.e. ACE-I / ARB, (B) beta blockers, (C) CCBs and (D) thiazide-type diuretics. No intervention (NI) was also included as a comparator. Details of this analysis are shown in appendix x. Since 2006 the cost of antihypertensive drugs has decreased; in particular the cost of CCBs and ARBs. The GDG decided that it would be informative to rerun the cost-effectiveness analysis as part of the 2011 update with updated costs. The base case analysis modelled the results for 65-year-old men and women with 2% CVD risk, 1% HF risk and 1.1% diabetes risk. Sensitivity analysis undertaken in 2006 were also rerun to evaluate whether and how the results varied by age, sex, and by varying the risks of CVD, HF and diabetes. The GDG noted that the clinical trial evidence on which the model is based included relatively few younger (under 55) people, so speculative sensitivity analyses were conducted to explore how different assumptions about treatment effects might impact on the costeffectiveness results for younger (under 45) people. The top line conclusion from this analysis is that treating hypertension is highly cost-effective. Treatment resulted in improved health outcomes (higher QALYs) and remarkably, with most of the drug classes in the model, actually resulted in overall cost savings when compared to no treatment. This cost saving is due to the fact that the reduction in cardiovascular events led to savings that offset the relatively low cost of antihypertensive medication. The GDG noted that this conclusion is based on the use of low cost generic drugs. Another important conclusion is that for most people, CCBs were found to be the most cost-effective treatment option for initial treatment of primary hypertension. Indeed, unlike the analysis in 2006, CCBs are now cost saving when compared to no intervention. The GDG noted another key difference from the 2006 analysis is that the absolute difference in costs between ACE/ARB, CCBs and thiazide-type diuretics is now much smaller than it was in 2006. The difference is QALYs between these drugs is also fairly small. Just as in 2006, beta-blockers are ruled out by simple dominance, however now all other treatments are estimated to be both cheaper and more effective further justifying the decision not to recommend beta-blockers as a preferred initial therapy for primary hypertension. The GDG then reviewed the cost-effectiveness analysis in various sub-groups and noted that when compared to the 2006 analysis, CCBs are most cost effective in a greater number of scenarios. The GDG noted that the sub-group analysis of cost-effectiveness was particularly sensitive to the relative effects of drug therapy on the prevention of diabetes and heart failure. The model predicts that for people at low to intermediate risk of heart failure, CCBs are the most cost-effective option because
Update 2011
260
Hypertension (partial update) Pharmacological interventions they are associated with a low risk of developing diabetes, especially when compared to thiazide type diuretics, and they also have a good effectiveness profile across the range of other CVD risks. Conversely, when people are judged to be at a high risk of developing heart failure, thiazide-type diuretics were estimated to be the most cost-effective option, provided that they do not also have a high risk of diabetes. For people with a high risk of both heart failure and diabetes, ACE inhibitors or ARBs may be the most cost-effective option. The GDG noted that the applicability of this data to people under the age of 55 is uncertain, since it is based on trial data from mostly older people. Furthermore, although the model was robust to a variety of sensitivity analyses, there remains uncertainty about the size of some treatment effects, which translates into uncertainty about the relative cost-effectiveness of the drugs. The GDG considered the implications of the cost-effectiveness analysis with regard to the preferred treatment strategy for hypertension. Most people with primary hypertension are a low-to intermediate risk of heart failure and have an increased risk of developing diabetes, this suggests that CCBs would be the most cost-effective step 1 therapy for most people aged over 55 years. The caveat to this conclusion is that the risk of heart failure increases with increasing age, especially in the elderly (i.e. 80 years) in whom a thiazide-like diuretic would be a more cost effective treatment. Moreover, some people might not tolerate a CCB or may have evidence of oedema that might benefit from the preferred used of a thiazide-type diuretic. The GDG concluded that the cost-effectiveness analysis demonstrated that CCBs are the most costeffective initial therapy for most people aged >55 years with primary hypertension, and indeed, cost saving when compared to no intervention. It was considered that the evidence supporting this conclusion was stronger than in 2006. In addition the GDG discussions around this recommendation highlighted new data demonstrating; i) that CCBs appear to be the most effective treatment option to suppress blood pressure variability, which in turn appears to be an independent predictor of cardiovascular disease risk in people with treated hypertension (see below); and ii) that new evidence suggests that for treatment at step 2, the combination of A + C will usually be preferred to A + D, thereby impacting on the preferred choice of therapy for step 1 treatment (see section below step 2 treatment). Consequently, the GDG recommended that a CCB should be the preferred initial therapy for people with primary hypertension and aged >55 years. A thiazide-like diuretic (i.e. chlortalidone or inadapamide) are considered a suitable alternative for those who cannot tolerate a CCB or who have developed, or are at high risk of developing heart failure. Blood Pressure Variability and the impact of Antihypertensive therapy: Just after the scope for this guideline update had been finalised, a series of analyses were published showing that excessive variability in blood pressure is an independent risk factor for cardiovascular events, over and above the effect of the level of blood pressure itself. Furthermore, a systematic review of previous trials suggested that different classes of antihypertensive medications varied in their capacity to influence blood pressure variability. The GDG decided to review this data as part of this update (see Appendix F.1). The GDG noted that blood pressure variability can be measured in a number of ways but is perhaps most easily understood when expressed at the standard deviation (SD) around the mean of a number of blood pressure readings. The series of blood pressure readings may have been taken repeatedly at a single clinic visit, or an analysis of the variation between clinic visits, or across a series of measurements recorded by ABPM. Put simply, two people could have the same mean blood pressure but a different SD value for multiple readings, reflecting differences in blood pressure variability. This can be expressed as systolic or diastolic pressure variability. The studies reviewed by the GDG involved a series of retrospective analyses of clinical trial data (see appendix x). Review or these studies showed that variability in systolic blood pressure when measured visit-to-visit was a strong predictor of stroke, independent of mean systolic blood pressure. Moreover, in people with treated hypertension, a higher residual blood pressure variability is associated with a higher risk of vascular events. The GDG noted that it was unclear if blood
Update 2011
261
Hypertension (partial update) Pharmacological interventions pressure variability was causally related to clinical outcomes, or a marker of more severe underlying vascular disease. Furthermore, blood pressure is highly variable and although less so when measured under standardised conditions, it is unclear what the boundaries of normal versus abnormal variability would be in usual clinical practice. The GDG agreed that whatever the underlying mechanisms, systolic blood pressure variability appears to be an important independent predictor of clinical outcomes. The GDG also reviewed data from a systematic review and meta-analysis which examined the effect of different classes of blood pressure treatment on blood pressure variability in trials. This analysis revealed that blood pressure variability was most effectively reduced by CCBs, closely followed by thiazide-type diuretics. The analysis also showed that beta-blockers were the least effective and may actually increase blood pressure variability. Having considered these findings on blood pressure variability the GDG concluded that those most at risk of having increased systolic blood pressure variability, i.e. older hypertensive people, will already be treated with the most effective drug classes to suppress systolic blood pressure variability, i.e. a CCB (or a thiazide-like diuretic if a CCB is not indicated or tolerated) as step 1 therapy, according to the recommendations in this guideline update. The GDG concluded that the updated guidance recommends the best available evidence-based treatment options to suppress blood pressure variability in people with hypertension. Step two therapy: Many people with treated hypertension will require more than one drug to control their blood pressure. For people whose blood pressure is not controlled by step 1 treatment, i.e. A in younger adults (55years) or C or D in people aged >55yrs, the 2006 pharmacological update of this guideline recommended that step 2 therapy should be a combination of A + C or A + D. the choice of which combination was solely dictated by whether the patient was commenced on treatment with C or D at step 1. This reflected the fact that at the time of the 2006 update, there was no published data to better inform the discussion about whether there was a preferred combination for most people at step 2. For this 2011 update of the guideline, one RCT 296 was found which prospectively examined the effect of A + C versus A + D on clinical outcomes in the ACCOMPLISH trial. This study compared treatment with the ACE-i benazepril + the CCB amlodipine vs. the ACE-i benazepril + the thiazide diuretic hydrochlorothiazide in 11,506 people with hypertension, for a follow-up of 24 months. The GDG discussed the evidence which showed that ACE+CCB was significantly more effective at preventing MI when compared to ACEi + diuretic. Study withdrawal was also significantly lower in patients randomised to treatment with the combination of ACEi+CCB. The other clinical outcomes were not significantly different between groups but all numerically favoured the ACEi + CCB combination. The GDG noted that the ACCOMPLISH trial was stopped earlier than planned because the primary composite outcome was significantly in favour of the ACEi + CCB. Thus, the study had inadequate power to address individual cardiovascular outcomes. There was no quality of life data identified. The GDG concluded that the combination of ACEi+CCB had a treatment advantage over ACEi+diuretic. However, the GDG noted that this conclusion is based on a single large study. The GDG also noted that the ACEi used in this study, i.e.benazepril, is not used in the UK but concluded that there was unlikely to be an important difference between benazepril and other ACEi. Likewise, the GDG considered it likely that the results with the ACEi + CCB would be replicated with an ARB + CCB. The GDG also considered the black people of African or Caribean origin, ACEi are associated with an increased risk of developing angioedema which can be life threatening. Although the incidence of this adverse of ACEi in back people of African or Caribean origin is low, the GDG suggested that an ARB in preference to an ACEi should be considered for such patients when step 2
Update 2011
262
Hypertension (partial update) Pharmacological interventions treatment in required. The GDG concluded that this data from the ACCOMPLISH trial, taken together with the updated cost-effectiveness analysis and the data on blood pressure variability, all favour the combination of A + C versus A +D with the caveat that the differences between C and D in each of these areas of analysis, whilst usually favouring C, was not large. The GDG emphasised that whilst a CCB should usually be preferred versus thiazide-like diuretic as step 1 and step 2 therapy for most people, a thiazide-like diuretic is a highly effective alternative and is preferred in people with evidence or, or at high risk of developing heart failure. The GDG recommended that A + C should be the preferred step 2 therapy for most patients. A+D is an alternative step 2 treatment in those intolerant of a CCB or in those with a high risk of heart failure. Step 3 Treatment for Hypertension: The GDG did not formally review new evidence for step 3 treatment for the 2011 update. However, the GDG discussed the implications of the recommendations for step 1 and 2 treatments with regard to step 3 treatment. The GDG concluded that it follows from the evidence reviews cited above that the recommended step 3 treatment should be; A (ACEi or ARB) + CCB + D (thiazide-like diuretic, i.e. chlothalidone or indapamide). Resistant hypertension: (step 4 treatment) The GDG decided that the term resistant hypertension should be applied to people requiring step 4 treatment and defined resistant hypertension as follows; Definition of Resistant Hypertension: A person with resistant hypertension is someone who has confirmed hypertension and in whom clinic blood pressure is not controlled (<140/90mmHg) despite treatment with a rational combination of optimum or best tolerated doses of three antihypertensive drugs (usually A+C+D). The GDG noted that poor compliance with therapy and white coat hypertension could each manifest as apparent resistance to drug treatment and should be considered. Secondary causes for hypertension should also be reconsidered in people with resistant hypertension and discussion with a specialist may be required to address some of these issues. Based on health survey for England data, the GDG estimated that resistant hypertension is likely to affect approximately 500,000 people with treated hypertension in the U.K. and thus represents an important clinical problem. These people will be older and often have established cardiovascular disease, diabetes or CKD and thus, be at high cardiovascular risk. From a cardiovascular risk perspective, such people potentially have much to gain in terms of absolute benefit from further blood pressure lowering. The GDG noted that the treatment of resistant hypertension has not been studied in detail, in part because few drugs are developed that are specifically targeted at resistant hypertension. There is as a consequence, a paucity of data upon which to base guidance for the treatment of resistant hypertension. For the 2006 pharmacological update of this guideline, there was no formal evidence review for step 4 treatment and the GDG cautiously recommended a range of options that included; further diuretic therapy, alpha blockers or beta blockers. For this 2011 update the literature was searched for all years and all study types were included. Populations which were exclusively diabetic or had chronic kidney disease were excluded. The data search failed to indentify a single head-to-head RCT that met our search criteria. Six studies did meet the search criteria, however, these were all retrospective cohort studies i.e. post-hoc analyses of studies in which patients had been treated with four or more antihypertensive therapies. The GDG noted that all of these studies evaluated the use of low doses of spironolactone (an aldosterone antagonist), usually 25mg o.d. Together, the review of this data suggested that low dose spironolactone was effective in resistant hypertension based on the surrogate outcome of blood
Update 2011
263
Hypertension (partial update) Pharmacological interventions pressure lowering. There was no data on other clinical outcomes. It is unclear from this very limited data whether spironolactone is always the most effective treatment option for every patient with resistant hypertension. Furthermore, the GDG noted that spironolactone is not licensed for the treatment of hypertension in the U.K. but this does not preclude its use. Not all people are able to tolerate spironolactone, the main adverse effect being the development of nipple tenderness and/or gynaecomastia in males. Another important consideration is that spironolactone is a potassium sparing diuretic and may cause hyperkalaemia, especially when combined with an ACE-inhibitor or ARB, as will be the case for most people with resistant hypertension treated according to the algorithm recommended by this guideline. The GDG considered this to be a very important safety issue. Where reported, the studies that have used spironolactone for the treatment of resistant hypertension have not used it when the baseline potassium level exceeded 5.00mol/L, and spironolcatone was used with caution in people with a reduced eGFR. The GDG discussed these safety aspects and recommended that in primary care, low dose spironolactone should only be considered for the treatment of resistant hypertension when the blood potassium level is <4.5mmol/L. Particular caution is advised in people with a reduced GFR as they are at increased risk of hyperkaelemia and renal function should be monitored closely in all patients receiving sprinolactone. Blood potassium, sodium and creatinine values should be checked approximately 2 weeks after treatment initiation and perdiodically thereafter. The GDG also highlighted that patients should be advised to discontinue spironolactone treatment if they become significantly dehydrated due to illness such as vomiting and/or diaorrhea. The GDG recognised that the emphasis of too many caveats and concerns might limit the use of what can be a very effective drug in the setting of resistant hypertension. Nevertheless, care is needed to monitor patients when treatment regimens become increasingly complex. The GDG discussed the potential use of other drug classes for resistant hypertension and noted that treatments such as higher doses of thiazide type diuretics, alpha blockers and beta blockers have been used as add-on therapy in clinical trials at step 2 and 3 but not necessarily at step 4. The GDG concluded that this provides some evidence for the potential effectiveness of these other treatment options as add-on therapy. The GDG also considered alternative further diuretic therapy to spironolactone if this was deemed inappropriate treatment because of an elevated baseline potassium level or concerns about renal function. The GDG concluded that If blood potassium levels are higher than 4.5 mmol/l, then higher-dose thiazide-like diuretic treatment may be considered as an alternative. The GDG also discussed newer therapies such as the direct renin inhibitor aliskiren but concluded that there was insufficient evidence of its effectiveness to determine its suitability for use in resistant hypertension. In summary, the GDG concluded that resistant hypertension is an important clinical problem that has been poorly studied with regard to the underlying causes and the most effective treatment options. Clinicians should consider referral of people with resistant hypertension for specialist advice/evaluation especially those who are younger and those with complex comorbidities. The best evidence, albeit weak evidence, suggests that low dose spironolactone (e.g. 25mg o.d.), when safe to use and when tolerated, can be an effective means of further lowering blood pressure. It is unclear if this is the optimal treatment for most people with resistant hypertension or whether other treatment options would be more effective in most or some cases. When use of spironolactone is not possible or not tolerated, then higher dose thiazide-like diuretic, alpha blockers or beta blockers are suitable alternatives for step 4 treatment, with the caveat that the evidence base is very limited and careful monitoring of electrolytes and renal function is essential. The GDG recognised the need of more research in this area.
Update 2011
12.10 Recommendations
40.Where possible, recommend treatment with drugs taken only once a day. [2004]
264
Hypertension (partial update) Pharmacological interventions 41.Prescribe non-proprietary drugs where these are appropriate and minimise cost. [2004] 42.Offer people with isolated systolic hypertension (systolic BP 160 mmHg or more) the same treatment as people with both raised systolic and diastolic blood pressure. [2004] 43.Offer people aged 80 years and over the same antihypertensive drug treatment as people aged 5580 years, taking into account any comorbidities. [new 2011] 44.Offer antihypertensive drug treatment to women of child-bearing potential in line with the recommendations on Management of pregnancy with chronic hypertension and Breastfeeding in Hypertension in pregnancy (NICE clinical guideline 107). [2010] Step 1 treatment 45.Offer people aged under 55 years step 1 antihypertensive treatment with an angiotensinconverting enzyme (ACE) inhibitor or a low-cost angiotensin-II receptor blocker (ARB). If an ACE inhibitor is prescribed and is not tolerated (for example, because of cough), offer a low-cost ARB. [new 2011] 46.Do not combine an ACE inhibitor with an ARB to treat hypertension. [new 2011] 47.Offer step 1 antihypertensive treatment with a calcium-channel blocker (CCB) to people aged over 55 years and to black people of African or Caribbean family origin of any age. If a CCB is not suitable, for example because of oedema or intolerance, or if there is evidence of heart failure or a high risk of heart failure, offer a thiazide-like diuretic. [new 2011] 48.If a diuretic is to be initiated or changed, offer a thiazide-like diuretic, such as chlortalidone (12.5 mg25.0 mg once daily) or indapamide (1.5mg slow release or 2.5 mg once daily) in preference to a conventional thiazide diuretic such as bendroflumethiazide or hydrochlorothiazide. [new 2011] 49.For people who are already having treatment with bendroflumethiazide or hydrochlorothiazide and whose blood pressure is stable and well controlled, continue treatment with the bendroflumethiazide or hydrochlorothiazide. [new 2011] 50.Beta-blockers are not a preferred initial therapy for hypertension. However, beta-blockers may be considered in younger people, particularly: those with an intolerance or contraindication to ACE inhibitors and angiotensin-II receptor antagonists or women of child-bearing potential or people with evidence of increased sympathetic drive. [2006] 51.If therapy is initiated with a beta-blocker and a second drug is required, add a calcium-channel blocker rather than a thiazide-like diuretic to reduce the persons risk of developing diabetes. [2006] Step 2 treatment 52.If blood pressure is not controlled by Step 1 treatment, offer step 2 treatment with a CCB in combination with either an ACE inhibitor or an ARB j. [new 2011]
j
Update 2011CCC
Choose a low-cost ARB.
265
Hypertension (partial update) Pharmacological interventions 53.If a CCB is not suitable for step 2 treatment, for example because of oedema or intolerance, or if there is evidence of heart failure or a high risk of heart failure, offer a thiazide-like diuretic. [new 2011] 54.For black people of African or Caribbean family origin, consider an ARB k in preference to an ACE inhibitor, in combination with a CCB. [new 2011] Step 3 treatment 55.Before considering step 3 treatment, review medication to ensure step 2 treatment is at optimal or best tolerated doses. [new 2011] 56.If treatment with three drugs is required, the combination of ACE inhibitor or angiotensin-II receptor blocker, calcium-channel blocker and thiazide-like diuretic should be used. [2006] Step 4 treatment 57.Regard clinic blood pressure that remains higher than 140/90 mmHg after treatment with the optimal or best tolerated doses of an ACE inhibitor or an ARB plus a CCB plus a diuretic as resistant hypertension, and consider adding a fourth antihypertensive drug and/or seeking expert advice. [new 2011] 58.For treatment of resistant hypertension at step 4: Consider further diuretic therapy with low-dose spironolactone (25 mg once daily) l. If the blood potassium level is 4.5 mmol/l or lower. Use particular caution in people with a reduced estimated glomerular filtration rate because they have an increased risk of hyperkaelemia. Consider higher-dose thiazide-like diuretic treatment if the blood potassium level is higher than 4.5 mmol/l. [new 2011] 59.When using further diuretic therapy for resistant hypertension at step 4, monitor blood sodium and potassium and renal function within 1 month and repeat as required thereafter. [new 2011] 60.If further diuretic therapy for resistant hypertension at step 4 is not tolerated, or is contraindicated or ineffective, consider an alpha- or beta-blocker. [new 2011] 61.If blood pressure remains uncontrolled with the optimal or maximum tolerated doses of four drugs, seek expert advice if it has not yet been obtained. [new 2011]
Update 2011
12.11 Research recommendations

6. In adults with hypertension, which drug treatment (diuretic therapy versus other step 4 treatments) is the most clinically and cost effective for step 4 treatment? Although this guideline provides recommendations on the use of further diuretic therapy for treatment at step 4 (resistant hypertension), they are largely based on post-hoc observational data from clinical trials. More data are needed to compare further diuretic therapies, for example a potassium-sparing diuretic with a higher-dose thiazide-like diuretic, and to compare diuretic therapy with alternative treatment options at step 4 to define whether further diuretic therapy is the best option.
k l
Choose a low-cost ARB. At the time of publication (August 2011), spironolactone did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented.
266
Hypertension (partial update) Patients perspectives
13 Patients perspectives
13.1 Introduction
A published survey that examined the views of 452 hypertensive patients in one urban GP practice illustrated the range of feelings surrounding the taking of antihypertensive medications. There was a 77% response rate among patients invited to participate71. Four in every five people taking part in the study said they had reservations about taking antihypertensives. Over a third of patients reported experiencing current or previous side effects from blood pressure lowering medication and nearly 40% were concerned by the potential harm caused by the long term use of such drugs. Thirtysix percent of responders wondered if they still needed blood pressure lowering medication and twothirds would prefer non-drug therapy. The most commonly cited reasons for taking antihypertensive medications were 'to achieve some good results' (92%), 'because of what happens at the doctors' (87%) and 'because it feels reassuring' (68%). Before starting on tablets to treat high blood pressure, patients often weighed the potential benefits against reservations in the context of a personal framework. Information available on the DIPEx website (www.dipex.org) was summarised and discussed by the guideline development group. The DIPEx web site reflects patients' experiences of serious illness, aiming to share experiences, provide patient friendly information, answer common questions and provide information on relevant organisations and support groups to patients, family and friends, carers and health professionals. The hypertension module contains transcribed interviews from 4050 people who have experienced hypertension and can be viewed as transcripts, video or audio clips of individuals, or collated information on specific topics. The modules are produced by an advisory panel of patients, health professionals and social scientists with relevant expertise. Below is a summary of patients' accounts of discovery, treatment and living with hypertension.
13.2 Discovering hypertension

The route to diagnosis of hypertension was varied, with some patients detected during routine screening whilst others were identified after a specific event, for example a transient ischaemic attack (TIA), or following a consultation for a specific problem, for example dizziness or chest pain. Many patients perceived stress as a major causative factor, even to the extent that they would blame stresses in their lives of which they had previously been unaware. Other factors which they linked to hypertension were family history, genetic make-up, race, personality traits and specific habits such as alcohol consumption, smoking and salt intake. Patients reported a degree of frustration when they had eliminated factors they believed to contribute to their hypertension only to find that their blood pressure remained unchanged. Many of those interviewed felt that they had not been given sufficient information regarding the cause of their hypertension. Attitudes were influenced by patients' background knowledge about hypertension and whether they were asymptomatic at diagnosis. Some patients exhibited a positive attitude, feeling that detection gave them the opportunity to modify their lifestyle and for their hypertension to be monitored and treated to prevent long term disease. Others felt that their hypertension might have been detected earlier if doctors had been more vigilant.
13.3 Treatment
Patients voiced a great deal of concern over the issue of long term medication, highlighting potential side effects and the cost and need for regular prescriptions as major worries. Many patients reported
267
Hypertension (partial update) Patients perspectives no problems with antihypertensive drugs, but others had experienced a variety of side effects. Patients were most concerned about taking beta-blockers and these were perceived as having a higher side effect profile. ACEi and calcium-channel blockers were more favoured. Some patients found it difficult to accept side effects of blood pressure lowering medication when they were asymptomatic. In particular, drugs which led to impotence were considered unacceptable. Compliance to medication was also an issue, and many reported that they found it difficult to remember to take tablets. Some patients accepted that taking tablets was just part of everyday life, whilst others felt it to be a constant reminder of living with disease. Patients often felt under pressure from family members or health care professionals to be compliant and selecting the right combination of tablets often led to anxiety as patients were changed from one medication to another. In attempts to avoid or delay drug therapy, a proportion of patients wanted to try lifestyle measures or complementary therapies as an initial alternative to blood pressure lowering drugs.
13.4 Living with hypertension

Many patients were unsure of what it meant to have a diagnosis of hypertension - how serious was it? The increased risk of stroke and heart disease led some to focus on personal mortality, and to worry about dependants or financial issues if such events were to occur. Some patients reported that nothing really changed whilst others now viewed themselves as unhealthy or even experienced denial. Patients were anxious as they found it difficult to regulate their behaviour, particularly as they did not have changing symptoms, so as not to further increase their risks of cardiovascular disease. Others reported symptoms that they thought were related to hypertension such as headache, dizziness and visual problems. Often side effects of tablets were attributed to disease. Most patients made some attempt to incorporate lifestyle changes, such as restricting salt intake, increasing exercise and reducing stress. Patients often felt they wanted advice from health care professionals to avoid 'self-harm' and reported feelings of guilt and frustration if targets were not achieved. In general, patients welcomed information provided by general practitioners; some felt doctors did not provide enough information and looked for other sources such as the web, media or medical magazines. Others felt doctors pitched information - both the amount and content - at just the right level. A minority of patients felt that the greater their understanding about high blood pressure, the more that they had to worry about. Other patients found that people's accounts of living with hypertension were a valuable source of reassurance; however, they acknowledged that speaking openly about this was often difficult. Some expressed the view that having hypertension was a very private issue, rarely discussed, but felt that talking did provide much needed support and welcomed sites such as DIPEx as a forum in which to share their experiences.
13.5 Education and adherence

13.5.1 Compliance with Prescribed Antihypertensive Medication
It is estimated that between 5080% of patients with hypertension do not take all of their prescribed medication377,518. This has implications for the successful management of hypertension with poor adherence to medications linked to inadequately controlled blood pressure273. Understanding patient's reasons for not taking medications and implementing effective strategies to overcome barriers to taking prescribed medication is therefore a crucial aspect in the management of hypertension. Compliance is used variably as a term within the literature, referring sometimes to the constant neglect of treatment346, 344 and sometimes to a range of behaviours including delay in dosing, skipping a dose, longer lapses in dosing and over compliance when extra doses are taken620. It has
268
Hypertension (partial update) Patients perspectives been argued that recognizing these differences in compliance patterns is valuable in working with patients on improving their adherence to prescribed drug regimens620. Compliance has also been challenged as a concept because of its implied paternalism and failure to see patients as active, intentional and responsible participants in their health care management346, 344. Increasingly the term concordance is used within the literature, implying a more interactive and participatory approach to drug prescribing518. Not only is it important that drug regimens are adhered to in order to control blood pressure but it has also been suggested that partial compliance and erratic patterns of dosing may do more transient harm than any overall beneficial effect of treatment143. For example abrupt discontinuation of medications may lead to rebound hypertension with elevated blood pressure. Variability in blood pressure caused by abrupt changes in drug taking patterns has been linked to certain kinds of target organ damage such as pulmonary congestion and a consequent deterioration of congestive heart failure143. Therefore strategies to improve adherence also need to address the need to maintain regular and consistent patterns of drug usage. There are many factors that influence patients' decisions not to take their drugs as prescribed70,267. Factors most pertinent for patients suffering from hypertension include the asymptomatic nature of the disease. A condition without symptoms combined with the possibly unpleasant side effects of treatment may contribute to a patient's decision to stop or reduce their medication83. The long term nature of the treatment is also a factor that can lead to poorer compliance. Drug complexity, poor instructions, poor provider-patient relationships and patient's disagreement about their need for treatment may also serve as a reason for non-adherence to drug regimens267. A wide range of interventions have been developed to try and help patients follow their prescribed drug regimens. These have included simplified dosing, educational interventions, telephone and computer assisted monitoring, family interventions, increased convenience of care with provision of care at the work site, and a team approach with increased involvement of a community nurse and/or a community pharmacist267,518. Two systematic reviews have sought to assess the effectiveness of these interventions267,292. One looked specifically at the relationship between daily dose frequency and adherence to antihypertensive medication292. In a meta-analysis of data from eight studies it was found that the average adherence rate was significantly higher for patients with once daily dosing compared taking those taking multiple daily doses (91% vs. 83%). Adherence rates were also significantly higher for patients taking once daily doses compared with twice daily doses (93% vs. 87%). The difference in adherence rates between twice daily and multiple daily dosing was not significant. Simplifying dosing regimens to once daily use appears to promote compliance. However it is insufficient on its own to result in adequate compliance and the medical consequences may be graver for patients failing to adhere to once daily regimens, since missing one dose will result in missing the total daily dose. A narrative review of a wide range of interventions designed to increase compliance with prescribed drug regimens across a range of chronic disease entities found that half were associated with a statistically significant increase in medication adherence but that many were too small to show an effect. However they concluded that even the most effective interventions did not lead to large improvements in adherence and treatment outcomes267. Whilst they may not result in large improvements in adherence to prescribed drug treatments it would appear that improving patient education, providing counselling, involving families and other members of the health care team can all have a positive impact. Qualitative research methods have also contributed to an understanding of how patients weigh up their reservations about treatment against different reasons for taking treatment: this involves positive experiences with doctors, perceived benefits of medication and pragmatic considerations70. Patients will balance reservations and reasons differently. Greater adherence to drug treatment might be achieved if health care professionals asked patients how they perceived the advantages and disadvantages of taking
269
Hypertension (partial update) Patients perspectives medication and listened to their reservations, their reasons for taking medication and the balance between the two.
13.5.2
Implementing lifestyle measures

