PAD Guideline 2018

Lower limb peripheral arterial
disease
Diagnosis and management
NICE Clinical Guideline 147

Methods, evidence and recommendations
August 2012
This guideline was updated by a standing committee in February 2018 and 2 new
recommendations were added on diagnosing peripheral arterial disease in people with
diabetes. The recommendations are in section 1.3 of the guidance. The evidence for these
recommendations is in evidence reviews A: determining the diagnosis and severity of
peripheral arterial disease in people with diabetes.
Update information
October 2018: The antiplatelet therapy link in recommendation 1.2.1 was updated. This change
Final version
can be seen in the short version of the guideline at http://www.nice.org.uk/guidance/cg147
Commissioned by the National Institute for
Health and Clinical Excellence
PAD
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 2012
© National Clinical Guideline Centre - 2012
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.
Final version
PAD
Contents
Contents
Guideline development group members ....................................................................................... 9
Guideline development group - members ...................................................................................... 9
Guideline development group – co-optee ...................................................................................... 9
National clinical guideline centre technical team ........................................................................... 9
Acknowledgments ......................................................................................................................10
Abbreviations .............................................................................................................................11
1 Introduction ........................................................................................................................12
1.1 Background and incidence .................................................................................................. 12
1.2 Specific risk factors.............................................................................................................. 12
1.3 Definitions and classification systems ................................................................................ 13
1.3.1 Classification of PAD based on symptom severity............................................... 13
1.3.2 Classification of PAD based on anatomical distribution of disease ..................... 13
1.3.3 Issues surrounding definitions of PAD ................................................................. 14
1.4 Initial management ............................................................................................................. 14
1.5 Secondary care .................................................................................................................... 14
1.6 Importance to the NHS ....................................................................................................... 15
2 Development of the guideline ..............................................................................................16
2.1 What is a NICE clinical guideline? ....................................................................................... 16
2.2 Remit ................................................................................................................................... 16
2.3 Who developed this guideline? .......................................................................................... 17
2.4 What this guideline covers .................................................................................................. 17
2.5 What this guideline does not cover .................................................................................... 17
2.6 Relationships between the guideline and other NICE guidance ......................................... 17
3 Methods ..............................................................................................................................20
3.1 Developing the review questions and outcomes................................................................ 20
3.1.1 Clinical outcomes not considered........................................................................ 23
3.1.2 Health related quality of life ................................................................................ 24
3.2 Searching for evidence ........................................................................................................ 24
3.2.1 Clinical literature search ...................................................................................... 24
3.2.2 Health economic literature search ...................................................................... 25
3.3 Evidence of effectiveness.................................................................................................... 25
3.3.1 Inclusion/exclusion .............................................................................................. 26
3.3.2 Methods of combining clinical studies ................................................................ 26
3.3.3 Type of studies ..................................................................................................... 27
3.3.4 Types of analysis .................................................................................................. 27
3.3.5 Appraising the quality of evidence by outcomes ................................................ 28
Final version
4
PAD
Contents
3.3.6 Grading the quality of clinical evidence............................................................... 29

3.3.7 Study limitations .................................................................................................. 29
3.3.8 Inconsistency ....................................................................................................... 30
3.3.9 Indirectness ......................................................................................................... 30
3.3.10 Imprecision .......................................................................................................... 30
3.4 Evidence of cost-effectiveness ............................................................................................ 32
3.4.1 Literature review ................................................................................................. 32
3.4.2 Undertaking new health economic analysis ........................................................ 34
3.4.3 Cost-effectiveness criteria ................................................................................... 35
3.4.4 In the absence of cost-effectiveness evidence .................................................... 35
3.5 Developing recommendations ............................................................................................ 35
3.5.1 Research recommendations ................................................................................ 36
3.5.2 Validation process ............................................................................................... 36
3.5.3 Updating the guideline ........................................................................................ 36
3.5.4 Disclaimer ............................................................................................................ 36
3.5.5 Funding ................................................................................................................ 36
4 Guideline summary ..............................................................................................................37
4.1 Key priorities for implementation....................................................................................... 37
4.1.1 The recommendations identified as priorities for implementation are:............. 37
4.2 Full list of recommendations .............................................................................................. 39
4.2.1 Information requirements ................................................................................... 39
4.2.2 Secondary prevention of cardiovascular disease in people with peripheral
arterial disease..................................................................................................... 39
4.2.3 Diagnosis .............................................................................................................. 39
4.2.4 Imaging for revascularisation .............................................................................. 40
4.2.5 Management of intermittent claudication .......................................................... 40
4.2.6 Management of critical limb ischaemia .............................................................. 41
4.3 Key research recommendations ......................................................................................... 42
5 Information requirements for people with peripheral arterial disease...................................43
5.1 Introduction ........................................................................................................................ 43
5.1.1 Review question .................................................................................................. 43
5.1.2 Economic evidence .............................................................................................. 54
5.1.3 Evidence statements ........................................................................................... 54
5.1.4 Recommendations and link to evidence ............................................................. 54
5.1.5 Research recommendation ................................................................................. 57
6 Secondary prevention of cardiovascular disease in people with peripheral arterial disease....58
6.1 Introduction ........................................................................................................................ 58
6.1.1 Reducing cardiovascular risk ............................................................................... 58
6.1.2 Existing NICE guidance and recommendations ................................................... 59
Final version
5
PAD
Contents
6.2 Recommendation ................................................................................................................ 60

6.2.1 Key priority for implementation .......................................................................... 60
7 Diagnosis of peripheral arterial disease ................................................................................61
7.1 Introduction ........................................................................................................................ 61
7.2 Methods of diagnosis of peripheral arterial disease .......................................................... 61
7.2.1 Review question .................................................................................................. 61
7.3 Measuring the ankle brachial pressure index ..................................................................... 69
7.3.1 Review question .................................................................................................. 69
8 Imaging for revascularisation in peripheral arterial disease ...................................................75
8.1 Introduction ........................................................................................................................ 75
8.2 Review question .................................................................................................................. 75
8.2.1 Clinical evidence .................................................................................................. 76
9 Management of intermittent claudication ............................................................................94
9.1 Introduction ........................................................................................................................ 94
9.1.1 Role of exercise.................................................................................................... 94
9.1.2 Role of naftidrofuryl oxalate................................................................................ 94
9.1.3 Endovascular techniques ..................................................................................... 94
9.1.4 Bypass surgery ..................................................................................................... 95
9.2 Supervised exercise compared to unsupervised exercise .................................................. 96
9.2.1 Review question .................................................................................................. 96
9.2.2 Evidence statements ......................................................................................... 118
9.2.3 Recommendations and link to evidence ........................................................... 119
9.2.4 Research recommendation ............................................................................... 123
9.3 Naftidrofuryl oxalate ......................................................................................................... 123
9.3.1 Review question ................................................................................................ 123
9.4 Comparisons between treatment options: exercise, best medical treatment,
angioplasty and bypass surgery ........................................................................................ 127
9.4.1 Review question ................................................................................................ 127
9.4.2 Best medical treatment compared to best medical treatment with
angioplasty......................................................................................................... 127
9.4.3 Supervised exercise with best medical treatment compared to supervised
Final version
6
PAD
Contents
exercise, best medical treatment plus angioplasty ........................................... 132

9.4.4 Best medical treatment with angioplasty compared to best medical
treatment with angioplasty and supervised exercise for intermittent
claudication........................................................................................................ 139
9.4.5 Angioplasty compared to supervised exercise .................................................. 145
9.4.6 Bypass surgery compared to supervised exercise ............................................. 153
9.4.7 Angioplasty compared to endovascular techniques ......................................... 156
9.4.8 Economic evidence ............................................................................................ 162
9.4.9 Clinical evidence statements ............................................................................. 186
9.4.10 Economic evidence statements ......................................................................... 191
9.5 Angioplasty with selective stent placement compared to angioplasty with primary
stent placement ................................................................................................................ 194
9.5.1 Review question ................................................................................................ 194
9.6 Bare metal compared to drug eluting stents .................................................................... 211
9.6.1 Review question ................................................................................................ 211
9.7 Autologous vein compared to prosthetic bypass ............................................................. 216
9.7.1 Review question ................................................................................................ 216
10 Management of critical limb ischaemia .............................................................................. 223
10.1 Introduction: chapter overview ........................................................................................ 223
10.2 Angioplasty compared to bypass surgery ......................................................................... 223
10.2.1 Review question ................................................................................................ 223
10.2.4 Research recommendations .............................................................................. 237
10.3 Angioplasty with selective stent placement compared with angioplasty with primary
stent placement ................................................................................................................ 238
10.3.1 Review question ................................................................................................ 238
10.4 Bare metal compared to drug eluting stents .................................................................... 246
10.4.1 Review question ................................................................................................ 246
Final version
7
PAD
Contents

10.5 Autologous vein compared to prosthetic bypass ............................................................. 254
10.5.1 Review question ................................................................................................ 254
11 Management of ischaemic pain in critical limb ischaemia ................................................... 260
11.1 Introduction ...................................................................................................................... 260
11.2 Management options for pain in critical limb ischaemia.................................................. 260
11.2.1 Review question ................................................................................................ 260
12 Major amputation for critical limb ischaemia ..................................................................... 267
12.1 Introduction ...................................................................................................................... 267
12.2 Review question ................................................................................................................ 268
12.2.1 Clinical evidence ................................................................................................ 268
12.2.2 Economic evidence ............................................................................................ 270
13 Glossary ............................................................................................................................ 278
14 References ........................................................................................................................ 286
Appendices ............................................................................................................................... 299

Appendix A: Full appendices in separate document ................................................................... 299
Final version
8
PAD
Guideline development group members
Guideline development group members

Guideline development group - members
Name Role
Professor Jonathan Michaels Chair, Professor of Vascular Surgery, University of Sheffield
Mr Barry Attwood Patient and Carer Member
Mr Andrew Beech Chief Vascular Scientist, Nottingham University Hospital
Professor Andrew Bradbury Professor of Vascular Surgery, University of Birmingham
Professor Duncan Ettles Consultant Vascular Interventional Radiologist, Hull Royal Infirmary
Mr Martin Fox Vascular Podiatrist, Pennine Acute Hospital, Manchester
Professor Michael Flynn Consultant Physician in General Medicine, Diabetes and Endocrinology, Kent
and Canterbury Hospital
Ms Ammy Lam Principal Clinical Pharmacist, Barts and the London NHS Trust
Mr Peter Maufe Patient and Carer Member
Dr Ricky Mullis Physiotherapist, Senior Research Associate, Institute of Public Health,
University of Cambridge
Dr Anita Sharma General Practitioner, South Chadderton Health Centre, Oldham
Professor Cliff Shearman Professor of Vascular Surgery/Consultant Vascular Surgeon, University of
Southampton
Ms Hazel Trender Senior Vascular Nurse Specialist, Northern General Hospital, Sheffield
Mr Raman Uberoi Consultant Interventional Radiologist, John Radcliff Hospital, Oxford
Guideline development group – co-optee

Name Role
Dr Manohar Sharma Consultant in Pain Medicine and Anaesthesia, The Walton centre for
Neurology and Neurosurgery NHS Trust, Liverpool
National clinical guideline centre technical team

Name Role
Ms Sarah Bermingham Health Economist
Ms Jill Cobb Information Scientist
Dr Bernard Higgins Clinical Director
Dr Kate Kelley Guidelines Manager (from February 2011 until September 2011)
Ms Taryn Krause Senior Project Manager (until November 2010)
Miss Jennifer Layden Senior Project Manager/Research Fellow (from November 2010)
Dr Rachel O’Mahony Senior Research Fellow (from April 2012)
Ms Jill Parnham Guidelines Operations Director (until March 2011)
Ms Laura Sawyer Senior Health Economist (until January 2012)
Miss Katrina Sparrow Senior Research Fellow (until April 2012)
Final version
9
PAD
Acknowledgments
Acknowledgments
The development of this guideline was assisted by:
 Julie Brown, Research Fellow
 Mr Anthony Gildea, Guideline Co-ordinator, NICE
 Ms Lyn Knott, Editor, NICE
 Dr Zarif Jabbar-Lopez, Research Fellow, National Clinical Guideline Centre
 Ms Fatema Limbada, Project Co-ordinator, National Clinical Guideline Centre
 Mr Paul Miller, Senior Information Scientist, National Clinical Guideline Centre
 Mrs Jacoby Patterson, Systematic Reviewer
 Ms Rachael Patterson, Editor, NICE
 Dr Robbie Pitcher, Research Fellow, National Clinical Guideline Centre
 Mr Tim Reason, Health Economist, National Clinical Guideline Centre
 Ms Claire Turner, Guideline Manager, NICE
Final version
10
PAD
Abbreviations
Abbreviations
ABPI Ankle brachial pressure index
BMS Bare metal stents
BMT Best medical treatment
CCA Cost-consequences analysis
CE-MRA Contrast-enhanced magnetic resonance angiography
CLI Critical limb ischaemia
CEA Cost-effectiveness analysis
CI Confidence interval
CTA Computed tomography angiography
DES Drug eluting stents
DUS Duplex ultrasound scanning
EQ-5D EuroQol-5D
GDG Guideline development group
GRADE Grading of recommendations assessment, development and evaluation
HRQoL Health-related quality of life
HTA Health technology assessment or appraisal
IC Intermittent claudication
ICER Incremental cost-effectiveness ratio
INB Incremental net benefit
ITT Intention-to-treat analysis
MI Myocardial infarction
MID Minimal important difference
MWD Maximum walking distance
NCGC National clinical guideline centre
NICE National institute for health and clinical excellence
NNT Numbers needed to treat
NPV Negative predictive value
PAD Peripheral arterial disease
PC MRA Phase-contrast magnetic resonance angiography
PFWD Pain free walking distance
PICO Patient, intervention, comparison, outcome
PPV Positive predictive value
QALY Quality-adjusted life year
QoL Quality of life
RCT Randomised controlled trial
RR Relative risk
SFA Superficial femoral artery
TA Technology appraisal
TLR Target lesion revascularisation
TOF MRA Time of flight magnetic resonance angiography
UNG Understanding NICE guidance
Final version
11
PAD
Introduction
1 Introduction
1.1 Background and incidence
The most common initial symptom of lower limb peripheral arterial disease (known in the document
as peripheral arterial disease, PAD) is pain in the leg on walking known as intermittent claudication
(IC). The incidence of PAD increases with age. Population studies have found that about 20% of
people aged over 60 years have some degree of PAD.1 In the majority of those with IC the symptoms
remain stable but approximately 20% will progress to develop increasingly severe symptoms with the
development of critical limb ischaemia (CLI). Those with CLI are at significant risk of developing
irreversible ischaemic damage to the leg or foot if they do not receive appropriate treatment and this
may lead to the need for amputation. Overall approximately 1% to 2% of people with IC will
eventually undergo amputation,2 although the risk is higher (about 5%) in people with diabetes.3
The incidence of PAD is high among people who smoke, people with diabetes, and people with
coronary artery disease. Even in the absence of clinical symptoms the presence of PAD (as indicated
by reduced ankle brachial pressure index, ABPI) has been shown to identify people who are at
increased risk of cardiac and cerebrovascular morbidity and mortality.4
Many people will have undetected and asymptomatic PAD. In post-mortem studies, there is a
significant incidence of such disease that has never led to lifetime symptoms. The development of
symptoms will depend both upon the extent of disease and activity levels of the individual.
Of those presenting with IC over a 5-year period approximately 70 - 80% will remain with stable
claudication, 10 – 20% will go on to have worsening symptoms and 5 – 10% will go on to develop CLI.
Approximately 10 – 15% dies of cardiovascular causes within 5 years and a further 20% will have a
non-fatal cardiovascular event.5
Of those who develop CLI there is a high mortality with approximately 25% dying within a year and
about 1/3 will require a major lower limb amputation within a year.6
1.2 Specific risk factors

There are a number of associated risk factors. Some people may require investigation and treatment
for risk factors and associated diseases.
There is an association between diabetes and the development of PAD and there is a correlation
between the level of haemoglobin A1c and the level of increase in the risk of asymptomatic PAD.7
There is also evidence that those with diabetes who develop PAD have less favourable outcomes for
both the disease and its treatment. Asymptomatic PAD is common in people with diabetes. NICE has
produced a number of guidelines relating to the management of diabetes (see section 2.6),
particularly in relation to foot problems. This guideline is not intended to replace any current
recommendations within those guidelines.
Smoking is an important risk factor with the Edinburgh Artery Study1 suggesting that current smokers
are almost four times as likely to develop asymptomatic PAD as non-smokers.
As with other forms of cardiovascular disease there are also associations of PAD with hypertension
and dyslipidaemia.
Other associations with the prevalence or severity of PAD include raised homocysteine, chronic renal
insufficiency and various hyperviscosity and hypercoagulable states.
Final version
12
PAD
Introduction
1.3 Definitions and classification systems

Whilst there are a number of definitions and classification systems for PAD, these are not used
consistently in either clinical practice or research settings. The evidence reviewed within this
guideline often utilises different criteria, some of which are based on anatomical distribution of
disease and others on symptom severity. To address this, the guideline development group (GDG)
identified general definitions for PAD, IC and CLI (see Table 1).
Table 1: Guideline definitions of peripheral arterial disease

Term Definition
Peripheral arterial disease (PAD)  Infra-renal atherosclerosis
 Also known as Peripheral Arterial Occlusive Disease (PAOD) or
Peripheral Vascular Disease (PVD).
Asymptomatic PAD Clinically significant PAD without symptoms of intermittent
claudication or rest pain.
Intermittent claudication (IC) Walking (exercise) induced pain in the lower limbs caused by
diminished circulation.
Critical limb ischaemia (CLI) People with severely impaired circulation, at imminent risk of limb
loss without undergoing revascularisation.
Severe Limb Ischaemia
The term severe limb ischemia has been used in some research in preference to CLI to indicate those
people who are clinically thought to be at significant risk of limb loss due to their circulatory disease.
1.3.1 Classification of PAD based on symptom severity

Standard classifications such as the Fontaine or Rutherford scales8 are commonly used in research
settings and do not correlate well with the degree of disability experienced by patients. Both
categorise PAD in terms of symptoms (asymptomatic, intermittent claudication, ischemic rest pain or
ulceration and/or gangrene) and severity (mild, moderate or severe). The Fontaine classification is
based upon the distance that a person can walk before pain occurs (pain free walking distance,
PFWD) dividing into two groups based upon a PFWD of greater than or less than 200 metres. The
Rutherford classification uses three groups based upon a combination of the results of a treadmill
exercise test and ABPI values.
1.3.2 Classification of PAD based on anatomical distribution of disease

Treatment options and outcomes can be dependent on the anatomical distribution of disease.
However, the anatomical disease site may not correlate closely with symptoms experienced by the
patient. For the purposes of this guideline, the broad anatomical definitions in Table 2 have been
used.
Table 2: Broad anatomical definitions of peripheral arterial disease

Arterial segment Main site of blood flow limitation
Aorto-iliac Above the groin
Femoro-popliteal Between the groin and the knee
Infra-geniculate Below the knee
There are more complex classifications dealing with the anatomical distribution and extent of arterial
occlusive disease and TASC definitions6 are quite widely quoted, particularly in research studies. The
TASC classification gives some indication of the site, extent and distribution of disease.
Final version
13
PAD
Introduction
Other terms relating to the anatomical distribution include infra-inguinal to describe disease
anywhere below the groin and tibial or peroneal to describe the specific vessels below the knee.
Arterial disease to the lower limbs often affects more than one site and there may be short discrete
narrowings or more extensive disease with long or multiple segments of occluded arteries.
1.3.3 Issues surrounding definitions of PAD
1.3.3.1 Ankle Brachial Pressure Index (APBI)
Various definitions and classifications often use ABPI as an indicator of disease severity, with the use
of a threshold value for ABPI of below 0.5 for CLI and <0.9 for PAD. There are, however, a significant
group of people, particularly those with diabetes mellitus, who may have significant impairment of
the circulation, non-healing ulceration of infection and be at significant risk of limb loss, but who do
not fall strictly within these definitions of ABPI.
1.3.3.2 Use of classifications in clinical settings
Whilst such classifications may be helpful in a research setting they are rarely used in clinical practice
as they often correlate poorly with the level of disability experienced due to IC or CLI. For example,
someone who has an active job or leisure activities that involve significant walking may be very
disabled despite falling into the milder categories on such a classification. Other people, who would
be classified as severe on such scales, may find that their symptoms have little impact if they have a
largely sedentary lifestyle. In practice a term such as “lifestyle limiting claudication” is often more
helpful in representing the individual impact of the condition.
1.4 Initial management

Mild symptoms are generally managed in primary care, with referral to secondary care when
symptoms do not resolve or deteriorate. There are several treatment options for people with IC. This
includes advice to exercise, management of cardiovascular risk factors for example, aspirin or statins)
and vasoactive drug treatment for example, naftidrofuryl oxalate).
There is considerable variation in the provision of these treatment options. Whilst supervised
exercise programmes can improve walking distance and quality of life, access to such programmes is
variable, and many are not funded by the NHS. Treatments for secondary prevention are less
commonly offered to people with PAD than for those with other cardiac and cerebrovascular risk
factors.
1.5 Secondary care

People with severe symptoms that are inadequately controlled are often referred to secondary care
for assessment for endovascular (such as angioplasty or stenting), surgical revascularisation and
amputation. In recent years, there has been a move away from invasive investigation by catheter
angiography to non-invasive investigation by duplex ultrasonography, magnetic resonance
angiography or computed tomography angiography. Treadmill walking tests and segmental pressures
are other commonly used investigations.
The risks and outcomes of these procedures vary according to the nature of the procedure, the
presenting symptoms, comorbidities, and the site and extent of the disease. However, the current
trend is toward less invasive treatment.
Final version
14
PAD
Introduction
1.6 Importance to the NHS

PAD is a marker for an increased risk of potentially preventable cardiovascular events even when it is
asymptomatic. If it becomes symptomatic it can lead to significant impairment of quality of life
through limiting mobility and in its more severe manifestations may lead to severe pain, ulceration
and gangrene and is the largest single cause of lower limb amputation in the UK.
The management of PAD of the lower limb remains controversial and treatments range from
watchful waiting, through medical management, exercise training, endovascular treatment or
surgical reconstruction. Rapid changes in diagnostic methods, endovascular treatments and vascular
services, associated with the emergence of new subspecialities in surgery and vascular radiology, has
resulted in considerable uncertainty and variation in practice. This guideline aims to resolve that
uncertainty and variation.
Final version
15
PAD
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
 information for the public (‘understanding NICE guidance’ or UNG) is written using suitable
language for people without specialist medical knowledge.
This version is the full version. The other versions can be downloaded from NICE at www.nice.org.uk
2.2 Remit
NICE received the remit for this guideline from the Department of Health. They commissioned the
NCGC to produce a clinical guideline on the diagnosis and management of lower limb peripheral
arterial disease.
Final version
16
PAD
2.3 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 GDG
membership and acknowledgements).
NICE funds the NCGC and thus supported the development of this guideline. The GDG was convened
by the NCGC and chaired by Professor Jonathan Michaels in accordance with guidance from the NICE.
The group met every 6-8 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 (see Appendix B). At all
subsequent GDG meetings, members declared arising conflicts of interest, which were also recorded.
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.
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-analyses and cost-effectiveness analyses where appropriate
and drafted the guideline in collaboration with the GDG.
2.4 What this guideline covers

This guideline covers adults presenting with symptoms of lower limb peripheral arterial disease. The
key clinical areas covered in this guideline were:
 Information requirements for people with peripheral arterial disease
 Secondary prevention measures
 Diagnosis of peripheral arterial disease
 Imaging for revascularisation
 Management of intermittent claudication through exercise, drug treatment, angioplasty, stenting
and bypass surgery
 Management of critical limb ischaemia through angioplasty, stenting and bypass surgery
 Management of pain associated with critical limb ischaemia
 Major amputation for critical limb ischaemia.
For further details please refer to the scope in Appendix A and review questions in section 3.1.
2.5 What this guideline does not cover

This guideline does not cover the following:
 Children and young people
 Screening of asymptomatic PAD
 Adults who have acute ischaemia of the lower limb.
2.6 Relationships between the guideline and other NICE guidance

Related NICE Health Technology Appraisals:
Final version
17
PAD
 Cilostazol, naftidrofyryl oxalate, pentoxifylline and inositol nicotinate for the treatment of
intermittent claudication in people with peripheral arterial disease. NICE technology appraisal
guidance 223 (2011). Available from http://publications.nice.org.uk/cilostazol-naftidrofuryl-
oxalate-pentoxifylline-and-inositol-nicotinate-for-the-treatment-of-ta223.
 Clopidogrel and modified-release dipyridamole for the prevention of occlusive vascular events.
NICE technology appraisal guidance 210 (2010). Available from
http://guidance.nice.org.uk/TA210.
 Spinal cord stimulation for chronic pain of neuropathic or ischaemic origin. NICE technology
appraisal guidance 159 (2008). Available from http://publications.nice.org.uk/spinal-cord-
stimulation-for-chronic-pain-of-neuropathic-or-ischaemic-origin-ta159.
 Ezetimibe for the treatment of primary (heterozygous-familial and non-familial)
hypercholesterolaemia. NICE technology appraisal guidance 132 (2007). Available from
http://publications.nice.org.uk/ezetimibe-for-the-treatment-of-primary-heterozygous-familial-
and-non-familial-ta132.
 Varenicline for smoking cessation. NICE technology appraisal guidance 123 (2007). Available from
http://publications.nice.org.uk/varenicline-for-smoking-cessation-ta123.
 Statins for the prevention of cardiovascular events. NICE technology appraisal guidance 94 (2006).
Available from http://publications.nice.org.uk/statins-for-the-prevention-of-cardiovascular-
events-ta94.
 Guidance on the use of patient-education models for diabetes. NICE technology appraisal
guidance 60 (2003). Available from http://guidance.nice.org.uk/TA60
Related NICE Interventional Procedures:

 Endovascular stent-grafting for popliteal aneurysms. NICE interventional procedure guidance
IPG390 (2011). Available from http://publications.nice.org.uk/endovascular-stent-grafting-of-
popliteal-aneurysms-ipg390.
 Percutaneous atherectomy of femoro-popliteal arterial lesions with plague incision devices. NICE
intervention procedure guidance IPG380 (2010). Available from
http://publications.nice.org.uk/percutaneous-atherectomy-of-femoropopliteal-arterial-lesions-
with-plaque-excision-devices-ipg380.
Related NICE Clinical Guidelines:

 Opioids in palliative care. NICE clinical guideline 140 (2012). Available from
http://publications.nice.org.uk/opioids-in-palliative-care-safe-and-effective-prescribing-of-strong-
opioids-for-pain-in-palliative-cg140.
 Patient experience in adult NHS services. NICE clinical guideline 138 (2012). Available from
http://publications.nice.org.uk/patient-experience-in-adult-nhs-services-improving-the-
experience-of-care-for-people-using-adult-cg138.
 Hypertension: Clinical management of primary hypertension in adults. NICE clinical guideline
CG127 (2011). Available from http://publications.nice.org.uk/hypertension-cg127.
 Diabetic foot problems - inpatient management of people with diabetic foot ulcers and infection.
NICE clinical guideline CG119 (2011). Available from http://publications.nice.org.uk/diabetic-foot-
problems-cg119.
 Medicines adherence. NICE clinical guideline 76 (2009). Available from
http://publications.nice.org.uk/medicines-adherence-cg76.
 Lipid modification. NICE clinical guideline 67 (2008). Available from
http://publications.nice.org.uk/lipid-modification-cg67.
 Type 2 diabetes. NICE clinical guideline 66 (2008). Available from
http://www.nice.org.uk/guidance/index.jsp?action=byID&o=11983
Final version
18
PAD
 Obesity. NICE clinical guideline 43 (2006). Available from http://publications.nice.org.uk/obesity-

cg43.
 Type 1 diabetes. NICE clinical guideline 15 (2004). Available from
http://publications.nice.org.uk/type-1-diabetes-cg15.
 Type 2 diabetes – footcare. NICE clinical guideline 10 (2004). Available from
http://publications.nice.org.uk/type-2-diabetes-cg10.
Related NICE Public Health Guidance:

 Preventing type 2 diabetes: population and community-level interventions in high-risk groups and
the general population. NICE public health guidance 35 (2011). Available from
http://publications.nice.org.uk/preventing-type-2-diabetes-population-and-community-level-
interventions-in-high-risk-groups-and-the-ph35.
 Prevention of cardiovascular disease. NICE public health guidance 25 (2010). Available from
http://publications.nice.org.uk/prevention-of-cardiovascular-disease-ph25.
 Preventing the uptake of smoking by children and young people. NICE public health guidance 14
(2008). Available from http://publications.nice.org.uk/mass-media-and-point-of-sales-measures-
to-prevent-the-uptake-of-smoking-by-children-and-young-ph14.
 Promoting physical activity in the workplace. NICE public health guidance 13 (2008). Available
from http://publications.nice.org.uk/workplace-health-promotion-how-to-encourage-employees-
to-be-physically-active-ph13.
 Smoking cessation services. NICE public health guidance 10 (2008). Available from
http://publications.nice.org.uk/smoking-cessation-services-in-primary-care-pharmacies-local-
authorities-and-workplaces-ph10.
 Physical activity and the environment. NICE public health guidance 8 (2008). Available from
http://publications.nice.org.uk/promoting-and-creating-built-or-natural-environments-that-
encourage-and-support-physical-activity-ph8.
 Four commonly used methods to increase physical activity. NICE public health guidance 2 (2006).
Available from http://publications.nice.org.uk/four-commonly-used-methods-to-increase-
physical-activity-brief-interventions-in-primary-care-ph2.
 Brief interventions and referral for smoking cessation in primary care and other settings. NICE
public health guidance 1 (2006). Available from http://publications.nice.org.uk/brief-
interventions-and-referral-for-smoking-cessation-in-primary-care-and-other-settings-ph1.
Final version
19
PAD
Methods
3 Methods
This chapter sets out in detail the methods used to generate the recommendations that are
presented in the subsequent chapters. This guidance was developed in accordance with the methods
outlined in the NICE Guidelines Manual 2009.9
3.1 Developing the review questions and outcomes

Review questions were developed in a PICO framework (patient/population, 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 (see Table 3). This
was to guide the literature searching process and to facilitate the development of recommendations
by the GDG. They were drafted by the NCGC technical team and refined and validated by the GDG.
The review questions were based on the key clinical areas identified in the scope (Appendix A). The
review question protocols can be found in Appendix C. The review questions and outcome measures
examined are presented in Table 3.
Table 3: List of guideline review questions

Chapter Review questions Outcomes
Chapter 5 What are peoples’ experiences of living with PAD and  Experiences of living
preferences for information requirements for PAD? with PAD
 Information people
with PAD wanted or
found useful
 If there are specific
information
requirements for
people with PAD
 If information received
changed the
perception of PAD
Chapter 7 In people with suspected PAD, is ABPI as an adjunct to clinical  Specificity
Section 7.2 assessment better than clinical assessment alone or ABPI  Sensitivity
alone, in determining the diagnosis and severity of PAD?
 Negative predictive
value
 Positive predictive
value
 Positive likelihood
ratio
 Negative likelihood
ratio
 Reproducibility.
Chapter 7 In people with suspected PAD undergoing ABPI, do different  Specificity
Section 7.3 methods result in different diagnostic accuracy?  Sensitivity
value
value
ratio
Final version
20
PAD
Methods
ratio
 Inter- and intra-
operative reliability
 Applicability.
Chapter 8 What is most clinical and cost-effective method of assessment  Specificity
of PAD (intermittent claudication and critical limb ischemia)?  Sensitivity
value
value
ratio
ratio.
Chapter 9 What is the clinical and cost effectiveness of supervised  Amputation free
Section 9.2 exercise therapy compared to unsupervised exercise therapy survival (all)
for the treatment of PAD in adults with intermittent  CV events
claudication?
 Quality of life
 Walking distance (all)
 Adverse events
 Exercise at follow up
 Withdrawal rates from
exercise programme
 Change in ABPI
Chapter 9 What is the clinical and cost effectiveness of naftidrofuryl  Mortality
Section 9.3 oxalate compared to exercise therapy, angioplasty or stents for  Amputation free
the treatment of PAD in adults with intermittent claudication? survival (all)
 Quality of life
 Adverse events
 Re-intervention rates
 Change in ABPI
Chapter 9 What is the clinical and cost effectiveness of endovascular or  Amputation free
Section 9.4 surgical techniques compared to or in combination with survival (all)
exercise or usual care for the treatment of PAD in adults with  CV events
intermittent claudication?
 Quality of life
 Adverse events
 Exercise at follow up
 Withdrawal rates
 Change in ABPI
Chapter 9 What is the clinical and cost effectiveness of angioplasty  Mortality
Section 9.4.7 compared to bypass surgery for the treatment of PAD in adults  Amputation free
with intermittent claudication? survival (all)
 Quality of life
 Adverse events
Final version
21
PAD
Methods
 Change in ABPI
Chapter 9 What is the clinical and cost effectiveness of angioplasty with  Mortality
Section 9.5 selective stent placement compared to angioplasty with  Amputation free
primary stent placement for the treatment of PAD in adults survival (all)
with intermittent claudication?
 Quality of life
 Adverse events
 Change in ABPI
Chapter 9 What is the clinical and cost effectiveness of bare metal stents  Mortality
Section 9.6 compared to drug eluting stents for the treatment of PAD in  Amputation free
adults with intermittent claudication? survival (all)
 Quality of life
 Adverse events
 Change in ABPI
Chapter 9 What is the clinical and cost effectiveness of autologous vein  Mortality
Section 9.7 compared to prosthetic bypass for the treatment of PAD in  Amputation free
adults with intermittent claudication? survival (all)
 Quality of life
 Adverse events
 Change in ABPI
Chapter 10 What is the clinical and cost effectiveness of angioplasty  Mortality
Section 10.2 compared to bypass surgery compared to amputation for the  Amputation free
treatment of PAD in adults with critical limb ischaemia? survival (all)
 Quality of life
 Adverse events
 Change in ABPI
Chapter 10 What is the clinical and cost effectiveness of angioplasty with  Mortality
Section 10.3 selective stent placement compared to angioplasty with  Amputation free
primary stent placement for the treatment of PAD in adults survival (all)
with critical limb ischaemia?
 Quality of life
 Adverse events
 Change in ABPI
Chapter10 What is the clinical and cost effectiveness of bare metal stents  Mortality
Section 10.4 compared to drug eluting stents for the treatment of PAD in  Amputation free
adults with critical limb ischaemia? survival (all)
 Quality of life
 Adverse events
 Change in ABPI
Chapter 10 What is the clinical and cost effectiveness of autologous vein  Mortality
Section 10.5 compared to prosthetic bypass for the treatment of PAD in  Amputation free
Final version
22
PAD
Methods
adults with critical limb ischaemia? survival (all)