Lifestyle interventions such as weight reducing diets, lowering salt intake, exercise, alcohol reduction and relaxation therapy can reduce blood pressure and it is recommended that patients are given advice to promote such lifestyle changes. However, it is recognised that lifestyle changes are difficult to adopt and their effectiveness is often limited. The concept of compliance has now evolved to encompass 'an active, intentional and responsible process whereby patients work to maintain their health in collaboration with health care personnel' rather than simply patients' adherence to instructions344. Many factors are thought to influence adherence including age, sex, education, understanding and disease perspectives, the mode of delivering advice and the type of health system647. Adherence may be improved by good communication between patients and health professionals addressing knowledge about disease, active involvement of patients in decisions, setting achievable goals and good family and community support344,358,647. Adherence with lifestyle modifications, especially dietary changes, is lower than with antihypertensive drug therapy by between 13% and 76%109. Few studies specifically address this issue and most research on adherence to lifestyle advice examines strategies to reduce cardiovascular risk. Important issues to consider are the characteristics of the 'information provider', the 'information receiver', the 'information itself' and the dissemination strategy. Who should give it? In many instances, lifestyle advice is given by nurses who manage clinics for the secondary prevention of coronary heart disease. These nurse-led initiatives have been shown to be effective at modifying lifestyle behaviours, reducing blood pressure, monitoring medication and ultimately in reducing mortality112,417. The regular follow-up provided by these clinics may help compliance358. The Department of Health has provided guidance for general practitioners and practice nurses who wish to refer patients to facilities such as leisure centres or gyms for supervised exercise programmes173. How should it be given? Advice alone is less effective than specifically adapted programmes supported by written and audiovisual material109,605. Material tailored to meet the educational and cultural needs of the population it is targeting has also been shown to be effective342. Who should receive it? Targeting of advice to higher risk populations is thought to be more clinically and cost effective. A systematic review of 18 trials examining the effects of multiple risk factor interventions (stopping smoking, exercise, dietary control, weight control, antihypertensive drugs and cholesterol lowering drugs) in the primary prevention of coronary heart disease in middle aged adults showed little overall effect on mortality. However, it was noted that hypertensive 'high risk' patients were more likely to benefit from counselling, education and effective drugs and thus targeting health education to this group might be of some value186. What are the most successful strategies for information delivery? A review of 46 studies on compliance with drug therapy and lifestyle modifications in cardiovascular risk reduction identified the following effective strategies; behavioural skill training, self monitoring, telephone/mail contact, self-efficacy enhancement and external cognitive aids358. A review of compliance with low salt diets suggested that successful interventions require specific goals,
270
Hypertension (partial update) Patients perspectives delegation of responsibilities, in-depth patient assessment, behavioural motivation, implementation plans, repetitive education and extensive monitoring376. Delivering programmes through specific channels, for example community based projects may increase effectiveness358.
13.5.3
Recommendations
62.Provide appropriate guidance and materials about the benefits of drugs and the unwanted side effects sometimes experienced in order to help people make informed choices. [2004] 63.People vary in their attitudes to their hypertension and their experience of treatment. It may be helpful to provide details of patient organisations that provide useful forums to share views and information. [2004] 64.Provide an annual review of care to monitor blood pressure, provide people with support and discuss their lifestyle, symptoms and medication. [2004] 65.Because evidence supporting interventions to increase adherence is inconclusive, only use interventions to overcome practical problems associated with non-adherence if a specific need is identified. Target the intervention to the need. Interventions might include: suggesting that patients record their medicine-taking encouraging patients to monitor their condition simplifying the dosing regimen using alternative packaging for the medicine using a multi-compartment medicines system. (This recommendation is taken from Medicines adherence, NICE 408 clinical guideline 76). [new 2011]
Update 2011
271
Hypertension (partial update) Reference list
14 Reference list
1 Effects of treatment on morbidity in hypertension. II. Results in patients with diastolic blood pressure averaging 90 through 114 mm Hg. JAMA. 1970; 213(7):1143-1152. 2 Effect of antihypertensive treatment on stroke recurrence. Hypertension-Stroke Cooperative Study Group. JAMA. 1974; 229(4):409-418. 3 Evaluation of drug treatment in mild hypertension: VA-NHLBI feasibility trial. Plan and preliminary results of a two-year feasibility trial for a multicenter intervention study to evaluate the benefits versus the disadvantages of treating mild hypertension. Prepared for the Veterans Administration-National Heart, Lung, and Blood Institute Study Group for Evaluating Treatment in Mild Hypertension. Annals of the New York Academy of Sciences. 1978; 304:267-292. 4 The Australian therapeutic trial in mild hypertension. Report by the Management Committee. Lancet. 1980; 1(8181):1261-1267. 5 Hydrochlorothiazide and bendroflumethiazide in low doses--a comparative trial. Acta Pharmacologica Et Toxicologica. 1984; 54 Suppl 1:47-51. 6 An international trial of antihypertensive therapy in elderly patients. Objectives, protocol and organization. European Working Party on High Blood Pressure in the Elderly (EWPHE). Archives Internationales De Pharmacodynamie Et De Therapie. 1985; 275(2):300-334. 7 Cardiovascular risk and risk factors in a randomized trial of treatment based on the betablocker oxprenolol: the International Prospective Primary Prevention Study in Hypertension (IPPPSH). The IPPPSH Collaborative Group. Journal of Hypertension. 1985; 3(4):379-392. 8 MRC trial of treatment of mild hypertension: principal results. Medical Research Council Working Party. BMJ. 1985; 291(6488):97-104. 9 Course of blood pressure in mild hypertensives after withdrawal of long term antihypertensive treatment. Medical Research Council Working Party on Mild Hypertension. BMJ. 1986; 293(6553):988-992. 10 Effects of replacing sodium intake in subjects on a low sodium diet: a crossover study. Australian National Health & Medical Research Council Dietary Salt Study Management Committee. Clinical and Experimental Hypertension Part A: Theory and Practice. 1989; 11(5-6):1011-1024. 11 Fall in blood pressure with modest reduction in dietary salt intake in mild hypertension. Australian National Health and Medical Research Council Dietary Salt Study Management Committee. Lancet. 1989; 1(8635):399-402. 12 Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. The SOLVD Investigators. New England Journal of Medicine. 1991; 325(5):293-302. 13 Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in the Elderly Program (SHEP). SHEP Cooperative Research Group. JAMA. 1991; 265(24):3255-3264. 14 Acute Pain Management: Operative or Medical Procedures and Trauma. Rockville, MD: Agency for Health Care Policy and Research Publications; 1992 15 Medical Research Council trial of treatment of hypertension in older adults: principal results. MRC Working Party. BMJ. 1992; 304(6824):405-412.
272
Hypertension (partial update) Reference list 16 Systolic hypertension in the elderly: Chinese trial (syst-China). Interim report. Zhonghua Xin Xue Guan Bing Za Zhi. 1992; 20(5):270-5, 323. 17 Systolic hypertension in the elderly: Chinese trial (Syst-China)--second interim report. Zhonghua Xin Xue Guan Bing Za Zhi. 1993; 21(3):135-7, 185. 18 The Oslo Diet and Exercise Study (ODES): design and objectives. Controlled Clinical Trials. 1993; 14(3):229-243. 19 Trial of secondary prevention with atenolol after transient ischemic attack or nondisabling ischemic stroke. The Dutch TIA Trial Study Group. Stroke. 1993; 24(4):543-548. 20 Post-stroke antihypertensive treatment study. A preliminary result. PATS Collaborating Group. Chinese Medical Journal. 1995; 108(9):710-717. 21 BS EN 1060-1: 1996 Specification for Non-Invasive Sphygmomanometers. General Requirements. 15-6-1996. 22 Clinical and Experimental Pheochromocytoma. Cambridge, MA.: Blackwell; 1996
23 Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and other Societies on coronary prevention. European Heart Journal. 1998; 19(10):1434-1503. 24 Health Survey for England: Cardiovascular Disease. The Stationary Office; 1999
25 Effects of ramipril on cardiovascular and microvascular outcomes in people with diabetes mellitus: results of the HOPE study and MICRO-HOPE substudy. Heart Outcomes Prevention Evaluation Study Investigators. Lancet. 2000; 355(9200):253-259. 26 27 British National Formulary, Issue 44, March 2003. BMJ Books and Pharmaceutical Press; 2001 Mulrow CD (editors). Evidence Based Hypertension. London: BMJ; 2001
28 Craig R, Mindell J (editors). Health Survey for England 2006. Volume 1: Cardiovascular Disease and Risk Factors in Adults. Leeds: The Information Centre; 2008. Available from: http://www.ic.nhs.uk/webfiles/publications/HSE06/HSE%2006%20report%20VOL%201%20v2.pdf 29 Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS). Hypertension Research. 2008; 31(12):2115-2127. 30 NHS Supply Chain catalogue. Available from: http://www.supplychain.nhs.uk/portal/page/portal/Public/NHS-CAT Last accessed on: 17 February 2010. 31 Achmon J, Granek M, Golomb M, Hart J. Behavioral treatment of essential hypertension: a comparison between cognitive therapy and biofeedback of heart rate. Psychosomatic Medicine. 1989; 51(2):152-164. 32 Adsett CA, Bellissimo A, Mitchell A, Wilczynski N, Haynes RB. Behavioral and physiological effects of a beta blocker and relaxation therapy on mild hypertensives. Psychosomatic Medicine. 1989; 51(5):523-536. 33 Agras WS, Southam MA, Taylor CB. Long-term persistence of relaxation-induced blood pressure lowering during the working day. Journal of Consulting and Clinical Psychology. 1983; 51(5):792-794.
273
Hypertension (partial update) Reference list 34 Agras WS, Taylor CB, Kraemer HC, Southam MA, Schneider JA. Relaxation training for essential hypertension at the worksite: II. The poorly controlled hypertensive. Psychosomatic Medicine. 1987; 49(3):264-273. 35 Aitchison F, Page A. Diagnostic imaging of renal artery stenosis. Journal of Human Hypertension. 1999; 13(9):595-603. 36 Aivazyan TA, Zaitsev VP, Salenko BB, Yurenev AP, Patrusheva IF. Efficacy of relaxation techniques in hypertensive patients. Health Psychology. 1988; 7 Suppl:193-200. 37 Alam S, Johnson AG. A meta-analysis of randomised controlled trials (RCT) among healthy normotensive and essential hypertensive elderly patients to determine the effect of high salt (NaCl) diet of blood pressure. Journal of Human Hypertension. 1999; 13(6):367-374. 38 Alderman MH, Davis TK, Gerber LM, Robb M. Antihypertensive drug therapy withdrawal in a general population. Archives of Internal Medicine. 1986; 146(7):1309-1311. 39 Alem M, Milia P, Muir S, Lees K, Walters M. Comparison of the effects of diuretics on blood pressure and arterial stiffness in patients with stroke. Journal of Stroke and Cerebrovascular Diseases. 2008; 17(6):373-377. 40 American Heart Association. 2000 Heart and Stroke Statistical Update. American Heart Association, 1999. 41 American National Standard for Electronic or Automated Sphymomanometers. ANSI/AAMI SP10 1987. Arlington, VA: Association for the Advancement of Medical Instrumentation, 1983. 42 Amery A, Birkenhager W, Brixko P, Bulpitt C, Clement D, Deruyttere M, De SA, Dollery C, Fagard R, Forette F, . Mortality and morbidity results from the European Working Party on High Blood Pressure in the Elderly trial. Lancet. 1985; 1(8442):1349-1354. 43 Amery A, Birkenhager W, Bulpitt CJ, Clement D, De Leeuw P, Dollery CT. Syst-Eur: a multicenter trial on the treatment of isolated systolic hypertension in the elderly objectives, protocol, and organisation. Ageing. 1991; 3:287-302. 44 Anderson KM, Odell PM, Wilson PW, Kannel WB. Cardiovascular disease risk profiles. American Heart Journal. 1991; 121(1 Pt 2):293-298. 45 Anderssen S, Holme I, Urdal P, Hjermann I. Diet and exercise intervention have favourable effects on blood pressure in mild hypertensives: the Oslo Diet and Exercise Study (ODES). Blood Pressure. 1995; 4(6):343-349. 46 Antivalle M, Lattuada S, Salvaggio A, Paravicini M, Rindi M, Libretti A. Placebo effect and adaptation to noninvasive monitoring of BP. Journal of Human Hypertension. 1990; 4(6):633-637. 47 Applegate WB, Miller ST, Elam JT, Cushman WC, el Derwi D., Brewer A, Graney MJ. Nonpharmacologic intervention to reduce blood pressure in older patients with mild hypertension. Archives of Internal Medicine. 1992; 152(6):1162-1166. 48 Ara, R and Brennan, A. Economic Evaluation of Sibutramine for the Treatment of Obesity in Adults Without Other Co-Morbidities in the UK. School of Health and Related Research (ScHARR), University of Sheffield, 2004. 49 Arima H, Chalmers J, Woodward M, Anderson C, Rodgers A, Davis S, Macmahon S, Neal B. Lower target blood pressures are safe and effective for the prevention of recurrent stroke: The PROGRESS trial. Journal of Hypertension. 2006; 24(6):1201-1208.
274
Hypertension (partial update) Reference list 50 Arima H, Tanizaki Y, Yonemoto K, Doi Y, Ninomiya T, Hata J, Fukuhara M, Matsumura K, Iida M, Kiyohara Y. Impact of blood pressure levels on different types of stroke: the Hisayama study. Journal of Hypertension. 2009; 27(12):2437-2443. 51 Armitage P, Berry G, Matthews JNS.Statistical Methods in Medical Research. 4th ed. Oxford: Blackwell Science; 2002 52 Asagami T, Kushiro T, Inoue J, Kanmatsuse K. Long-term reproducibility and usefulness of daytime recording of noninvasive 24-hour ambulatory blood pressure monitoring in borderline hypertension: a two-year follow-up study. Clinical and Experimental Hypertension (New York). 1996; 18(5):637-657. 53 Asayama K, Ohkubo T, Kikuya M, Obara T, Metoki H, Inoue R, Hara A, Hirose T, Hoshi H, Hashimoto J, Totsune K, Satoh H, Imai Y. Prediction of stroke by home "morning" versus "evening" blood pressure values: the Ohasama study. Hypertension. 2006; 48(4):737-743. 54 Asayama K, Ohkubo T, Yoshida S, Suzuki K, Metoki H, Harada A, Murakami Y, Ohashi Y, Ueshima H, Imai Y, Japan Arteriosclerosis Longitudinal Study (JALS) group. Stroke risk and antihypertensive drug treatment in the general population: the Japan arteriosclerosis longitudinal study. Journal of Hypertension. 2009; 27(2):357-364. 55 ASCOT Steering Committee. Age-Stratified Analysis of Blood Pressure Responses. Personal communication: 06 56 Asmar R, Safar M, Queneau P. Evaluation of the placebo effect and reproducibility of blood pressure measurement in hypertension. American Journal of Hypertension. 2001; 14(6 I):546-552. 57 Assmann G, Cullen P, Evers T, Petzinna D, Schulte H. Importance of arterial pulse pressure as a predictor of coronary heart disease risk in PROCAM. European Heart Journal. 2005; 26(20):21202126. 58 Ayman D, Goldshine AD. Blood pressure determination by patients with essential hypertension: the difference between clinic and home readings before treatment. American Journal of the Medical Sciences. 1940; 200(4):465-474. 59 Bamford J, Sandercock P, Dennis M, Warlow C, Jones L, McPherson K, Vessey M, Fowler G, Molyneux A, Hughes T. A prospective study of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Project 1981-86. 1. Methodology, demography and incident cases of first-ever stroke. Journal of Neurology, Neurosurgery & Psychiatry. 1988; 51(11):1373-1380. 60 Barengo NC, Hu G, Kastarinen M, Antikainen R, Tuomilehto J. The effects of awareness, treatment and control of hypertension on future stroke incidence in a community-based population study in Finland. Journal of Hypertension. 2009; 27(7):1459-1465. 61 Barengo NC, Kastarinen M, Antikainen R, Nissinen A, Tuomilehto J. The effects of awareness, treatment and control of hypertension on cardiovascular and all-cause mortality in a communitybased population. Journal of Human Hypertension. 2009; 23(12):808-816. 62 Bayo J, Cos FX, Roca C, Dalfo A, Martin-Baranera MM, Albert B. Home blood pressure selfmonitoring: Diagnostic performance in white-coat hypertension. Blood Pressure Monitoring. 2006; 11(2):47-52. 63 Beckett NS, Peters R, Fletcher AE, Staessen JA, Liu L, Dumitrascu D, Stoyanovsky V, Antikainen RL, Nikitin Y, Anderson C, Belhani A, Forette F, Rajkumar C, Thijs L, Banya W, Bulpitt CJ, HYVET Study Group. Treatment of hypertension in patients 80 years of age or older. New England Journal of Medicine. 2008; 358(18):1887-1898.
275
Hypertension (partial update) Reference list 64 Beevers G, Lip G, O'Brien E. ABC of hypertension: Blood pressure measurement. Part II. BMJ. 2001; 322(7293):1043-1047. 65 Beevers G, Lip GY, O'Brien E. ABC of hypertension. Blood pressure measurement. Part Isphygmomanometry: factors common to all techniques. BMJ. 2001; 322(7292):981-985. 66 Beevers G, Lip GY, O'Brien E. ABC of hypertension: The pathophysiology of hypertension. BMJ. 2001; 322(7291):912-916. 67 Bejan-Angoulvant T, Saadatian-Elahi M, Wright JM, Schron EB, Lindholm LH, Fagard R, Staessen JA, Gueyffier F. Treatment of hypertension in patients 80 years and older: The lower the better? A meta-analysis of randomized controlled trials. Journal of Hypertension. 2010; 28(7):13661372. 68 Benetos A, Thomas F, Bean KE, Guize L. Why cardiovascular mortality is higher in treated hypertensives versus subjects of the same age, in the general population. Journal of Hypertension. 2003; 21(9):1635-1640. 69 Bennett P, Wallace L, Carroll D, Smith N. Treating Type A behaviours and mild hypertension in middle-aged men. Journal of Psychosomatic Research. 1991; 35(2-3):209-223. 70 Benson J, Britten N. Patients' decisions about whether or not to take antihypertensive drugs: qualitative study. BMJ. 2002; 325(7369):873. 71 Benson J, Britten N. Patients' views about taking antihypertensive drugs: questionnaire study. BMJ. 2003; 326(7402):1314-1315. 72 Berglund A, Andersson OK, Berglund G, Fagerberg B. Antihypertensive effect of diet compared with drug treatment in obese men with mild hypertension. BMJ. 1989; 299(6697):480-485. 73 Bhatnagar D, Anand IS, Durrington PN, Patel DJ, Wander GS, Mackness MI, Creed F, Tomenson B, Chandrashekhar Y, Winterbotham M, others. Coronary risk factors in people from the Indian subcontinent living in west London and their siblings in India. Lancet. 1995; 345(8947):405409. 74 Bhopal R, Unwin N, White M, Yallop J, Walker L, Alberti KG, Harland J, Patel S, Ahmad N, Turner C, Watson B, Kaur D, Kulkarni A, Laker M, Tavridou A. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study. BMJ. 1999; 319(7204):215-220. 75 Bing RF, Briggs RS, Burden AC, Russell GI, Swales JD, Thurston H. Reversible hypertension and hypothyroidism. Clinical Endocrinology. 1980; 13(4):339-342. 76 Bing RF, Russell GI, Swales JD, Thurston H. Indapamide and bendrofluazide: A comparison in the management of essential hypertension. British Journal of Clinical Pharmacology. 1981; 12(6):883886. 77 Bjorklund K, Lind L, Zethelius B, Berglund L, Lithell H. Prognostic significance of 24-h ambulatory blood pressure characteristics for cardiovascular morbidity in a population of elderly men. Journal of Hypertension. 2004; 22(9):1691-1697. 78 Black HR, Elliott WJ, Grandits G, Grambsch P, Lucente T, White WB, Neaton JD, Grimm RH, Jr., Hansson L, Lacourciere Y, Muller J, Sleight P, Weber MA, Williams G, Wittes J, Zanchetti A, Anders RJ, CONVINCE Research Group. Principal results of the Controlled Onset Verapamil Investigation of Cardiovascular End Points (CONVINCE) trial. JAMA. 2003; 289(16):2073-2082. 79 Black HR, Elliott WJ, Neaton JD, Grandits G, Grambsch P, Grimm RH, Hansson L, Lacourciere Y, Muller J, Sleight P, Weber MA, White WB, Williams G, Wittes J, Zanchetti A, Fakouhi TD. Rationale
276
Hypertension (partial update) Reference list and design for the controlled ONset Verapamil INvestigation of Cardiovascular Endpoints (CONVINCE) Trial. Controlled Clinical Trials. 1998; 19:370-390. 80 Blanchard EB, Miller ST, Abel GG, Haynes MR, Wicker R. Evaluation of biofeedback in the treatment of borderline essential hypertension. Journal of Applied Behavior Analysis. 1979; 12(1):99109. 81 Bland JM, Jones DR, Bennett S, Haines AP, MacFarlane AJ. Is the clinical trial evidence about new drugs statistically adequate? British Journal of Clinical Pharmacology. 1985; 19:155-160. 82 Blood Pressure Lowering Treatment Trialists' Collaboration, Turnbull F, Neal B, Ninomiya T, Algert C, Arima H, Barzi F, Bulpitt C, Chalmers J, Fagard R, Gleason A, Heritier S, Li N, Perkovic V, Woodward M, Macmahon S. Effects of different regimens to lower blood pressure on major cardiovascular events in older and younger adults: meta-analysis of randomised trials. BMJ. 2008; 336(7653):1121-1123. 83 Bloom BS. Daily regimen and compliance with treatment. BMJ. 2001; 323(7314):647.
84 Blumenthal JA, Sherwood A, Gullette EC, Babyak M, Waugh R, Georgiades A, Craighead LW, Tweedy D, Feinglos M, Appelbaum M, Hayano J, Hinderliter A. Exercise and weight loss reduce blood pressure in men and women with mild hypertension: effects on cardiovascular, metabolic, and hemodynamic functioning. Archives of Internal Medicine. 2000; 160(13):1947-1958. 85 Blumenthal JA, Siegel WC, Appelbaum M. Failure of exercise to reduce blood pressure in patients with mild hypertension. Results of a randomized controlled trial. JAMA. 1991; 266(15):20982104. 86 Bobrie G, Chatellier G, Genes N, Clerson P, Vaur L, Vaisse B, Menard J, Mallion JM. Cardiovascular prognosis of "masked hypertension" detected by blood pressure self-measurement in elderly treated hypertensive patients. JAMA. 2004; 291(11):1342-1349. 87 Bobrie G, Genes N, Vaur L, Clerson P, Vaisse B, Mallion JM, Chatellier G. Is "isolated home" hypertension as opposed to "isolated office" hypertension a sign of greater cardiovascular risk? Archives of Internal Medicine. 2001; 161(18):2205-2211. 88 Boggia J, Li Y, Thijs L, Hansen TW, Kikuya M, Bjorklund-Bodegard K, Richart T, Ohkubo T, Kuznetsova T, Torp-Pedersen C, Lind L, Ibsen H, Imai Y, Wang J, Sandoya E, O'Brien E, Staessen JA, International Database on Ambulatory blood pressure monitoring in relation to Cardiovascular Outcomes (IDACO) investigators. Prognostic accuracy of day versus night ambulatory blood pressure: a cohort study. Lancet. 2007; 370(9594):1219-1229. 89 Borghi C, Dormi A, L'Italien G, Lapuerta P, Franklin SS, Collatina S, Gaddi A. The relationship between systolic blood pressure and cardiovascular risk--results of the Brisighella Heart Study. Journal of Clinical Hypertension. 2003; 5(1):47-52. 90 Borhani NO, Mercuri M, Borhani PA, Buckalew VM, Canossa-Terris M, Carr AA, Kappagoda MD, Rocco MV, Schnaper HW, Sowers JR, Bond MG. Final outcome results of the Multicenter Isradipine Diuretic Atherosclerosis Study (MIDAS). JAMA. 1996; 276(10):785-791. 91 Borrello G, Mastroroberto P, Curcio F, Chello M, Zofrea S, Mazza ML. The effects of magnesium oxide on mild essential hypertension and quality of life. Current Therapeutic Research. 1996; 57(10):767-774. 92 Bosley F, Allen TW. Stress management training for hypertensives: cognitive and physiological effects. Journal of Behavioral Medicine. 1989; 12(1):77-89.
277
Hypertension (partial update) Reference list 93 Bowlus WE, Langford HG. A comparison of the antihypertensive effect of chlorthalidone and hydrochlorthiazide. Clinical Pharmacology and Therapeutics. 1964; 5:708-711. 94 Brandao SA, Izar CO, Fischer SM, Santos AO, Monteiro CM, Povoa RM, Helfenstein T, Carvalho AC, Monteiro AM, Ramos E, Gidlund M, Figueiredo Neto AM, Fonseca FA. Early increase in autoantibodies against human oxidized low-density lipoprotein in hypertensive patients after blood pressure control. American Journal of Hypertension. 2010; 23(2):208-214. 95 Brauer AP, Horlick L, Nelson E, Farquhar JW, Agras WS. Relaxation therapy for essential hypertension: a Veterans Administration Outpatient study. Journal of Behavioral Medicine. 1979; 2(1):21-29. 96 Bray EP, Holder R, Mant J, McManus RJ. Does self-monitoring reduce blood pressure? Metaanalysis with meta-regression of randomized controlled trials. Annals of Medicine. 2010; 42(5):371386. 97 Brenner BM, Cooper ME, de ZD, Keane WF, Mitch WE, Parving HH, Remuzzi G, Snapinn SM, Zhang Z, Shahinfar S. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. New England Journal of Medicine. 2001; 345(12):861-869. 98 Brenner H, Gefeller O. Variation of sensitivity, specificity, likelihood ratios and predictive values with disease prevalence. Stat Med. 1997; 16(9):981-991. 99 British Hypertension Society.Blood Pressure Measurement CD-ROM. BMJ; 1999
100 British Hypertension Society. Validated devices suitable for home use. Available from: http://www.bhsoc.org/blood_pressure_list.stm Last accessed on: 11 November 2010. 101 Britton KA, Gaziano JM, Djousse L. Normal systolic blood pressure and risk of heart failure in US male physicians. European Journal of Heart Failure. 2009; 11(12):1129-1134. 102 Bronnum-Hansen H, Davidsen M, Thorvaldsen P. Long-term survival and causes of death after stroke. Stroke. 2001; 32(9):2131-2136. 103 Bronnum-Hansen H, Jorgensen T, Davidsen M, Madsen M, Osler M, Gerdes LU, Schroll M. Survival and cause of death after myocardial infarction: the Danish MONICA study. Journal of Clinical Epidemiology. 2001; 54(12):1244-1250. 104 Brook RD. Home blood pressure: accuracy is independent of monitoring schedules. American Journal of Hypertension. 2000; 13(6 Pt 1):625-631. 105 Brown MJ, Palmer CR, Castaigne A, de Leeuw PW, Mancia G, Rosenthal T. Morbidity and mortality in patients randomised to double-blind treatment with a long-acting calcium-channel blocker or diuretic in the International Nifedipine GITS study: Intervention as a Goal in Hypertension Treatment (INSIGHT). Lancet. 2000; 356(5):366-372. 106 Brown MJ, Palmer CR, Castaigne A, de Leeuw PW, Mancia G, Rosenthal T. Principal results from the International Nifedipine GITS Study: Intervention as a Goal in Hypertension Treatment (INSIGHT). European Heart Journal. 2001; 3(Suppl B):B20-B26. 107 Bulpitt CJ, Ferrier G, Lewis PJ, Daymond M, Bulpitt PF, Dollery CT. Potassium supplementation fails to lower blood pressure in hypertensive patients receiving a potassium losing diuretic. Annals of Clinical Research. 1985; 17(4):126-130. 108 Burgess E, Lewanczuk R, Bolli P, Chockalingam A, Cutler H, Taylor G, Hamet P. Lifestyle modifications to prevent and control hypertension. 6. Recommendations on potassium, magnesium and calcium. Canadian Hypertension Society, Canadian Coalition for High Blood Pressure Prevention
278
Hypertension (partial update) Reference list and Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke Foundation of Canada. Canadian Medical Association Journal. 1999; 160(9 Suppl):S35-S45. 109 Burke LE, Dunbar-Jacob JM, Hill MN. Compliance with cardiovascular disease prevention strategies: a review of the research. Annals of Behavioral Medicine. 1997; 19(3):239-263. 110 Burt VL, Whelton P, Roccella EJ, Brown C, Cutler JA, Higgins M, Horan MJ, Labarthe D. Prevalence of hypertension in the US adult population. Results from the Third National Health and Nutrition Examination Survey, 1988-1991. Hypertension. 1995; 25(3):305-313. 111 Calvo C, Hermida RC, Ayala DE, Lopez JE, Fernandez JR, Dominguez MJ, Mojon A, Covelo M. The 'ABPM effect' gradually decreases but does not disappear in successive sessions of ambulatory monitoring. Journal of Hypertension. 2003; 21(12):2265-2273. 112 Campbell NC, Ritchie LD, Thain J, Deans HG, Rawles JM, Squair JL. Secondary prevention in coronary heart disease: a randomised trial of nurse led clinics in primary care. Heart. 1998; 80(5):447-452. 113 Campbell NR, Burgess E, Choi BC, Taylor G, Wilson E, Cleroux J, Fodor JG, Leiter LA, Spence D. Lifestyle modifications to prevent and control hypertension. 1. Methods and an overview of the Canadian recommendations. Canadian Hypertension Society, Canadian Coalition for High Blood Pressure Prevention and Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke Foundation of Canada. Canadian Medical Association Journal. 1999; 160(9 Suppl):S1-S6. 114 Campbell P, Ghuman N, Wakefield D, Wolfson L, White WB. Long-term reproducibility of ambulatory blood pressure is superior to office blood pressure in the very elderly. Journal of Human Hypertension. 2010; 24(11):749-754. 115 Canino E, Cardona R, Monsalve P, Perez AF, Lopez B, Fragachan F. A behavioral treatment program as a therapy in the control of primary hypertension. Acta Cientifica Venezolana. 1994; 45(1):23-30. 116 Cappuccio FP, Cook DG, Atkinson RW, Strazzullo P. Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in south London. Heart. 1997; 78(6):555-563. 117 Cappuccio FP, MacGregor GA. Does potassium supplementation lower blood pressure? A meta-analysis of published trials. Journal of Hypertension. 1991; 9(5):465-473. 118 Cappuccio FP, Siani A, Strazzullo P. Oral calcium supplementation and blood pressure: an overview of randomized controlled trials. Journal of Hypertension. 1989; 7(12):941-946. 119 Carlsson AC, Theobald H, Hellnius M, Wndell PE. Cardiovascular and total mortality in men and women with different blood pressure levels -A 26year followup. Blood Pressure. 2009; 18(3):105110. 120 Carson MA, Hathaway A, Tuohey JP, McKay BM. The effect of a relaxation technique on coronary risk factors. Behavioral Medicine. 1988; 14(2):71-77. 121 Cavallini MC, Roman MJ, Blank SG, Pini R, Pickering TG, Devereux RB. Association of the auscultatory gap with vascular disease in hypertensive patients. Annals of Internal Medicine. 1996; 124(10):877-883. 122 Celentano A, Galderisi M, Tammaro P, Mureddu GF, Garofalo M, de Divitiis O. Effects on casual and 24-h ambulatory blood pressure of slow-release nicardipine and chlorthalidone in arterial essential hypertension: double-blind, crossover study. International Journal of Clinical Pharmacology, Therapy, and Toxicology. 1990; 28(5):190-196.
279
Hypertension (partial update) Reference list 123 Celis H, De CP, Fagard R, Thijs L, Staessen JA. For how many days should blood pressure be measured at home in older patients before steady levels are obtained? Journal of Human Hypertension. 1997; 11(10):673-677. 124 Celis H, Yodfat Y, Thijs L, Clement D, Cozic J, de Cort P, Forette F, Grgoire M, Heyrman J, Stibbe G, van den Haute M, Staessen J, Fagard R, The Syst-Eur Investigators. Antihypertensive therapy in older patients with isolated systolic hypertension: the Syst-Eur experience in general practice. Family Practice. 1996; 13(2):138-143. 125 Chalmers J, Morgan T, Doyle A, Dickson B, Hopper J, Mathews J, Matthews G, Moulds R, Myers J, Nowson C, others. Australian National Health and Medical Research Council dietary salt study in mild hypertension. Journal of Hypertension - Supplement. 1986; 4(6):S629-S637. 126 Chapman N, Dobson J, Wilson S, Dahlof B, Sever PS, Wedel H, Poulter NR, AngloScandinavian Cardiac Outcomes Trial Investigators. Effect of spironolactone on blood pressure in subjects with resistant hypertension. Hypertension. 2007; 49(4):839-845. 127 Chaturvedi N, McKeigue PM, Marmot MG. Resting and ambulatory blood pressure differences in Afro-Caribbeans and Europeans. Hypertension. 1993; 22(1):90-96. 128 Chesney MA, Black GW, Swan GE, Ward MM. Relaxation training for essential hypertension at the worksite: I. The untreated mild hypertensive. Psychosomatic Medicine. 1987; 49(3):250-263. 129 Chien KL, Huang PJ, Chen MF, Chiang FT, Lai LP, Lee YT. Assessment of quality of life in a double-blind, randomized clinical trial of imidapril and captopril for hypertensive Chinese in Taiwan. Cardiovascular Drugs and Therapy. 2002; 16(3):221-226. 130 Chou T. Wake up and smell the coffee. Caffeine, coffee, and the medical consequences. Western Journal of Medicine. 1992; 157(5):544-553. 131 Clement DL, De Buyzere ML, De Bacquer DA, de Leeuw PW, Duprez DA, Fagard RH, Gheeraert PJ, Missault LH, Braun JJ, Six RO, Van Der Niepen P, O'Brien E, Office versus Ambulatory Pressure Study Investigators. Prognostic value of ambulatory blood-pressure recordings in patients with treated hypertension. New England Journal of Medicine. 2003; 348(24):2407-2415. 132 Cleroux J, Feldman RD, Petrella RJ. Lifestyle modifications to prevent and control hypertension. 4. Recommendations on physical exercise training. Canadian Hypertension Society, Canadian Coalition for High Blood Pressure Prevention and Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke Foundation of Canada. Canadian Medical Association Journal. 1999; 160(9 Suppl):S21-S28. 133 Coats AJS, Radaelli A, Clark SJ, Conway J, Sleight P. The influence of ambulatory blood pressure monitoring on the design and interpretation of trials in hypertension. Journal of Hypertension. 1992; 10(4):385-391. 134 Coca A, Messerli FH, Benetos A, Zhou Q, Champion A, Cooper-DeHoff RM, Pepine CJ. Predicting stroke risk in hypertensive patients with coronary artery disease: a report from the INVEST. Stroke. 2008; 39(2):343-348. 135 Collins R, Peto R, Macmahon S, Hebert P, Fiebach NH, Eberlein KA, Godwin J, Qizilbash N, Taylor JO, Hennekens CH. Blood pressure, stroke, and coronary heart disease. Part 2, Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet. 1990; 335(8693):827-838. 136 Conen D, Ridker PM, Buring JE, Glynn RJ. Risk of cardiovascular events among women with high normal blood pressure or blood pressure progression: prospective cohort study. BMJ. 2007; 335(7617):432.
280
Hypertension (partial update) Reference list 137 Conen D, Tschudi P, Martina B. Twenty-four hour ambulatory blood pressure for the management of antihypertensive treatment: a randomized controlled trial. Journal of Human Hypertension. 2009; 23(2):122-129. 138 Conlin PR, Chow D, Miller ER, III, Svetkey LP, Lin PH, Harsha DW, Moore TJ, Sacks FM, Appel LJ. The effect of dietary patterns on blood pressure control in hypertensive patients: results from the Dietary Approaches to Stop Hypertension (DASH) trial. American Journal of Hypertension. 2000; 13(9):949-955. 139 Cook TD, Campbell DT.Quasi-Experimentation: Design and Analysis Issues for Field Settings. Chicago: Rand McNally; 1979 140 Coope J, Warrender TS. Randomised trial of treatment of hypertension in elderly patients in primary care. BMJ. 1986; 293(6555):1145-1151. 141 Costa FV, Ambrosioni E, Montebugnoli L, Paccaloni L, Vasconi L, Magnani B. Effects of a lowsalt diet and of acute salt loading on blood pressure and intralymphocytic sodium concentration in young subjects with borderline hypertension. Clinical Science. 1981; 61 Suppl 7:21s-23s. 142 Cottier C, Shapiro K, Julius S. Treatment of mild hypertension with progressive muscle relaxation. Predictive value of indexes of sympathetic tone. Archives of Internal Medicine. 1984; 144(10):1954-1958. 143 Cramer JA. Consequences of intermittent treatment for hypertension: the case for medication compliance and persistence. American Journal of Managed Care. 1998; 4(11):1563-1568. 144 Croft PR, Brigg D, Smith S, Harrison CB, Branthwaite A, Collins MF. How useful is weight reduction in the management of hypertension? Journal of the Royal College of General Practitioners. 1986; 36(291):445-448. 145 Cruickshank JK, Beevers DG, Osbourne VL, Haynes RA, Corlett JC, Selby S. Heart attack, stroke, diabetes, and hypertension in West Indians, Asians, and whites in Birmingham, England. BMJ. 1980; 281(6248):1108. 146 Cupples ME, McKnight A. Randomised controlled trial of health promotion in general practice for patients at high cardiovascular risk. BMJ. 1994; 309(6960):993-996. 147 Curtin F, Altman DG, Elbourne D. Meta-analysis combining parallel and cross-over clinical trials. Statistics in Medicine. 2002; 21(15):2131-2144. 148 Cushman WC, Cutler JA, Hanna E, Bingham SF, Follmann D, Harford T, Dubbert P, Allender PS, Dufour M, Collins JF, Walsh SM, Kirk GF, Burg M, Felicetta JV, Hamilton BP, Katz LA, Perry HM, Jr., Willenbring ML, Lakshman R, Hamburger RJ. Prevention and Treatment of Hypertension Study (PATHS): effects of an alcohol treatment program on blood pressure. Archives of Internal Medicine. 1998; 158(11):1197-1207. 149 Cushman WC, Langford HG. Randomised controlled trial of potassium chloride versus placebo in mildly hypertensive blacks and whites. Circulation. 1988; 78:Suppl II-370. 150 Cuspidi C, Macca G, Michev I, Salerno M, Fusi V, Severgnini B, Corti C, Meani S, Valerio C, Magrini F, Zanchetti A. Short-term reproducibility of nocturnal non-dipping pattern in recently diagnosed essential hypertensives. Blood Pressure. 2002; 11(2):79-83. 151 Cuspidi C, Meani S, Sala C, Valerio C, Fusi V, Zanchetti A, Mancia G. How reliable is isolated clinical hypertension defined by a single 24-h ambulatory blood pressure monitoring? Journal of Hypertension. 2007; 25(2):315-320.
281
Hypertension (partial update) Reference list 152 Cutler JA, Follmann D, Allender PS. Randomized trials of sodium reduction: an overview. American Journal of Clinical Nutrition. 1997; 65(2 Suppl):643S-651S. 153 Cutler JA, Follmann D, Elliott P, Suh I. An overview of randomized trials of sodium reduction and blood pressure. Hypertension. 1991; 17(1 Suppl):I27-I33. 154 Dahlf B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristiansson K, Lederballe-Pedersen O, Lindholm LH, Nieminen M, Omvik S, Wedel H. Cardiovascular morbidity and mortality in the losartan intervention for endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002; 359:995-1003. 155 Dahlf B, Hansson L, Lindholm LH, Schersten B, Wester PO, Ekbom T, Hedner T, de Faire U. STOP-Hypertension-2: a prospective intervention trial of newer versus older treatment alternatives in old patients with hypertension. Blood Pressure. 1993; 2:136-141. 156 Dahlof B, Lindholm LH, Hansson L, Schersten B, Ekbom T, Wester PO. Morbidity and mortality in the Swedish Trial in Old Patients with Hypertension (STOP-Hypertension). Lancet. 1991; 338(8778):1281-1285. 157 Dahlof B, Sever PS, Poulter NR, Wedel H, Beevers DG, Caulfield M, Collins R, Kjeldsen SE, Kristinsson A, McInnes GT, Mehlsen J, Nieminen M, O'Brien E, Ostergren J, ASCOT Investigators. Prevention of cardiovascular events with an antihypertensive regimen of amlodipine adding perindopril as required versus atenolol adding bendroflumethiazide as required, in the AngloScandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm (ASCOT-BPLA): a multicentre randomised controlled trial. Lancet. 2005; 366(9489):895-906. 158 Davy BM, Melby CL, Beske SD, Ho RC, Davrath LR, Davy KP. Oat consumption does not affect resting casual and ambulatory 24-h arterial blood pressure in men with high-normal blood pressure to stage I hypertension. Journal of Nutrition. 2002; 132(3):394-398. 159 Dawes MG, Coats AJ, Juszczak E. Daytime ambulatory systolic blood pressure is more effective at predicting mortality than clinic blood pressure. Blood Pressure Monitoring. 2006; 11(3):111-118. 160 de Hoon JN, Vanmolkot FH, van den Ven LL, Van Bortel LM. Quality of life comparison between bisoprolol and nifedipine retard in hypertension. Cardiovascular Drugs and Therapy. 1997; 11(3):465-471. 161 de Lame PA, Droussin AM, Thomson M, Verhaest L, Wallace S. The effects of enalapril on hypertension and quality of life. A large multicenter study in Belgium. Acta Cardiologica. 1989; 44(4):289-302. 162 De Plaen JF, Detry JM. Hemodynamic effects of physical training in established arterial hypertension. Acta Cardiologica. 1980; 35(3):179-188. 163 de Souza F., Muxfeldt E, Fiszman R, Salles G. Efficacy of spironolactone therapy in patients with true resistant hypertension. Hypertension. 2010; 55(1):147-152. 164 Deary AJ, Schumann AL, Murfet H, Haydock SF, Foo RS, Brown MJ. Double-blind, placebocontrolled crossover comparison of five classes of antihypertensive drugs. Journal of Hypertension. 2002; 20(4):771-777. 165 Deckers JW, Goedhart DM, Boersma E, Briggs A, Bertrand M, Ferrari R, Remme WJ, Fox K, Simoons ML. Treatment benefit by perindopril in patients with stable coronary artery disease at different levels of risk. European Heart Journal. 2006; 27(7):796-801.
282
Hypertension (partial update) Reference list 166 Degl'innocenti A, Elmfeldt D, Hofman A, Lithell H, Olofsson B, Skoog I, Trenkwalder P, Zanchetti A, Wiklund I. Health-related quality of life during treatment of elderly patients with hypertension: results from the Study on COgnition and Prognosis in the Elderly (SCOPE). Journal of Human Hypertension. 2004; 18(4):239-245. 167 Den Hond E, Celis H, Fagard R, Keary L, Leeman M, O'Brien E, Vandenhoven G, Staessen JA. Self-measured versus ambulatory blood pressure in the diagnosis of hypertension. Journal of Hypertension. 2003; 21(4):717-722. 168 Denardo SJ, Gong Y, Nichols WW, Messerli FH, Bavry AA, Cooper-DeHoff RM, Handberg EM, Champion A, Pepine CJ. Blood pressure and outcomes in very old hypertensive coronary artery disease patients: an INVEST substudy. American Journal of Medicine. 2010; 123(8):719-726. 169 Dennis M, Bamford J, Sandercock P, Warlow C. Prognosis of transient ischemic attacks in the Oxfordshire Community Stroke Project. Stroke. 1990; 21(6):848-853. 170 Dennis MS, Bamford JM, Sandercock PA, Warlow CP. Incidence of transient ischemic attacks in Oxfordshire, England. Stroke. 1989; 20(3):333-339. 171 Department of Health. Blood Pressure Measurement Devices - Mercury and Non-Mercury. (MDA DB2000(03)). London: Department of Health, 2000. 172 Department of Health. Medical Devices and Equipment Management: Repair and Maintenance Provision. (MDA DB2000(02)). London: Department of Health, 2000. 173 Department of Health. Exercise Referral Systems: a National Quality Assessment Framework. London: 2001. 174 Department of Health. NHS reference costs 2008-09. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH _111591 175 DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clinical Trials. 1986; 7(3):177-188. 176 Devereux RB, Dahlof B, Gerdts E, Boman K, Nieminen MS, Papademetriou V, Rokkedal J, Harris KE, Edelman JM, Wachtell K. Regression of hypertensive left ventricular hypertrophy by losartan compared with atenolol: the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) trial. Circulation. 2004; 110(11):1456-1462. 177 Dickerson JE, Hingorani AD, Ashby MJ, Palmer CR, Brown MJ. Optimisation of antihypertensive treatment by crossover rotation of four major classes. Lancet. 1999; 353(9169):2008-2013. 178 Dolan E, Stanton A, Thijs L, Hinedi K, Atkins N, McClory S, Den Hond E, McCormack P, Staessen JA, O'Brien E. Superiority of ambulatory over clinic blood pressure measurement in predicting mortality: the Dublin outcome study. Hypertension. 2005; 46(1):156-161. 179 Dolfgoff S, SCHREK R, BALLARD GP, BAKER LA. Tobacco smoking as an etiologic factor in disease. 2. Coronary disease and hypertension. Angiology. 1952; 3(4):323-334. 180 Doll R, HILL AB. Smoking and carcinoma of the lung; preliminary report. BMJ. 1950; 2(4682):739-748. 181 Doll R, Peto R, Hall E, Wheatley K, Gray R. Mortality in relation to consumption of alcohol: 13 years' observations on male British doctors. BMJ. 1994; 309(6959):911-918.
283
Hypertension (partial update) Reference list 182 Dosh SA. The diagnosis of essential and secondary hypertension in adults. Journal of Family Practice. 2001; 50(8):707-712. 183 Drummond MF, Sculpher M, Torrance G, O'Brien B, Stoddart G.Methods for the Economic Evaluation of Health Care Programmes. 3rd ed. Oxford: OUP; 2005. Available from: http://www.oup.com/us/catalog/general/subject/Medicine/PublicHealth/?view=usa&ci=978019852 9453 184 Duncan JJ, Farr JE, Upton SJ, Hagan RD, Oglesby ME, Blair SN. The effects of aerobic exercise on plasma catecholamines and blood pressure in patients with mild essential hypertension. JAMA. 1985; 254(18):2609-2613. 185 Dzau VJ, Herrmann HC. Hormonal control of angiotensinogen production. Life Sciences. 1982; 30(7-8):577-584. 186 Ebrahim S, Smith GD. Systematic review of randomised controlled trials of multiple risk factor interventions for preventing coronary heart disease. BMJ. 1997; 314(7095):1666-1674. 187 Ebrahim S, Smith GD. Lowering blood pressure: a systematic review of sustained effects of non-pharmacological interventions. Journal of Public Health Medicine. 1998; 20(4):441-448. 188 Eccles M, Mason J. How to develop cost-conscious guidelines. Health Technology Assessment. 2001; 5(16) 189 Egger M, Jni P, Bartlett C, Holenstein F, Stern J. How important are comprehensive literature searches and the assessment of trial quality in systematic reviews? Empirical study. Health Technology Assessment. 2003; 7(1) 190 Eguchi K, Hoshide S, Hoshide Y, Ishikawa S, Shimada K, Kario K. Reproducibility of ambulatory blood pressure in treated and untreated hypertensive patients. Journal of Hypertension. 2010; 28(5):918-924. 191 Eguchi K, Kuruvilla S, Ogedegbe G, Gerin W, Schwartz JE, Pickering TG. What is the optimal interval between successive home blood pressure readings using an automated oscillometric device? Journal of Hypertension. 2009; 27(6):1172-1177. 192 Eisenberg DM, Delbanco TL, Berkey CS, Kaptchuk TJ, Kupelnick B, Kuhl J, Chalmers TC. Cognitive behavioral techniques for hypertension: are they effective? Annals of Internal Medicine. 1993; 118(12):964-972. 193 Elbourne D, Altman DG, Higgins JPT, Curtin F, Worthington HV, vail A. Meta-analyses involving cross-over trials: methodological issues. International Journal of Epidemiology. 2002; 31(1):140-149. 194 Elliott WJ, Weber RR, Murphy MB. A double-blind, randomized, placebo-controlled comparison of the metabolic effects of low-dose hydrochlorothiazide and indapamide. Journal of Clinical Pharmacology. 1991; 31(8):751-757. 195 Emeriau JP, Knauf H, Pujadas JO, Calvo-Gomez C, Abate G, Leonetti G, Chastang C, European S, I. A comparison of indapamide SR 1.5 mg with both amlodipine 5 mg and hydrochlorothiazide 25 mg in elderly hypertensive patients: a randomized double-blind controlled study. Journal of Hypertension. 2001; 19(2):343-350. 196 Enstrom I, Pennert K. Does it matter whether ambulatory blood pressure is recorded during a work day or a non-work day? Journal of Hypertension. 1996; 14(5):565-569. 197 Eriksson S, Olofsson B, Wester P. Atenolol in Secondary Prevention after Stroke. Cerebrovascular Diseases. 1995; 5(1):21-25.
284
Hypertension (partial update) Reference list 198 Ernst ME, Carter BL, Goerdt CJ, Steffensmeier JJG, Phillips BB, Zimmerman MB, Bergus GR. Comparative antihypertensive effects of hydrochlorothiazide and chlorthalidone on ambulatory and office blood pressure. Hypertension. 2006; 47(3):352-358. 199 Ernst ME, Carter BL, Zheng S, Grimm RH, Jr. Meta-analysis of dose-response characteristics of hydrochlorothiazide and chlorthalidone: effects on systolic blood pressure and potassium. American Journal of Hypertension. 2010; 23(4):440-446. 200 Ernst ME, Weber CA, Dawson JD, O'Connor MA, Lin W, Carter BL, Bergus GR. How well does a shortened time interval characterize results of a full ambulatory blood pressure monitoring session? Journal of Clinical Hypertension. 2008; 10(6):431-435. 201 Espeland MA, Whelton PK, Kostis JB, Bahnson JL, Ettinger WH, Cutler JA, Appel LJ, Kumanyika S, Farmer D, Elam J, Wilson AC, Applegate WB. Predictors and mediators of successful long-term withdrawal from antihypertensive medications. TONE Cooperative Research Group. Trial of Nonpharmacologic Interventions in the Elderly. Archives of Family Medicine. 1999; 8(3):228-236. 202 Esposti LD, Di Martino M., Saragoni S, Sgreccia A, Capone A, Buda S, Esposti ED. Pharmacoeconomics of antihypertensive drug treatment: an analysis of how long patients remain on various antihypertensive therapies. Journal of Clinical Hypertension. 2004; 6:76-84. 203 Ewald S, vor dem EJ, Uen S, Neikes F, Vetter H, Mengden T. Relationship between the frequency of blood pressure self-measurement and blood pressure reduction with antihypertensive therapy : results of the OLMETEL (OLMEsartan TELemonitoring blood pressure) study. Clinical Drug Investigation. 2006; 26(8):439-446. 204 Fagard RH. The role of exercise in blood pressure control: supportive evidence. Journal of Hypertension. 1995; 13(11):1223-1227. 205 Fagard RH. Physical activity in the prevention and treatment of hypertension in the obese. Medicine and Science in Sports and Exercise. 1999; 31(11 Suppl):S624-S630. 206 Fagard RH, Cornelissen VA. Incidence of cardiovascular events in white-coat, masked and sustained hypertension versus true normotension: A meta-analysis. Journal of Hypertension. 2007; 25(11):2193-2198. 207 Fagard RH, Staessen JA. Treatment of isolated systolic hypertension in the elderly: the SystEur trial. Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Clinical and Experimental Hypertension. 1999; 21(5&6):491-497. 208 Fagard RH, Staessen JA, Thijs L, Bulpitt CJ, Clement D, de Leeuw PW, Jaaskivi M, Mancia G, O'Brien E, Palatini P, Tuomilehto J, Webster J, Systolic Hypertension in Europe Trial Investigators. Relationship between ambulatory blood pressure and follow-up clinic blood pressure in elderly patients with systolic hypertension. Journal of Hypertension. 2004; 22(1):81-87. 209 Fagard RH, Staessen JA, Thijs L, Celis H, Bulpitt CJ, de Leeuw PW, Leonetti G, Tuomilehto J, Yodfat Y. On-treatment diastolic blood pressure and prognosis in systolic hypertension. Archives of Internal Medicine. 2007; 167(17):1884-1891. 210 Fagard RH, Thijs L, Staessen JA, Clement DL, De Buyzere ML, De Bacquer DA. Prognostic significance of ambulatory blood pressure in hypertensive patients with history of cardiovascular disease. Blood Pressure Monitoring. 2008; 13(6):325-332. 211 Fagard RH, Van Den Broeke C, de Cort P. Prognostic significance of blood pressure measured in the office, at home and during ambulatory monitoring in older patients in general practice. Journal of Human Hypertension. 2005; 19(10):801-807.
285
Hypertension (partial update) Reference list 212 Fagerberg B, Andersson OK, Persson B, Hedner T. Reactivity to norepinephrine and effect of sodium on blood pressure during weight loss. Hypertension. 1985; 7(4):586-592. 213 Fang X-H, Zhang X-H, Yang Q-D, Dai X-Y, Su F-Z, Rao M-L, Wu S-P, Du X-L, Wang W-Z, Li S-C. Subtype hypertension and risk of stroke in middle-aged and older chinese: A 10-year follow-up study. Stroke. 2006; 37(1):38-43. 214 Feldman RD, Campbell N, Larochelle P, Bolli P, Burgess E, Carruthers G, Floras J, Haynes B, Honos G, Leenen F, Leiter LA, Logan AG, Myers J, Spence D, Zarnke KB. Lifestyle modifications to prevent and control hypertension. Canadian Medical Association Journal. 1999; 161(Supplement):S1S22. 215 Ferrara LA, Iannuzzi R, Castaldo A, Iannuzzi A, Dello RA, Mancini M. Long-term magnesium supplementation in essential hypertension. Cardiology. 1992; 81(1):25-33. 216 Finnerty FA, Jr. A double-blind study of chlorthalidone and hydrochlorothiazide in an outpatient population of moderate hypertensives. Angiology. 1976; 27(12):738-744. 217 Fitzgerald PA. Endocrinology. Current Medical Diagnosis and Treatment, 40th edn. New York: Lange/McGraw-Hill, 2001: 1088-1160. 218 Fletcher AE, Bulpitt CJ, Chase DM, Collins WC, Furberg CD, Goggin TK, Hewett AJ, Neiss AM. Quality of life with three antihypertensive treatments. Cilazapril, atenolol, nifedipine. Hypertension. 1992; 19(6 Pt 1):499-507. 219 Flores L, Recasens M, Gomis R, Esmatjes E. White coat hypertension in type 1 diabetic patients without nephropathy. American Journal of Hypertension. 2000; 13(5 Pt 1):560-563. 220 Fodor JG, Whitmore B, Leenen F, Larochelle P. Lifestyle modifications to prevent and control hypertension. 5. Recommendations on dietary salt. Canadian Hypertension Society, Canadian Coalition for High Blood Pressure Prevention and Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke Foundation of Canada. Canadian Medical Association Journal. 1999; 160(9 Suppl):S29-S34. 221 Frankel BL, Patel DJ, Horwitz D, Friedewald WT, Gaarder KR. Treatment of hypertension with biofeedback and relaxation techniques. Psychosomatic Medicine. 1978; 40(4):276-293. 222 Franklin SS, Wachtell K, Papademetriou V, Olsen MH, Devereux RB, Fyhrquist F, Ibsen H, Kjeldsen SE, Dahlof B. Cardiovascular morbidity and mortality in hypertensive patients with lower versus higher risk: a LIFE substudy. Hypertension. 2005; 46(3):492-499. 223 Fravel MA, Ernst ME, Weber CA, Dawson JD, Carter BL, Bergus GR. Influence of patient characteristics on success of ambulatory blood pressure monitoring. Pharmacotherapy. 2008; 28(11):1341-1347. 224 Freemantle N, Urdahl H, Eastaugh J, Hobbs FD. What is the place of beta-blockade in patients who have experienced a myocardial infarction with preserved left ventricular function? Evidence and (mis)interpretation. Progress in Cardiovascular Diseases. 2002; 44(4):243-250. 225 Fukunaga H, Ohkubo T, Kobayashi M, Tamaki Y, Kikuya M, Obara T, Nakagawa M, Hara A, Asayama K, Metoki H, Inoue R, Hashimoto J, K, Imai Y. Cost-effectiveness of the introduction of home blood pressure measurement in patients with office hypertension. Journal of Hypertension. 2008; 26(4):685-690. 226 Gaddam K, Pimenta E, Thomas SJ, Cofield SS, Oparil S, Harding SM, Calhoun DA. Spironolactone reduces severity of obstructive sleep apnoea in patients with resistant hypertension: a preliminary report. Journal of Human Hypertension. 2010; 24(8):532-537.
286
Hypertension (partial update) Reference list 227 Galloe AM, Graudal N, Moller J, Bro H, Jorgensen M, Christensen HR. Effect of oral calcium supplementation on blood pressure in patients with previously untreated hypertension: a randomised, double-blind, placebo-controlled, crossover study. Journal of Human Hypertension. 1993; 7(1):43-45. 228 Garcia-Vera MP, Sanz J. How many self-measured blood pressure readings are needed to estimate hypertensive patients' "true" blood pressure? Journal of Behavioral Medicine. 1999; 22(1):93-113. 229 Geleijnse JM, Witteman JC, Bak AA, den Breeijen JH, Grobbee DE. Reduction in blood pressure with a low sodium, high potassium, high magnesium salt in older subjects with mild to moderate hypertension. BMJ. 1994; 309(6952):436-440. 230 Gibbs CR, Beevers DG, Lip GY. The management of hypertensive disease in black patients. QJM. 1999; 92(4):187-192. 231 Gillett PA, White AT, Caserta MS. Effect of exercise and/or fitness education on fitness in older, sedentary, obese women. Journal of Aging and Physical Activity. 1996; 4:42-45. 232 Goldstein IB, Shapiro D, Thananopavarn C, Sambhi MP. Comparison of drug and behavioral treatments of essential hypertension. Health Psychology. 1982; 1(1):7-26. 233 Gong L, Zhang W, Zhu Y, Zhu J, Kong D, Page V, Ghadirian P, LeLorier J, Hamet P. Shanghai trial of nifedipine in the elderly (STONE). Journal of Hypertension. 1996; 14(10):1237-1245. 234 Goodacre S, Nicholl J, Dixon S, Cross E, Angelini K, Arnold J, Revill S, Locker T, Capewell SJ, Quinney D, Campbell S, Morris F. Randomised controlled trial and economic evaluation of a chest pain observation unit compared with routine care. British Medical Journal. 2004; 328(7434):254. 235 Gordon NF, Scott CB, Levine BD. Comparison of single versus multiple lifestyle interventions: are the antihypertensive effects of exercise training and diet-induced weight loss additive? American Journal of Cardiology. 1997; 79(6):763-767. 236 Gorton T. Blood pressure of adults by age and sex, United States. Vital and Health Statistics Series 11, Data From the National Health Survey. 1964; 11:1-40. 237 Gosse P, Cipriano C, Bemurat L, Mas D, Lemetayer P, N'Tela G, Clementy J. Prognostic significance of blood pressure measured on rising. Journal of Human Hypertension. 2001; 15(6):413417. 238 Graudal NA, Galloe AM, Garred P. Effects of sodium restriction on blood pressure, renin, aldosterone, catecholamines, cholesterols, and triglyceride: a meta-analysis. JAMA. 1998; 279(17):1383-1391. 239 Griffith LE, Guyatt GH, Cook RJ, Bucher HC, Cook DJ. The influence of dietary and nondietary calcium supplementation on blood pressure: an updated metaanalysis of randomized controlled trials. American Journal of Hypertension. 1999; 12(1 Pt 1):84-92. 240 Grimm RH, Jr., Neaton JD, Elmer PJ, Svendsen KH, Levin J, Segal M, Holland L, Witte LJ, Clearman DR, Kofron P, . The influence of oral potassium chloride on blood pressure in hypertensive men on a low-sodium diet. New England Journal of Medicine. 1990; 322(9):569-574. 241 Grobbee DE, Hofman A. Does sodium restriction lower blood pressure? BMJ. 1986; 293(6538):27-29. 242 Grobbee DE, Hofman A. Effect of calcium supplementation on diastolic blood pressure in young people with mild hypertension. Lancet. 1986; 2(8509):703-707.
287
Hypertension (partial update) Reference list 243 Gudmundsson LS, Johannsson M, Thorgeirsson G, Sigfusson N, Sigvaldason H, Witteman JCM. Hypertension control as predictor of mortality in treated men and women, followed for up to 30 years. Cardiovascular Drugs and Therapy. 2005; 19(3):227-235. 244 Gustavsen PH, Hoegholm A, Bang LE, Kristensen KS. White coat hypertension is a cardiovascular risk factor: A 10-year follow-up study. Journal of Human Hypertension. 2003; 17(12):811-817. 245 Hafner RJ. Psychological treatment of essential hypertension: a controlled comparison of meditation and meditation plus biofeedback. Biofeedback and Self Regulation. 1982; 7(3):305-316. 246 Hagberg JM, Montain SJ, Martin WH, III, Ehsani AA. Effect of exercise training in 60- to 69year-old persons with essential hypertension. American Journal of Cardiology. 1989; 64(5):348-353. 247 Haider AW, Larson MG, Franklin SS, Levy D. Systolic blood pressure, diastolic blood pressure, and pulse pressure as predictors of risk for congestive heart failure in the Framingham Heart Study. Annals of Internal Medicine. 2003; 138(1):10-16. 248 Halbert JA, Silagy CA, Finucane P, Withers RT, Hamdorf PA, Andrews GR. The effectiveness of exercise training in lowering blood pressure: a meta-analysis of randomised controlled trials of 4 weeks or longer. Journal of Human Hypertension. 1997; 11(10):641-649. 249 Halbert JA, Silagy CA, Finucane PM, Withers RT, Hamdorf PA. Physical activity and cardiovascular risk factors: effect of advice from an exercise specialist in Australian general practice. Medical Journal of Australia. 2000; 173(2):84-87. 250 Hall WD. Resistant hypertension, secondary hypertension, and hypertensive crises. Cardiology Clinics. 2002; 20(2):281-289. 251 Hamet P. The evaluation of the scientific evidence for a relationship between calcium and hypertension. Journal of Nutrition. 1995; 125(2 Suppl):311S-400S. 252 Hammond IW, Devereux RB, Alderman MH, Lutas EM, Spitzer MC, Crowley JS, Laragh JH. The prevalence and correlates of echocardiographic left ventricular hypertrophy among employed patients with uncomplicated hypertension. Journal of the American College of Cardiology. 1986; 7(3):639-650. 253 Hansen TW, Jeppesen J, Rasmussen S, Ibsen H, Torp-Pedersen C. Ambulatory blood pressure and mortality: a population-based study. Hypertension. 2005; 45(4):499-504. 254 Hansen TW, Kikuya M, Thijs L, Bjorklund-Bodegard K, Kuznetsova T, Ohkubo T, Richart T, Torp-Pedersen C, Lind L, Jeppesen J, Ibsen H, Imai Y, Staessen JA, IDACO I. Prognostic superiority of daytime ambulatory over conventional blood pressure in four populations: a meta-analysis of 7,030 individuals. Journal of Hypertension. 2007; 25(8):1554-1564. 255 Hansson L. Results of the STOP-Hypertension-2 trial. Blood Pressure. 2000; 9(Suppl 2):17-20.
256 Hansson L, Hedner T, Lindholm L, et al. The Captopril Prevention Project (CAPPP) in Hypertension: baseline data and current status. Blood Pressure. 1997; 6:365-367. 257 Hansson L, Hedner T, Lund-Johansen P, Kjeldsen SE, Lindholm LH, Syvertsen JO, Lanke J, de Faire U, Dahlf B, Karlberg BE. Randomised trial of effects of calcium antagonists compared with diuretics and beta-blockers on cardiovascular morbidity and mortality in hypertension: the Nordic Diltiazem (NORDIL) study. Lancet. 2000; 356:359-365. 258 Hansson L, Lindholm LH, Ekbom T, Dahlf B, Lanke J, Schersten B, Wester PO, Hedner T. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality
288
Hypertension (partial update) Reference list and morbidity the Swedish Trial in Old Patients with Hypertension-2 study. Lancet. 1999; 354:17511756. 259 Hansson L, Lindholm LH, Niskanen L, Lanke J, Hedner T, Niklason A, Luomanmaki K, Dahlof B, de FU, Morlin C, Karlberg BE, Wester PO, Bjorck JE. Effect of angiotensin-converting-enzyme inhibition compared with conventional therapy on cardiovascular morbidity and mortality in hypertension: the Captopril Prevention Project (CAPPP) randomised trial. Lancet. 1999; 353(9153):611-616. 260 Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S, Menard J, Rahn KH, Wedel H, Westerling S. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group. Lancet. 1998; 351(9118):1755-1762. 261 Harris KA, Holly RG. Physiological response to circuit weight training in borderline hypertensive subjects. Medicine and Science in Sports and Exercise. 1987; 19(3):246-252. 262 Harsha DW, Lin PH, Obarzanek E, Karanja NM, Moore TJ, Caballero B. Dietary Approaches to Stop Hypertension: a summary of study results. DASH Collaborative Research Group. Journal of the American Dietetic Association. 1999; 99(8 Suppl):S35-S39. 263 Hart CL, Smith GD, Hole DJ, Hawthorne VM. Alcohol consumption and mortality from all causes, coronary heart disease, and stroke: results from a prospective cohort study of scottish men with 21 years of follow up. BMJ. 1999; 318(7200):1725-1729. 264 Harvard CEA Registry. Cost Effectiveness Analysis (CEA) Registry. 1997. Tufts-New England Medical Center. https://research.tufts-nemc.org/cear4/Default.aspx 265 Hatch JP, Klatt KD, Supik JD, Rios N, Fisher JG, Bauer RL, Shimotsu GW. Combined behavioral and pharmacological treatment of essential hypertension. Biofeedback and Self Regulation. 1985; 10(2):119-138. 266 Hatt PY, Leblond JB. A comparative study of the activity of a new agent, indapamide, in essential arterial hypertension. Current Medical Research and Opinion. 1975; 3(3):138-144. 267 Haynes RB, McDonald H, Garg AX, Montague P. Interventions for helping patients to follow prescriptions for medications. Cochrane Database of Systematic Reviews. 2002;(2):CD000011. 268 He FJ, MacGregor GA. Effect of modest salt reduction on blood pressure: a meta-analysis of randomized trials. Implications for public health. Journal of Human Hypertension. 2002; 16(11):761770. 269 Head GA, Mihailidou AS, Duggan KA, Beilin LJ, Berry N, Brown MA, Bune AJ, Cowley D, Chalmers JP, Howe PRC, Hodgson J, Ludbrook J, Mangoni AA, McGrath BP, Nelson MR, Sharman JE, Stowasser M. Definition of ambulatory blood pressure targets for diagnosis and treatment of hypertension in relation to clinic blood pressure: Prospective cohort study. BMJ. 2010; 340(7751):849. 270 Henderson DG, Schierup J, Schodt T. Effect of magnesium supplementation on blood pressure and electrolyte concentrations in hypertensive patients receiving long term diuretic treatment. BMJ. 1986; 293(6548):664-665. 271 Hermida RC, Calvo C, Ayala DE, Fernandez JR, Ruilope LM, Lopez JE. Evaluation of the extent and duration of the "ABPM effect" in hypertensive patients. Journal of the American College of Cardiology. 2002; 40(4):710-717.
289
Hypertension (partial update) Reference list 272 Hernandez-del RR, Martin-Baranera M, Sobrino J, Gorostidi M, Vinyoles E, Sierra C, Segura J, Coca A, Ruilope LM, Spanish Society of Hypertension Ambulatory Blood Pressure Monitoring Registry Investigators. Reproducibility of the circadian blood pressure pattern in 24-h versus 48-h recordings: the Spanish Ambulatory Blood Pressure Monitoring Registry. Journal of Hypertension. 2007; 25(12):2406-2412. 273 Hershey JC, Morton BG, Davis JB, Reichgott MJ. Patient compliance with antihypertensive medication. American Journal of Public Health. 1980; 70(10):1081-1089. 274 Hinderliter A, Sherwood A, Gullette EC, Babyak M, Waugh R, Georgiades A, Blumenthal JA. Reduction of left ventricular hypertrophy after exercise and weight loss in overweight patients with mild hypertension. Archives of Internal Medicine. 2002; 162(12):1333-1339. 275 Hodgkinson J, Mant J, Martin U, Guo B, Hobbs FDR, Deeks JJ, Heneghan C, Roberts N, McManus RJ. Relative effectiveness of clinic and home blood pressure monitoring compared with ambulatory blood pressure monitoring in diagnosis of hypertension: systematic review. BMJ. 2011; 342:d3621 276 Hoelscher TJ, Lichstein KL, Fischer SM, Hegarty TB. Relaxation treatment of hypertension: Do home relaxation tapes enhance treatment outcome? Behavior Therapy. 1987; 18(1):33-37. 277 Hoelscher TJ, Lichstein KL, Rosenthal TL. Home relaxation practice in hypertension treatment: objective assessment and compliance induction. Journal of Consulting and Clinical Psychology. 1986; 54(2):217-221. 278 Hollenberg NK, Williams GH, Anderson R, Akhras KS, Bittman RM, Krause SL. Symptoms and the distress they cause: comparison of an aldosterone antagonist and a calcium channel blocking agent in patients with systolic hypertension. Archives of Internal Medicine. 2003; 163(13):1543-1548. 279 Hooper L, Bartlett C, Davey SG, Ebrahim S. Systematic review of long term effects of advice to reduce dietary salt in adults. BMJ. 2002; 325(7365):628. 280 Hosohata K, Saito S, Ohkubo T, Kikuya M, Obara T, Kato T, Totsune K, Miura Y, Arakawa K, Fujishima M, Fujii J, Fukiyama K, Hisamichi S, Iimura O, Ishii M, Omae T, Saruta T, Yoshinaga K, Abe I, Abukawa T, Ashida T, Dohba N, Etoh T, Fujimura A, Gotoh T, Hama H, Hano T, Hayashi H, Hayashida N, Hayashi M, Hiramori K, Hirai Y, Hirata Y, Hiwada K, Hora K, Ichikawa S, Imaizumi T, Ishikawa K, Ito I, Iwaoka D, Kanamasa K, Katagiri T, Katayama S, Kawano Y, Kida H, Kimura G, Kitaoka H, Kobayashi S, Kohara K, Kojima S, Komuro I, Kumagai H, Kusano E, Kushiro T, Kuwajima I, Maruyama Y, Masani F, Matsubara H, Matsubara T, Matsumoto M, Matsuoka H, Matsuura H, Mishima Y, Miura M, Miyamori I, Murakami H, Muratani H, Nakao K, Naruse M, Nishio I, Ogihara T, Ohta M, Ohtsuka K, Ohuchi Y, Oikawa S, Okabe M, Okumura K, Saitoh I, Saitoh H, Sakata T, Saku K, Sasaki H, Senda S, Shimada K, Shimamoto K, Shiomi T, Shirato K, Takada M, Takeda K, Takeda N, Takeshita A, Takishita S, Toba K, Tochikubo O, Tomoike H, Ueno Y, Umemura S, Urata H, Yamada K, Yamaguchi T, Yamashina A, Yoshimura M, Satoh H, Oka Y, Katagiri H, Kondo Y, Shishido H, Kohinata A, Kanno Y, Ikeda H, Takahashi T, Kimura A, Funahashi J, Takada N, Hanazawa T, Osugi M, Hashimoto J, Miyamura T, Masuda Y, Tanaka S, Koizumi M, Ando T, Tsuchida A, Nanba M, Yonekura S, Shiroishi M, Nakagawa M, Oda K, Kanahara I, Hirota N, Tsutsui M, Anzai J, Hosoda S, Shiiki M, Takagawa Y, Iida M, Tada M, Noto T, Tanaka H, Tajima G, Kawamorita K, Komai K, Seki H, Omoto A, Takahashi H, Kitabayashi A, Kimura M, Kimura Y, Takahashi K, Kawamorita Y, Emura Y, Ishikawa Y, Nagai K, Nagai S, Ito T, Nihei K, Yamada S, Yamamoto N, Suzuki Y, Ito H, Ito M, Fujita T, Tominaga S, Suzuki K, Ishibashi K, Ando Y, Sato S, Morikawa H, Kashima S, Nakayama D, Hayashi Y, Ohta K, Metoki H, Fukami K, Hayashi T, Kishi M, Tajima J, Seino J, Hoshi H, Imai Y, Kimura H, Mori R, Matsuo K, Tanno Y, Shibasaki A, Yamagishi T, Ohtomo E, Ohtsuka Y, Sasaki S, Seino M, Kurosawa K, Kyogoku S, Ito K, Ono Y, Watanabe S, Hiwatashi N, Yagi C, Unakami H, Asayama K, Mouri T, Watanabe T, Kikuchi K, Maruyama N, Sone M, Iwamoto M, Naganuma S, Mashiko H, Ishii H, Takaya Y, Kamimoto M, Shirai A, Watanabe R, Tohyama Y, Araki
290
Hypertension (partial update) Reference list F, Sakuma H, Amada Y, Sato E, Techigawara M, Nakamura K, Nakayama T, Hasegawa K, Minami J, Kumagai Y, Cho T, Okamoto K, Iguchi T, Honzawa T, Koitabashi T, Horikoshi H, Nagao T, Yoshimatsu H, Seki K, Akiyama N, Suga H, Matsunaga R, Sinozaki T, Ishimaru Y, Yagi N, Takikawa H, Hukutome T, Nakajima C, Sato H, Simizu J, Arai M, Koide H, Takada H, Umezu T, Aoki H, Nakamoto H, Komahashi K, Sekiya S, Sugimoto H, Arai T, Inokuma S, Funazaki T. Progress report on the Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure (HOMED-BP) study: Status at February 2004. Clinical and Experimental Hypertension. 2007; 29(1):69-81. 281 Hulley SB, Furberg CD, Gurland B, McDonald R, Perry HM, Schnaper HW. Systolic Hypertension in the Elderly Program (SHEP): antihypertensive efficacy of chlorthalidone. American Journal of Cardiology. 1985; 56(15):913-920. 282 Ichihara A, Hayashi M, Koura Y, Tada Y, Hirota N, Saruta T. Long-term effects of intensive blood-pressure lowering on arterial wall stiffness in hypertensive patients. American Journal of Hypertension. 2003; 16(11 Pt 1):959-965. 283 Imai Y, Satoh H, Nagai K, Sakuma M, Sakuma H, Minami N, Munakata M, Hashimoto J, Yamagishi T, Watanabe N, . Characteristics of a community-based distribution of home blood pressure in Ohasama in northern Japan. Journal of Hypertension. 1993; 11(12):1441-1449. 284 Ingelsson E, Bjorklund-Bodegard K, Lind L, Arnlov J, Sundstrom J. Diurnal blood pressure pattern and risk of congestive heart failure. JAMA. 2006; 295(24):2859-2866. 285 Inoue R, Ohkubo T, Kikuya M, Metoki H, Asayama K, Obara T, Hirose T, Hara A, Hoshi H, Hashimoto J, Totsune K, Satoh H, Kondo Y, Imai Y. Stroke risk in systolic and combined systolic and diastolic hypertension determined using ambulatory blood pressure. The Ohasama study. American Journal of Hypertension. 2007; 20(10):1125-1131. 286 Institute for Clinical Systems Improvement. Hypertension Diagnosis and Treatment. Bloomington, MN: Institute for Clinical Systems Improvement, 2000. 287 International Centre for Alcohol Policies. What Is a "Standard Drink"? (5). Washington, DC: International Centre for Alcohol Policies, 1998. 288 Irvine MJ, Johnston DW, Jenner DA, Marie GV. Relaxation and stress management in the treatment of essential hypertension. Journal of Psychosomatic Research. 1986; 30(4):437-450. 289 Irvine MJ, Logan AG. Relaxation behavior therapy as sole treatment for mild hypertension. Psychosomatic Medicine. 1991; 53(6):587-597. 290 Ishikawa J, Carroll DJ, Kuruvilla S, Schwartz JE, Pickering TG. Changes in home versus clinic blood pressure with antihypertensive treatments: a meta-analysis. Hypertension. 2008; 52:856-864. 291 Ishikawa S, Kario K, Kayaba K, Gotoh T, Nago N, Nakamura Y, Tsutsumi A, Kajii E. Continued high risk of stroke in treated hypertensives in a general population: The Jichi medical school cohort study. Hypertension Research. 2008; 31(6):1125-1133. 292 Iskedjian M, Einarson TR, MacKeigan LD, Shear N, Addis A, Mittmann N, Ilersich AL. Relationship between daily dose frequency and adherence to antihypertensive pharmacotherapy: evidence from a meta-analysis. Clinical Therapeutics. 2002; 24(2):302-316. 293 Jacob RG, Chesney MA, Williams DM, Ding Y, Shapiro AP. Relaxation therapy for hypertension: design effects and treatment effects. Annals of Behavioral Medicine. 1991; 13(1):5-17. 294 Jacob RG, Fortmann SP, Kraemer HC, Farquhar JW, Agras WS. Combining behavioral treatments to reduce blood pressure. A controlled outcome study. Behavior Modification. 1985; 9(1):32-53.
291
Hypertension (partial update) Reference list 295 Jalkanen L. The effect of a weight reduction program on cardiovascular risk factors among overweight hypertensives in primary health care. Scandinavian Journal of Social Medicine. 1991; 19(1):66-71. 296 Jamerson K, Weber MA, Bakris GL, Dahlof B, Pitt B, Shi V, Hester A, Gupte J, Gatlin M, Velazquez EJ, ACCOMPLISH Trial Investigators. Benazepril plus amlodipine or hydrochlorothiazide for hypertension in high-risk patients. New England Journal of Medicine. 2008; 359(23):2417-2428. 297 James JE. Is habitual caffeine use a preventable cardiovascular risk factor? Lancet. 1997; 349(9047):279-281. 298 JATOS Study Group. The Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients (JATOS): protocol, patient characteristics, and blood pressure during the first 12 months. Hypertension Research. 2005; 28(6):513-520. 299 Jee SH, He J, Whelton PK, Suh I, Klag MJ. The effect of chronic coffee drinking on blood pressure: a meta-analysis of controlled clinical trials. Hypertension. 1999; 33(2):647-652. 300 Jee SH, Miller ER, III, Guallar E, Singh VK, Appel LJ, Klag MJ. The effect of magnesium supplementation on blood pressure: a meta-analysis of randomized clinical trials. American Journal of Hypertension. 2002; 15(8):691-696. 301 Johannesson M. The cost-effectiveness of th switch towards more expensive antihypertensive drugs. Health Policy. 1994; 28(1):1-13. 302 Johansson JK, Niiranen TJ, Puukka PJ, Jula AM. Optimal schedule for home blood pressure monitoring based on a clinical approach. Journal of Hypertension. 2010; 28(2):259-264. 303 Johnson KA, Partsch DJ, Rippole LL, McVey DM. Reliability of self-reported blood pressure measurements. Archives of Internal Medicine. 1999; 159(22):2689-2693. 304 Johnston DW, Gold A, Kentish J, Smith D, Vallance P, Shah D, Leach G, Robinson B. Effect of stress management on blood pressure in mild primary hypertension. BMJ. 1993; 306(6883):963-966. 305 Johnston ME, Gibson ES, Terry CW, Haynes RB, Taylor DW, Gafni A, Sicurella JI, Sackett DL. Effects of labelling on income, work and social function among hypertensive employees. Journal of Chronic Diseases. 1984; 37(6):417-423. 306 Joint Formulary Committee.British National Formulary: BNF 60. 60th ed. London: UK: BMJ Group and Pharmaceutical Press; 2010. Available from: http://www.bnf.org.uk 307 Jonsson B, Carides GW, Burke TA, Dasbach EJ, Lindholm LH, Dahlof B. Cost effectiveness of losartan in patients with hypertension and LVH: an economic evaluation for Sweden of the LIFE trial. Journal of Hypertension. 2005; 23(7):1425-1431. 308 Jonsson B, Hansson L, Stalhammar NO. Health economics in the hypertension optimal treatment (HOT) study: costs and cost-effectiveness of intensive blood pressure lowering and lowdose aspirin in patients with hypertension. Journal of Internal Medicine. 2003; 253:472-480. 309 Jula A, Ronnemaa T, Rastas M, Karvetti RL, Maki J. Long-term nopharmacological treatment for mild to moderate hypertension. Journal of Internal Medicine. 1990; 227(6):413-421. 310 Jula A, Ronnemaa T, Tikkanen I, Karanko H. Responses of atrial natriuretic factor to long-term sodium restriction in mild to moderate hypertension. Journal of Internal Medicine. 1992; 231(5):521529.
292
Hypertension (partial update) Reference list 311 Jula AM, Karanko HM. Effects on left ventricular hypertrophy of long-term nonpharmacological treatment with sodium restriction in mild-to-moderate essential hypertension. Circulation. 1994; 89(3):1023-1031. 312 Julius S, Kjeldsen SE, Weber M, Brunner HR, Ekman S, Hansson L, Hua T, Laragh J, McInnes GT, Mitchell L, Plat F, Schork A, Smith B, Zanchetti A, VALUE trial group. Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amlodipine: the VALUE randomised trial. Lancet. 2004; 363(9426):2022-2031. 313 Kagiyama S, Fukuhara M, Ansai T, Matsumura K, Soh I, Takata Y, Sonoki K, Awano S, Takehara T, Iida M. Association between blood pressure and mortality in 80-year-old subjects from a populationbased prospective study in Japan. Hypertension Research. 2008; 31(2):265-270. 314 Kaplan NM. Other Forms of Secondary Hypertension. In: Kaplan NM, Lieberman E (eds), Clinical Hypertension, 7th edition edn. Baltimore: Williams & Wilkins, 1998: 315 Kawabe H, Saito I. Correlation of repeated measurements of home blood pressure on one occasion and diagnosis of hypertension: study by measurement over seven consecutive days. Clinical and Experimental Hypertension. 2008; 30(1):79-85. 316 Kawabe H, Saito I, Saruta T. Influence of repeated measurement on one occasion, on successive days, and on workdays on home blood pressure values. Clinical and Experimental Hypertension. 2005; 27(2-3):215-222. 317 Kawano Y, Matsuoka H, Takishita S, Omae T. Effects of magnesium supplementation in hypertensive patients: assessment by office, home, and ambulatory blood pressures. Hypertension. 1998; 32(2):260-265. 318 Kawano Y, Yoshimi H, Matsuoka H, Takishita S, Omae T. Calcium supplementation in patients with essential hypertension: assessment by office, home and ambulatory blood pressure. Journal of Hypertension. 1998; 16(11):1693-1699. 319 Kelley GA. Aerobic exercise and resting blood pressure among women: a meta-analysis. Preventive Medicine. 1999; 28(3):264-275. 320 Kelley GA, Kelley KS. Aerobic exercise and resting blood pressure in women: a meta-analytic review of controlled clinical trials. Journal of Women's Health & Gender-Based Medicine. 1999; 8(6):787-803. 321 Kelley GA, Kelley KS. Progressive resistance exercise and resting blood pressure : A metaanalysis of randomized controlled trials. Hypertension. 2000; 35(3):838-843. 322 Kelley GA, Kelley KS, Tran ZV. Walking and resting blood pressure in adults: a meta-analysis. Preventive Medicine. 2001; 33(2 Pt 1):120-127. 323 Kelley GA, Sharpe KK. Aerobic exercise and resting blood pressure in older adults: a metaanalytic review of randomized controlled trials. Journals of Gerontology Series A, Biological Sciences and Medical Sciences. 2001; 56(5):M298-M303. 324 Khattar RS, Swales JD, Banfield A, Dore C, Senior R, Lahiri A. Prediction of coronary and cerebrovascular morbidity and mortality by direct continuous ambulatory blood pressure monitoring in essential hypertension. Circulation. 1999; 100(10):1071-1076. 325 Khattar RS, Swales JD, Dore C, Senior R, Lahiri A. Effect of aging on the prognostic significance of ambulatory systolic, diastolic, and pulse pressure in essential hypertension. Circulation. 2001; 104(7):783-789.
293
Hypertension (partial update) Reference list 326 Kikuya M, Hansen TW, Thijs L, Bjorklund-Bodegard K, Kuznetsova T, Ohkubo T, Richart T, Torp-Pedersen C, Lind L, Ibsen H, Imai Y, Staessen JA, International Database on Ambulatory Blood Pressure Monitoring in relation to Cardiovascular Outcomes Investigators. Diagnostic thresholds for ambulatory blood pressure monitoring based on 10-year cardiovascular risk. Circulation. 2007; 115(16):2145-2152. 327 Kinoshita A, Urata H, Tanabe Y, Ikeda M, Tanaka H, Shindo M, Arakawa K. What types of hypertensives respond better to mild exercise therapy? Journal of Hypertension - Supplement. 1988; 6(4):S631-S633. 328 Kjeldsen SE, Lyle PA, Kizer JR, Dahlof B, Devereux RB, Julius S, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Snapinn SM, Harris KE, Wedel H, LIFE Study Group. The effects of losartan compared to atenolol on stroke in patients with isolated systolic hypertension and left ventricular hypertrophy. The LIFE study. Journal of Clinical Hypertension. 2005; 7(3):152-158. 329 Klein A. Thyroid Hormone and High Blood Pressure. In: Laragh JH, Brener BM, Kaplan NM (eds), Endocrine Mechanism in Hypertension, New York: Raven Press, 1989: 61-79. 330 Koehler E, Brown E, Haneuse J-PA. On the Assessment of Monte Carlo Error in SimulationBased Statistical Analyses. The American Statistician. 2009; 63(2):155-162. 331 Kokubo Y, Kamide K, Okamura T, Watanabe M, Higashiyama A, Kawanishi K, Okayama A, Kawano Y. Impact of high-normal blood pressure on the risk of cardiovascular disease in a Japanese urban cohort: the Suita study. Hypertension. 2008; 52(4):652-659. 332 Kono S, Kushiro T, Hirata Y, Hamada C, Takahashi A, Yoshida Y. Class of antihypertensive drugs, blood pressure status, and risk of cardiovascular disease in hypertensive patients: A casecontrol study in Japan. Hypertension Research. 2005; 28(10):811-817. 333 Koopman H, Deville W, van Eijk JT, Donker AJ, Spreeuwenberg C. Diet or diuretic? Treatment of newly diagnosed mild to moderate hypertension in the elderly. Journal of Human Hypertension. 1997; 11(12):807-812. 334 Korhonen M, Kastarinen M, Uusitupa M, Puska P, Nissinen A. The effect of intensified diet counseling on the diet of hypertensive subjects in primary health care: a 2-year open randomized controlled trial of lifestyle intervention against hypertension in eastern Finland. Preventive Medicine. 2003; 36(1):8-16. 335 Kostis JB, Davis BR, Cutler J, Grimm RH, Jr., Berge KG, Cohen JD, Lacy CR, Perry HM, Jr., Blaufox MD, Wassertheil-Smoller S, Black HR, Schron E, Berkson DM, Curb JD, Smith WM, McDonald R, Applegate WB. Prevention of heart failure by antihypertensive drug treatment in older persons with isolated systolic hypertension. SHEP Cooperative Research Group. JAMA. 1997; 278(3):212-216. 336 Kostis JB, Espeland MA, Appel L, Johnson KC, Pierce J, Wofford JL. Does withdrawal of antihypertensive medication increase the risk of cardiovascular events? Trial of Nonpharmacologic Interventions in the Elderly (TONE) Cooperative Research Group. American Journal of Cardiology. 1998; 82(12):1501-1508. 337 Kostis JB, Rosen RC, Brondolo E, Taska L, Smith DE, Wilson AC. Superiority of nonpharmacologic therapy compared to propranolol and placebo in men with mild hypertension: a randomized, prospective trial. American Heart Journal. 1992; 123(2):466-474. 338 Krakoff LR. Cost-effectiveness of ambulatory blood pressure: a reanalysis. Hypertension. 2006; 47(1):29-34.
294
Hypertension (partial update) Reference list 339 Kreeft JH, Langlois S, Ogilvie RI. Comparative trial of indapamide and hydrochlorothiazide in essential hypertension, with forearm plethysmography. Journal of Cardiovascular Pharmacology. 1984; 6(4):622-626. 340 Kshirsagar AV, Carpenter M, Bang H, Wyatt SB, Colindres RE. Blood pressure usually considered normal is associated with an elevated risk of cardiovascular disease. American Journal of Medicine. 2006; 119(2):133-141. 341 Kukkonen K, Rauramaa R, Voutilainen E, Lansimies E. Physical training of middle-aged men with borderline hypertension. Annals of Clinical Research. 1982; 14 Suppl 34:139-145. 342 Kumanyika SK, Adams-Campbell L, Van HB, Ten Have TR, Treu JA, Askov E, Williams J, Achterberg C, Zaghloul S, Monsegu D, Bright M, Stoy DB, Malone-Jackson M, Mooney D, Deiling S, Caulfield J. Outcomes of a cardiovascular nutrition counseling program in African-Americans with elevated blood pressure or cholesterol level. Journal of the American Dietetic Association. 1999; 99(11):1380-1391. 343 Kuwajima I, Kuramoto K, Ogihara T, Iimura O, Abe K, Saruta T, et al. Tolerability and safety of a calcium channel blocker in comparison with a diuretic in the treatment of elderly patients with hypertension: Secondary analysis of the NICS-EH. Hypertension Research. 2001; 24:475-480. 344 Kyngas H, Lahdenpera T. Compliance of patients with hypertension and associated factors. Journal of Advanced Nursing. 1999; 29(4):832-839. 345 LaGrone R, Jeffrey TB, Ferguson CL. Effects of education and relaxation training with essential hypertension patients. Journal of Clinical Psychology. 1988; 44(2):271-276. 346 Lahdenpera T, Kyngas H. Compliance and its evaluation in patients with hypertension. Journal of Clinical Nursing. 2000; 9:826-833. 347 Lane DA, Shah S, Beevers DG. Low-dose spironolactone in the management of resistant hypertension: A surveillance study. Journal of Hypertension. 2007; 25(4):891-894. 348 Lang T, Nicaud V, Darne B, Rueff B. Improving hypertension control among exssive alcohol drinkers: a randomised controlled trial in France. The Walpa Group. Journal of Epidemiology and Community Health. 1995; 49(610):616. 349 Langford HG, Blaufox MD, Oberman A, Hawkins CM, Curb JD, Cutter GR, Wassertheil-Smoller S, Pressel S, Babcock C, Abernethy JD, . Dietary therapy slows the return of hypertension after stopping prolonged medication. JAMA. 1985; 253(5):657-664. 350 Law MR, Frost CD, Wald NJ. By how much does dietary salt reduction lower blood pressure? III--Analysis of data from trials of salt reduction. BMJ. 1991; 302(6780):819-824. 351 Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugs in the prevention of cardiovascular disease: Meta-analysis of 147 randomised trials in the context of expectations from prospective epidemiological studies. BMJ. 2009; 338(7705):1245. 352 Law MR, Morris JK and Wald NJ. Unpublished Data From the Published Study: Law MR, Morris JK, Wald NJ. Use of Blood Pressure Lowering Drugs in the Prevention of Cardiovascular Disease: Meta-Analysis of 147 Randomised Trials in the Context of Expectations From Prospective Epidemiological Studies. BMJ. 2009; 338(7705):1245. Personal communication: 22/11/10 353 Lede RL, Voto LS, Orti J, Margulies M. Agreement between different frequencies of measurements in ambulatory blood pressure monitoring. Journal of Obstetrics and Gynaecology. 1997; 17(4):337-339.
295
Hypertension (partial update) Reference list 354 Leenen FH, Nwachuku CE, Black HR, Cushman WC, Davis BR, Simpson LM, Alderman MH, Atlas SA, Basile JN, Cuyjet AB, Dart R, Felicetta JV, Grimm RH, Haywood LJ, Jafri SZ, Proschan MA, Thadani U, Whelton PK, Wright JT, Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial Collaborative Research Group. Clinical events in high-risk hypertensive patients randomly assigned to calcium channel blocker versus angiotensin-converting enzyme inhibitor in the antihypertensive and lipid-lowering treatment to prevent heart attack trial. Hypertension. 2006; 48(3):374-384. 355 Leiter LA, Abbott D, Campbell NR, Mendelson R, Ogilvie RI, Chockalingam A. Lifestyle modifications to prevent and control hypertension. 2. Recommendations on obesity and weight loss. Canadian Hypertension Society, Canadian Coalition for High Blood Pressure Prevention and Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke Foundation of Canada. Canadian Medical Association Journal. 1999; 160(9 Suppl):S7-12. 356 Leren P, Helgeland A. Oslo Hypertension Study. Drugs. 1986; 31 Suppl 1:41-45.
357 Levin ML, Goldstein H, Gerhardt PR. Cancer and tobacco smoking; a preliminary report. JAMA. 1950; 143(4):336-338. 358 Levine DM, Cohen JD, Dustan HP, Falkner B, Flora JA, Lefebvre RC, Morisky DE, Oberman A, Pickering TG, Roccella EJ, . Behavior changes and the prevention of high blood pressure. Workshop II. AHA Prevention Conference III. Behavior change and compliance: keys to improving cardiovascular health. Circulation. 1993; 88(3):1387-1390. 359 Levinson PD, Khatri IM, Freis ED. Persistence of normal BP after withdrawal of drug treatment in mild hypertension. Archives of Internal Medicine. 1982; 142(13):2265-2268. 360 Levy A, Lightman SL. Diagnosis and management of pituitary tumours. BMJ. 1994; 308(6936):1087-1091. 361 Lewington S, Clarke R, Qizilbash N, Peto R, Collins R. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002; 360(9349):1903-1913. 362 Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, Ritz E, Atkins RC, Rohde R, Raz I. Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. New England Journal of Medicine. 2001; 345(12):851-860. 363 Li Y, Boggia J, Thijs L, Hansen TW, Kikuya M, Bjorklund-Bodegard K, Richart T, Ohkubo T, Kuznetsova T, Torp-Pedersen C, Lind L, Ibsen H, Imai Y, Wang J, Sandoya E, O'Brien E, Staessen JA, International Database on Ambulatory Blood Pressure Monitoring in relation to Cardiovascular Outcomes Investigators. Is blood pressure during the night more predictive of cardiovascular outcome than during the day? Blood Pressure Monitoring. 2008; 13(3):145-147. 364 Lim PO, Rodgers P, Cardale K, Watson AD, MacDonald TM. Potentially high prevalence of primary aldosteronism in a primary-care population. Lancet. 1999; 353(9146):40. 365 Lind L, Lithell H, Pollare T, Ljunghall S. Blood pressure response during long-term treatment with magnesium is dependent on magnesium status. A double-blind, placebo-controlled study in essential hypertension and in subjects with high-normal blood pressure. American Journal of Hypertension. 1991; 4(8):674-679. 366 Linden W, Chambers L. Clinical effectiveness of non-drug treatment for hypertension: a meta-analysis. Annals of Behavioral Medicine. 1994; 16:35-45. 367 Linden W, Lenz JW, Con AH. Individualized stress management for primary hypertension: a randomized trial. Archives of Internal Medicine. 2001; 161(8):1071-1080.
296
Hypertension (partial update) Reference list 368 Lindholm LH, Hansson L, Ekbom T, Dahlf B, Lanke J, Linjer E, et al. Comparison of antihypertensive treatments in preventing cardiovascular events in elderly diabetic patients: results from the Swedish trial in old patients with hypertension -2. Journal of Hypertension. 2000; 18:16711675. 369 Lindholm LH, Ibsen H, Borch-Johnsen K, Olsen MH, Wachtell K, Dahlof B, Devereux RB, Beevers G, de Faire U, Fyhrquist F, Julius S, Kjeldsen SE, Kristianson K, Lederballe-Pedersen O, Nieminen MS, Omvik P, Oparil S, Wedel H, Aurup P, Edelman JM, Snapinn S, For the LIFE study group. Risk of new-onset diabetes in the Losartan Intervention For Endpoint reduction in hypertension study. Journal of Hypertension. 2002; 20(9):1879-1886. 370 Lindholm LH, Ibsen H, Dahlf B, Devereux RB, Beevers G, de Faire U, et al. Cardiovascular morbidity and mortality in patients with diabetes in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against Atenolol. Lancet. 2002; 359:1004-1010. 371 Lithell H, Hansson L, Skoog I, Elmfeldt D, Hofman A, Olofsson B, Trenkwalder P, Zanchetti A. The Study on Cognition and Prognosis in the Elderly (SCOPE): principal results of a randomized double-blind intervention trial. Journal of Hypertension. 2003; 21(5):875-886. 372 Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, Mant D. Comparison of agreement between different measures of blood pressure in primary care and daytime ambulatory blood pressure. BMJ. 2002; 325(7358):254. 373 Liu L, Wang JG, Gong L, Liu G, Staessen JA. Comparison of active treatment and placebo in older Chinese patients with isolated systolic hypertension. Systolic Hypertension in China (Syst-China) Collaborative Group. Journal of Hypertension. 1998; 16(12 Pt 1):1823-1829. 374 Lorgelly P, Siatis I, Brooks A, Slinn B, Millar-Craig MW, Donnelly R, Manning G. Is ambulatory blood pressure monitoring cost-effective in the routine surveillance of treated hypertension patients in primary care? British Journal of General Practice. 2003; 53:794-796. 375 Luengo-Fernandez R, Gray AM, Rothwell PM. Costs of stroke using patient-level data: a critical review of the literature. Stroke. 2009; 40(2):e18-e23. 376 Luft FC, Morris CD, Weinburger MH. Compliance to a low sodium diet. American Journal of Clinical Nutrition. 1997; 65:S698-S703. 377 Luscher TF, Vetter H, Siegenthaler W, Vetter W. Compliance in hypertension: facts and concepts. Journal of Hypertension - Supplement. 1985; 3(1):S3-S9. 378 Lyle RM. Does baseline serum total calcium level influence the blood pressure response to calcium supplementation? A double-blind study. Netherlands Journal of Medicine. 1992; 41(1-2):4855. 379 Macmahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet. 1990; 335(8692):765-774. 380 MacMahon SW, Macdonald GJ, Bernstein L, Andrews G, Blacket RB. A randomized controlled trial of weight reduction and metoprolol in the treatment of hypertension in young overweight patients. Clinical and Experimental Pharmacology and Physiology. 1985; 12(3):267-271. 381 MacMahon SW, Macdonald GJ, Bernstein L, Andrews G, Blacket RB. Comparison of weight reduction with metoprolol in treatment of hypertension in young overweight patients. Lancet. 1985; 1(8440):1233-1236.
297
Hypertension (partial update) Reference list 382 Maheswaran R, Beevers M, Beevers DG. Effectiveness of advice to reduce alcohol consumption in hypertensive patients. Hypertension. 1992; 19(1):79-84. 383 Mahmud A, Mahgoub M, Hall M, Feely J. Does aldosterone-to-renin ratio predict the antihypertensive effect of the aldosterone antagonist spironolactone? American Journal of Hypertension. 2005; 18(12):1631-1635. 384 Malacco E, Mancia G, Rappelli A, Menotti A, Zuccaro MS, Coppini A, Shell I. Treatment of isolated systolic hypertension: the SHELL study results. Blood Pressure. 2003; 12(3):160-167. 385 Mallion JM, Genes N, Vaur L, Clerson P, Vaisse B, Bobrie G, Chatellier G. Blood pressure levels, risk factors and antihypertensive treatments: lessons from the SHEAF study. Journal of Human Hypertension. 2001; 15(12):841-848. 386 Mancia G, Omboni S, Parati G, Trazzi S, Mutti E. Limited reproducibility of hourly blood pressure values obtained by ambulatory blood pressure monitoring: Implications for studies on antihypertensive drugs. Journal of Hypertension. 1992; 10(12):1531-1535. 387 Mancia G, Parati G. Office compared with ambulatory blood pressure in assessing response to anti hypertensive treatment: a meta-analysis. Journal of Hypertension. 2004; 22(3):435-445. 388 Mandle CL, Jacobs SC, Arcari PM, Domar AD. The efficacy of relaxation response interventions with adult patients: a review of the literature. Journal of Cardiovascular Nursing. 1996; 10(3):4-26. 389 Mansoor GA, McCabe EJ, White WB. Long-term reproducibility of ambulatory blood pressure. Journal of Hypertension. 1994; 12(6):703-708. 390 Mar J, Pastor R, Abasolo R, Ruiz de GR. Ambulatory blood pressure monitoring and diagnostic errors in hypertension: a Bayesian approach. Medical Decision Making. 1998; 18(4):429-435. 391 Martin JE, Dubbert PM, Cushman WC. Controlled trial of aerobic exercise in hypertension. Circulation. 1990; 81(5):1560-1567. 392 Mason, J. and Eccles, M. Guideline Recommendation an Evidence Grading (GREG): a New Grading Method for Clinical Guideline Development Groups. Volume Report 109. Newcastle upon Tyne: Centre for Health Services Research, University of Newcastle upon Tyne, 2003. 393 Mason J, Eccles M, Freemantle N, Drummond M. A framework for incorporating costeffectiveness in evidence-based clinical practice guidelines. Health Policy. 1999; 47(1):37-52. 394 Materson BJ, Reda DJ, Cushman WC, Massie BM, Freis ED, Kochar MS, Hamburger RJ, Fye C, Lakshman R, Gottdiener J. Single-drug therapy for hypertension in men. A comparison of six antihypertensive agents with placebo. The Department of Veterans Affairs Cooperative Study Group on Antihypertensive Agents. New England Journal of Medicine. 1993; 328(13):914-921. 395 Maxwell MH, Kushiro T, Dornfeld LP, Tuck ML, Waks AU. BP changes in obese hypertensive subjects during rapid weight loss. Comparison of restricted v unchanged salt intake. Archives of Internal Medicine. 1984; 144(8):1581-1584. 396 McCarron DA, Morris CD. Blood pressure response to oral calcium in persons with mild to moderate hypertension. A randomized, double-blind, placebo-controlled, crossover trial. Annals of Internal Medicine. 1985; 103(6 ( Pt 1)):825-831. 397 McGrady A. Effects of group relaxation training and thermal biofeedback on blood pressure and related physiological and psychological variables in essential hypertension. Biofeedback and Self Regulation. 1994; 19(1):51-66.
298
Hypertension (partial update) Reference list 398 McGrady AV, Yonker R, Tan SY, Fine TH, Woerner M. The effect of biofeedback-assisted relaxation training on blood pressure and selected biochemical parameters in patients with essential hypertension. Biofeedback and Self Regulation. 1981; 6(3):343-353. 399 McKeigue PM, Marmot MG, Syndercombe Court YD, Cottier DE, Rahman S, Riemersma RA. Diabetes, hyperinsulinaemia, and coronary risk factors in Bangladeshis in east London. British Heart Journal. 1988; 60(5):390-396. 400 McKeigue PM, Shah B, Marmot MG. Relation of central obesity and insulin resistance with high diabetes prevalence and cardiovascular risk in South Asians. Lancet. 1991; 337(8738):382-386. 401 McManus RJ, Mant J, Bray EP, Holder R, Jones MI, Greenfield S, Kaambwa B, Banting M, Bryan S, Little P, Williams B, Hobbs FD. Telemonitoring and self-management in the control of hypertension (TASMINH2): a randomised controlled trial. Lancet. 2010; 376(9736):163-172. 402 Medical Research Council Working Party. MRC trial of treatment of mild hypertension: principal results. BMJ. 1985; 291(6488):97-104. 403 Melsop KA, Boothroyd DB, Hlatky MA. Quality of life and time trade-off utility measures in patients with coronary artery disease. American Heart Journal. 2003; 145(1):36-41. 404 Mesquita-Bastos J, Bertoquini S, Polonia J. Cardiovascular prognostic value of ambulatory blood pressure monitoring in a Portuguese hypertensive population followed up for 8.2 years. Blood Pressure Monitoring. 2010; 15(5):240-246. 405 Metoki H, Ohkubo T, Kikuya M, Asayama K, Obara T, Hara A, Hirose T, Hashimoto J, Totsune K, Hoshi H, Satoh H, Imai Y. Prognostic significance of night-time, early morning, and daytime blood pressures on the risk of cerebrovascular and cardiovascular mortality: the Ohasama Study. Journal of Hypertension. 2006; 24(9):1841-1848. 406 Metz JA, Stern JS, Kris-Etherton P, Reusser ME, Morris CD, Hatton DC, Oparil S, Haynes RB, Resnick LM, Pi-Sunyer FX, Clark S, Chester L, McMahon M, Snyder GW, McCarron DA. A randomized trial of improved weight loss with a prepared meal plan in overweight and obese patients: impact on cardiovascular risk reduction. Archives of Internal Medicine. 2000; 160(14):2150-2158. 407 Midgley JP, Matthew AG, Greenwood CM, Logan AG. Effect of reduced dietary sodium on blood pressure: a meta-analysis of randomized controlled trials. JAMA. 1996; 275(20):1590-1597. 408 Miller ER, III, Erlinger TP, Young DR, Jehn M, Charleston J, Rhodes D, Wasan SK, Appel LJ. Results of the Diet, Exercise, and Weight Loss Intervention Trial (DEW-IT). Hypertension. 2002; 40(5):612-618. 409 Miller JM, Miller JM. Diabetes mellitus and hypertension in black and white populations. Southern Medical Journal. 1986; 79(10):1229. 410 Mills CA, Porter MM. Tobacco smoking habits and cancer of the mouth and respiratory system. Cancer Research. 1950; 10(9):539-542. 411 Mitchell A, Haynes RB, Adsett CA, Bellissimo A, Wilczynski N. The likelihood of remaining normotensive following antihypertensive drug withdrawal. Journal of General Internal Medicine. 1989; 4(3):221-225. 412 Montgomery AA, Fahey T, Ben-Shlomo Y, Harding J. The influence of absolute cardiovascular risk, patient utilities, and costs on the decision to treat hypertension: a Markov decision analysis. Journal of Hypertension. 2003; 21(9):1753-1759. 413 Morgan T, Hopper J, Anderson A, Carricks L, Jones E, Johns J. Can drug therapy be stopped in elderly hypertensive patients? Cardiology in the Elderly. 1994; 2:119-125.
299
Hypertension (partial update) Reference list 414 Morris CD, Karanja N, McCarron DA. Dietary versus supplemental calcium to reduce blood pressure. Clinical Research. 1988; 36:A139. 415 Mulrow CD, Chiquette E, Angel L, Cornell J, Summerbell C, Anagnostelis B, Grimm R, Jr., Brand MB. Dieting to reduce body weight for controlling hypertension in adults. Cochrane Database of Systematic Reviews. 2000;(2):CD000484. 416 Murakami S, Otsuka K, Kubo Y, Shinagawa M, Yamanaka T, Ohkawa S-I, Kitaura Y. Repeated ambulatory monitoring reveals a Monday morning surge in blood pressure in a community-dwelling population. American Journal of Hypertension. 2004; 17(12):1179-1183. 417 Murchie P, Campbell NC, Ritchie LD, Simpson JA, Thain J. Secondary prevention clinics for coronary heart disease: four year follow up of a randomised controlled trial in primary care. BMJ. 2003; 326(7380):84. 418 Murugesan R, Govindarajulu N, Bera TK. Effect of selected yogic practices on the management of hypertension. Indian Journal of Physiology and Pharmacology. 2000; 44(2):207-210. 419 Musini VM, Tejani AM, Bassett K, Wright JM. Pharmacotherapy for hypertension in the elderly. Cochrane Database of Systematic Reviews. 2009; Issue 4:CD000028. 420 Musso NR, Vergassola C, Barone C, Lotti G. Ambulatory blood pressure monitoring: how reproducible is it? American Journal of Hypertension. 1997; 10(8):936-939. 421 Myers MG, Reeves RA, Oh PI, Joyner CD. Overtreatment of hypertension in the community? American Journal of Hypertension. 1996; 9(5):419-425. 422 National Centre for Social Research and University College London.Department of Epidemiology and Public Health.Health Survey for England, 2006 [Computer File]. 3rd ed. Colchester: Essex: UK Data Archive [distributor]; 2010 423 National Clinical Guideline Centre. Unstable Angina and NSTEMI: the Early Management of Unstable Angina and Non-ST-Segment-Elevation Myocardial Infarction. (CG94). London: Royal College of Physicians, 2010. 424 National Clinical Guideline Centre for Acute and Chronic Conditions. Management of Stable Angina: NICE Guideline: Draft for Consultation, December 2010. 2010. London, National Clinical Guideline Centre for Acute and Chronic Conditions. http://www.nice.org.uk/nicemedia/live/11878/52141/52141.pdf 425 National Collaborating Centre for Chronic Conditions.Hypertension: Management in Adults in Primary Care: Pharmacological Update. London: Royal College of Physicians; 2006. Available from: http://guidance.nice.org.uk/CG34 426 National Collaborating Centre for Primary Care.Medicines Adherence - Involving Patients in Decisions About Prescribed Medicines and Supporting Adherence. National Clinical Guideline Number 76. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners; 2009. Available from: http://guidance.nice.org.uk/CG76 427 National Institute for Health and Clinical Excellence. Guide to the Methods of Technology Appraisal. London: National Institute for Health and Clinical Excellence, 2008. 428 National Institute for Health and Clinical Excellence. Lipid Modification: Cardiovascular Risk Assessment and the Modification of Blood Lipids for the Primary and Secondary Prevention of Cardiovascular Disease. London: National Institute for Health and Clinical Excellence, 2008. 429 National Institute for Health and Clinical Excellence. Social Value Judgements. Principles for the development of NICE guidance. Second edition. Available from:
300
Hypertension (partial update) Reference list http://www.nice.org.uk/media/C18/30/SVJ2PUBLICATION2008.pdf Last accessed on: 15 November 2010. 430 National Institute for Health and Clinical Excellence.The Guidelines Manual. London: National Institute for Health and Clinical Excellence; 2009. Available from: http://www.nice.org.uk/media/5F2/44/The_guidelines_manual_2009_-_All_chapters.pdf 431 National Institute for Health and Clinical Excellence. Chest Pain of Recent Onset: Assessment and Diagnosis of Recent Onset Chest Pain or Discomfort of Suspected Cardiac Origin. London: National Institute for Health and Clinical Excellence, 2010. 432 National Institute for Health and Clinical Excellence. Clopidogrel and Modified-Release Dipyridamole for the Prevention of Occlusive Vascular Events: Review of NICE Technology Appraisal Guidance 90. London: National Institute for Health and Clinical Excellence, 2010. 433 Nelson M. Author's reply to two responses. BMJ. 2002;
434 Nelson M, Reid C, Krum H, McNeil J. A systematic review of predictors of maintenance of normotension after withdrawal of antihypertensive drugs. American Journal of Hypertension. 2001; 14(2):98-105. 435 Nelson MR, Reid CM, Krum H, Muir T, Ryan P, McNeil JJ. Predictors of normotension on withdrawal of antihypertensive drugs in elderly patients: prospective study in second Australian national blood pressure study cohort. BMJ. 2002; 325(7368):815. 436 Newcastle Guideline Development and Research Unit. Hypertension: Management of Hypertension in Adults in Primary Care. London: National Institute for Clinical Excellence, 2004. 437 Newell-Price J, Trainer P, Besser M, Grossman A. The diagnosis and differential diagnosis of Cushing's syndrome and pseudo-Cushing's states. Endocrinology Review. 1998; 19(5):647-672. 438 Niiranen TJ, Hanninen MR, Johansson J, Reunanen A, Jula AM. Home-measured blood pressure is a stronger predictor of cardiovascular risk than office blood pressure: the Finn-Home study. Hypertension. 2010; 55(6):1346-1351. 439 Niiranen TJ, Kantola IM, Vesalainen R, Johansson J, Ruuska MJ. A comparison of home measurement and ambulatory monitoring of blood pressure in the adjustment of antihypertensive treatment. American Journal of Hypertension. 2006; 19(5):468-474. 440 North of England Evidence-based Guidelines Development Project. Primary Care Management of Secondary Prophylaxis for Patients Who Have Experienced a Myocardial Infarction: Drug Treatment, Cardiac Rehabilitation and Dietary Manipulation. Newcastle: University of Newcastle, Centre for Health Services Research, 2001. 441 North of England Hypertension Guideline Development Group. Essential Hypertension: Managing Adult Patients in Primary Care. Newcastle upon Tyne: Cetnre for Health Services Research, 2004. 442 Nowson C, Morgan T. Effect of calcium carbonate on blood pressure in normotensive and hypertensive people. Hypertension. 1989; 13(6 Pt 1):630-639. 443 Nowson CA, Morgan TO. Magnesium supplementation in mild hypertensive patients on a moderately low sodium diet. Clinical and Experimental Pharmacology and Physiology. 1989; 16(4):299-302. 444 Nurminen ML, Niittynen L, Korpela R, Vapaatalo H. Coffee, caffeine and blood pressure: a critical review. European Journal of Clinical Nutrition. 1999; 53(11):831-839.
301
Hypertension (partial update) Reference list 445 O'Brien E. Replacing the mercury sphygmomanometer. Requires clinicians to demand better automated devices. BMJ. 2000; 320(7238):815-816. 446 O'Brien E. State of the market for devices for blood pressure measurement. Blood Pressure Monitoring. 2001; 6(6):281-286. 447 O'Brien E, Atkins N. A comparison of the British Hypertension Society and Association for the Advancement of Medical Instrumentation protocols for validating blood pressure measuring devices: can the two be reconciled? Journal of Hypertension. 1994; 12(9):1089-1094. 448 O'Brien E, Beevers G, Lip G. ABC of hypertension: Blood pressure measurement. Part III. BMJ. 2001; 322(7294):1110-1114. 449 O'Brien E, Beevers G, Lip GY. ABC of hypertension: Blood pressure measurement. Part IVautomated sphygmomanometry: self blood pressure measurement. BMJ. 2001; 322(7295):11671170. 450 O'Brien E, Coats A, Owens P, Petrie J, Padfield PL, Littler WA, de SM, Mee F. Use and interpretation of ambulatory blood pressure monitoring: recommendations of the British hypertension society. BMJ. 2000; 320(7242):1128-1134. 451 O'Brien E, Petrie J, Littler W, de SM, Padfield PL, Altman DG, Bland M, Coats A, Atkins N. An outline of the revised British Hypertension Society protocol for the evaluation of blood pressure measuring devices. Journal of Hypertension. 1993; 11(6):677-679. 452 O'Brien E, Waeber B, Parati G, Staessen J, Myers MG. Blood pressure measuring devices: recommendations of the European Society of Hypertension. BMJ. 2001; 322(7285):531-536. 453 O'Malley K, McCormack P, O'Brien ET. Isolated systolic hypertension: data from the European Working Party on High Blood Pressure in the Elderly. Journal of Hypertension - Supplement. 1988; 6(1):S105-S108. 454 Obara F, Saitoh S, Takagi S, Shimamoto K. Influence of hypertension on the incidence of cardiovascular disease in two rural communities in Japan: the Tanno-Sobetsu [corrected] study. Hypertension Research - Clinical and Experimental. 2007; 30(8):677-682. 455 Obel AO. Placebo-controlled trial of potassium supplements in black patients with mild essential hypertension. Journal of Cardiovascular Pharmacology. 1989; 14(2):294-296. 456 Octavio JA, Contreras J, Amair P, Octavio B, Fabiano D, Moleiro F, Omboni S, Groppelli A, Bilo G, Mancia G, Parati G. Time-weighted vs. conventional quantification of 24-h average systolic and diastolic ambulatory blood pressures. Journal of Hypertension. 2010; 28(3):459-464. 457 Office for National Statistics. Mortality Statistcs: Cause: Review of the Registrar General on Deaths by Cause, Sex and Age in England and Wales, 2000. London: Office of National Statistics, 2001. 458 Office for National Statistics. Life Tables (Online Edition). 2011. London, Office for National Statistics. 459 Office for National Statistics (ONS). England and Wales, Interim Life Tables, 1980-82 to 200709. Available from: http://www.statistics.gov.uk/StatBase/Product.asp?vlnk=14459 Last accessed on: 21 October 2010. 460 Office for National Statistics (ONS). Mortality statistics: deaths registered in 2008. Available from: http://www.statistics.gov.uk/statbase/product.asp?vlnk=15096 Last accessed on: 21 October 2010.
302
Hypertension (partial update) Reference list 461 Ogedegbe G, Pickering TG, Clemow L, Chaplin W, Spruill TM, Albanese GM, Eguchi K, Burg M, Gerin W. The misdiagnosis of hypertension: The role of patient anxiety. Archives of Internal Medicine. 2008; 168(22):2459-2465. 462 Ogihara T, Saruta T, Rakugi H, Fujimoto A, Ueshima K, Yasuno S, Oba K, Takeda K, Higaki J, Nakao K. Relationship between the achieved blood pressure and the incidence of cardiovascular events in Japanese hypertensive patients with complications: A sub-analysis of the CASE-J trial. Hypertension Research. 2009; 32(4):248-254. 463 Ogihara T, Saruta T, Rakugi H, Matsuoka H, Shimamoto K, Shimada K, Imai Y, Kikuchi K, Ito S, Eto T, Kimura G, Imaizumi T, Takishita S, Ueshima H, Valsartan in Elderly Isolated Systolic Hypertension Study Group. Target blood pressure for treatment of isolated systolic hypertension in the elderly: valsartan in elderly isolated systolic hypertension study. Hypertension. 2010; 56(2):196202. 464 Ohkubo T, Asayama K, Kikuya M, Metoki H, Hoshi H, Hashimoto J, Totsune K, Satoh H, Imai Y, Ohasama S. How many times should blood pressure be measured at home for better prediction of stroke risk? Ten-year follow-up results from the Ohasama study. Journal of Hypertension. 2004; 22(6):1099-1104. 465 Ohkubo T, Imai Y, Tsuji I, Nagai K, Watanabe N, Minami N, Itoh O, Bando T, Sakuma M, Fukao A, Satoh H, Hisamichi S, Abe K. Prediction of mortality by ambulatory blood pressure monitoring versus screening blood pressure measurements: a pilot study in Ohasama. Journal of Hypertension. 1997; 15(4):357-364. 466 Okayama A, Kadowaki T, Okamura T, Hayakawa T, Ueshima H, NIPPON DATA. Age-specific effects of systolic and diastolic blood pressures on mortality due to cardiovascular diseases among Japanese men (NIPPON DATA80). Journal of Hypertension. 2006; 24(3):459-462. 467 74. Onusko E. Diagnosing secondary hypertension. American Family Physician. 2003; 67(1):67-
468 Organisation for Economic Co-operation and Development (OECD). OECD Stat Extracts: purchasing power parities for GDP. Available from: http://stats.oecd.org/Index.aspx?datasetcode=SNA_TABLE4 Last accessed on: 10 November 2010. 469 Orth DN. Cushing's syndrome. New England Journal of Medicine. 1995; 332(12):791-803.
470 Overlack A, Conrad H, Stumpe KO. The influence of oral potassium citrate/bicarbonate on blood pressure in essential hypertension during unrestricted salt intake. Klinische Wochenschrift. 1991; 69 Suppl 25:79-83. 471 Pagliari CP, Grimshaw JM, Eccles M. The potential influence of small group processes on guideline development. Journal of Evaluation in Clinical Practice. 2001; 7(2):165-173. 472 Palatini P, Julius S, Collatina S, Rappelli S, Staessen J, Pessina AC. Optimizing the assessment of the elderly patient with borderline hypertension: the Hypertension and Ambulatory Recording in the OLD (HAROLD) study. Aging. 1997; 9(5):365-371. 473 Palatini P, Mormino P, Canali C, Santonastaso M, De VG, Zanata G, Pessina AC. Factors affecting ambulatory blood pressure reproducibility. Results of the HARVEST Trial. Hypertension and Ambulatory Recording Venetia Study. Hypertension. 1994; 23(2):211-216. 474 Palmer, S, Sculpher, M, Philips, Z, Robinson, M, and et al. A Cost Effectiveness Model Comparing Alternative Management Strategies for the Use of Glycoprotein IIB/IIIA Anatagonists in Non-ST-Elevation Acute Conorary Syndrome. York: Center for Health Economics, 2002.
303
Hypertension (partial update) Reference list 475 Paolisso G, Di MG, Cozzolino D, Salvatore T, D'Amore A, Lama D, Varricchio M, D'Onofrio F. Chronic magnesium administration enhances oxidative glucose metabolism in thiazide treated hypertensive patients. American Journal of Hypertension. 1992; 5(10):681-686. 476 Parati G, Omboni S, Albini F, Piantoni L, Giuliano A, Revera M, Illyes M, Mancia G, TeleBPCare Study Group. Home blood pressure telemonitoring improves hypertension control in general practice. The TeleBPCare study. Journal of Hypertension. 2009; 27(1):198-203. 477 Patel C, Marmot M. Can general practitioners use training in relaxation and management of stress to reduce mild hypertension? BMJ. 1988; 296(6614):21-24. 478 Patel C, Marmot MG, Terry DJ. Controlled trial of biofeedback-aided behavioural methods in reducing mild hypertension. BMJ. 1981; 282(6281):2005-2008. 479 Patel C, Marmot MG, Terry DJ, Carruthers M, Hunt B, Patel M. Trial of relaxation in reducing coronary risk: four year follow up. BMJ. 1985; 290(6475):1103-1106. 480 Payne R. Excel worksheet for calculating cardiovascular risk. Available from: http://cvrisk.mvm.ed.ac.uk/calculator/excelcalc.htm 481 Pepine CJ, Handberg EM, Cooper-DeHoff RM, Marks RG, Kowey P, Messerli FH, Mancia G, Cangiano JL, Garcia-Barreto D, Keltai M, Erdine S, Bristol HA, Kolb HR, Bakris GL, Cohen JD, Parmley WW. A calcium antagonist vs a non-calcium antagonist hypertension treatment strategy for patients with coronary artery disease. The International Verapamil-Trandolapril Study (INVEST): a randomized controlled trial. JAMA. 2003; 290(21):2805-2816. 482 Perloff D, Grim C, Flack J, Frohlich ED, Hill M, McDonald M, Morgenstern BZ. Human blood pressure determination by sphygmomanometry. Circulation. 1993; 88(5 Pt 1):2460-2470. 483 Perry HM, Davis BR, Price TR, Applegate WB, Fields WS, Guralnik JM, Kuller L, Pressel S, Stamler J, Probsfieldt JL. Effect of treating isolated systolic hypertension on the risk of developing various types and subtypes of stroke: the Systolic Hypertension in the Elderly Program (SHEP). JAMA. 2000; 284:465-471. 484 Perry HM, McDonald RH, Hulley SB, Smith WM, Furberg CD, Greenlick MR. Systolic Hypertension in the Elderly Program. Pilot Study (SHEP-PS): morbidity and mortality experience. Journal of Hypertension. 1986; 4(6):S21-S23. 485 Perry HM, Jr., Smith WM, McDonald RH, Black D, Cutler JA, Furberg CD, Greenlick MR, Kuller LH, Schnaper HW, Schoenberger JA, . Morbidity and mortality in the Systolic Hypertension in the Elderly Program (SHEP) pilot study. Stroke. 1989; 20(1):4-13. 486 Personal Social Services Research Unit. Unit costs of health and social care 2010. Available from: http://www.pssru.ac.uk/uc/uc2010contents.htm Last accessed on: 23 December 10 A.D. 487 Petersen LJ, Rudnicki M, Hojsted J. Long-term oral calcium supplementation reduces diastolic blood pressure in end stage renal disease. A randomized, double-blind, placebo controlled study. International Journal of Artificial Organs. 1994; 17(1):37-40. 488 Peto R, Darby S, Deo H, Silcocks P, Whitley E, Doll R. Smoking, smoking cessation, and lung cancer in the UK since 1950: combination of national statistics with two case-control studies. BMJ. 2000; 321(7257):323-329. 489 Petrella RJ. How effective is exercise training for the treatment of hypertension? Clinical Journal of Sport Medicine. 1998; 8(3):224-231.
304
Hypertension (partial update) Reference list 490 Pickering T. Recommendations for the use of home (self) and ambulatory blood pressure monitoring. American Society of Hypertension Ad Hoc Panel. American Journal of Hypertension. 1996; 9(1):1-11. 491 Pickering T, Schwartz J, Verdecchia P, Imai Y, Kario K, Eguchi K, Pierdomenico S, Ohkubo T, Wing L. Prediction of strokes versus cardiac events by ambulatory monitoring of blood pressure: results from an international database. Blood Pressure Monitoring. 2007; 12(6):397-399. 492 Piller LB, Ford CE, Davis BR, Nwachuku C, Black HR, Oparil S, Retta TM, Probstfield JL, ALLHAT Collaborative Research Group. Incidence and predictors of angioedema in elderly hypertensive patients at high risk for cardiovascular disease: a report from the Antihypertensive and LipidLowering Treatment to Prevent Heart Attack Trial (ALLHAT). Journal of Clinical Hypertension. 2006; 8(9):649-656. 493 Plante GE, Dessurault DL. Hypertension in elderly patients. A comparative study between indapamide and hydrochlorothiazide. American Journal of Medicine. 1988; 84(1B):98-103. 494 Plante GE, Robillard C. Indapamide in the treatment of essential arterial hypertension: results of a controlled study. Current Medical Research and Opinion. 1983; 8 Suppl 3:59-66. 495 Poppitt SD, Keogh GF, Prentice AM, Williams DE, Sonnemans HM, Valk EE, Robinson E, Wareham NJ. Long-term effects of ad libitum low-fat, high-carbohydrate diets on body weight and serum lipids in overweight subjects with metabolic syndrome. American Journal of Clinical Nutrition. 2002; 75(1):11-20. 496 Poulter NR, Khaw KT, Hopwood BE, Mugambi M, Peart WS, Rose G, Sever PS. The Kenyan Luo migration study: observations on the initiation of a rise in blood pressure. BMJ. 1990; 300(6730):967972. 497 Power C, Rodgers B, Hope S. U-shaped relation for alcohol consumption and health in early adulthood and implications for mortality. Lancet. 1998; 352(9131):877. 498 Prisco D, Paniccia R, Bandinelli B, Filippini M, Francalanci I, Giusti B, Giurlani L, Gensini GF, Abbate R, Neri Serneri GG. Effect of medium-term supplementation with a moderate dose of n-3 polyunsaturated fatty acids on blood pressure in mild hypertensive patients. Thrombosis Research. 1998; 91(3):105-112. 499 Pritchard DA, Hyndman J, Taba F. Nutritional counselling in general practice: a cost effective analysis. Journal of Epidemiology and Community Health. 1999; 53(5):311-316. 500 PROGRESS Collaborative Group. Randomised trial of a perindopril-based blood-pressurelowering regimen among 6,105 individuals with previous stroke or transient ischaemic attack. Lancet. 2001; 358(9287):1033-1041. 501 Psaty BM, Smith NL, Siscovick DS, Koepsell TD, Weiss NS, Heckbert SR, Lemaitre RN, Wagner EH, Furberg CD. Health outcomes associated with antihypertensive therapies used as first-line agents. A systematic review and meta-analysis. JAMA. 1997; 277(9):739-745. 502 Puddey IB, Parker M, Beilin LJ, Vandongen R, Masarei JR. Effects of alcohol and caloric restrictions on blood pressure and serum lipids in overweight men. Hypertension. 1992; 20(4):533541. 503 Rakic D, Rumboldt Z, Bagatin J, Polic S. Effects of four antihypertensive monotherapies on cardiac mass and function in hypertensive patients with left ventricular hypertrophy: randomized prospective study. Croatian Medical Journal. 2002; 43(6):672-679.
305
Hypertension (partial update) Reference list 504 Raleigh VS. Diabetes and hypertension in britain's ethnic minorities: implications for the future of renal services. BMJ. 1997; 314(7075):209. 505 Ramsay L, Williams B, Johnston G, MacGregor G, Poston L, Potter J, Poulter N, Russell G. Guidelines for management of hypertension: report of the third working party of the British Hypertension Society. Journal of Human Hypertension. 1999; 13(9):569-592. 506 Redon J, Campos C, Narciso ML, Rodicio JL, Pascual JM, Ruilope LM. Prognostic value of ambulatory blood pressure monitoring in refractory hypertension: a prospective study. Hypertension. 1998; 31(2):712-718. 507 Reims HM, Oparil S, Kjeldsen SE, Devereux RB, Julius S, Brady WE, Fyhrquist F, Ibsen H, Lindholm LH, Omvik P, Wedel H, Beevers G, de Faire U, Kristianson K, Lederballe-Pedersen O, Nieminen MS, Dahlof B, LIFE Study Group. Losartan benefits over atenolol in non-smoking hypertensive patients with left ventricular hypertrophy: the LIFE study. Blood Pressure. 2004; 13(6):376-384. 508 Reisin E, Abel R, Modan M, Silverberg DS, Eliahou HE, Modan B. Effect of weight loss without salt restriction on the reduction of blood pressure in overweight hypertensive patients. New England Journal of Medicine. 1978; 298(1):1-6. 509 Richards AM, Nicholls MG, Espiner EA, Ikram H, Turner JG, Brownlie BE. Hypertension in hypothyroidism: arterial pressure and hormone relationships. Clinical and Experimental Hypertension Part A: Theory and Practice. 1985; 7(11):1499-1514. 510 Rivas M, Garay RP, Escanero JF, Cia P, Jr., Cia P, Alda JO. Soy milk lowers blood pressure in men and women with mild to moderate essential hypertension. Journal of Nutrition. 2002; 132(7):1900-1902. 511 Rodilla E, Costa JA, Perez-Lahiguera F, Baldo E, Gonzalez C, Pascual JM. Spironolactone and doxazosin treatment in patients with resistant hypertension. Revista Espanola De Cardiologia. 2009; 62(2):158-166. 512 Rodriguez Roca GC, onso Moreno FJ, Garcia JA, Vega A, Llisterri Caro JL, Barrios A, V, Segura FA, Clemente = Lirola E et al. Cost-effectiveness of ambulatory blood pressure monitoring in the follow-up of hypertension. Blood Pressure. 2006; 15(1):27-36. 513 Rogers MW, Probst MM, Gruber JJ, Berger R, Boone JB, Jr. Differential effects of exercise training intensity on blood pressure and cardiovascular responses to stress in borderline hypertensive humans. Journal of Hypertension. 1996; 14(11):1369-1375. 514 Rosei EA, Dal Pal C, Leonetti G, Magnani B, Pessina A, Zanchetti A. Clinical results of the Verapamil in Hypertension and Atherosclerosis Study. Journal of Hypertension. 1997; 15:1337-1344. 515 Rosengren A, Wilhelmsen L, Hagman M, Wedel H. Natural history of myocardial infarction and angina pectoris in a general population sample of middle-aged men: a 16-year follow-up of the Primary Prevention Study, Goteborg, Sweden. Journal of Internal Medicine. 1998; 244(6):495-505. 516 Rosmarin PC. Coffee and coronary heart disease: a review. Progress in Cardiovascular Diseases. 1989; 32(3):239-245. 517 Roth RP, Cantekin EI, Bluestone CD, Welch RM, Cho YW. Nasal decongestant activity of pseudoephedrine. Annals of Otology, Rhinology and Laryngology. 1977; 86(2 pt. 1):235-242. 518 Royal Pharmaceutical Society of Great Britian Working Party. Partnership in Medicine Taking: a Consultative Document. London: Royal Pharmaceutical Society of Great Britain and Mercke Sharpe and Dohme, 1996.
306
Hypertension (partial update) Reference list 519 Sacks FM, Brown LE, Appel L, Borhani NO, Evans D, Whelton P. Combinations of potassium, calcium, and magnesium supplements in hypertension. Hypertension. 1995; 26(6 Pt 1):950-956. 520 Sacks FM, Obarzanek E, Windhauser MM, Svetkey LP, Vollmer WM, McCullough M, Karanja N, Lin PH, Steele P, Proschan MA, . Rationale and design of the Dietary Approaches to Stop Hypertension trial (DASH). A multicenter controlled-feeding study of dietary patterns to lower blood pressure. Annals of Epidemiology. 1995; 5(2):108-118. 521 Sairenchi T, Iso H, Irie F, Fukasawa N, Yamagishi K, Kanashiki M, Saito Y, Ota H, Nose T. Agespecific relationship between blood pressure and the risk of total and cardiovascular mortality in Japanese men and women. Hypertension Research - Clinical and Experimental. 2005; 28(11):901-909. 522 Saito I, Kobayashi M, Matsushita Y, Saruta T. Pharmacoeconomical evaluation of combination therapy for lifetime hypertension treatment in Japan. Journal of the Medical Association of Japan. 2005; 48(12):574-585. 523 Sakuma M, Imai Y, Tsuji I, Nagai K, Ohkubo T, Watanabe N, Sakuma H, Satoh H, Hisamichi S. Predictive value of home blood pressure measurement in relation to stroke morbidity: a populationbased pilot study in Ohasama, Japan. Hypertension Research. 1997; 20(3):167-174. 524 Sareli P, Radevski IV, Valtchanova ZP, Libhaber E, Candy GP, Den HE, Libhaber C, Skudicky D, Wang JG, Staessen JA. Efficacy of different drug classes used to initiate antihypertensive treatment in black subjects: results of a randomized trial in Johannesburg, South Africa. Archives of Internal Medicine. 2001; 161(7):965-971. 525 Schein MH, Gavish B, Herz M, Rosner-Kahana D, Naveh P, Knishkowy B, Zlotnikov E, Ben-Zvi N, Melmed RN. Treating hypertension with a device that slows and regularises breathing: a randomised, double-blind controlled study. Journal of Human Hypertension. 2001; 15(4):271-278. 526 Schillaci G, Verdecchia P, Porcellati C, Cuccurullo O, Cosco C, Perticone F. Continuous relation between left ventricular mass and cardiovascular risk in essential hypertension. Hypertension. 2000; 35(2):580-586. 527 Schillaci G, Verdecchia P, Zampi I, Battistelli M, Bartoccini C, Porcellati C. Non-invasive ambulatory BP monitoring during the night: Randomised comparison of different reading intervals. Journal of Human Hypertension. 1994; 8(1):23-27. 528 Schrader J, Luders S, Kulschewski A, Hammersen F, Plate K, Berger J, Zidek W, Dominiak P, Diener HC. Morbidity and Mortality After Stroke, Eprosartan Compared with Nitrendipine for Secondary Prevention: principal results of a prospective randomized controlled study (MOSES). Stroke. 2005; 36(6):1218-1226. 529 Schwander B, Gradl B, Zollner Y, Lindgren P, Diener HC, Luders S, Schrader J, Villar FA, Greiner W, Jonsson B. Cost-utility analysis of eprosartan compared to enalapril in primary prevention and nitrendipine in secondary prevention in Europe--the HEALTH model. Value in Health. 2009; 12(6):857-871. 530 Schwartz GL, Turner ST, Moore JH, Sing CF. Effect of time of day on intraindividual variability in ambulatory blood pressure. American Journal of Hypertension. 2000; 13(11):1203-1209. 531 Schwartz GL, Turner ST, Moore JH, Sing CF. Predictors of interindividual variation in ambulatory blood pressure and their time or activity dependence. American Journal of Hypertension. 2000; 13(1 I):52-60. 532 Scientific AdvisoryCommittee on Nutrition. Salt and Health. Norwich: The Stationery Office, 2003.
307
Hypertension (partial update) Reference list 533 Seer P, Raeburn JM. Meditation training and essential hypertension: a methodological study. Journal of Behavioral Medicine. 1980; 3(1):59-71. 534 Sega R, Facchetti R, Bombelli M, Cesana G, Corrao G, Grassi G, Mancia G. Prognostic value of ambulatory and home blood pressures compared with office blood pressure in the general population: follow-up results from the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA) study. Circulation. 2005; 111(14):1777-1783. 535 Semple PF. Investigation. ABC of Hypertension: Articles From the British Medical Journal, London: British Medical Asssociation, 1981: 38-41. 536 SHEP Cooperative Research Group. Rationale and design of a randomized clinical trial on prevention of stroke in isolated systolic hypertension. Journal of Clinical Epidemiology. 1988; 41(12):1197-1208. 537 SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA. 1991; 265(24):3255-3264. 538 Sheps SG, Bailey KR, Zachariah PK. Short-term (six hour), ambulatory blood pressure monitoring. Journal of Human Hypertension. 1994; 8(12):873-878. 539 Shimamoto K, Fujita T, Ito S, Naritomi H, Ogihara T, Shimada K, Tanaka H, Yoshiike N, HEALTH Study Committee. Impact of blood pressure control on cardiovascular events in 26,512 Japanese hypertensive patients: the Japan Hypertension Evaluation with Angiotensin II Antagonist Losartan Therapy (J-HEALTH) study, a prospective nationwide observational study. Hypertension Research - Clinical and Experimental. 2008; 31(3):469-478. 540 Shimbo D, Kuruvilla S, Haas D, Pickering TG, Schwartz JE, Gerin W. Preventing misdiagnosis of ambulatory hypertension: algorithm using office and home blood pressures. Journal of Hypertension. 2009; 27(9):1775-1783. 541 Shinagawa M, Otsuka K, Murakami S, Kubo Y, Cornelissen G, Matsubayashi K, Yano S, Mitsutake G, Yasaka K, Halberg F. Seven-day (24-h) ambulatory blood pressure monitoring, selfreported depression and quality of life scores. Blood Pressure Monitoring. 2002; 7(1):69-76. 542 Shulman NB, Hall WD. Renal vascular disease in African-Americans and other racial minorities. Circulation. 1991; 83(4):1477-1479. 543 Siani A, Strazzullo P, Russo L, Guglielmi S, Iacoviello L, Ferrara LA, Mancini M. Controlled trial of long term oral potassium supplements in patients with mild hypertension. BMJ. 1987; 294(6585):1453-1456. 544 Silman AJ, Locke C, Mitchell P, Humpherson P. Evaluation of the effectiveness of a low sodium diet in the treatment of mild to moderate hypertension. Lancet. 1983; 1(8335):1179-1182. 545 Singh RB, Niaz MA, Bishnoi I, Singh U, Begum R, Rastogi SS. Effect of low energy diet and weight loss on major risk factors, central obesity and associated disturbances in patients with essential hypertension. Journal of Human Hypertension. 1995; 9(5):355-362. 546 Sleight P, Redon J, Verdecchia P, Mancia G, Gao P, Fagard R, Schumacher H, Weber M, Bhm M, Williams B, Pogue J, Koon T, Yusuf S, ONTARGET i. Prognostic value of blood pressure in patients with high vascular risk in the Ongoing Telmisartan Alone and in combination with Ramipril Global Endpoint Trial study. Journal of Hypertension. 2009; 27(7):1360-1369. 547 Smith MB, Feldman W. Over-the-counter cold medications. A critical review of clinical trials between 1950 and 1991. JAMA. 1993; 269(17):2258-2263.
308
Hypertension (partial update) Reference list 548 Smith WM. Treatment of mild hypertension: results of a ten-year intervention trial. Circulation Research. 1977; 40(5 Suppl 1):I98-105. 549 Solomon SD, Verma A, Desai A, Hassanein A, Izzo J, Oparil S, Lacourciere Y, Lee J, Seifu Y, Hilkert RJ, Rocha R, Pitt B, Exforge Intensive Control of Hypertension to Evaluate Efficacy in Diastolic Dysfunction Investigators. Effect of intensive versus standard blood pressure lowering on diastolic function in patients with uncontrolled hypertension and diastolic dysfunction. Hypertension. 2010; 55(2):241-248. 550 Spence JD, Barnett PA, Linden W, Ramsden V, Taenzer P. Lifestyle modifications to prevent and control hypertension. 7. Recommendations on stress management. Canadian Hypertension Society, Canadian Coalition for High Blood Pressure Prevention and Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke Foundation of Canada. Canadian Medical Association Journal. 1999; 160(9 Suppl):S46-S50. 551 Spence JD, Huff M, Barnett PA. Effects of indapamide versus hydrochlorothiazide on plasma lipids and lipoproteins in hypertensive patients: a direct comparison. Canadian Journal of Clinical Pharmacology. 2000; 7(1):32-37. 552 Spinar J, Vitovec J, Soucek M, Dusek L, Pavlik T, Invesigators CORD. CORD: COmparsion of Recommended Doses of ACE inhibitors and angiotensin II receptor blockers. Vnitrni Lekarstvi. 2009; 55(5):481-488. 553 Sprafka JM, Strickland D, Gomez-Marin O, Prineas RJ. The effect of cuff size on blood pressure measurement in adults. Epidemiology. 1991; 2(3):214-217. 554 Staessen JA, Den HE, Celis H, Fagard R, Keary L, G, O'Brien ET. Antihypertensive treatment based on blood pressure measurement at home or in the physician's office: a randomized controlled trial. JAMA. 2004; 291(8):955-964. 555 Staessen JA, Fagard R, Thijs L, Celis H, Arabidze GG, Birkenhager WH, Bulpitt CJ, De Leeuw P, Dollery CT, Fletcher AE, Forette F, Leonetti G, Nachev C, O'Brien ET, Rosenfeld J, Rodicio JL, Tuomilehto J, Zanchetti A. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet. 1997; 350(9080):757-764. 556 Staessen JA, Thijs L. Development of diagnostic thresholds for automated self-measurement of blood pressure in adults. First International Consensus Conference on Blood Pressure SelfMeasurement. Blood Pressure Monitoring. 2000; 5(2):101-109. 557 Staessen JA, Thijs L, Fagard R, O'Brien ET, Clement D, de Leeuw PW, Mancia G, Nachev C, Palatini P, Parati G, Tuomilehto J, Webster J. Predicting cardiovascular risk using conventional vs ambulatory blood pressure in older patients with systolic hypertension. Systolic Hypertension in Europe Trial Investigators. JAMA. 1999; 282(6):539-546. 558 Staessen JA, Wang JG, Thijs L, Celis H, Gasowski J, Fagard RH. Use of dihydropyridines for antihypertensive treatment in older patients: Evidence from the systolic hypertension in Europe trial. Journal of Clinical & Basic Cardiology. 2000; 3:15-21. 559 Staffileno BA, Braun LT, Rosenson RS. The accumulative effects of physical activity in hypertensive post-menopausal women. Journal of Cardiovascular Risk. 2001; 8(5):283-290. 560 Stafilas PC, Sarafidis PA, Lasaridis AN, Aletras VH, Niakas D. An economic evaluation of the 2003 European Society of Hypertension-European Society of Cardiology guidelines for the management of mild-to-moderate hypertension in Greece. American Journal of Hypertension. 2005; 18(9):1233-1240.
309
Hypertension (partial update) Reference list 561 Stamler R, Stamler J, Grimm R, Gosch F, Dyer A, Berman R, Civinelli J, Elmer P, Fishman J, Van HN, . Trial on control of hypertension by nutritional means: three-year results. Journal of Hypertension - Supplement. 1984; 2(3):S167-S170. 562 Stenehjem AE, Os I. Reproducibility of blood pressure variability, white-coat effect and dipping pattern in untreated, uncomplicated and newly diagnosed essential hypertension. Blood Pressure. 2004; 13(4):214-224. 563 Stergiou GS, Baibas NM, Gantzarou AP, Skeva II, Kalkana CB, Roussias LG, Mountokalakis TD. Reproducibility of home, ambulatory, and clinic blood pressure: implications for the design of trials for the assessment of antihypertensive drug efficacy. American Journal of Hypertension. 2002; 15(2 Pt 1):101-104. 564 Stergiou GS, Baibas NM, Kalogeropoulos PG. Cardiovascular risk prediction based on home blood pressure measurement: the Didima study. Journal of Hypertension. 2007; 25(8):1590-1596. 565 Stergiou GS, Nasothimiou EG, Kalogeropoulos PG, Pantazis N, Baibas NM. The optimal home blood pressure monitoring schedule based on the Didima outcome study. Journal of Human Hypertension. 2010; 24(3):158-164. 566 Stergiou GS, Salgami EV, Tzamouranis DG, Roussias LG. Masked hypertension assessed by ambulatory blood pressure versus home blood pressure monitoring: is it the same phenomenon? American Journal of Hypertension. 2005; 18(6):772-778. 567 Stergiou GS, Skeva II, Baibas NM, Kalkana CB, Roussias LG, Mountokalakis TD. Diagnosis of hypertension using home or ambulatory blood pressure monitoring: comparison with the conventional strategy based on repeated clinic blood pressure measurements. Journal of Hypertension. 2000; 18(12):1745-1751. 568 Stergiou GS, Skeva II, Zourbaki AS, Mountokalakis TD. Self-monitoring of blood pressure at home: how many measurements are needed? Journal of Hypertension. 1998; 16(6):725-731. 569 Stergiou GS, Zourbaki AS, Skeva II, Mountokalakis TD. White coat effect detected using selfmonitoring of blood pressure at home: comparison with ambulatory blood pressure. American Journal of Hypertension. 1998; 11(7):820-827. 570 Stewart PM. Mineralocorticoid hypertension. Lancet. 1999; 353(9161):1341-1347.
571 Strazzullo P, Siani A, Gugliemi S, Di CA, Galletti F, Cirillo M, Mancini M. Controlled trial of long-term oral calcium supplementation in essential hypertension. Hypertension. 1986; 8(11):10841088. 572 Streeten DH, Anderson GH, Jr., Howland T, Chiang R, Smulyan H. Effects of thyroid function on blood pressure. Recognition of hypothyroid hypertension. Hypertension. 1988; 11(1):78-83. 573 Suarez C, Del AC, Garcia-Polo I. Ambulatory blood pressure monitoring: is the daytime period enough for making clinical decisions? Blood Pressure Monitoring. 2003; 8(6):267-270. 574 Sutcliffe SJ, Fox KF, Wood DA, Sutcliffe A, Stock K, Wright M, Akhras F, Langford E. Incidence of coronary heart disease in a health authority in London: review of a community register. British Medical Journal. 2003; 326(7379):20. 575 Suter E, Marti B, Tschopp A, Wanner HU, Wenk C, Gutzwiller F. Effects of self-monitored jogging on physical fitness, blood pressure and serum lipids: a controlled study in sedentary middleaged men. International Journal of Sports Medicine. 1990; 11(6):425-432.
310
Hypertension (partial update) Reference list 576 Suzuki Y, Kuwajima I, Aono T, Kanemaru A, Nishinaga M, Shibata H, Ozawa T. Prognostic value of nighttime blood pressure in the elderly: a prospective study of 24-hour blood pressure. Hypertension Research - Clinical and Experimental. 2000; 23(4):323-330. 577 Svetkey LP, Simons-Morton D, Vollmer WM, Appel LJ, Conlin PR, Ryan DH, Ard J, Kennedy BM. Effects of dietary patterns on blood pressure: subgroup analysis of the Dietary Approaches to Stop Hypertension (DASH) randomized clinical trial. Archives of Internal Medicine. 1999; 159(3):285293. 578 Svetkey LP, Yarger WE, Feussner JR, DeLong E, Klotman PE. Double-blind, placebo-controlled trial of potassium chloride in the treatment of mild hypertension. Hypertension. 1987; 9(5):444-450. 579 Swales JD.Manual of Hypertension. Blackwell Science; 1995
580 Szucs TD, Waeber B, Tomonaga Y. Cost-effectiveness of antihypertensive treatment in patients 80 years of age or older in Switzerland: an analysis of the HYVET study from a Swiss perspective. Journal of Human Hypertension. 2010; 24(2):117-123. 581 Takagi Y, Fukase M, Takata S, Fujimi T, Fujita T. Calcium treatment of essential hypertension in elderly patients evaluated by 24 H monitoring. American Journal of Hypertension. 1991; 4(10 Pt 1):836-839. 582 Takata Y, Yoshizumi T, Ito Y, Ueno M, Tsukashima A, Iwase M, Kobayashi K, Fujishima M. Comparison of withdrawing antihypertensive therapy between diuretics and angiotensin converting enzyme inhibitors in essential hypertensives. American Heart Journal. 1992; 124(6):1574-1580. 583 Tanabe Y, Urata H, Kiyonaga A, Ikeda M, Tanaka H, Shindo M, Arakawa K. Changes in serum concentrations of taurine and other amino acids in clinical antihypertensive exercise therapy. Clinical and Experimental Hypertension Part A: Theory and Practice. 1989; 11(1):149-165. 584 Tanji JL, Lew EY, Wong GY, Treguboff C, Ward JA, Amsterdam EA. Dietary calcium supplementation as a treatment for mild hypertension. Journal of the American Board of Family Practice. 1991; 4(3):145-150. 585 Taylor AH, Doust J, Webborn N. Randomised controlled trial to examine the effects of a GP exercise referral programme in Hailsham, East Sussex, on modifiable coronary heart disease risk factors. Journal of Epidemiology and Community Health. 1998; 52(9):595-601. 586 Taylor CB, Farquhar JW, Nelson E, Agras S. Relaxation therapy and high blood pressure. Archives of General Psychiatry. 1977; 34(3):339-342. 587 Tedesco MA, Natale F, Calabro R. Effects of monotherapy and combination therapy on blood pressure control and target organ damage: a randomized prospective intervention study in a large population of hypertensive patients. Journal of Clinical Hypertension. 2006; 8(9):634-641. 588 Tengs TO, Lin TH. A meta-analysis of quality-of-life estimates for stroke. Pharmacoeconomics. 2003; 21(3):191-200. 589 The ALLHAT Officers and Co-ordinators for the ALLHAT Collaborative Research Group. Major cardiovascular events in hypertensive patients randomized to doxazosin vs chlorthalidone: The antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT). JAMA. 2000; 283:1967-1975. 590 The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic. JAMA. 2002; 288(23):2981-2997.
311
Hypertension (partial update) Reference list 591 The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHATLLT). Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to Pravastatin vs usual care. JAMA. 2002; 288(23):2998-3007. 592 The CAPPP group. The Captopril Prevention Project: a prospective intervention trial of angiotensin converting enzyme inhibition in the treatment of hypertension. Journal of Hypertension. 1990; 8:985-990. 593 The NHS Information Centre Prescribing Support Unit. NHS Prescription Cost Analysis 2009. The Health and Social Care Information Centre, 2009. 594 The Nordic Diltiazem Study Group. A prospective intervention trials of calcium antagonist therapy in hypertension. Blood Pressure. 1993; 2(4):312-321. 595 Thijs L, Amery A, Clement D, Cox J, De CP, Fagard R, Fowler G, Guo C, Mancia G, Marin R, O'Brien E, O'Malley K, Palatini P, Parati G, Petrie J, Ravogli A, Rosenfeld J, Staessen J, Webster J. Ambulatory blood pressure monitoring in elderly patients with isolated systolic hypertension. Journal of Hypertension. 1992; 10(7):693-699. 596 Thun MJ, Peto R, Lopez AD, Monaco JH, Henley SJ, Heath CW, Jr., Doll R. Alcohol consumption and mortality among middle-aged and elderly U.S. adults. New England Journal of Medicine. 1997; 337(24):1705-1714. 597 Thurm RH, Smith WM. On restting of "Barostats" in hypertensive patients. JAMA. 1967; 201:301-304. 598 Toal CB, Mahon WA, Barnes C, Burelle D. Nifedipine gastrointestinal therapeutic system (GITS) for hypertensive patients in a primary care setting: results of the Extended Release Adalat Canadian Trial (EXACT). Clinical Therapeutics. 1997; 19(5):924-935. 599 Torjesen PA, Birkeland KI, Anderssen SA, Hjermann I, Holme I, Urdal P. Lifestyle changes may reverse development of the insulin resistance syndrome. The Oslo Diet and Exercise Study: a randomized trial. Diabetes Care. 1997; 20(1):26-31. 600 Trazzi S, Mutti E, Frattola A, Imholz B, Parati G, Mancia G. Reproducibility of non-invasive and intra-arterial blood pressure monitoring: implications for studies on antihypertensive treatment. Journal of Hypertension. 1991; 9(2):115-119. 601 Tresukosol D, Sriyudhasak O. Amlodipine and hydrochlorothiazide for isolated systolic hypertension in the Thai elderly. Siriraj Medical Journal. 2005; 57(9):374-379. 602 Trudel X, Brisson C, Larocque B, Milot A. Masked hypertension: different blood pressure measurement methodology and risk factors in a working population. Journal of Hypertension. 2009; 27(8):1560-1567. 603 Ungar A, Pepe G, Monami M, Lambertucci L, Torrini M, Baldasseroni S, Tarantini F, Marchionni N, Masotti G. Isolated ambulatory hypertension is common in outpatients referred to a hypertension centre. Journal of Human Hypertension. 2004; 18(12):897-903. 604 Urata H, Tanabe Y, Kiyonaga A, Ikeda M, Tanaka H, Shindo M, Arakawa K. Antihypertensive and volume-depleting effects of mild exercise on essential hypertension. Hypertension. 1987; 9(3):245-252. 605 Uusitupa M, Louheranta A, Lindstrom J, Valle T, Sundvall J, Eriksson J, Tuomilehto J. The Finnish Diabetes Prevention Study. British Journal of Nutrition. 2000; 83 Suppl 1:S137-S142.
312
Hypertension (partial update) Reference list 606 Vaccarino V, Berger AK, Abramson J, Black HR, Setaro JF, Davey JA, Krumholz HM. Pulse pressure and risk of cardiovascular events in the systolic hypertension in the elderly program. American Journal of Cardiology. 2001; 88(9):980-986. 607 Van Bortel LM, Bulpitt CJ, Fici F. Quality of life and antihypertensive effect with nebivolol and losartan. American Journal of Hypertension. 2005; 18(8):1060-1066. 608 van der Steen MS, Lenders JW, Graafsma SJ, den AJ, Thien T. Reproducibility of ambulatory blood pressure monitoring in daily practice. Journal of Human Hypertension. 1999; 13(5):303-308. 609 van Ittersum FJ, Ijzerman RG, Stehouwer CD, Donker AJ. Analysis of twenty-four-hour ambulatory blood pressure monitoring: what time period to assess blood pressures during waking and sleeping? Journal of Hypertension. 1995; 13(9):1053-1058. 610 Van Montfrans GA, Karemaker JM, Wieling W, Dunning AJ. Relaxation therapy and continuous ambulatory blood pressure in mild hypertension: a controlled study. BMJ. 1990; 300(6736):1368-1372. 611 Verberk WJ, Kroon AA, Kessels AG, de Leeuw PW. Home blood pressure measurement: a systematic review. Journal of the American College of Cardiology. 2005; 46:743-751. 612 Verberk WJ, Kroon AA, Kessels AG, Lenders JW, Thien T, Van Montfrans GA, Smit AJ, de Leeuw PW. The optimal scheme of self blood pressure measurement as determined from ambulatory blood pressure recordings. Journal of Hypertension. 2006; 24(8):1541-1548. 613 Verdecchia P, Reboldi G, Porcellati C, Schillaci G, Pede S, Bentivoglio M, Angeli F, Norgiolini S, Ambrosio G. Risk of cardiovascular disease in relation to achieved office and ambulatory blood pressure control in treated hypertensive subjects. Journal of the American College of Cardiology. 2002; 39(5):878-885. 614 Verdecchia P, Schillaci G, Borgioni C, Ciucci A, Pede S, Porcellati C. Ambulatory pulse pressure: a potent predictor of total cardiovascular risk in hypertension. Hypertension. 1998; 32(6):983-988. 615 Verdecchia P, Schillaci G, Borgioni C, Ciucci A, Porcellati C. Prognostic significance of the white coat effect. Hypertension. 1997; 29(6):1218-1224. 616 Verdecchia P, Staessen JA, Angeli F, de SG, Achilli A, Ganau A, Mureddu G, Pede S, Maggioni AP, Lucci D, Reboldi G, Cardio-Sis i. Usual versus tight control of systolic blood pressure in nondiabetic patients with hypertension (Cardio-Sis): an open-label randomised trial. Lancet. 2009; 374(9689):525-533. 617 Vogt TM, Appel LJ, Obarzanek E, Moore TJ, Vollmer WM, Svetkey LP, Sacks FM, Bray GA, Cutler JA, Windhauser MM, Lin PH, Karanja NM. Dietary Approaches to Stop Hypertension: rationale, design, and methods. DASH Collaborative Research Group. Journal of the American Dietetic Association. 1999; 99(8 Suppl):S12-S18. 618 Wachtell K, Hornestam B, Lehto M, Slotwiner DJ, Gerdts E, Olsen MH, Aurup P, Dahlof B, Ibsen H, Julius S, Kjeldsen SE, Lindholm LH, Nieminen MS, Rokkedal J, Devereux RB. Cardiovascular morbidity and mortality in hypertensive patients with a history of atrial fibrillation: The Losartan Intervention For End Point Reduction in Hypertension (LIFE) study. Journal of the American College of Cardiology. 2005; 45(5):705-711. 619 Wachtell K, Lehto M, Gerdts E, Olsen MH, Hornestam B, Dahlof B, Ibsen H, Julius S, Kjeldsen SE, Lindholm LH, Nieminen MS, Devereux RB. Angiotensin II receptor blockade reduces new-onset atrial fibrillation and subsequent stroke compared to atenolol: the Losartan Intervention For End
313
Hypertension (partial update) Reference list Point Reduction in Hypertension (LIFE) study. Journal of the American College of Cardiology. 2005; 45(5):712-719. 620 Waeber B, Brunner HR, Metry JM. Compliance with antihypertensive treatment: implications for practice. Blood Pressure. 1997; 6(6):326-331. 621 Walker AF, Marakis G, Morris AP, Robinson PA. Promising hypotensive effect of hawthorn extract: a randomized double-blind pilot study of mild, essential hypertension. Phytotherapy Research. 2002; 16(1):48-54. 622 Wallace JP, Park S, Zakutansky DW, Lehmkuhl LA, Jastremski CA. Time of day to monitor ambulatory blood pressure affects the outcome. Blood Pressure Monitoring. 2005; 10(1):43-50. 623 Wang JG, Staessen JA, Franklin SS, Fagard R, Gueyffier F. Systolic and diastolic blood pressure lowering as determinants of cardiovascular outcome. Hypertension. 2005; 45(5):907-913. 624 Wang JG, Staessen JA, Gong L, Liu L. Chinese trial on isolated systolic hypertension in the elderly. Systolic Hypertension in China (Syst-China) Collaborative Group. Archives of Internal Medicine. 2000; 160(2):211-220. 625 Ward, S. Statins for the Prevention of Coronary Events. London: National Insititute for Clinical Excellence, 2005. 626 Ward S, Lloyd JM, Pandor A, Holmes M, Ara R, Ryan A, Yeo W, Payne N. A systematic review and economic evaluation of statins for the prevention of coronary events. Health Technology Assessment. 2007; 11(14):1-iv. 627 Weinberger MH, Wagner UL, Fineberg NS. The blood pressure effects of calcium supplementation in humans of known sodium responsiveness. American Journal of Hypertension. 1993; 6(9):799-805. 628 Weir MR. Major outcomes in high risk hypertensive patients randomized to angiotensinconverting enzyme inhibitor or CCB vs diuretic: the Antihypertensive and Lipid-lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Current Hypertension Reports. 2003; 5(5):405-407. 629 Weitzman D, Goldbourt U. The significance of various blood pressure indices for long-term stroke, coronary heart disease, and all-cause mortality in men: the Israeli Ischemic Heart Disease study. Stroke. 2006; 37(2):358-363. 630 West R. Assessment of evidence versus consensus or prejudice. Journal of Epidemiology and Community Health. 1992; 46(4):321-322. 631 Whelton PK, Appel LJ, Espeland MA, Applegate WB, Ettinger WH, Jr., Kostis JB, Kumanyika S, Lacy CR, Johnson KC, Folmar S, Cutler JA. Sodium reduction and weight loss in the treatment of hypertension in older persons: a randomized controlled trial of nonpharmacologic interventions in the elderly (TONE). TONE Collaborative Research Group. JAMA. 1998; 279(11):839-846. 632 Whelton PK, He J. Potassium in preventing and treating high blood pressure. Seminars in Nephrology. 1999; 19(5):494-499. 633 Whelton PK, He J, Cutler JA, Brancati FL, Appel LJ, Follmann D, Klag MJ. Effects of oral potassium on blood pressure. Meta-analysis of randomized controlled clinical trials. JAMA. 1997; 277(20):1624-1632. 634 Whelton SP, Chin A, Xin X, He J. Effect of aerobic exercise on blood pressure: a meta-analysis of randomized, controlled trials. Annals of Internal Medicine. 2002; 136(7):493-503.
314
Hypertension (partial update) Reference list 635 White WB. Ambulatory blood pressure as a predictor of target organ disease and outcome in the hypertensive patient. Blood Pressure Monitoring. 1999; 4(3-4):181-184. 636 White WB, Dey HM, Schulman P. Assessment of the daily blood pressure load as a determinant of cardiac function in patients with mild-to-moderate hypertension. American Heart Journal. 1989; 118(4):782-795. 637 Whitehead A. Meta-Analysis of Controlled Clinical Trials. Chichester, UK: Wiley, 2002: 192.
638 Whitty CJ, Brunner EJ, Shipley MJ, Hemingway H, Marmot MG. Differences in biological risk factors for cardiovascular disease between three ethnic groups in the Whitehall II study. Atherosclerosis. 1999; 142(2):279-286. 639 Widimsky J. Treatment of very elderly hypertensives significantly reduces total mortality and the risk of death from stroke. Results of the HYVET trial. Cor Et Vasa. 2008; 50(9):354-357. 640 Wikstrand J, Warnold I, Olsson G, Tuomilehto J, Elmfeldt D, Berglund G. Primary prevention with Metoprolol in patients with hypertension: mortality results from the MAPHY study. JAMA. 1988; 259(13):1976-1982. 641 Wilhelmsen L, Berglund G, Elmfeldt D, Fitzsimons T, Holzgreve H, Hosie J, Hrnkvist PE, Pennert K, Tuomilehto J, Wedel H. Beta Blockers versus diuretics in hypertensive men: main results from the HAPPHY Trial. Journal of Hypertension. 1987; 5(5):561-572. 642 Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JF, Sever PS, Thom SM. Guidelines for management of hypertension: report of the fourth working party of the British Hypertension Society, 2004-BHS IV. Journal of Human Hypertension. 2004; 18(3):139-185. 643 Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998; 97(18):1837-1847. 644 Wing LM, Reid CM, Ryan P, Beilin LJ, Brown MA, Jennings GL, Johnston CI, McNeil JJ, Macdonald GJ, Marley JE, Morgan TO, West MJ. A comparison of outcomes with angiotensinconverting--enzyme inhibitors and diuretics for hypertension in the elderly. New England Journal of Medicine. 2003; 348(7):583-592. 645 Wirell MP, Wester PO, Stegmayr BG. Nutritional dose of magnesium in hypertensive patients on beta blockers lowers systolic blood pressure: a double-blind, cross-over study. Journal of Internal Medicine. 1994; 236(2):189-195. 646 Witteman JC, Grobbee DE, Derkx FH, Bouillon R, de Bruijn AM, Hofman A. Reduction of blood pressure with oral magnesium supplementation in women with mild to moderate hypertension. American Journal of Clinical Nutrition. 1994; 60(1):129-135. 647 World Health Organisation. Adherence to Long Term Therapies: Evidence for Action. WHO, 2003. 648 Wynder EL, Graham EA. Tobacco smoking as a possible etiologic factor in bronchiogenic carcinoma; a study of 684 proved cases. JAMA. 1950; 143(4):329-336. 649 Youman P, Wilson K, Harraf F, Kalra L. The economic burden of stroke in the United Kingdom. Pharmacoeconomics. 2003; 21 Suppl 1:43-50. 650 Yui Y, Sumiyoshi T, Kodama K, Hirayama A, Nonogi H, Kanmatsuse K, Origasa H, Iimura O, Ishii M, Saruta T, Arakawa K, Hosoda S, Kawai C. Nifedipine retard was as effective as angiotensin converting enzyme inhibitors in preventing cardiac events in high-risk hypertensive patients with diabetes and coronary artery disease: the Japan Multicenter Investigation for Cardiovascular
315
Hypertension (partial update) Reference list Diseases-B (JMIC-B) subgroup analysis. Hypertension Research - Clinical and Experimental. 2004; 27(7):449-456. 651 Yui Y, Sumiyoshi T, Kodama K, Hirayama A, Nonogi H, Kanmatsuse K, Origasa H, Iimura O, Ishii M, Saruta T, Arakawa K, Hosoda S, Kawai C, Japan Multicenter Investigation for Cardiovascular Diseases-. Comparison of nifedipine retard with angiotensin converting enzyme inhibitors in Japanese hypertensive patients with coronary artery disease: the Japan Multicenter Investigation for Cardiovascular Diseases-B (JMIC-B) randomized trial. Hypertension Research - Clinical and Experimental. 2004; 27(3):181-191. 652 Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensinconverting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. New England Journal of Medicine. 2000; 342(3):145-153. 653 Yusuf S, Teo K, Pogue J, Dyal L, Copland I, Schumacher H, Dagenais G, Sleight P, Anderson C. Telmisartan, ramipril, or both in patients at high risk for vascular events. New England Journal of Medicine. 2008; 358(15):1547-1559. 654 Zakopoulos NA, Nanas SN, Lekakis JP, Vemmos KN, Kotsis VT, Pitiriga VC, Stamatelopoulos SF, Moulopoulos SD. Reproducibility of ambulatory blood pressure measurements in essential hypertension. Blood Pressure Monitoring. 2001; 6(1):41-45. 655 Zanchetti A. Bottom blood pressure or bottom cardiovascular risk? how far can cardiovascular risk be reduced? Journal of Hypertension. 2009; 27(8):1509-1520. 656 Zanchetti A, Bond MG, Hennig M, Neiss A, Mancia G, Dal Palu C, Hansson L, Magnani B, Rahn KH, Reid JL, Rodicio J, Safar M, Eckes L, Rizzini P, European Lacidipine Study on Atherosclerosis investigators. Calcium antagonist lacidipine slows down progression of asymptomatic carotid atherosclerosis: principal results of the European Lacidipine Study on Atherosclerosis (ELSA), a randomized, double-blind, long-term trial. Circulation. 2002; 106(19):2422-2427. 657 Zanchetti A, Crepaldi G, Bond MG, Gallus G, Veglia F, Mancia G, Ventura A, Baggio G, Sampieri L, Rubba P, Sperti G, Magni A, PHYLLIS I. Different effects of antihypertensive regimens based on fosinopril or hydrochlorothiazide with or without lipid lowering by pravastatin on progression of asymptomatic carotid atherosclerosis: principal results of PHYLLIS--a randomized double-blind trial. Stroke. 2004; 35(12):2807-2812. 658 Zanchetti A, Rosei EA, et al. The Verapamil in Hypertension and Atherosclerosis Study (VHAS): results of long-term randomized treatment with either verapamil or chlorthalidone on carotid intimamedia thickness. Journal of Hypertension. 1998; 16:1667-1676. 659 Zemel PC, Zemel MB, Urberg M, Douglas FL, Geiser R, Sowers JR. Metabolic and hemodynamic effects of magnesium supplementation in patients with essential hypertension. American Journal of Clinical Nutrition. 1990; 51(4):665-669. 660 Zoccali C, Mallamaci F, Delfino D, Ciccarelli M, Parlongo S, Iellamo D, Moscato D, Maggiore Q. Double-blind randomized, crossover trial of calcium supplementation in essential hypertension. Journal of Hypertension. 1988; 6(6):451-455. 661 Zurawski RM, Smith TW, Houston BK. Stress management for essential hypertension: comparison with a minimally effective treatment, predictors of response to treatment, and effects on reactivity. Journal of Psychosomatic Research. 1987; 31(4):453-462. 662 Zyczynski TM, Leidy NK, Kong BW, Helaszek CT, Michelson EL. Effects of candesartan cilexetil on health-related quality of life in black patients with systemic hypertension in the ABC Trial. Heart Disease. 2000; 2(6):400-406.
316
Hypertension (partial update) Glossary
15 Glossary
Term Ambulatory blood pressure monitoring (ABPM) Abstract Aerobic exercise Algorithm (in guidelines) Allocation concealment Definition A technique for measuring BP while an individual goes about their normal daily activities Summary of a study, which may be published alone or as an introduction to a full scientific paper. Exercise requiring increased oxygen A flow chart of the clinical decision pathway described in the guideline, where decision points are represented with boxes, linked with arrows. The process used to prevent advance knowledge of group assignment in a RCT. The allocation process should be impervious to any influence by the individual making the allocation, by being administered by someone who is not responsible for recruiting participants. A strangling pain in the chest due to reduced blood flowing to the heart muscles Drug used to lower blood pressure The degree to which the results of an observation, study or review are likely to hold true in a particular clinical practice setting. Sub-section of individuals within a study who receive one particular intervention, for example placebo arm A variation in the normal rhythm of the heart Statistical relationship between two or more events, characteristics or other variables. The relationship may or may not be causal. Examination of the internal organs by listening to the sound produced The initial set of measurements at the beginning of a study (after run-in period where applicable), with which subsequent results are compared. A study that investigates the effects of an intervention by measuring particular characteristics of a population both before and after taking the intervention, and assessing any change that occurs. Systematic (as opposed to random) deviation of the results of a study from the true results that is caused by the way the study is designed or conducted. Sight or sound information letting the individual know how an aspect of their body is functioning Keeping the study participants, caregivers, researchers and outcome assessors unaware about the interventions to which the participants have been allocated in a study. Force exerted by blood against the walls of blood vessels A substance which acts as a stimulant, found in coffee and tea An element necessary for normal body function; most of our calcium intake comes from milk and milk products A unit of heat, used as a measure of energy supplied by food Disease affecting the heart or blood vessels Someone other than a health professional who is involved in caring for a person with a medical condition. Comparative observational study in which the investigator selects individuals who have experienced an event (For example, developed a disease) and others who have not (controls), and then collects data to determine previous exposure to a possible cause.
Angina pectoris: Antihypertensive Applicability Arm (of a clinical study) Arrhythmia Association Auscultation Baseline Before-and-after study
Bias Biofeedback Blinding
Blood pressure Caffeine Calcium Calorie Cardiovascular Disease Carer (caregiver) Case-control study
317