 Quality of life
 Adverse events
 Change in ABPI
Chapter 11 What is the clinical and cost effectiveness of chemical  Mortality
sympathectomy, opiates, gabapentin, pregabalin or tricyclic  Quality of life
antidepressants compared to each other in any combination
 Adverse events
for the management of pain in adults with critical limb
ischemia?  Pain measures
 Duration of pain
control
 Patient satisfaction
Chapter 12 What are the clinical indications for major amputation for the  Clinical indications for
management of pain in people with critical limb ischemia and major amputation
does major amputation improve the quality of life in people  Quality of life before
with critical limb ischemia? and after scores (both
must be reported)
3.1.1 Clinical outcomes not considered
Patency
The final scope for this guideline identified graft and vessel patency (primary and secondary) as an
outcome to be considered in the clinical and cost effectiveness evidence reviews. The use of patency
as an outcome measure for PAD was discussed by the GDG at length. The GDG were aware that it has
been used in many clinical trials as a surrogate endpoint for studies of treatments for PAD,
particularly endovascular treatments. The GDG were of the opinion that patency was not a good
surrogate outcome and should not therefore be included as an outcome for most comparisons.
The major concern was that the usefulness of patency as an outcome depended upon clear evidence
to make the link between patency and clinical outcomes of relevance to people with PAD. The GDG
noted that some treatments that are known to have an effect upon symptoms in people with PAD
have no effect upon patency. Their clinical experience and knowledge of the literature suggests that
it is common for people to develop recurrent symptoms despite a patent segment of vessel or to
develop re-stenosis or re-occlusion without having recurrent symptoms. They therefore considered
that the results of treatment were far better measured by outcomes of relevance to patients such as
symptoms, quality of life and the need for further interventions.
Another consideration in respect to the use of patency as an outcome is the variability in definitions
used in the literature, which may be based upon different modalities of measurement or differing
degrees of narrowing. A threshold for degree of narrowing (e.g. 50% based upon a chosen imaging
modality) leads to the anomaly that changes in narrowing of a few percent close to the threshold
determine "success", but are likely to have little or no clinical significance. It was also noted that
patency focused specifically on technical outcomes for disease at a specific site in an artery, whereas
PAD often occurs at multiple sites. The GDG felt that outcomes that consider the impact of disease
and treatment on the limb or the patient are of greater relevance.
The use of patency as a surrogate outcome also leads to difficulties in undertaking comparisons with
other treatments, such as exercise or drug treatment, where an effect on patency is not to be
expected. Even where both treatments aim to increase the diameter of a vessel, such as in the
comparison of angioplasty and stents, the initial result that is expected may differ, and thus a
threshold for patency based upon a specific degree of stenosis may be less likely to reflect a clinically
significant change for one of the treatments. Furthermore, the difference in initial treatment may
Final version
23
PAD
Methods
mean that the clinical implications, for example in terms of the potential modalities and expected
outcome of retreatment, may be different.
The only situation where the GDG considered that patency would be a potentially useful outcome
was where the two treatments being compared were expected to have identical mechanical effects,
such as in comparing similar stents with and without drug elution. Even in this situation clinical
outcomes would be preferred where available and the usefulness of the surrogate outcome would
depend upon the availability of evidence to link this to clinical outcomes which, for the reasons
above, would need to be related to the specific treatment.
3.1.2 Health related quality of life

Two types of instrument are available for measuring health related quality of life: disease specific
and generic questionnaires. The former focuses on problems associated with individual diseases,
while the latter include questions that span a number of physical and emotional dimensions common
to all people. Generic measurements of quality of life can be further divided into two major classes:
health profiles and utility measures.
Several disease specific and generic health profiles have been used to measure quality of life in
people with IC. These include, but are not limited to: the SF-36; Nottingham Health Profile; Sickness
Impact Profile; Walking Impairment Questionnaire; and VascuQol.
Utility measures are designed to reflect preferences for different treatment processes and outcomes
and comprise the primary measure of effectiveness in cost-utility analyses. In cost-utility analyses,
measures of health benefit are valued in terms of quality adjusted life years (QALYs). The QALY is a
measure of a person’s length of life weighted by a valuation of their HRQoL over that period. The
utility weighting comprises two elements: the description of changes in HRQoL and an overall
valuation of that description. Generic utility measures include: the EQ-5D; HUI 2; and SF-6D.
The different methods of measuring quality of life are not mutually exclusive; each may be useful for
under certain circumstances and for specific purposes. Early in the guideline development process,
the GDG decided that they wished to inform the economic analyses with health related quality of life
obtained directly from the included clinical studies. Changes in disease specific functional disability
would be captured by including walking distance as an outcome. The NICE reference case10 specifies
that the EQ-5D is the preferred method of QALY measurement. Therefore, only EQ-5D values or
health state descriptions which could be mapped to EQ-5D were included as measures of health
related quality of life. Disease specific questionnaires and other generic health profiles were not
included as outcomes in the review.
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 2009.9 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. Where possible,
searches were restricted to articles published in English language. All searches were conducted on
core databases, MEDLINE, Embase, Cinahl and The Cochrane Library. In addition, PsychInfo database
was used for the patient information review question. All searches were updated on the 9th January
2012. No papers after this date were considered.
Final version
24
PAD
Methods
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 D.
During the scoping stage, a search was conducted for guidelines and reports on the websites listed
below and on 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 randomised controlled trials comparing the
effectiveness of drug eluting stents to bare metal stents for the treatment of peripheral arterial
disease 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.
3.2.2 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 people with peripheral arterial disease in the NHS economic evaluation
database (NHS EED), the Health Economic Evaluations Database (HEED) and health technology
assessment (HTA) databases with no date restrictions. Additionally, the search was run on MEDLINE
and Embase, with a specific economic filter, from 2010, 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. All searches were updated on
the 9th January 2012. No papers published after this date were considered.
3.3 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 C)
 Critically appraised relevant studies using the appropriate checklist as specified in the Guidelines
Manual 20099
 Extracted key information about the study’s methods and results into evidence tables (clinical
evidence tables are included in Appendix H)
 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
Final version
25
PAD
Methods
o Observational studies: data presented as a range of values in GRADE profiles

o Diagnostic studies: data presented as a range of values in adapted GRADE profiles
o Qualitative studies: each study summarised in adapted GRADE profiles.
3.3.1 Inclusion/exclusion
The inclusion/exclusion of studies was based on the review protocols (Appendix C). The GDG were
consulted about any uncertainty regarding inclusion/exclusion of selected studies.
3.3.2 Methods of combining clinical studies
3.3.2.1 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 binary outcomes: mortality,
amputation free survival, cardiovascular events, adverse events, re-intervention rates and
withdrawal rates. The continuous outcomes: quality of life, walking distance, exercise level at follow
up, change in ABPI pain measures, duration of pain control and patient satisfaction were analysed
using an inverse variance method for pooling weighted mean differences and where the studies had
different scales, standardised mean differences were used. Where reported, time-to-event data was
presented as a hazard ratio.
Three network meta-analyses were considered for the guideline. The three proposed networks were
for the outcome of walking distance in the IC population, mortality in the CLI population and
amputation free survival in the CLI population. None of the network meta-analyses were
methodologically possible to conduct due to lack of evidence to build complete networks for the
outcomes proposed.
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 if there
were differences, 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, more than 50% missing data 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 employed to provide a more conservative estimate of the effect.
For continuous outcomes, the means and standard deviations 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. When the only evidence was based on studies summarised results by only
presenting means this information was included in the GRADE tables without calculating the relative
and absolute effect.
For binary outcomes, absolute event rates were also calculated using the GRADEpro software using
event rate in the control arm of the pooled results.
Final version
26
PAD
Methods
3.3.2.2 Data synthesis for diagnostic test accuracy review
Evidence for diagnostic data was evaluated by study, using the Quality Assessment of Diagnostic
Accuracy Studies (QUADAS) checklists.
For diagnostic test accuracy studies, the following data were extracted, either directly from the study
report or calculated from other study data: components of the “2x2 table” (true positives, false
positives, false negatives and true negatives) and test accuracy parameters: sensitivity, specificity,
positive/negative predictive values and positive/negative likelihood ratios (there are other outcomes
that can be included such as area under curve (AUC for ROC curves) reproducibility, applicability and
inter and intra operative reliability). In cases where the outcomes were not reported, 2x2 tables were
constructed from raw data to allow calculation of accuracy measures.
Forest plots of sensitivity and specificity with their 95% confidence intervals were presented side-by-
side for individual studies using Cochrane Review Manager (RevMan5) software (for RevMan see
Appendix J).
When data from 5 or more studies were available, a diagnostic meta-analysis was carried out. To
show the differences between study results, pairs of sensitivity and specificity were plotted for each
study on one receiver operating characteristics (ROC) curve in Microsoft EXCEL software (for Excel
plots please see Appendix J). A ROC plot shows true positive rate (i.e. sensitivity) as a function of false
positive rate (i.e. 1 – specificity). Study results were pooled using the bivariate method for the direct
estimation of summary sensitivity and specificity using a random effects approach (in WinBUGS®
software - for the program code see Appendix J). This model also assesses the variability by
incorporating the precision by which sensitivity and specificity have been measured in each study. A
confidence ellipse is shown in the graph that indicates the confidence region around the summary
sensitivity / specificity point. A summary ROC curve is also presented. From the WinBUGS® output we
report the summary estimate of sensitivity and specificity (plus their 95% confidence intervals) as
well as between study variation measured as logit sensitivity and specificity as well as correlations
between the two measures of variation. The summary diagnostic odds ratio with its 95% confidence
interval is also reported.
3.3.3 Type of studies

For most intervention evidence reviews in this guideline, RCTs were included. Where the GDG
believed RCT data would not be appropriate this is detailed in the protocols in Appendix C. RCTs were
included as they are considered the most robust type of study design that could produce an unbiased
estimate of the intervention effects.
For diagnostic evidence reviews, diagnostic randomised controlled trials, diagnostic cohorts and case
controls studies were included in this guideline.
3.3.4 Types of analysis

Estimates of effect from individual studies were based available case analysis (ACA) where possible
or intention to treat (ITT) analysis if this was not possible. ITT analysis is where all participants that
were randomised are considered in the final analysis based on the intervention and control groups to
which they were originally assigned. It was assumed that participants in the trials lost to follow-up
did not experience the outcome of interest (categorical outcomes) and they would not considerably
change the average scores of their assigned groups (for continuous outcomes).
It is important to note that ITT analyses tend to bias the results towards no difference. ITT analysis is
a conservative approach to analyse the data, and therefore the effect may be smaller than in reality.
Final version
27
PAD
Methods
3.3.5 Appraising the quality of evidence by outcomes

The evidence for outcomes from the included RCTs and observational 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 presented as one table in the guideline
(called clinical evidence profiles). This includes the details of the quality assessment pooled outcome
data, and 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.
Each outcome was examined separately for the quality elements listed and defined in Table 4 and
each graded using the quality levels listed in Table 5 and Table 6. The main criteria considered in the
rating of these elements are discussed below (see section 3.3.6 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.
The GRADE toolbox is currently designed only for RCTs and observational studies but however, for
the purposes of this guideline, the quality assessment elements and outcome presentation was
adapted for diagnostic accuracy and qualitative studies.
Table 4: Description of quality elements in GRADE for intervention studies

Quality element Description
Limitations 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 Inconsistency refers to an unexplained heterogeneity of results.
Indirectness Indirectness refers to differences in study population, intervention, comparator and
outcomes between the available evidence and the review question, or
recommendation made.
Imprecision 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 Publication bias is a systematic underestimate or an overestimate of the underlying
beneficial or harmful effect due to the selective publication of studies.
Table 5: Levels of quality elements in GRADE

Level Description
None There are no serious issues with the evidence.
Serious The issues are serious enough to downgrade the outcome evidence by one level.
Very serious The issues are serious enough to downgrade the outcome evidence by two levels.
Table 6: Overall quality of outcome evidence in GRADE

Level Description
High Further research is very unlikely to change our confidence in the estimate of effect.
Moderate Further research is likely to have an important impact on our confidence in the estimate
Final version
28
PAD
Methods
of effect and may change the estimate.

Low 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.
Very low Any estimate of effect is very uncertain.
3.3.6 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, uncontrolled case series as LOW or VERY LOW.
2. The rating was then downgraded for the specified criteria: study limitations, inconsistency,
indirectness, imprecision and reporting bias. These criteria are detailed below (see Table 7).
Observational studies were upgraded if there was: a large magnitude of effect, dose-response
gradient, and if all plausible confounding would reduce a demonstrated effect or suggest a
spurious effect when results showed no effect. Each quality element considered to have “serious”
or “very serious” risk of bias were rated down -1 or -2 points respectively.
3. The downgraded/upgraded 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 3.3.10.
3.3.7 Study limitations

The main limitations for RCTs are listed in Table 7.
The GDG accepted that investigator blinding in surgical intervention studies was impossible and
participant blinding was also difficult to achieve in most situations. Nevertheless, open-label studies
for surgery were downgraded to maintain a consistent approach in quality rating across the
guideline.
Table 7: Study limitations of randomised controlled trials

Limitation Explanation
Allocation Those enrolling patients are aware of the group to which the next enrolled patient
concealment will be allocated (major problem in “pseudo” or “quasi” randomised trials with
allocation by day of week, birth date, chart number, etc).
Lack of blinding Patient, caregivers, those recording outcomes, those adjudicating outcomes, or data
analysts are aware of the arm to which patients are allocated.
Incomplete Loss to follow-up not accounted.
accounting of
patients and
outcome events
Selective outcome Reporting of some outcomes and not others on the basis of the results.
reporting
Other limitations For example:
 Stopping early for benefit observed in randomised trials, in particular in the absence
of adequate stopping rules
 Use of un-validated patient-reported outcomes
Final version
29
PAD
Methods
 Carry-over effects in cross-over trials

 Recruitment bias in cluster randomised trials.
3.3.8 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. In addition to the I-square and Chi square values, the
decision for downgrading was also dependent on factors such as whether the intervention is
associated with benefit in all other outcomes or whether the uncertainty about the magnitude of
benefit (or harm) of the outcome showing heterogeneity would influence the overall judgment about
net benefit or harm (across all outcomes).
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 gives a plausible
explanation of heterogeneity, the quality of evidence would not be downgraded.
3.3.9 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.
3.3.10 Imprecision
The minimal important difference (MID) in the outcome between the two groups were the main
criteria considered.
The thresholds of important benefits or harms, or the MID for an outcome are important
considerations for determining whether there is a “clinically important” difference between
intervention and control groups and in assessing imprecision. For continuous outcomes, the MID is
defined as “the smallest difference in score in the outcome of interest that informed patients or
informed proxies perceive as important, ether beneficial or harmful, and that would lead the patient
or clinician to consider a change in the management.11-14 An effect estimate larger than the MID is
considered to be “clinically important”.
The difference between two interventions, as observed in the studies, was compared against the
MID when considering whether the findings were of “clinical importance”; this is useful to guide
decisions. For example, if the effect size was small (less than the MID), this finding suggests that
there may not be enough difference to strongly recommend one intervention over the other based
on that outcome.
The criteria applied for imprecision are based on the confidence intervals for pooled or the best
estimate of effect as outlined in Table 8 and illustrated in Figure 1.
Table 9 presents the MID thresholds used for the specified outcomes for this guideline as specified by
the GDG.
Final version
30
PAD
Methods
Table 8: 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:
‘no serious imprecision’ Does not cross either of the two minimal important
difference (MID) thresholds (the threshold lines for
appreciable benefit or harm); defined as precise.
‘serious’ Crosses one of the two MID thresholds (appreciable
benefit or appreciable harm); defined as imprecise.
‘very serious’ Crosses both of the two MID thresholds (appreciable
benefit and appreciable harm); defined as imprecise.
Figure 1: Illustration of precise and imprecise outcomes based on the confidence interval
of outcomes in a forest plot
Source: Figure adapted from GRADEPro software.
The MIDs are the threshold for appreciable benefits and harms. The confidence intervals of the top
three points of the diagram were considered precise because the upper and lower limits did not
cross the MID. Conversely, the bottom three points of the diagram were considered imprecise
because all of them crossed the MID and reduced our certainty of the results.
Final version
31
PAD
Methods
Table 9: Minimal important differences (MIDs) for the outcomes used in this guideline
Outcome MID
Mortality 1%
Amputation free survival 1%
CV events for people with IC 5%
Target lesion revascularisation 10%
Quality of life (EQ5D) Change of 0.05 (mean difference, continuous outcome)
Maximum walking distance Doubling in baseline distance (mean difference, continuous
outcome)
Pain free walking distance Doubling in baseline distance (mean difference, continuous
outcome)
Major adverse events 10%
Minor adverse events 10%
Re-intervention rate 10%
Change in ABPI Change of 0.15 (mean difference, continuous outcome)
Pain measures (as reported in papers) 0.5 standardised mean difference
Duration of pain 0.5 standardised mean difference
Patient satisfaction 0.5 standardised mean difference
3.4 Evidence of cost-effectiveness