Term Case-series Definition Report of a number of cases of a given disease, usually covering the course of the disease and the response to treatment. There is no comparison (control) group of patients. Stroke (part of the brain is damaged due to lack of oxygen) Narrowing of the arteries supplying blood to the brain The extent to which an intervention is active when studied under controlled research conditions. The extent to which an intervention produces an overall health benefit in routine clinical practice. A healthcare professional providing direct patient care, for example doctor, nurse or physiotherapist. The Cochrane Library consists of a regularly updated collection of evidencebased medicine databases including the Cochrane Database of Systematic Reviews (reviews of randomised controlled trials prepared by the Cochrane Collaboration). Describing mental processes A retrospective or prospective follow-up study. Groups of individuals to be followed up are defined on the basis of presence or absence of exposure to a suspected risk factor or intervention. A cohort study can be comparative, in which case two or more groups are selected on the basis of differences in their exposure to the agent of interest. Co-existence of more than one disease or an additional disease (other than that being studied or treated) in an individual. Similarity of the groups in characteristics likely to affect the study results (such as health status or age). This is a recent term whose meaning has changed. It was initially applied to the consultation process in which doctor and patient agree therapeutic decisions that incorporate their respective views, but now includes patient support in medicine taking as well as prescribing communication. Concordance reflects social values but does not address medicine-taking and may not lead to improved adherence. A range of values for an unknown population parameter with a stated confidence (conventionally 95%) that it contains the true value. The interval is calculated from sample data, and generally straddles the sample estimate. The confidence value means that if the method used to calculate the interval is repeated many times, then that proportion of intervals will actually contain the true value. In a study, confounding occurs when the effect of an intervention on an outcome is distorted as a result of an association between the population or intervention or outcome and another factor (the confounding variable) that can influence the outcome independently of the intervention under study. Techniques that aim to reach an agreement on a particular issue. Consensus methods may used when there is a lack of strong evidence on a particular topic. A group of patients recruited into a study that receives no treatment, a treatment of known effect, or a placebo (dummy treatment) - in order to provide a comparison for a group receiving an experimental treatment, such as a new drug. Heart disease due to narrowing of the arteries which provide the heart's blood supply; may manifest as angina or heart attack A type of economic evaluation where both costs and benefits of healthcare
Cerebrovascular accident Cerebrovascular disease Clinical efficacy Clinical effectiveness Clinician Cochrane Review
Cognitive Cohort study
Comorbidity Comparability Concordance
Confidence interval (CI)
Confounding
Consensus methods
Control group
Coronary heart disease Cost benefit analysis
318

Term Definition treatment are measured in the same monetary units. If benefits exceed costs, the evaluation would recommend providing the treatment. A type of economic evaluation where various health outcomes are reported in addition to cost for each intervention, but there is no overall measure of health gain. An economic study design in which consequences of different interventions are measured using a single outcome, usually in natural units (For example, life-years gained, deaths avoided, heart attacks avoided, cases detected). Alternative interventions are then compared in terms of cost per unit of effectiveness. An explicit mathematical framework, which is used to represent clinical decision problems and incorporate evidence from a variety of sources in order to estimate the costs and health outcomes. A form of cost-effectiveness analysis in which the units of effectiveness are quality-adjusted life-years (QALYs). The Bayesian equivalent of a confidence interval. An explicit quantitative approach to decision making under uncertainty, based on evidence from research. This evidence is translated into probabilities, and then into diagrams or decision trees which direct the clinician through a succession of possible scenarios, actions and outcomes. The lowest blood pressure during each heartbeat (e.g. 80 if blood pressure is 140/80 mmHg) Costs and perhaps benefits incurred today have a higher value than costs and benefits occurring in the future. Discounting health benefits reflects individual preference for benefits to be experienced in the present rather than the future. Discounting costs reflects individual preference for costs to be experienced in the future rather than the present. An intervention is said to be dominated if there is an alternative intervention that is both less costly and more effective. Change in the dose of a drug A participant who withdraws from a trial before the end. Comparative analysis of alternative health strategies (interventions or programmes) in terms of both their costs and consequences. The observed association between interventions and outcomes or a statistic to summarise the strength of the observed association.
Cost-consequences analysis (CCA) Cost-effectiveness analysis (CEA)
Cost-effectiveness model
Cost-utility analysis (CUA) Credible Interval Decision analysis
Diastolic blood pressure Discounting
Dominance Dose titration Drop-out Economic evaluation Effect (as in effect measure, treatment effect, estimate of effect, effect size) Effectiveness Efficacy Epidemiological study
See Clinical effectiveness. See Clinical efficacy. The study of a disease within a population, defining its incidence and prevalence and examining the roles of external influences (For example, infection, diet) and interventions. A standardise instrument used to measure a health outcome. It provides a single index value for health status. High blood pressure which is not due to a known underlying disease Over 21 units/week for men; over 14 units/week for women Over 5 cups/day Information on which a decision or guidance is based. Evidence is obtained
EQ-5D (EuroQol-5D) Essential hypertension Excessive alcohol consumption Excessive coffee consumption Evidence
319