The GDG is required to make decisions based on the best available evidence of both clinical and cost
effectiveness. Guideline recommendations should be based on the expected costs of the treatment
options in relation to their expected health benefits (that is, their ‘cost effectiveness’), rather than on
the total cost or resource impact of implementing them.9 Thus, if the evidence suggests that an
intervention provides significant health benefits at an acceptable cost per patient treated, it should
be recommended even if it would be expensive to implement across the whole population.
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.
3.4.1 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 20099
 Extracted key information about the study’s methods and results into evidence tables (included in
Appendix I)
 Generated summaries of the evidence in NICE economic evidence profiles (included in the
relevant chapter write-ups) – see below for details.
Final version
32
PAD
Methods
3.4.1.1 Inclusion/exclusion
Full economic evaluations (studies comparing costs and health consequences of alternative courses
of action: cost–utility, cost-effectiveness, cost-benefit and cost-consequence analyses) and
comparative costing studies that addressed the review question in the relevant population were
considered potentially includable as economic evidence.
Studies that only reported cost per hospital (not per patient), or only reported average cost
effectiveness without disaggregated costs and effects, were excluded. 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, except for American studies, which were considered
‘partially applicable’).
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 not have been included. Where
exclusions occurred on this basis, this is noted in the relevant section and included in the list of
excluded studies in Appendix F.
For more details about the assessment of applicability and methodological quality see the economic
evaluation checklist.9
When no relevant economic analysis was identified in the economic literature review, relevant UK
NHS unit costs related to the compared interventions were presented to the GDG to inform the
possible economic implication of the recommendation to make.
3.4.1.2 NICE economic evidence profiles
The NICE economic evidence profile has been used to summarise cost and cost-effectiveness
estimates (see Table 10). The economic evidence profile includes an assessment of applicability and
methodological quality for each economic study, with footnotes indicating the reasons for each
assessment. These assessments were made by the health economist using the economic evaluation
checklist from the Guidelines Manual 2009.9 It also shows incremental costs, incremental outcomes
(for example, QALYs) and the incremental cost-effectiveness ratio, as well as information about the
assessment of uncertainty in the analysis.
Several of the pair wise clinical comparisons conducted in the IC population concerned the same
decision question. Due to the nature of the question and the difficulty of considering multiple-
comparator evaluations in a pair wise context, the clinical and economic evidence for these
questions were presented in separate sections.
All costs converted into 2009/10 pounds sterling using the appropriate purchasing power parity.15
Table 10: Content of NICE economic profile

Item Description
Study First author name, reference, date of study publication and country perspective.
Applicability 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.
Final version
33
PAD
Methods
Limitations 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.
Other comments Particular issues that should be considered when interpreting the study.
Incremental cost The mean cost associated with one strategy minus the mean cost of a comparator
strategy.
Incremental effects The mean QALYs (or other selected measure of health outcome) associated with
one strategy minus the mean QALYs of a comparator strategy.
Cost effectiveness Incremental cost-effectiveness ratio (ICER): the incremental cost divided by the
incremental effects (i.e. QALYs gained).
Uncertainty 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.
(a) Applicability and limitations were assessed using the economic evaluation checklist from The Guidelines Manual 9
3.4.2 Undertaking new health economic analysis

As well as reviewing the published economic literature for each review question, as described above,
new economic analysis was undertaken by the health economist in priority selected areas. Priority
areas for new health economic analysis were agreed by the GDG after formation of the review
questions and consideration of the available health economic evidence.
The GDG identified the treatment of IC using exercise and endovascular interventions as the highest
priority areas for original economic modelling. Specifically, these areas include the cost effectiveness
of supervised compared to unsupervised exercise, and exercise compared to angioplasty for the
treatment of IC.
The following general principles were adhered to in developing the cost-effectiveness analysis:
 Methods were consistent with the NICE reference case10
 The GDG was involved in the design of the model, selection of inputs and interpretation of the
results
 Model inputs were based on the systematic review of the clinical literature supplemented with
other published data sources where possible
 When published data was not available GDG expert opinion was used to populate the model
 Model inputs and assumptions were reported fully and transparently
 The results were subject to sensitivity analysis and limitations were discussed
 The model was peer-reviewed by another health economist at the NCGC.
Additional data for the analysis was identified as required through additional literature searches
undertaken by the health economist and in 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.
Full methods for the original health economic analyses undertaken for this guideline are described in
Appendices K and L.
Final version
34
PAD
Methods
3.4.3 Cost-effectiveness criteria

NICE’s 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.9,16
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 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 ‘recommendations and link to evidence’
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’.16
3.4.4 In the absence of cost-effectiveness evidence

When no relevant published studies were found, and a new analysis was not prioritised, the GDG
made a qualitative judgement about cost effectiveness by considering expected differences in
resource use between comparators and relevant UK NHS unit costs alongside the results of the
clinical review of effectiveness evidence.
3.5 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 Appendices H and I
 Summary of clinical and economic evidence and quality (as presented in chapters 5-12)
 Forest plots, diagnostic meta-analysis and summary ROC curves (Appendix J)
 A description of the methods and results of the cost-effectiveness analysis undertaken for the
guideline (Appendix K and L).
Recommendations were drafted on the basis of the GDG’s interpretation of the available evidence,
taking into account the balance of benefits, harms and costs. When clinical and economic evidence
was of poor quality, conflicting or absent, the GDG drafted recommendations based on their expert
opinion. The considerations for making consensus based recommendations include the balance
between potential harms and benefits, economic or implications compared to the benefits, current
practices, recommendations made in other relevant guidelines, patient preferences and equality
issues. The consensus recommendations were done through discussions in the GDG. The GDG may
also consider whether the uncertainty is sufficient to justify delaying making a recommendation to
await further research, taking into account the potential harm of failing to make a clear
recommendation (see section 3.5.1).
The main considerations specific to each recommendation are outlined in the recommendations and
link to evidence section following the clinical and economic evidence reviews.
Final version
35
PAD
Methods
3.5.1 Research recommendations

When areas were identified for which good evidence was lacking, the GDG 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.
3.5.2 Validation process

The guidance is subject to a six 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.
3.5.3 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 NICE’s Guidance executive
whether the evidence base has progressed significantly to alter the guideline recommendations and
warrant an update.
3.5.4 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.
3.5.5 Funding
The National Clinical Guideline Centre was commissioned by the National Institute for Health and
Clinical Excellence to undertake the work on this guideline.
Final version
36
PAD
Guideline summary
4 Guideline summary
4.1 Key priorities for implementation
From the full set of recommendations, the GDG selected 9 key priorities for implementation. They
selected recommendations that would:
 Have a high impact on outcomes that are important to patients
 Have a high impact on reducing variation in care and outcomes
 Lead to a more efficient use of NHS resources
 Promote patient choice
 Promote equality.
In addition to this, the GDG also considered which recommendations were particularly likely to
benefit from implementation support. They considered whether a recommendation:
 Relates to an intervention that is not part of routine care
 Requires changes in service delivery
 Requires retraining of staff of the development of new skills and competencies
 Highlights the need for practice change
 Affects an needs to be implemented across a number of agencies or settings (complex
interactions)
 May be viewed as potentially contentious, or difficult to implement for other reasons.
The reasons that each of these recommendations was chosen are shown in the table linking the
evidence to the recommendation in the relevant chapter.
4.1.1 The recommendations identified as priorities for implementation are:
Information requirement for people with peripheral arterial disease

 Offer all people with peripheral arterial disease oral and written information about their
condition. Discuss it with them so they can share decision-making, and understand the course of
the disease and what they can do to help prevent disease progression. Information should
include:
o the causes of their symptoms and the severity of their disease
o the risks of limb loss and/or cardiovascular events associated with peripheral arterial disease
o the key modifiable risk factors, such as smoking, control of diabetes, hyperlipidaemia, diet,
body weight and exercise (see also recommendation on secondary prevention of
cardiovascular disease)
o how to manage pain
o all relevant treatment options, including the risks and benefits of each
o how they can access support for dealing with depression and anxiety.
Ensure that information, tailored to the individual needs of the person, is available at diagnosis and
subsequently as required, to allow people to make decisions throughout the course of their
treatment.
Final version
37
PAD
Guideline summary
Secondary prevention of cardiovascular disease in people with peripheral arterial disease

 Offer all people with peripheral arterial disease information, advice, support and treatment
regarding the secondary prevention of cardiovascular disease, in line with published NICE
guidance (see section 2.6) on:
o smoking cessation
o diet, weight management and exercise
o lipid modification and statin therapy
o the prevention, diagnosis and management of diabetes
o the prevention, diagnosis and management of high blood pressure
o antiplatelet therapy.
Diagnosis
 Assess people for the presence of peripheral arterial disease if they:
o have symptoms suggestive of peripheral arterial disease or
o have diabetes, non-healing wounds on the legs or feet or unexplained leg pain or
o are being considered for interventions to the leg or foot or
o need to use compression hosiery.
 Assess people with suspected peripheral arterial disease by:

o asking about the presence and severity of possible symptoms of intermittent claudication and
critical limb ischaemia
o examining the legs and feet for evidence of critical limb ischaemia, for example ulceration
o examining the femoral, popliteal and foot pulses
o measuring the ankle brachial pressure index (see recommendation below).
 Measure the ankle brachial pressure index in the following way:

o The person should be resting and supine if possible.
o Record systolic blood pressure with an appropriately sized cuff in both arms and in the
posterior tibial, dorsalis pedis and, where possible, peroneal arteries.
o Take measurements manually using a Doppler probe of suitable frequency in preference to an
automated system.
o Document the nature of the Doppler ultrasound signals in the foot arteries.
o Calculate the index in each leg by dividing the highest ankle pressure by the highest arm
pressure.
Imaging for revascularisation

 Offer contrast-enhanced magnetic resonance angiography to people with peripheral arterial
disease who need further imaging (after duplex ultrasound) before considering revascularisation.
Management of intermittent claudication

 Offer a supervised exercise programme to all people with intermittent claudication.
Management of critical limb ischaemia

 Ensure that all people with critical limb ischaemia are assessed by a vascular multidisciplinary
team before treatment decisions are made.
Final version
38
PAD
Guideline summary
 Do not offer major amputation to people with critical limb ischaemia unless all options for
revascularisation have been considered by a vascular multidisciplinary team.
4.2 Full list of recommendations

4.2.1 Information requirements
1. Offer all people with peripheral arterial disease oral and written information about their
condition. Discuss it with them so they can share decision-making, and understand the course of
the disease and what they can do to help prevent disease progression. Information should
include:
o the causes of their symptoms and the severity of their disease
o the risks of limb loss and/or cardiovascular events associated with peripheral arterial disease
o the key modifiable risk factors, such as smoking, control of diabetes, hyperlipidaemia, diet,
body weight and exercise (see also recommendation 3)
o how to manage pain
o all relevant treatment options, including the risks and benefits of each
o how they can access support for dealing with depression and anxiety.
Ensure that information, tailored to the individual needs of the person, is available at diagnosis and
subsequently as required, to allow people to make decisions throughout the course of their
treatment.
2. NICE has produced guidance on the components of good patient experience in adult NHS services.
Follow the recommendations in Patient experience in adult NHS services (NICE clinical guideline
138).
4.2.2 Secondary prevention of cardiovascular disease in people with peripheral arterial disease
3. Offer all people with peripheral arterial disease information, advice, support and treatment
regarding the secondary prevention of cardiovascular disease, in line with published NICE
guidance (see 'Related NICE guidance'; section 2.6) on:
o smoking cessation
o diet, weight management and exercise
o lipid modification and statin therapy
o the prevention, diagnosis and management of diabetes
o the prevention, diagnosis and management of high blood pressure
o antiplatelet therapy.
4.2.3 Diagnosis
4. Assess people for the presence of peripheral arterial disease if they:
o have symptoms suggestive of peripheral arterial disease or
o have diabetes, non-healing wounds on the legs or feet or unexplained leg pain or
o are being considered for interventions to the leg or foot or
o need to use compression hosiery.
5. Assess people with suspected peripheral arterial disease by:
Final version
39
PAD
Guideline summary
o asking about the presence and severity of possible symptoms of intermittent claudication and
critical limb ischaemia
o examining the legs and feet for evidence of critical limb ischaemia, for example ulceration
o examining the femoral, popliteal and foot pulses
o measuring the ankle brachial pressure index (see recommendation 6).
6. Measure the ankle brachial pressure index in the following way:

o The person should be resting and supine if possible.
o Record systolic blood pressure with an appropriately sized cuff in both arms and in the
posterior tibial, dorsalis pedis and, where possible, peroneal arteries.
o Take measurements manually using a Doppler probe of suitable frequency in preference to an
automated system.
o Document the nature of the Doppler ultrasound signals in the foot arteries.
o Calculate the index in each leg by dividing the highest ankle pressure by the highest arm
pressure.
4.2.4 Imaging for revascularisation

7. Offer duplex ultrasound as first-line imaging to all people with peripheral arterial disease for
whom revascularisation is being considered.
8. Offer contrast-enhanced magnetic resonance angiography to people with peripheral arterial
disease who need further imaging (after duplex ultrasound) before considering revascularisation.
9. Offer computed tomography angiography to people with peripheral arterial disease who need
further imaging (after duplex ultrasound) if contrast-enhanced magnetic resonance angiography is
contraindicated or not tolerated.
4.2.5 Management of intermittent claudication
4.2.5.1 Supervised exercise programme

10.Offer a supervised exercise programme to all people with intermittent claudication.
11.Consider providing a supervised exercise programme for people with intermittent claudication
which involves:
o 2 hours of supervised exercise a week for a 3-month period
o encouraging people to exercise to the point of maximal pain.
4.2.5.2 Angioplasty and stenting

12.Offer angioplasty for treating people with intermittent claudication only when:
o advice on the benefits of modifying risk factors has been reinforced (see recommendation 3)
and
o a supervised exercise programme has not led to a satisfactory improvement in symptoms and
o imaging has confirmed that angioplasty is suitable for the person.
13.Do not offer primary stent placement for treating people with intermittent claudication caused by
aorto-iliac disease (except complete occlusion) or femoro-popliteal disease.
Final version
40
PAD
Guideline summary
14.Consider primary stent placement for treating people with intermittent claudication caused by
complete aorto-iliac occlusion (rather than stenosis).
15.Use bare metal stents when stenting is used for treating people with intermittent claudication.
4.2.5.3 Bypass surgery and graft types

16.Offer bypass surgery for treating people with severe lifestyle-limiting intermittent claudication
only when:
o angioplasty has been unsuccessful or is unsuitable and
o imaging has confirmed that bypass surgery is appropriate for the person.
17.Use an autologous vein whenever possible for people with intermittent claudication having infra-
inguinal bypass surgery.
4.2.5.4 Naftidrofuryl oxalate

18.Consider naftidrofuryl oxalate for treating people with intermittent claudication, starting with the
least costly preparation, only when:
o supervised exercise has not led to satisfactory improvement and
o the person prefers not to be referred for consideration of angioplasty or bypass surgery.
Review progress after 3-6 months and discontinue naftidrofuryl oxalate if there has been no
symptomatic benefit.
4.2.6 Management of critical limb ischaemia