Term Definition from a range of sources including randomised controlled trials, observational studies, expert opinion (of clinical professionals and/or patients). Explicit standards used to decide which studies should be excluded from consideration as potential sources of evidence. Criteria that define who is not eligible to participate in a clinical study. If Option A is both more clinically effective than Option B and has a lower cost per unit of effect, when both are compared with a do-nothing alternative then Option A is said to have extended dominance over Option B. Option A is therefore more efficient and should be preferred, other things remaining equal. In data analysis, predicting the value of a parameter outside the range of observed values. Observation over a period of time of an individual, group or initially defined population whose appropriate characteristics have been assessed in order to observe changes in health status or health-related variables. The extent to which the results of a study based on measurement in a particular patient population and/or a specific context hold true for another population and/or in a different context. In this instance, this is the degree to which the guideline recommendation is applicable across both geographical and contextual settings. For instance, guidelines that suggest substituting one form of labour for another should acknowledge that these costs might vary across the country. GRADE / GRADE profile A system developed by the GRADE Working Group to address the shortcomings of present grading systems in healthcare. The GRADE system uses a common, sensible and transparent approach to grading the quality of evidence. The results of applying the GRADE system to clinical trial data are displayed in a table known as a GRADE profile. Adverse effects of an intervention. The study of the allocation of scarce resources among alternative healthcare treatments. Health economists are concerned with both increasing the average level of health in the population and improving the distribution of health. A combination of an individuals physical, mental and social well-being; not merely the absence of disease. Reduction in the heart's pumping efficiency, leading to accumulation of fluid in the lungs and body, causing fatigue, breathlessness and leg swelling Or lack of The term is used in meta-analyses and systematic reviews when the results or estimates of effects of treatment from separate studies seem to be very different in terms of the size of treatment effects or even to the extent that some indicate beneficial and others suggest adverse treatment effects. Such results may occur as a result of differences between studies in terms of the patient populations, outcome measures, definition of variables or duration of follow-up. High blood pressure Results are imprecise when studies include relatively few patients and few events and thus have wide confidence intervals around the estimate of effect. Explicit criteria used to decide which studies should be considered as potential sources of evidence. The analysis of additional costs and additional clinical outcomes with different interventions.
Exclusion criteria (literature review) Exclusion criteria (clinical study) Extended dominance
Extrapolation Follow-up
Generalisability
Gold standard See Reference standard.
Harms Health economics
Health-related quality of life (HRQoL) Heart failure Heterogeneity homogeneity.
Hypertension Imprecision Inclusion criteria (literature review) Incremental analysis
320

Term Incremental cost Incremental costeffectiveness ratio (ICER) Incremental net benefit (INB) Definition The mean cost per patient associated with an intervention minus the mean cost per patient associated with a comparator intervention. The difference in the mean costs in the population of interest divided by the differences in the mean outcomes in the population of interest for one treatment compared with another. The value (usually in monetary terms) of an intervention net of its cost compared with a comparator intervention. The INB can be calculated for a given cost-effectiveness (willingness to pay) threshold. If the threshold is 20,000 per QALY gained then the INB is calculated as: (20,000 x QALYs gained) Incremental cost. The available evidence is different to the review question being addressed, in terms of PICO (population, intervention, comparison and outcome). A strategy for analysing data from a randomised controlled trial. All participants are included in the arm to which they were allocated, whether or not they received (or completed) the intervention given to that arm. Intention-to-treat analysis prevents bias caused by the loss of participants, which may disrupt the baseline equivalence established by randomisation and which may reflect non-adherence to the protocol. Healthcare action intended to benefit the patient, for example, drug treatment, surgical procedure, psychological therapy. The period of time during a surgical procedure. See Coronary heart disease A statistical measure of inter-rater agreement that takes into account the agreement occurring by chance. The total number of days a participant stays in hospital. See Product licence. A measure to change a participant's behaviour in order to improve their health (e.g. exercise to reduce blood pressure) Mean average years of life gained per person as a result of the intervention compared with an alternative intervention. The likelihood ratio combines information about the sensitivity and specificity. It tells you how much a positive or negative result changes the likelihood that a patient would have the disease. The likelihood ratio of a positive test result (LR+) is sensitivity divided by 1- specificity. Drugs used to lower the level of fats in the blood Residential care in a home that may include skilled nursing care and help with everyday activities. This includes nursing homes and residential homes. The loss of participants during the course of a study. An element necessary for normal body function; found in food A method for estimating long-term costs and effects for recurrent or chronic conditions, based on health states and the probability of transition between them within a given time period (cycle). A statistical technique for combining (pooling) the results of a number of studies that address the same question and report on the same outcomes to produce a summary result. The aim is to derive more precise and clear information from a large data pool. It is generally more reliably likely to confirm or refute a hypothesis than the individual trials. Use of only one drug (rather than two or more) A statistical model for analysis of the relationship between two or more predictor (independent) variables and the outcome (dependent) variable.
Indirectness Intention to treat analysis (ITT)
Intervention Intraoperative Ischaemic heart disease Kappa statistic Length of stay Licence Lifestyle intervention Life-years gained Likelihood ratio
Lipid lowering drugs Long-term care Loss to follow-up Magnesium Markov model
Meta-analysis
Monotherapy Multivariate model
321