19.Ensure that all people with critical limb ischaemia are assessed by a vascular multidisciplinary
team before treatment decisions are made.
4.2.6.1 Revascularisation
20.Offer angioplasty or bypass surgery for treating people with critical limb ischaemia who require
revascularisation, taking into account factors including:
o comorbidities
o pattern of disease
o availability of a vein
o patient preference.
21.Do not offer primary stent placement for treating people with critical limb ischaemia caused by
aorto-iliac disease (except complete occlusion) or femoro-popliteal disease.
22.Consider primary stent placement for treating people with critical limb ischaemia caused by
complete aorto-iliac occlusion (rather than stenosis).
23.Use bare metal stents when stenting is used for treating people with critical limb ischaemia.
24.Use an autologous vein whenever possible for people with critical limb ischaemia having infra-
inguinal bypass surgery.
Final version
41
PAD
Guideline summary
4.2.6.2 Management of critical limb ischaemic pain

25.Offer paracetamol, and either weak or strong opioids depending on the severity of pain, to people
with critical limb ischaemic pain.
26.Offer drugs such as laxatives and anti-emetics to manage the adverse effects of strong opioids, in
line with the person's needs and preferences.
27.Refer people with critical limb ischaemic pain to a specialist pain management service if any of
the following apply:
o their pain is not adequately controlled and revascularisation is inappropriate or impossible.
o ongoing high doses of opioids are required for pain control
o pain persists after revascularisation or amputation.
28.Do not offer chemical sympathectomy to people with critical limb ischaemic pain, except in the
context of a clinical trial.
4.2.6.3 Major amputation

29.Do not offer major amputation to people with critical limb ischaemia unless all options for
revascularisation have been considered by a vascular multidisciplinary team.
4.3 Key research recommendations

 What is the clinical and cost effectiveness of a ‘bypass surgery first’ strategy compared with an
‘angioplasty first’ strategy for treating people with critical limb ischaemia caused by disease of the
infra-geniculate (below the knee) arteries?
 What is the clinical and cost effectiveness of supervised exercise programmes compared with
unsupervised exercise for treating people with intermittent claudication, taking into account the
effects on long-term outcomes and continuing levels of exercise?
 What is the effect of people’s attitudes and beliefs about their peripheral arterial disease on the
management and outcome of their condition?
 What is the clinical and cost effectiveness of selective stent placement compared with angioplasty
plus primary stent placement for treating people with critical limb ischaemia caused by disease of
the infra-geniculate arteries?
 What is the clinical and cost effectiveness of chemical sympathectomy in comparison with other
methods of pain control for managing critical limb ischaemic pain?
Final version
42
PAD
Information requirements for people with peripheral arterial disease
5 Information requirements for people with

peripheral arterial disease
5.1 Introduction
Peripheral arterial disease (PAD) is a chronic condition for which the person will require ongoing
support and guidance. It is important that the person receives information relevant to their stage of
disease that will enable them to make an informed decision about the treatment that is available and
the lifestyle choices that may affect the outcome.
People with PAD need to recognise that lifestyle factors e.g. exercise levels, smoking and diet, will
have an impact on disease progression and severity (see chapter 6 for further information and
recommendations). This information is needed from the time of diagnosis but needs to be offered in
a fashion appropriate to the person’s background as cultural and social factors have a large influence
not just on lifestyle but also on response that will be made to any proposed changes. The patient’s
baseline understanding must be established and their attitude to any current proposed treatment
should be sensitively explored.
The resources available for changing lifestyle will include not only consultation with healthcare
professionals but voluntary workshops, self help groups and if possible friends and family. The
information might include both written and verbal and if appropriate and available, audio and visual
material.
5.1.1 Review question

What are people’s experiences of living with PAD and people’s preferences for information
requirements?
The GDG were interested in identifying people’s experiences of living with PAD and any specific
information requirements. A qualitative literature search was undertaken, there were no study
design filters placed on the search.
5.1.1.1 Clinical evidence
Four qualitative studies17-20 were identified. Information from the studies was further synthesised
into themes (see Table 11) and has been summarised in modified clinical evidence profiles (see Table
12, Table 13, Table 14, Table 15, Table 16).
Table 11: Themes from qualitative studies on peoples’ experiences of peripheral arterial disease
and their information requirements
Main theme Sub-themes
Impact of disease  Disease severity
 Pain
 Physical function/physical symptoms
 Mental health/emotional function
 Social/role function
 Sense of self
Perceptions and beliefs No sub-themes
Needs and concerns  Physical
 Mental health/emotional
Final version
43
PAD
 Social
 Support
 Information
Expectations No sub-themes
Strategies for adaption/improvement/scoping No sub-themes
Final version
44
PAD
Table 12: Evidence profile: Theme 1 – Impact of peripheral arterial disease

No. of studies Design Sample Themes(a) Quality assessment(b)
Sub theme: Disease severity
119 1:1 interviews N=38 PAD  Many people expressed both positive and negative feelings  High quality
 Those with more severe disease expressed more negative feelings  Transferable to population
addressed
Sub-theme: Pain
317,19,20 1:1 interviews N=9 post-surgery17  Pain was a common outcome for most people17,19,20  High quality
 Pain was mainly pre-operative17  Transferable to population
N=38 PAD19
 Pain resulted in: addressed
o cramping, aching, burning, fatigue19
N=24 PAD20
o sleep disturbance17
o loss of quality of life17

Sub-theme: Physical function / physical symptoms
317,19,20 1:1 interviews N=9 post-surgery17 Effects on physical function/physical symptoms included:  High quality
 Altered sensation (e.g. coldness/deadness of limb) 17  Transferable to population
N=38 PAD19
 Non-healing wounds20 addressed
N=24 PAD20  Carrying a hard-to-bear physical burden and struggling for relief20
 Restricted mobility/walking impairment/walking slowly for short
distances (compromising independence and physical activities at
home or work, recreational “becoming an invalid”), quality of life,
social and emotional function, accomplishing goals17,19,20
 Fatigue (sleep disturbance, lack of energy)20
Sub-theme: Mental health / emotional function
217,20 1:1 interviews N=9 post-surgery17  Carrying a hard-to-bear emotional burden and struggling for relief20  High quality
 Emotional change (often due to lifestyle changes or health status):  Transferable to population
N=24 PAD20 o depression17,20 addressed
o mood and temper influenced by pain20
o having to ask for help20
o despair20
Final version
45
PAD
o powerlessness/feeling useless (sometimes due to direct

effects of condition and treatment modalities) 17,20
Sub-theme: Social/role function
317,19,20 1:1 interviews N=9 post-surgery17  Impact on/changed interaction with relationships and friends17,20  High quality
 Carrying a hard-to-bear social burden and struggling for relief20  Transferable to population
N=38 PAD19  Isolation and loss of independence (restricting freedom, loneliness, addressed
missing previous activities and social activities, loss of interest)17,19,20
N=24 PAD20  Limitation in social and role functioning:19
o inadequacy (slowing down friends or family)
o being a burden to family (other people having to bear
responsibility for supporting the family)
o role and employment limitations (threat of job loss; need to
change jobs; loss of opportunity for promotion)
o homemakers expressed inability to fulfil role (including parenting)
Sub-theme: Sense of self
217,19 1:1 interviews N=9 post-surgery17 Compromise of self:  High quality
 Compromising sense of wholeness19  Transferable to population
N=38 PAD19  Premature aging19 addressed
 Feeling abnormal (sense of shame)19
 Unfulfilled desire19
 Loss of sense of self (“who they are”; loss of the person they used to
be, having to give up activities and independence)17,19
(a) Clarification: not all participants reported in the study sample contributed to the themes.
(b) Quality assessment included study limitations, indirectness (transferability) and other considerations.
Table 13: Evidence profile: Theme 2 – Perception and beliefs

No. of studies Design Sample Themes emerged(a) Quality assessment(b)
117 1:1 interviews N=9 Causes of illness:  High quality
Post-surgery  Role of chance in getting illness in the first place  Transferable to population
 Mostly external factors identified as causes of patients’ health addressed
problems (1 person identified responsibility due to smoking).
117 1:1 interviews N=9 Treatment and recovery:  High quality
Final version
46
PAD
Post-surgery  Role of chance in getting access to treatment and whether  Transferable to population
treatment is successful addressed
 Perceived a lack of control over course of illness; treatment not
guaranteed to work
 Believed their best chance of recovery lay in the hands of others
and their own role mostly limited to playing by the rules (e.g.
modifying lifestyle factors, partly so that medical staff haven’t
wasted their time)
 Some stopped smoking (their side of the “bargain” with medical
staff)
 Some continued smoking as much as before (disbelieving that
smoking caused their condition)
 Some reduced smoking but did not stop altogether, accepting that
smoking caused their condition but denying (to themselves or
others) that they continued to smoke (e.g. smoking in secret,
avoiding the subject, convincing themselves that smoking
occasionally did not matter).
118 Questionnaire N=60 Dissatisfaction with body structure and function (particularly women).  Low quality
interviews Pre-surgery  Poor or no report of study
design, data collection and
most elements of validity,
analysis and synthesis
methods
 Transferable to population
addressed
118 Questionnaire N=60  Twice as many women as men perceived themselves to be in  Low quality
interviews Pre-surgery control in the hospital while twice as many men as women felt  Poor or no report of study
lonely and cut off from normal family support design, data collection and
 Men were three times more likely to have financial worries due to most elements of validity,
reduction in income analysis and synthesis
methods
addressed
118 Questionnaire N=60 Older patients (vs. younger, ≤64 years) perceived:  Low quality
Final version
47
PAD
interviews Pre-surgery  Less need to follow a special diet  Poor or no report of study
 Demonstrated less awareness of the negative relationship between design, data collection and
smoking and circulatory pathology most elements of validity,
methods
addressed
118 Questionnaire N=60 Most did not relate perceived benefits of dietary management and  Low quality
interviews Pre-surgery regular foot care to vascular disease and were not following these  Poor or no report of study
practices. design, data collection and
methods
addressed
Table 14: Evidence profile: Theme 3 – Needs and concerns

Sub-theme: Physical
218,19 Questionnaire N=60 pre-surgery18 Physiological needs (smoking):  High quality(d); Low quality18
interviews &  Most considered it important to decrease or quit smoking (fear of  Poor or no report of study
1:1 interviews N=38 PAD(d) lung cancer rather than vascular disease)18 design, data collection and
 Only 26% had actually stopped18 most elements of validity,
 Addiction (patients recognised smoking as a serious issue but some analysis and synthesis
were still unable to quit even after being confronted with potential methods18
loss of limb or life)(d)  Transferable to population
addressed
317,19,20 1:1 interviews N=9 post-surgery(e) Concerns were mainly physical. The greatest and most frequent  High quality
personal concerns were:  Transferable to population
N=38 PAD(d)  Fears relating to: increased pain(e), loss of function(d), addressed
amputation(d)(e), death(d), taking pills and unwanted effects(f)
N=24 PAD(f)  Treatment or operation failure(e)
Final version
48
PAD
 Hospitalisation(e)
 Need for surgery(e)
Sub-theme: Mental health / emotional
118 Questionnaire N=60 pre-surgery18 Psychosocial needs:  Low quality
interviews  Difficulties coping with alterations in self-concept and role function  Poor or no report of study
were closely related design, data collection and
 Most people were unhappy with changes that had occurred with most elements of validity,
the progression of their disease: felt useless, frustration and analysis and synthesis
depression with their situation and with their perceived inability to methods
cope with it.  Transferable to population
 Less than half of people felt themselves to be in control during addressed
hospitalisation.
 People perceived a need to have a sense of control over life / the
future.
 Anxiety about the effect of surgery on disease progression (more
than about hospitalisation itself).
Sub-theme: Social
218,19 Questionnaire N=60 pre-surgery18  Social needs/concerns:  High quality(d); Low quality18
interviews o Loneliness and separation from families18  Poor or no report of study
1:1 interviews N=38 PAD(d) o Loss of independence(d) design, data collection and
methods18
addressed
Sub-theme: Support
118 Questionnaire N=60 Need support for:  Low quality
interviews Pre-surgery  Difficulties coping with negatively perceived changes in self-concept  Poor or no report of study
 Alterations in role relationships design, data collection and
 Anxiety about the effect of surgery on disease progression most elements of validity,
 General operative support measures
methods
Most people found the following useful:
Final version
49
PAD
 Pre-operative passive support measures (wanted a friendly, positive addressed

atmosphere and emphasised the importance of considering
patients as people, not just individuals with a particular disease
condition).
 Physical and emotional support in pre-operative period (women)
 Emotional support in pre-operative period (men)
 General nursing support (older people vs younger, ≤64 years)
Investigator identified the following needs for support:

 Active emotional support by nurses
 Fostering sense of control
 Reducing anxiety
 Enhancing family support
Sub-theme: Information
118 Questionnaire N=60 Investigator identified a need for information on preventive health  Low quality
interviews Pre-surgery behaviours (diet, smoking, foot care, use of analgesics)  Poor or no report of study
design, data collection and
methods
addressed
218,19 Questionnaire N=60 People identified needs for information on:  High quality(d); Low quality18
interviews Pre-surgery18  Pre-operative information (to decrease anxiety, but many did not  Poor or no report of study
1:1 interviews wish to know “too much” and some desired no information at all. design, data collection and
N=38 PAD(d) Older patients desired less pre-operative information than younger most elements of validity,
patients (≤64 years) 18 analysis and synthesis
 Aortographic procedures under local anaesthetic (felt they had not methods18
been adequately prepared and experienced discomfort) 18  Transferable to population
 Knowledge of side-effects of analgesics (many people taking large addressed
amounts) 18
 Knowledge of disease and importance of risk factor management (d)
 Lack of control(d)
Final version
50
PAD
119 1:1 Interviews N=38 Delay in diagnosis and frustration with management of disease:  High quality
 Person’s delay due to not recognising symptoms (e.g. thinking it  Transferable to population
PAD was a normal part of aging) addressed
 Clinician delay (e.g. going to several doctors before getting
diagnosis)
Table 15: Evidence profile: Theme 4 – Expectations of people with peripheral arterial disease
117 1:1 Interviews N=9 Cause and management of illness:  High quality
Post-surgery  The “acute” style of management of PAD led to unrealistic  Transferable to population
expectations, and gave rise to powerlessness. addressed
Participation in decisions someone else’s problem:

 Little evidence of participation in decisions over whether or not to
have surgery (accepting medical advice; faith in medical system;
expecting “clear results” and surgery to be a cure; sick role; external
locus of control).
Prior to surgery:
 Expectations were unrealistic and positive (e.g. belief operation
would get things “back to normal” and “that would be it”).
After surgery:
 When it became apparent that surgery had not restored their
function as much as they hoped, expectations were tempered by
realism expressed positively (e.g. “it’s done what it’s meant to do
really”) or negatively (“I can’t see me getting any better”)
 Concerned and disappointed when pain persisted after they
expected to have recovered (may be related to unrealistic hope in
the power of medicine to alleviate symptoms and focus on surgery
as a cure)
 Expected pain to be considerable in the early post-operative period
Final version
51
PAD
but then to reduce rapidly and not recur.

Table 16: Evidence profile: Theme 5 – Strategies for adaptation/improvement/coping

217,20 1:1 Interviews N=9 post-surgery(c) Acceptance:  High quality
 Being realistic, facing up to problems, lowering expectations)(c)  Transferable to population
N=24 PAD(d)  Trying to create sense of normality(c) addressed
 Adjusting to changed social relationships(c)
 Dealing with role changes(c)
 Reorientation (adjusting activities, taking on new interests [e.g.
reading] to compensate for loss of old ones, positive thinking [e.g.
others worse off])(d)
 Resignation (being realistic, facing up to problems, lowering
expectations, giving responsibility to healthcare professionals)(d)
 Struggling against loss of independence (modifying routines to
maintain some control [e.g. walking where they could rest]) (c)(d)
 Struggling to not accept limitations but live as normally as possible
(e.g. exercises and keeping in good shape)(d)
217,19 1:1 Interviews N=9 post-surgery(c) Control:  High quality
 Tried to maintain control of factors within their remit; maintaining  Transferable to population
N=38 PAD(e) independence (e.g. shopping)(c) addressed
 Adaptation to the effects of the disease and demonstration of
resiliency (adjustment, flexibility)(e)
217,20 1:1 Interviews N=9 post-surgery(c) Adaptations to physical limitations:  High quality
 To deal with pain pre-operatively medication and alteration of  Transferable to population
N=24 PAD(d) activity (but had little effect)(c) addressed
 Learned by trial and error(c)
 Allowed for day-to-day variations in ability(c)
 Prioritising activities and carrying them out efficiently with suitable
resting places(c)
 Relieving pain and promoting circulation (pain unpredictable;
Final version
52
PAD
analgesics used, changing position of leg; distracting activities [e.g.

TV])(d)
 Managing non-healing wounds (looking after wounds, trying
different bandages, letting professionals take care of wounds)(d)
(c) Gibson, 199817
(d) Wann-Hansson et al, 200520
(e) Treat-Jacobson, 200219
Final version
53
PAD
5.1.2 Economic evidence

No cost effectiveness evidence was identified for this question.
5.1.3 Evidence statements
5.1.3.1 Clinical
Four qualitative studies of high to low quality17-20 with a total of 131 participants, showed the
following findings about the experiences and information requirements of people with PAD:
 Pain, restricted mobility/walking impairment, depression, anxiety and sleep disturbance were
problems for many people with PAD.
 PAD had a major impact on people’s sense of self (who they were) and limitations on their social
and role functions (feelings of isolation, loss of independence, burden to friends and family,
missing out on previous activities and social activities, limitations on work).
 People with PAD did not feel in control of their illness, many did not believe that modifying
lifestyle (including diet and smoking) would help their condition, and often felt that treatment
may not work.
 Their concerns were mainly ‘physical’ and many had fears of: increased pain, loss of function,
amputation, failure of operations or other treatment.
 Psychosocial concerns and needs included: loss of independence, loneliness and separation from
families, feeling out of control and difficulties coping.
 Most people with PAD and investigators felt support was needed for: coping with negative
changes (e.g. control, self, role relationships, family support, anxiety about surgery), and found
that pre-operative support measures, physical and emotional support and general nursing
support was useful.
 People with PAD had unrealistic expectations of the management of PAD and the results of
surgery (particularly on pain and function - which led to feelings of powerlessness), and expected
that it would be a cure.
 People often experienced a delay in diagnosis (due to not recognising symptoms and perceived
clinician delays), and expressed a need for information on a number of areas including: disease
and risk factor management, preventative health behaviours, aortographic procedures, lack of
control, pre-operative information to reduce anxiety, and the adverse events of analgesics.
 Strategies people adopted for coping with / adapting to living with PAD included:
o acceptance
o re-orientation (e.g. finding new activities they could do and modifying routines and activities to
maintain some feeling of control and independence).
5.1.3.2 Economic
5.1.4 Recommendations and link to evidence
1. Offer all people with peripheral arterial disease oral and written
information about their condition. Discuss it with them so they
can share decision-making, and understand the course of the
disease and what they can do to help prevent disease
Recommendations progression. Information should include:
Final version
54
PAD
 the causes of their symptoms and the severity of their disease

 the risks of limb loss and/or cardiovascular events associated
with peripheral arterial disease
 the key modifiable risk factors, such as smoking, control of
diabetes, hyperlipidaemia, diet, body weight and exercise (see
also recommendation 3)
 how to manage pain
 all relevant treatment options, including the risks and benefits
of each
 how they can access support for dealing with depression and
anxiety.
Ensure that information, tailored to the individual needs of

the person, is available at diagnosis and subsequently as
required, to allow people to make decisions throughout the
course of their treatment.
2. NICE has produced guidance on the components of good patient

experience in adult NHS services. Follow the recommendations in
Patient experience in adult NHS services (NICE clinical guideline
138).
Relative values of different The aim of the evidence review was to identify:
outcomes  Experiences of living with PAD
 Information people with PAD wanted or found useful
 If there are specific information requirements for people with PAD
 If information received changed the perception of the disease.
A number of important themes emerged from the qualitative evidence review

on patient information needs and requirements. In particular, the review
highlighted that people with PAD require:
 Psychosocial support as well as medical treatment. Such support would
address issues with coping, depression and anxiety as well as perceptions
and beliefs around the disease management.
 Encouraging autonomy and shared decision making.
 Managing expectations through ensuring that patients have realistic
expectations and understanding of PAD.
Trade off between clinical Information needs and requirements will change during the course of the
benefits and harms disease and must be tailored to this.
Healthcare professionals must be aware of the impact of information on

patients. This may have a negative impact or may be mis-understood.
However, if delivered adequately it should aid in understanding the disease,
encourage self-management and involvement in the decision-making.
Economic considerations The GDG discussed patient information in the context of routine healthcare
practice. It was expected that any impact on time and resource use would be
minimal and would likely be offset by an improvement in quality of life.
Quality of evidence The evidence reviewed was either high or low quality by GRADE criteria. A
Final version
55
PAD
number of the studies were 1:1 interviews, which is considered high quality
within qualitative research. The studies reported tended to be small in sample
size. In addition, the samples consisted of people at different stages in the
treatment pathway and degree of disease severity, which may affect their
responses to questionnaires. Therefore, there has to be some caution in
attributing themes reported to all patients with PAD.
Other considerations The GDG developed the recommendations based on the clinical evidence
presented. The GDG agreed that the NICE Patient Experience (publication to be
confirmed) guideline contains general recommendations around
communication, treating the patient as an individual that healthcare
professionals should follow. Information should be available in a variety of
formats (including written and verbal information) and translations should be
available where appropriate. In addition to the Patient Experience guideline
recommendations, the GDG identified recommendations specific to people
with PAD.
Information requirements
From the clinical evidence review, the GDG discussed several aspects of
information required by people with PAD and this information informed the
recommendations. In particular,
 Addressing the disease stage and severity – is it intermittent claudication or
critical limb ischaemia? What is the prognosis?
 Lifestyle and preventative behaviours - the clinical evidence review did
highlight that there is evidence that perception and beliefs need to be
challenged. For example, some people did not believe that modifying
behaviours such as smoking, diet and exercise would impact on the disease.
This behaviour should be explored and patient educated in the importance
of these factors and the benefits of lifestyle change
 Cardiovascular risk factors - the GDG patient members highlighted that
patients are unlikely to ask or be aware of CV risk factors associated with
PAD. Therefore, it is important that the healthcare professionals inform and
reinforce this information
 Understanding of the disease process
 Restricted mobility and walking impairment.
Psychosocial aspects of PAD

There was some discussion about around the psychosocial aspects of PAD
particularly in relation to experiences of pain, loss of control, and depression
and anxiety. This may be related to the belief systems some people hold and
by changing attitudes towards PAD may alleviate stress and negative emotions.
The GDG patient members also highlighted that some patients may not be
aware that they may experience negative emotions. Some of the GDG had the
view that depression may not be routinely sought in people with PAD and that
this should be considered. NICE have produced a guideline on “Depression in
adults with a chronic physical condition”, which healthcare professionals can
consult as an additional resource when dealing with depression in people with
PAD.21 It was noted that primary care healthcare professionals do undertake
some screening of mood and anxiety of people with chronic conditions.
Other discussions
It is important to give simple summary information to patients and then assess
the impact of the information on the individual. For example, is it having an
effect on behaviours and coping styles. From this the healthcare professional
can challenge any negative beliefs.
Final version
56
PAD
There was a discussion around setting individualised care plans. This did not
emerge as a theme from the clinical review and was not a specific review
question. The GDG emphasised that all people with PAD should be fully
involved in all decision making. The NICE guideline on “Patient Experience in
adult NHS services” (CG138) contains further recommendations relating to
this, therefore the GDG did not make a specific recommendation.
The GDG felt that it was not appropriate to prepare a standardised patient
leaflet but to give some clear recommendations about what should be
included in patient discussions.
Concern was expressed that people are not always given enough consultation
time to discuss the diagnosis and treatment fully. This has been covered within
the Patient Experience guideline but was re-emphasised by the GDG.
Key priority for implementation

The GDG identified recommendation 1 about the information requirements for
people with PAD as a key priority for implementation. The reason for selecting
this recommendation for prioritisation was that there is variability in the
information given. In particular, patients may not be given sufficient
information on the benefits of secondary prevention of cardiovascular risk
factors. By highlighting this as a key priority, variation in care and outcomes
will be reduced. This recommendation also promotes patient choice.
5.1.5 Research recommendation
1. What is the effect of people’s attitudes and beliefs about their peripheral arterial disease on
the management and outcome of their condition?
Why this is important
The evidence reviewed suggested that, among people with peripheral arterial disease, there is a lack
of understanding of the causes of the disease, a lack of belief that lifestyle interventions will have a
positive impact on disease outcomes, and unrealistic expectations of the outcome of surgical
interventions. Much of the research has been conducted on the subpopulation of people with
peripheral arterial disease who have been referred for surgical intervention, but little evidence is
available for the majority of people diagnosed with peripheral arterial disease in a primary care
setting. Research is required to further investigate attitudes and beliefs in relation to peripheral
arterial disease, interventions that might influence these and how these may have an impact on
behavioural changes in relation to risk factors for peripheral arterial disease, attitudes to
intervention and clinical outcomes.
Final version
57
PAD
6 Secondary prevention of cardiovascular disease

in people with peripheral arterial disease
6.1 Introduction
Peripheral arterial disease (PAD) is strongly associated with cardiovascular disease. The modifiable
and non modifiable risk factors for PAD are shared with those for cardiovascular disease. Many
individuals with PAD will have evidence of cardiovascular disease, and people diagnosed with PAD
are at high risk of further cardiovascular events such as stroke and myocardial infarction. The severity
of PAD is a prognostic indicator of cardiovascular risk, those with the most severe symptoms faring
worse. In people with CLI, the cardiovascular mortality rate is even higher, with a one in five
mortality rate within one year of diagnosis. Although less marked even the asymptomatic group have
an increased cardiovascular risk. This observation has led to the main focus of treatment shifting to
address cardiovascular risk in people with PAD by attempting to modify their risk factors. There is
some qualitative evidence that people with PAD do not associate their symptoms with negative
behaviours such as smoking or poor diet18,19 and often have the perception that disease management
is outwith their control.17 Chapter 5 provides further information on patients beliefs, expectations
and coping with PAD. Whilst clinicians recognise and have well established protocols for the
management of risk factors in cardiovascular disease these are less well recognised and acted on in
PAD.22
There is a paucity of evidence to address risk factor modification specifically in PAD and available
evidence is usually related to subgroup analysis.23-25 Nevertheless, the strong association of PAD and
cardiovascular disease and common shared risk factors justifies extrapolation to PAD using
information from other conditions associated with atherosclerosis.
6.1.1 Reducing cardiovascular risk
6.1.1.1 Smoking
Smoking is the most important risk factor for the development of PAD and even passive smoking
increases cardiovascular risk. Excess cardiovascular risk is halved within one year of cessation and is
the same as non-smokers within 5 years in those patients that successfully give up smoking. There is
no strong evidence for the benefits of smoking cessation to the limb but some observational studies
have suggested an improvement in walking distance and a reduction in amputation rates. Smoking
cessation advice when combined with nicotine replacement therapy, for example, improves quit
rates to around 30%.
6.1.1.2 Diabetes
Diabetes is an important risk factor for PAD and the incidence and prevalence of PAD increases with
duration of both Type 1 and Type 2 diabetes.26 The effects of diabetes are compounded by later
presentation with more extensive disease27 neuropathy and risk of infection. The risk of amputation
is significantly greater in a diabetic population and over 50% of all amputations occur in people with
diabetes. No trials have been set up to examined the role of improved glycaemic control in relation
to PAD. There is evidence that improved glycaemic control influences cardiovascular disease
progression.28
Final version
58
PAD
6.1.1.3 Cholesterol management
There is overwhelming evidence for the benefits of lowering cholesterol in patients with PAD. In the
Heart Protection Study (2002),23 people with PAD with a total cholesterol over 3.5 mmol/l who took
simvastatin (a HMG-CoA reductase inhibitor) had a 17.6% reduction in cardiovascular events
compared to those on placebo. There was also a reduction in the subsequent need for both cardiac
and non-cardiac revascularisation procedures. Based on these results, nearly all people with PAD
should be prescribed statin therapy. There is also emerging evidence that statins have a direct effect
on atherosclerotic plaque, stabilising it and possibly causing plaque regression in high doses.
6.1.1.4 Hypertension
Up to 24% of the adult population are hypertensive and hypertension is associated with a 3 fold
increase risk of PAD, as well as being a strongly associated with stroke and myocardial infarction.
Treatment of hypertension will reduce stroke rates by 38% and cardiovascular deaths by 14%. In the
Heart Outcomes Study29, the angiotensin converting enzyme inhibitor, ramipril, demonstrated an
advantage in reducing cardiovascular events, even in those patients whose blood pressure was not
elevated. However, there are potential problems with widespread use of ramapril in people with PAD
as many will have renal artery disease. At present in those with PAD and hypertension ramipril
should be considered as the first line treatment but there is not enough evidence to suggest
widespread use in the non-hypertensive patients.
6.1.1.5 Anti-platelet agents
The Antithrombotic Trialists’ Collaboration27 meta-analysis found that antiplatelet agents

(predominantly aspirin, a cyclo-oxygenase inhibitor) reduced the risk of cardiovascular events by 23%
in people with PAD. 75mg was as effective as higher doses. Approximately 20% of patients are unable
to take aspirin largely due to gastrointestinal disturbance and it is emerging that a similar proportion
of patients have aspirin resistance. In these patients usual doses of aspirin do not have the normal
effect on patients. In these patients clopidogrel should be used. Clopidogrel is a theopyridine
derivative that blocks ADP induced platelet activity. In the Caprie study, clopidogrel (was shown to
further reduce cardiovascular events compared to aspirin (particularly in the PAD group) with a
relative risk reduction of 8.7%. The NICE TA 21030 recommends clopidogrel as first line option.
Combination therapy of aspirin and clopidogrel should be considered very carefully. In the Charisma
study patients on both drugs had a significantly greater risk of bleeding complications which overall
exceeded any apparent benefit.
6.1.1.6 Weight management and exercise
A number of other life style changes should be advocated. Weight reduction and regular exercise
have proven cardiovascular benefit. The role of exercise for intermittent claudication is discussed in
chapter 9. They also have a positive effect on other risk factors. Omega 3 fatty acids (fish oils) appear
to have some beneficial effects but their clinical role in PAD has not been established. Likewise
antioxidants and other dietary additives have not been demonstrated to be of benefit.
6.1.2 Existing NICE guidance and recommendations

The GDG recognised that there are existing NICE recommendations covering many of the aspects for
the secondary prevention of cardiovascular disease, which were relevant for people with PAD.
Therefore, the GDG agreed that no further evidence review was required and that recommendations
for PAD should follow existing NICE guidance.
Final version
59
PAD
6.2 Recommendation
3. Offer all people with peripheral arterial disease information,
advice, support and treatment regarding the secondary
prevention of cardiovascular disease, in line with published NICE
guidance (see 'Related NICE guidance'; section 2.6) on:
 smoking cessation
 diet, weight management and exercise
 lipid modification and statin therapy
 the prevention, diagnosis and management of diabetes
 the prevention, diagnosis and management of high blood
pressure
Recommendation  antiplatelet therapy.
6.2.1 Key priority for implementation

The GDG identified this recommendation as a key priority for implementation. The GDG were of the
opinion that guidance relating to cardiovascular disease and secondary prevention is given in an
inconsistent way to people with PAD. The appropriate and consistent application of information on
secondary prevention of cardiovascular disease is likely to have high impact on patient outcomes and
reducing variation in care and outcome.
Final version
60
PAD
Diagnosis of peripheral arterial disease
7 Diagnosis of peripheral arterial disease

7.1 Introduction
People with suspected PAD most commonly present with pain in the leg muscle brought on by
exertion. They may also present with other leg and foot symptoms such as rest pain, foot ulcers or
tissue loss. PAD can be found in asymptomatic patients attending, for example, a general
examination or diabetic foot screening. They will most likely present to GP's, nurses or allied health
professionals in primary care.
The diagnosis of PAD is based on a good clinical history and a clinical examination including the
palpation of femoral, popliteal and pedal pulses, and when this is done by an experienced clinician
additional diagnostic tests may well be unnecessary. A readily available test which is often performed
is the ankle brachial pressure index (ABPI), which is simply the measurement of resting systolic ankle
blood pressure divided by the systolic brachial pressure. An ABPI ratio of <0.9 is an indicator of PAD.
However, a normal resting ABPI (>0.9) does not exclude its presence. The measurement of ABPI is
user dependant and has its limitations when used in patient with swollen limbs or where arterial wall
calcification is present, such as in some of the diabetic population. The GDG therefore wished to
assess the utility of measurement of ABPI in the diagnostic work-up of suspected PAD.
Other forms of imaging are sometimes utilised to diagnose PAD and are able to delineate the site and
severity of arterial lesions producing the signs and symptoms of PAD, but they are usually not
necessary for diagnosis per se. This guideline did not consider imaging for diagnosis but did review
the evidence for its role in assessment for revascularisation (see chapter 8).
7.2 Methods of diagnosis of peripheral arterial disease