Term Negative predictive value (NPV) [In screening/diagnostic tests:] Normotension Number needed to treat (NNT) Observational study Definition A measure of the usefulness of a screening/diagnostic test. It is the proportion of those with a negative test result who do not have the disease, and can be interpreted as the probability that a negative test result is correct. Blood pressure that is within the normal range The number of patients that who on average must be treated to prevent a single occurrence of the outcome of interest. Retrospective or prospective study in which the investigator observes the natural course of events with or without control groups; for example, cohort studies and casecontrol studies. A measure of treatment effectiveness. The odds of an event happening in the treatment group, expressed as a proportion of the odds of it happening in the control group. The 'odds' is the ratio of events to non-events. The loss of other health care programmes displaced by investment in or introduction of another intervention. This may be best measured by the health benefits that could have been achieved had the money been spent on the next best alternative healthcare intervention. The measurement of blood pressure using an electronic device rather than by listening to Korotkoff sounds (auscultation) Measure of the possible results that may stem from exposure to a preventive or therapeutic intervention. Outcome measures may be intermediate endpoints or they can be final endpoints. See Intermediate outcome. The probability that an observed difference could have occurred by chance, assuming that there is in fact no underlying difference between the means of the observations. If the probability is less than 1 in 20, the P value is less than 0.05; a result with a P value of less than 0.05 is conventionally considered to be statistically significant. The period from admission through surgery until discharge, encompassing the pre-operative and post-operative periods. Narrowing of the arteries providing circulation to the legs An inactive and physically identical medication or procedure used as a comparator in controlled clinical trials. The use or prescription of multiple medications. In screening/diagnostic tests: A measure of the usefulness of a screening/diagnostic test. It is the proportion of those with a positive test result who have the disease, and can be interpreted as the probability that a positive test result is correct. Pertaining to the period after patients leave the operating theatre, following surgery. For diagnostic tests. The proportion of patients with that particular test result who have the target disorder. An element necessary for normal body function; found in food The ability to demonstrate an association when one exists. Power is related to sample size; the larger the sample size, the greater the power and the lower the risk that a possible association could be missed. The period before surgery commences. For diagnostic tests. The proportion of people with the target disorder in the population at risk at a specific time point or time interval. Prevalence may depend on how a disorder is diagnosed. Healthcare delivered to patients outside hospitals. Primary care covers a range of services provided by general practitioners, nurses, dentists, pharmacists,
Odds ratio
Opportunity cost
Oscilllometry Outcome
P-value
Perioperative Peripheral vascular disease Placebo Polypharmacy Positive predictive value (PPV)
Postoperative Post-test probability Potassium Power (statistical)
Preoperative Pre-test probability
Primary care
322