In people with suspected PAD, is ABPI as an adjunct to clinical assessment better than clinical
assessment alone or ABPI alone, in determining the diagnosis and severity of PAD?
A literature search was conducted for diagnostic studies that compared the diagnostic accuracy of
clinical assessment, ABPI or ABPI with clinical assessment, to the reference standard of imaging in
people with suspected PAD.
Suspected PAD was described as symptoms of intermittent claudication (IC), leg ulcers, common foot
problems or having cardiovascular risk factors; indirect populations (such as a general population
without suspected PAD) and emergency settings were excluded.
Five studies31-35 were identified which addressed the question and were included in the review.
Specifically:
 Two studies compared manual ABPI using Doppler to angiography31,32
 Two studies compared manual ABPI using Doppler to duplex ultrasound33,34
 One study compared manual ABPI without Doppler to angiography31
 One study compared automated oscillometric method of ABPI to angiography35
 Two studies32,33 considered patients with diabetes. No diagnostic studies were identified
comparing ABPI and clinical assessment or clinical assessment alone to imaging. None of studies
Final version
61
PAD
reported on subgroups for people with diabetes, asymptomatic PAD or people with renal
failure/advanced renal disease for the outcomes.
The studies are summarised in the clinical evidence profiles below (Table 17, Table 18 and Table 19).
See also the full study evidence tables in Appendix H. Diagnostic forest plots are presented in
Appendix J.
Final version
62
PAD
Table 17: Clinical evidence profile: Manual ankle brachial pressure index using Doppler compared to imaging
Quality Assessment Summary of Findings
No of Design No of Sensitivity Specificity PPV NPV
consideration
Inconsistency
Indirectness
Imprecision
Risk of bias
studies patients
Other
Quality
Reference standard – Angiography; ABPI cut-off <1.0
131 Cross 20 No serious No serious No serious Serious(a) None 92% 80% NR NR MODERATE
sectional risk of bias inconsistency indirectness
study
Reference standard – Angiography; ABPI cut-off <0.5; people with diabetes
132 Cross 106 No serious No serious No serious serious(a) None 36% 86% 67% 64% MODERATE
study
Reference standard – Angiography; ABPI cut-off <0.7; patients with diabetes
132 Cross 106 No serious No serious No serious Serious(a) None 59% 67% 58% 68% MODERATE
study
Reference standard – Angiography; ABPI cut-off <0.9; patients with diabetes
study
Reference standard – duplex ultrasound; ABPI cut-off <0.9 patients with diabetes
133 Cross 100 No serious No serious No serious Serious(a) None 70.6% 88.5% 94.1% 53.4% MODERATE
study
Reference standard – duplex ultrasound; lower ankle pressure ABPI cut-off <0.9
study
Final version
63
PAD
Reference standard – duplex ultrasound; higher ankle pressure ABPI cut-off ≥0.9
study
(a) No confidence intervals reported.
Abbreviations: PPV=Positive predictive value; NPV=Negative predictive value; NR=Not reported.
Table 18: Clinical evidence profile: Manual ABPI without Doppler compared to angiography
Quality Assessment Summary of findings
consideration
Inconsistency
Indirectness
Imprecision
Risk of bias
studies patients
Other
Quality
Reference standard – Angiography; ABPI cut off <1.0
study
Table 19: Clinical evidence profile: Automated oscillometric ABPI compared to angiography
consideration
Inconsistency
Indirectness
Imprecision
Risk of bias
studies patients Quality
Other
Reference standard – Angiography
ABPI cut-off 0.53
135 Cross 298 No serious No serious No serious Serious(a) None 14.3% 100% NR NR MODERATE
study
Final version
64
PAD
ABPI cut-off 0.9

study
ABPI cut-off 0.95
study
ABPI cut-off 1.12
study
Final version
65
PAD
7.2.1.2 Economic evidence
No cost effectiveness evidence was identified in the literature search. In the absence of published
economic evaluations, the GDG were asked to estimate the additional resource use associated with
obtaining a measure of ABPI compared to performing clinical assessment alone.
The GDG agreed that ABPI is typically performed by a practice nurse or podiatrist while taking a
clinical history. It may add between 5 and 15 minutes to the time required for the clinical
examination. In some instances patients may be referred to a different healthcare provider if they do
not have access to equipment or expertise. Clinicians may attend supervised training placements
which may be considered part of the overhead cost associated with ABPI. Similarly, the purchase of a
manual or automated device for measuring ABPI incurs a onetime cost to each healthcare centre.
With correct care, these devices have a lifespan of many years.
7.2.2.1 Clinical
Manual ABPI using Doppler compared to angiography

 Manual ABPI using Doppler with a cut-off < 1.0 had a sensitivity of 92% and specificity of 80%
compared to the reference standard of angiography [1 study, 20 participants, moderate quality
evidence]31
 Manual ABPI using Doppler with a cut-off <0.5 had a sensitivity of 36%; specificity of 86%; positive
predictive value of 67% and negative predictive value of 64% compared to the reference standard
of angiography [1 study, 106 participants, moderate quality evidence]32
Manual ABPI using Doppler compared to duplex ultrasound

 Manual ABPI using Doppler with a cut-off <0.9 had a sensitivity of 70.6%; specificity of 88.5%;
positive predictive value of 94.1% and negative predictive value of 53.4% compared to the
reference standard of duplex ultrasound [1 study, 100 participants, moderate quality evidence]33
 Manual ABPI using Doppler with the lower ankle measurement cut-off <0.9 had a sensitivity of
89%; specificity of 93%; positive predictive value of 93% and negative predictive value of 88%
compared to the reference standard of duplex ultrasound [1 study, 216 participants, moderate
quality evidence]34
 Manual ABPI using Doppler with the higher ankle measurement cut-off ≥0.9 had a sensitivity of
68%; specificity of 99%; positive predictive value of 99% and negative predictive value of 74%
compared to the reference standard of duplex ultrasound [1 study, 216 participants, moderate
quality evidence]34
Manual ABPI without Doppler compared to angiography

 Manual ABPI without Doppler with cut off <1.0 had a sensitivity of 100% and specificity of 40%
compared to the reference standard of angiography [1 study, 20 participants, moderate quality
evidence]31
Final version
66
PAD
Automated oscillometric ABPI without Doppler compared to angiography

 Automated oscillometric method of ABPI with cut-off 0.53 had a sensitivity of 14.3% and
specificity of 100% compared to the reference standard of angiography [1 study, 298 participants,
moderate quality evidence]35
 Automated oscillometric method of ABPI with cut-off 0.9 had a sensitivity of 76% and specificity
of 90% compared to the reference standard of angiography [1 study, 298 participants, moderate
quality evidence]35
quality evidence]35
quality evidence]35
7.2.2.2 Economic
4. Assess people for the presence of peripheral arterial disease if

they:
 have symptoms suggestive of peripheral arterial disease or
 have diabetes, non-healing wounds on the legs or feet or
unexplained leg pain or
 are being considered for interventions to the leg or foot or
 need to use compression hosiery.
5. Assess people with suspected peripheral arterial disease by:

 asking about the presence and severity of possible symptoms of
intermittent claudication and critical limb ischaemia
 examining the legs and feet for evidence of critical limb
ischaemia, for example ulceration
 examining the femoral, popliteal and foot pulses
 measuring the ankle brachial pressure index (see
Recommendations recommendation 6).
Relative values of different No evidence was found comparing clinical assessment to ABPI. There was
outcomes evidence relating to the accuracy of ABPI in comparison to imaging for the
diagnosis of PAD. The GDG considered that the most important outcomes were
sensitivity and negative predictive values since their main concern was to avoid
missing any people with peripheral arterial disease. An ABPI of 0.9 therefore
appears to be the best indicator for PAD. The values seen varied between the
available studies, and also inevitably varied when different cut-off values of
ABPI were used.
The GDG concluded that the studies offered some support for the use of ABPI,
since it shows acceptable predictive values when compared to the results of
the less readily available “gold standard” imaging techniques. Unfortunately
none of the trials addressed the key question posed by the GDG, of the added
value of ABPI to a careful clinical examination.
Final version
67
PAD
Trade off between clinical ABPI is a non-invasive test and there are no recognised dangers of correct use
benefits and harms of equipment. It is important that healthcare professionals are appropriately
trained as failure to correctly measure ABPI may result in a mis-diagnosis,
thereby delaying referral or treatment.
Economic considerations The GDG considered the additional resources required for obtaining an ABPI
compared to clinical examination alone. Based on expert opinion, the GDG
thought that the incremental resource requirements associated with
measuring ABPI were small compared to the benefit of accurately identifying
people with suspected PAD. Accurate diagnosis would be expected to improve
quality of life and save costs by ensuring that patients are managed
appropriately.
Quality of evidence The evidence relating to the diagnostic accuracy of ABPI was deemed to be of
moderate quality by GRADE criteria, based on the QUADAS checklist. A
diagnostic meta-analysis was not undertaken as 4 or more studies are
required, as described in the methodology chapter of the guideline. No studies
were found comparing clinical assessment alone, or clinical assessment with
ABPI, to imaging as a reference standard. The GDG noted the variation in the
studies in baseline patient characteristics and healthcare settings, where the
diagnostic tests were performed. However, the GDG did not feel these
differences biased the results of the studies.
Other considerations The GDG made the recommendations based on consensus. It is currently
common practice for patients to be misdiagnosed, referred for treatment
when they do not have PAD, or for referral to be delayed due to incorrect
diagnosis. It was the opinion of GDG that making a diagnosis of PAD requires
three forms of assessment.
 Structured questioning about the symptoms of PAD is required. This not only
aids in the diagnosis but can also be used to indicate whether referral to a
specialist service is required i.e. referral for revascularisation if symptoms
are severe and lifestyle limiting. There are a number of valid PAD
questionnaires available, which can be used. The GDG did not make a
recommendation on a specific questionnaire, as an assessment of these was
not specified in the scope of the guideline and was not part of the evidence
review.
 Careful examination of the peripheral pulses is an important and basic
principle in diagnosing PAD. The examination should also include an
assessment of the signs which might be associated with critical limb
ischaemia, such as temperature of the limb, hair loss, or ulceration.
 Measuring ABPI – this is discussed further in the next section.
In the opinion of the GDG, neither clinical history, examination nor ABPI alone
is sufficient to diagnose PAD. All three methods in combination would be
useful. However, even in combination they are not universally accurate, and
clinicians should use their judgement and refer if necessary.
The GDG were also aware that peripheral arterial disease might be detected
without being the primary cause for presentation, and when symptoms are
absent or minimal. For example, absent pulses might be detected during an
assessment for some other form of leg surgery, or an abnormal ABPI might be
round when assessing for compression hosiery. The GDG agreed that there
were a group of people, as defined in the recommendation, for whom it was
important to maintain a high index of suspicion of the presence of PAD and
Final version
68
PAD
that it was important to assess them fully to diagnose or exclude the presence
of PAD in line with the subsequent recommendations. The recommendations
which follow in this guideline would equally apply in such cases.

The GDG identified recommendation 4 as a key priority for implementation as
failure to suspect and identify PAD in these groups of people, particularly if
invasive procedures are carried out or compression applied, may lead to
serious complications and adverse outcomes.
The GDG identified recommendation 5 as a key priority for implementation as

the diagnosis of PAD is currently subject to considerable variability, in
particular to the extent in which clinician’s measure pulses and ABPI. As such,
this recommendation would have the potential to improve the accurate
diagnosis of PAD, timely referral and thereby, improving patient outcomes.
7.3 Measuring the ankle brachial pressure index

In people with suspected PAD undergoing ABPI, do different methods result in different diagnostic
accuracy?
A literature search was conducted for diagnostic studies that compared the effectiveness of different
techniques for taking an ABPI in people with suspected PAD. The comparisons of types of ABPI
considered were:
 Duration of the rest period prior to measurements
 Sitting versus lying down during measurement or seated Doppler ABPI measurement compared to
supine Doppler ABPI measurement
 Location of the cuff
 Higher compared to lower vessel measurement
 Automated compared to manual device.
No time limit was placed on the literature search, and there were no limitations on sample size.
Suspected PAD was described as symptoms of IC, leg ulcers, common foot problems or having
cardiovascular risk factors, indirect populations and emergency settings were excluded.
One study was identified which compared seated Doppler ABPI measurement to supine Doppler ABPI
measurement.36 Seated Doppler measurement of ABPI was described as ABPI measured in the seated
position after the supine measurements had been taken and additional 5 minute rest period applied.
Supine Doppler measurement of ABPI was described as ABPI measured after 10 minutes of rest in
the supine position.
No studies were found on the duration of the rest period prior to measurements or the location of
the cuff. None of the studies reported on subgroups for people with asymptomatic PAD, diabetes or
people with renal failure/advanced renal disease for the outcomes.
The study is summarised in the clinical GRADE evidence profile below (Table 20). See also the full
clinical evidence tables in Appendix H. Diagnostic forests plots are presented in Appendix J.
Final version
69
PAD
Table 20: Clinical evidence profile: Seated Doppler ABPI compared to supine Doppler ABPI
No. of Design No. of Risk of Inconsistency Indirectness Imprecision Other Correlation co-efficient
studies patients bias considerations
Quality
36 (a)
1 Cross 106 No serious No serious No serious Serious None 0.936 (p < 0.001) MODERATE
study
Final version
70
PAD
The GDG discussed the costs associated with manual and automatic ABPI devices. The acquisition
cost of automatic devices is typically two to three times greater than manual devices. The group
acknowledged that the manufacturers of these devices often claim that they are time saving and, by
extension, that they are cost saving. However, several GDG members had conducted informal
evaluations within their centres and found manual devices to be both more reliable and faster than
automatic devices; it was their clinical experience that automatic devices often fail to produce a valid
reading and cannot be used on a large proportion of people with suspected PAD.
7.3.2.1 Clinical
Seated compared to supine Doppler ABPI
The correlation co-efficient between seated and supine Doppler ABPI was 0.936 (p < 0.001) [1 study,
106 participants, moderate quality evidence]36
7.3.2.2 Economic
No cost effectiveness evidence was identified.
6. Measure the ankle brachial pressure index in the following way:

 The person should be resting and supine if possible.
 Record systolic blood pressure with an appropriately sized cuff
in both arms and in the posterior tibial, dorsalis pedis and,
where possible, peroneal arteries.
 Take measurements manually using a Doppler probe of suitable
frequency in preference to an automated system.
 Document the nature of the Doppler ultrasound signals in the
foot arteries.
 Calculate the index in each leg by dividing the highest ankle
Recommendation pressure by the highest arm pressure.
Relative values of different This evidence review was conducted to identify the best technique for ABPI
outcomes measurement. One paper was found comparing seated and supine
measurements of ABPI, but this only reported the correlation co-efficient.
Although a high correlation was observed, this does not give an indication as to
whether seated or supine measurement elicits a more accurate ABPI.
No outcome data was available for the duration of the rest period prior to
assessment, the optimal cuff location, higher compared to lower vessel
measurements.
Trade off between clinical No evidence was available on benefits versus harms for different methods for
benefits and harms ABPI, but this is unsurprising since the measurement is non-invasive and
unlikely to cause any harm if carried out correctly. The main concern is that the
Final version
71
PAD
ABPI is accurately measured to avoid a misdiagnosis.
Economic considerations The GDG discussed the resource use associated with manual and automatic
devices for measuring ABPI. Based on clinical experience, the GDG considered
hand held Doppler devices to be less expensive and more reliable than
automatic devices.
Quality of evidence The evidence reviewed for this question was rated as moderate quality by
GRADE criteria.
The recommendations were based on GDG consensus and clinical experience

as no definitive evidence was found. The GDG also extrapolated the clinical
evidence on automated compared to manual ABPI presented in section 7.2 to
inform their recommendation on taking ABPI measurements manually.
Other considerations The aim of the evidence review was to identify the techniques to measure
ABPI. Due to the lack of evidence, the GDG made consensus recommendations
on the standard method to measure ABPI. The following areas were discussed
and rationale for their inclusion into the recommendation given.
Seated compared to supine measurement

One study compared seated and supine ABPI measurements.36 However, the
study only reported the correlation co-efficient. The GDG agreed that the
person should be supine where possible when measuring ABPI. Lying supine
equalises the blood pressure in the brachial and lower limb systems. Where it
is not possible and the person is seated, the height difference between the
arm and ankle should be noted and the reading adjusted appropriately. This
adjustment is not being done in routine practice and is a reason for variation in
results. Taking a seated ABPI measurement in practice is sometimes necessary,
for example, in someone in a wheelchair with significant pain or mobility
problems.
Cuff size
The cuff should be placed on the calf. The cuff size is important in the
measurement of ABPI, as an incorrect or ill-fitting cuff will lead to an incorrect
ABPI measurement. No specific cuff size was recommended; the cuff needs to
fit comfortably around the patient’s limb. A range of cuff sizes should be
available when measuring ABPI.
Advice regarding appropriate cuff sizes for arm use is given in the NICE
guideline on Hypertension (CG127) and the GDG were of the view that similar
considerations apply to choice of cuff size for the lower limb. This guideline
quotes the following: “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.”
Rest period prior to ABPI measurement

No evidence was found to recommend a minimum rest period before taking
ABPI. The GDG agreed that the rest period should be long enough for blood
Final version
72
PAD
pressure to return to normal, but be practical for the running of clinics.
Measuring blood pressure

It is considered standard practice that blood pressure be measured in the arms
and legs using a Doppler probe. The GDG agreed that it was appropriate to
recommend this. Doppler probes may be of different frequencies as they are
used for different purposes. To adequately measure the peripheral vessels the
recommended range is between 7-10 MHz but that 8MHz is average. A range
of Doppler probes should be available when taking an ABPI.
Arteries to be measured
Although no evidence was reviewed to determine which arteries should be
measured, the GDG considered this an important aspect in the correct
measurement of ABPI and therefore in the diagnosis of PAD. The GDG
recognised that in most clinical practice only two arteries are measured.
However, it was noted that some people, particularly those with diabetes, may
only have a pulse in the peroneal artery in the foot. Therefore it is important to
attempt to measure all three arteries. The GDG recognised that it can be
difficult to identify all three, but felt that assessment should, where possible,
include the peroneal artery.
Manual compared to automated ABPI measurements

The evidence, as presented in section 7.2, whilst indirect supports the use of
manual ABPI measurement over automated and the GDG made a
recommendation to use a handheld manual Doppler based on this evidence. In
addition to the evidence, the GDG were of the opinion that the automated
methods are unreliable and do not always give an accurate reading.
Documenting ABPI measurements

The GDG emphasised that ABPI measurements should always be noted in
patient case-notes to allow for future comparison. In addition, the method (i.e.
lying down or sitting, level of cuff, length of rest period) should also be noted
along with any abnormal signals.
How to calculate ABPI

The GDG considered it necessary to recommend the method of calculation of
ABPI. Different values can be obtained depending on whether higher or lower
foot or arm pressures are taken. By stipulating a standard method variability
will be minimised and improve the diagnosis of PAD. A calculation on
measuring this is given below.a
Other considerations
It is important that people with tissue loss and/or painful limbs should still
have an ABPI measured. Falsely elevated ankle pressures can occur in diabetes
and renal failure, which should be borne in mind but should not preclude its
use.