Term Primary outcome Product licence Prognosis Definition opticians and other healthcare professionals. The outcome of greatest importance, usually the one in a study that the power calculation is based on. An authorisation from the MHRA to market a medicinal product. A probable course or outcome of a disease. Prognostic factors are patient or disease characteristics that influence the course. Good prognosis is associated with low rate of undesirable outcomes; poor prognosis is associated with a high rate of undesirable outcomes. A study in which people are entered into the research and then followed up over a period of time with future events recorded as they happen. This contrasts with studies that are retrospective. Also known as reporting bias. A bias caused by only a subset of all the relevant data being available. The publication of research can depend on the nature and direction of the study results. Studies in which an intervention is not found to be effective are sometimes not published. Because of this, systematic reviews that fail to include unpublished studies may overestimate the true effect of an intervention. In addition, a published report might present a biased set of results (e.g. only outcomes or sub-groups where a statistically significant difference was found. See Health-related quality of life. An index of survival that is adjusted to account for the patients quality of life during this time. QALYs have the advantage of incorporating changes in both quantity (longevity/mortality) and quality (morbidity, psychological, functional, social and other factors) of life. Used to measure benefits in costutility analysis. The QALYs gained are the mean QALYs associated with one treatment minus the mean QALYs associated with an alternative treatment. An abridged version of NICE guidance, which presents the key priorities for implementation and summarises the recommendations for the core clinical audience. Allocation of participants in a research study to two or more alternative groups using a chance procedure, such as computer-generated random numbers. This approach is used in an attempt to ensure there is an even distribution of participants with different characteristics between groups and thus reduce sources of bias. A comparative study in which participants are randomly allocated to intervention and control groups and followed up to examine differences in outcomes between the groups. A rapid irregular heartbeat See Randomised controlled trial. A graphical method of assessing the accuracy of a diagnostic test. Sensitivity Is plotted against 1-specificity. A perfect test will have a positive, vertical linear slope starting at the origin. A good test will be somewhere close to this ideal. The test that is considered to be the best available method to establish the presence or absence of the outcome this may not be the one that is routinely used in practice. The number of times more likely or less likely an event is to happen in one group compared with another (calculated as the risk of the event in group A/the risk of the event in group B). Renin is an enzyme produced by the kidney and has an important role in hypertension. Renin converts a protein in the blood called angiotensinogen into angiotensin I. This is then turned into angiotensin II by angiotensin converting enzyme in the lungs. Angiotensin II reduces the size of the blood
Prospective study
Publication bias
Quality of life Quality-adjusted life year (QALY)
Quick Reference Guide
Randomisation
Randomised controlled trial (RCT) Rapid atrial fibrillation RCT Receiver operated characteristic (ROC) curve Reference standard
Relative risk (RR)
Renin-Angiotensin System
323

Term Definition vessels (increasing blood pressure) and triggers the release of a hormone called aldosterone. Aldosterone is responsible for the retention of water and salt (which further increase blood pressure). See publication bias. Someone whose blood pressure is not controlled to <140/90mmHg, despite optimal or best tolerated doses of third line treatment The likely impact in terms of finance, workforce or other NHS resources. A retrospective study deals with the present/ past and does not involve studying future events. This contrasts with studies that are prospective. In guideline development, this term refers to the questions about treatment and care that are formulated to guide the development of evidence-based recommendations. An outcome used to evaluate additional effects of the intervention deemed a priori as being less important than the primary outcomes. A systematic bias in selecting participants for study groups, so that the groups have differences in prognosis and/or therapeutic sensitivities at baseline. Randomisation (with concealed allocation) of patients protects against this bias. Sensitivity or recall rate is the proportion of true positives which are correctly identified as such. For example in diagnostic testing it is the proportion of true cases that the test detects. See the related term Specificity A means of representing uncertainty in the results of economic evaluations. Uncertainty may arise from missing data, imprecise estimates or methodological controversy. Sensitivity analysis also allows for exploring the generalisability of results to other settings. The analysis is repeated using different assumptions to examine the effect on the results. One-way simple sensitivity analysis (univariate analysis): each parameter is varied individually in order to isolate the consequences of each parameter on the results of the study. Multi-way simple sensitivity analysis (scenario analysis): two or more parameters are varied at the same time and the overall effect on the results is evaluated. Threshold sensitivity analysis: the critical value of parameters above or below which the conclusions of the study will change are identified. Probabilistic sensitivity analysis: probability distributions are assigned to the uncertain parameters and are incorporated into evaluation models based on decision analytical techniques (For example, Monte Carlo simulation). A result is deemed statistically significant if the probability of the result occurring by chance is less than 1 in 20 (p <0.05). The proportion of true negatives that a correctly identified as such. For example in diagnostic testing the specificity is the proportion of non-cases incorrectly diagnosed as cases. See related term Sensitivity. In terms of literature searching a highly specific search is generally narrow and aimed at picking up the key papers in a field and avoiding a wide range of papers. A device used to measure blood pressure Those with an interest in the use of the guideline. Stakeholders include manufacturers, sponsors, healthcare professionals, and patient and carer groups.
Reporting bias Resistant hypertension Resource implication Retrospective study Review question
Secondary outcome Selection bias
Sensitivity
Sensitivity analysis
Significance (statistical) Specificity
Sphygmomanometer Stakeholder
324

Term Stepped care Systematic review Definition A drug intervention where the dose of the drugs can be increased and/or other drugs could be added Research that summarises the evidence on a clearly formulated question according to a pre-defined protocol using systematic and explicit methods to identify, select and appraise relevant studies, and to extract, collate and report their findings. It may or may not use statistical meta-analysis. The peak blood pressure during each heartbeat (e.g. 140 if blood pressure is 140/80 mmHg) The time span over which costs and health outcomes are considered in a decision analysis or economic evaluation. The unwanted side-effects of drug treatment. These may vary from mild and/or self-limiting through to chronic and/or severe. Drugs are studied extensively before use in patients to understand (and avoid) the circumstances when they may become inappropriately toxic to patients. Temporary paralysis, numbness, speech difficulty or other neurological symptoms that start suddenly and recover within 24 hours Assigning a participant to a particular arm of the trial. Analysis which separately explores each variable in a data set. A measure of the strength of an individuals preference for a specific health state in relation to alternative health states. The utility scale assigns numerical values on a scale from 0 (death) to 1 (optimal or perfect health). Health states can be considered worse than death and thus have a negative value. Failure or refusal to take the assigned treatment (e.g. because of side effects or dislike of treatment)
Systolic blood pressure Time horizon Toxicity
Transient ischaemic attack Treatment allocation Univariate Utility
Withdrawal
325

NICE CG 127 Hypertension

Uploaded by

Copyright:

Available Formats

NICE CG 127 Hypertension

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NICE CG 127 Hypertension

Uploaded by

Copyright:

Available Formats

What is the document about?

What is the document about?

What is included in the contents section?

What is included in the contents section?

Update of clinical guidelines 18 and 34

Clinical Guideline 127

Commissioned by the National Institute for Health and Clinical Excellence

Hypertension (partial update) Contents

Hypertension (partial update) Contents

Evidence of cost-effectiveness ............................................................................................ 24

Developing recommendations ............................................................................................ 27

2004 Methods .....................................................................................................................29

2006 methods ..................................................................................................................... 34

Measuring blood pressure....................................................................................................46

Diagnosis of Hypertension ...................................................................................................52 7.1

Cost-effectiveness of clinic, home and ambulatory measurements................................... 61 7.3.1

Measurement protocols for diagnosing hypertension ....................................................... 68

8.6 8.7 9 9.1

Link from evidence to recommendations ......................................................................... 141

Blood pressure targets for treatment ............................................................................... 154

Recommendations ............................................................................................................ 171 Research Recommendation .............................................................................................. 172

12.3 2011 update: Pharmacological therapy for hypertension ................................................ 208

12.4 Cost-effectiveness analysis ............................................................................................... 238

14 Reference list..................................................................................................................... 272 15 Glossary ............................................................................................................................ 317

Hypertension (partial update)

Rationale for update

Hypertension (partial update)

Guideline development group members

Hypertension (partial update)

Hypertension (partial update)

Acronyms and abbreviations

Hypertension (partial update)

Hypertension (partial update)

Hypertension (partial update)

Hypertension (partial update)

2 Development of the guideline

Hypertension (partial update)

2.2 Who developed this guideline?

2.3 What this guideline covers

2.4 What this guideline does not cover

Hypertension (partial update)

2.5 Relationships between the guideline and other NICE guidance

Hypertension (partial update)

3.1 Developing the review questions and outcomes

3.2 Searching for evidence

Hypertension (partial update)

Health economic literature search

Call for evidence

Hypertension (partial update)

Methods of combining clinical studies

Appraising the quality of evidence by outcomes

Hypertension (partial update)

Quality element Limitations

Levels of quality elements in GRADE

Overall quality of outcome evidence in GRADE

Grading the quality of clinical evidence

Hypertension (partial update)

Study limitations of randomised controlled trials

Hypertension (partial update)

Hypertension (partial update)