The GDG identified this recommendation as a key priority for implementation
as the measurement of PAD is currently subject to considerable variability. As
a Corrected ankle pressure _ Measured ankle pressure – D*(.078), where D _ the vertical distance between the arm and
ankle cuffs (mm). This formula equates to a correction factor of 78 mm Hg per meter distance between the arm and
ankle cuffs from Validation of a method for determination of the ankle-brachial index in the seated position. Gornik, B
Garcia, Wolski, DC. Jones, KA Macdonald, RVT and A Fronek.
Final version
73
PAD
such, this recommendation would have the potential to improve the accurate
diagnosis of PAD, timely referral and thereby, improving patient outcomes.
Final version
74
PAD
Imaging for revascularisation in peripheral arterial disease
8 Imaging for revascularisation in peripheral

arterial disease
8.1 Introduction
In people with PAD when interventional treatment is being considered, further assessment by
diagnostic imaging is indicated. This is important as the extent and location of any narrowing
(stenosis) or blockage (occlusion) of the arteries to the legs will determine the treatments that may
be available or appropriate to improve the blood flow (revascularisation).
Available diagnostic imaging modalities include duplex ultrasound scanning (DUS), magnetic
resonance angiography (MRA), computed tomographic angiography (CTA) and digital subtraction
angiography (DSA). DUS and MRA resonance imaging offer the least invasive options and avoid the
use of ionising radiation. DUS offers the unique advantage of functional assessment of arterial
stenosis, but it is the most operator dependent of the available techniques. MRA imaging provides a
three dimensional map of the imaged vessels and is able to image the pelvic vessels with more
reproducibly than DUS. However, MRA may be contraindicated in some patients, for instance those
with pacemakers and advanced renal insufficiency. CTA and DSA both require injection of contrast
media, with attendant risks to renal function, and exposure to ionising radiation. DSA requires
insertion of a catheter usually via the femoral artery and is now infrequently performed as a primary
imaging modality.
The choice of imaging modality used will be influenced to some extent by local expertise and
availability of imaging equipment. In general terms, a less invasive and lower cost strategy is
preferred. The purpose of imaging people with PAD is to determine the severity and distribution of
the lesions affecting the arterial tree and to plan and optimise any therapeutic intervention. As a
result of imaging, some people may be excluded from further intervention while others may be
selected for surgical or endovascular management.
8.2 Review question

What is most clinical and cost-effective method of assessment of PAD (intermittent claudication and
critical limb ischemia)?
The GDG were interested in looking at pre-interventional assessment of stenosis and occlusion for
people with PAD. The review question for this clinical guideline updated part of the HTA “A
systematic review of duplex ultrasound, magnetic resonance angiography and computed
tomography angiography for the diagnosis and assessment of symptomatic, lower limb peripheral
arterial disease”.37 The HTA reviewed the diagnostic accuracy of DUS, MRA and CTA, alone or in
combination, for the assessment of lower limb peripheral arterial disease. The review question for
this guideline updated the HTA analysis on this objective. The HTA also addressed the impact of
assessment method on patient management and outcomes, studies of patient attitude, adverse
events and economic evaluations. However, these objectives were not addressed in the review.
The review followed the HTA protocol. A literature search was conducted updating the HTA search
from May 2005, for diagnostic cohort or case control studies that compared the effectiveness DUS,
MRA and CTA to the reference standard of digital subtraction angiography/arteriography (DSA) in
people with symptomatic PAD. Studies were included if they provided sufficient data to calculate a
2x2 table.
Final version
75
PAD
8.2.1 Clinical evidence

Seven new studies38-44 were identified which addressed the question and were added to the HTA
review.
A diagnostic meta-analysis (see Appendix J) was performed in outcomes with more than 4 studies per
comparison. Where there were less than 4 studies for an outcome, the data was presented as a
range of values or for single studies as the results with a 95% confidence interval. A modified GRADE
table has been used to present the data from the diagnostic studies (see Table 21, Table 22, Table 23,
Table 24). See also the full study evidence tables in Appendix H and the diagnostic forest plots and
ROC curves in Appendix J.
Final version
76
PAD
Table 21: Clinical evidence profile: Two dimensional PC magnetic resonance angiography (2D PC MRA) and 2D TOF magnetic resonance angiography (2D
TOF MRA) compared to digital subtraction angiography/arteriography (DSA)
Study characteristics Quality Assessment Summary of findings
No. of Design No. of Risk of bias Inconsistency Indirectness Imprecision Other Sensitivity Specificity
studies Segments considerations (%) (%)
Quality
2D PC MRA
Whole leg, 50-100% stenosis
137 Diagnostic 253 No serious No serious No serious Serious(a) None 97.6 (95% 73.7 (95% MODERATE
cohort, data risk of bias inconsistency indirectness CI 95.1, CI 51.2,
taken from 99.1) 88.2)
HTA
2D TOF MRA
Whole leg, 50-100% stenosis
537 Diagnostic 2668 No serious No serious No serious Serious(b) None 88.07 85.38 MODERATE
cohort, data risk of bias inconsistency indirectness
taken from
HTA
Whole leg, ≥70% stenosis
137 Diagnostic 206 No serious No serious No serious Serious(a) None 89.8 (95% 96.6 (95% MODERATE
cohort, data risk of bias inconsistency indirectness CI 79.5, CI 92.3,
HTA
Whole leg, occlusion
437 Diagnostic 2290 No serious No serious No serious Serious(c) None 76.9 to 100 85.1 to 98.3 MODERATE
taken from
HTA
Above knee
Final version
77
PAD
337 Diagnostic 800 No serious No serious No serious Serious(c) None 71.4 to 84.4 to 100 MODERATE
cohort, data risk of bias inconsistency indirectness 97.3
taken from
HTA
Below knee
337 Diagnostic 1823 No serious No serious No serious No serious None 92.53 94.73 HIGH
cohort, data risk of bias inconsistency indirectness imprecision
taken from
HTA
Foot
137 Diagnostic 33 No serious No serious No serious Very None 86.4 (95% 27.3 (95% CI LOW
cohort, data risk of bias inconsistency indirectness serious(d) CI 66.7, 9.7, 56.6)
taken from 95.3)
HTA
(a) Wide confidence intervals around specificity.
(b) Wide confidence around pooled effect (see appendix J).
(c) Range of values, no estimate of confidence in effect.
(d) Wide confidence intervals around sensitivity and specificity.
Table 22: Clinical evidence profile: Contrast-enhanced magnetic resonance angiography (CE MRA) compared to digital subtraction
angiography/arteriography (DSA)
studies segments considerations (%) (%) Quality
CE MRA
1037,40,42,43 Diagnostic 7710 No serious No serious No serious No serious None 94.96 96.37 HIGH
taken from
HTA
Final version
78
PAD
437 Diagnostic 2773 No serious No serious No serious Serious(a) None 90.9 to 100 96.2 to MODERATE
taken from
HTA
837,39,40 Diagnostic 6403 No serious No serious No serious No serious None 91.83 98.71 HIGH
taken from
HTA
Above knee, ≥50% stenosis
taken from
HTA
137 Diagnostic 576 No serious No serious No serious No serious None 90.5 (95% CI 97.3 (95% HIGH
cohort, data risk of bias inconsistency indirectness imprecision 83, 94.9) CI 95.4,
taken from 98.4)
HTA
Above knee, occlusion
cohort, data risk of bias inconsistency indirectness 100
taken from
HTA
Below knee, ≥50% stenosis
337 Diagnostic 721 No serious No serious No serious Serious(a) None 71.1 to 96.5 87 to 95.8 MODERATE
taken from
HTA
137 Diagnostic 298 No serious No serious No serious No serious None 91.3 (95% CI 93.7 (95% HIGH
cohort, data risk of bias inconsistency indirectness imprecision 83.8, 95.5) CI 89.5,
taken from 96.3)
HTA
Final version
79
PAD
Below knee, occlusion

237 Diagnostic 627 No serious No serious No serious Serious(a) None 86.2 to 95.2 92.9 to MODERATE
taken from
HTA
Foot
137 Diagnostic 286 No serious No serious No serious Serious (a) None 78.7 to 79.4 70.6 to MODERATE
taken from
HTA
(a) Range of values, no estimate of confidence in effect.
Table 23: Clinical evidence profile: Computed tomography angiography (CTA) compared to digital subtraction angiography/arteriography (DSA)
studies segments considerations (%) (%)
Quality
CTA
taken from
HTA
437,44 Diagnostic 9599 No serious No serious No serious Serious(a) None 87.4 to 99 97 to 98.5 MODERATE
taken from
HTA
taken from
HTA
Final version
80
PAD

337 Diagnostic 628 No serious No serious No serious Serious(a) None 94.2 to 91.2 to MODERATE
cohort, data risk of bias inconsistency indirectness 96.9 98.1
taken from
HTA
337,38 Diagnostic 1150 No serious No serious No serious Serious(a) None 99 to 100 99.4 to MODERATE
taken from
HTA
taken from
HTA
137 Diagnostic 390 No serious No serious No serious No serious None 89.5 (95% 73.9 (95% HIGH
cohort, data risk of bias inconsistency indirectness imprecision CI 84.1, CI 67.6,
HTA
taken from
HTA
(b) No confidence intervals.
Final version
81
PAD
Table 24: Clinical evidence profile: Duplex ultrasound scanning (DUS) compared to digital subtraction angiography/arteriography (DSA)
studies segments considerations (%) (%)
Quality
DUS
1137,40,42 Diagnostic 8335 No serious No serious No serious No serious None 89.7 95.64 HIGH
cohort, risk of bias inconsistency indirectness imprecision
data taken
from HTA
937,40 Diagnostic 7396 No serious No serious No serious No serious None 89.12 97.8 HIGH
taken from
HTA
Whole leg, other stenosis thresholds
437,41 Diagnostic 3021 No serious No serious No serious Serious(a) None 60.9 to 88 79 to MODERATE
taken from
HTA
taken from
HTA
taken from
HTA
Final version
82
PAD
taken from
HTA
Above knee, other stenosis thresholds
taken from
HTA
taken from
HTA
Below knee, occlusion
taken from
HTA
Below knee, other stenosis thresholds
taken from
HTA
Foot
137 Diagnostic 140 No serious No serious No serious Very None 64.3 (95% 80.4 (95% LOW
cohort, data risk of bias inconsistency indirectness serious(c) CI 53.6, CI 68.2,
HTA
(b) Wide confidence around pooled effect (see Appendix J).
(c) Wide confidence intervals around sensitivity and specificity.
Final version
83
PAD
Four studies were identified that evaluated comparators which were relevant to the review question:
a decision analytic model developed as part of the HTA by Collins et al 200745 and three trial-based
economic evaluations which have been published since the HTA cut-off date (May 2005).46-48 No
single study included all comparators. The results of both the HTA and studies included as part of the
guideline update search are summarised in the economic evidence profiles below (see Table 26).
The HTA by Collins 2007 concluded that DUS is the most cost effective choice for both whole leg and
below the knee imaging. The analysis shows that 2D TOF MRA is the most cost effective alternative
when imaging is confined to areas above the knee. However, this model is subject to several
potentially serious limitations:
 Treatment pathway
o The model assumed that people diagnosed with >50% stenosis could be treated with
angioplasty, bypass or amputation. The GDG did not consider bypass and amputation to be
appropriate first-line options for treatment of these patients. If only exercise and angioplasty
were considered, the consequences of an inaccurate test result would be different than those
predicted by the model.
o People with <50% stenosis were treated with medical management. The GDG considered
exercise therapy (supervised and unsupervised) to be the most appropriate treatment options
for this patient group. If these options had been included as an option in the model, the
consequences of an inaccurate test result would be different to those predicted by the model.
o The probability of undergoing each interventional treatment was estimated according to the
results of each included imaging study. This introduces a confounding factor into the model as
it is difficult to determine how much of the difference in cost and quality of life between the
different imaging procedures is due to the accuracy of the diagnostic test and how much is due
to the treatment pathway (which varies between tests).
o Similarly, the probability of reintervention and experiencing a change in treatment plan
differed for each imaging strategy. The effect of this is to further skew the results of the model
as the differences in initial treatment plans (according to diagnostic test) are amplified.
 Cost
o The costs included in this model were derived from estimates based largely on expert opinion
are very different to those derived from current NHS reference costs (Table 25). As a result,
the consequences of an inaccurate test result predicted by the HTA model may be different to
those which could be expected given current costs.
The three RCT-based studies included as part of the economic update search are pair-wise
comparisons of different imaging procedures.46-48 Each study reported costs that were adjusted to
take into account predictive baseline characteristics using multivariable linear and logistic regression.
Two47,48 also adjusted QALYs, allowing for an adjusted ICER to be calculated. Based on the results of
the adjusted data, CTA is less costly and more effective than both CE MRA and DSA. However, if the
unadjusted figures are used to calculate ICERs, CE MRA is more cost-effective than both CTA and
DUS, and DSA is more cost effective than CTA.
It is difficult to draw comparisons between the studies included in this update and the results of the
HTA. The RCT-based studies did not report sensitivity and specificity of each intervention, making it
impossible to compare the results of these studies to that of the HTA or current clinical review. In
addition, two of these studies included investment costs associated with imaging equipment in the
total cost of each strategy; this was not a cost considered by the HTA. Moreover, none of the studies
included as part of the update search specified the location of the imaging procedure (whole leg,
above knee or below knee) or reported results by subgroups.
Final version
84
PAD
This area was prioritised for new cost-effectiveness analysis subject to timing and data. Given the
time required to complete the network meta-analysis and the fact that the conclusions of the original
HTA were essentially unchanged, no new analysis was preformed for this question.
Current unit costs
In order to allow comparison to the costs used in the included studies, relevant UK unit costs are
provided below (Table 25).
Table 25: Cost of imaging procedures in the NHS (2010)

Imaging procedure National average unit cost Interquartile range
(a)
Contrast-enhanced MRA £229 £174 to £267(b)
Duplex ultrasound £90(c) £61 to £176(d)
Computed tomography angiography £146(e) £112 to £162(f)
Catheter angiography £679(g) £480 to £778(h)
Sources:
(a) RA03Z MRI of one area, pre and post contrast49
(b) RA03Z MRI scan of one area, pre and post contrast49
(c) RA26Z Ultrasound mobile scan/ intra-operative procedures 20 to 40 minutes49
(d) RA25Z Ultrasound mobile scan/intra-operative procedures less than 20 minutes & RA 27Z Ultrasound mobile scan /Intra-
operative procedures more than 40 minutes49
(e) RA12Z CT scan, two areas with contrast49
(f) RA 10Z CT scan, one area, pre and post contrast & RA 13Z CT scan, three areas with contrast49
(g) Day case HRG QZ 16C Diagnostic radiology and other transluminal procedures without CC49
(h) Day case HRG QZ16 A & QZ 16B49
Final version
85
PAD
Table 26: Economic evidence profile: Imaging for revascularisation in peripheral arterial disease
Incremental Incremental Cost
Study Limitations Applicability Other comments costs effects effectiveness Uncertainty
2D TOF MRA vs. CE MRA vs. DUS vs. CA
Collins 200745 Potentially Partially  Decision analytic model Whole leg
serious applicable(b)  Population: People with IC and CLI DUS was the DUS was DUS was the There was a 95%
limitations(a)
 Time Horizon: 1 year least costly of equally as dominant probability that DUS
 Costs: Diagnostic test costs (and all evaluated effective as CA strategy was the most cost
secondary CA for inconclusive test strategies and CE MRA, effective strategy at a
results), cost of treatment resulting in a threshold of £20, 000.
(angioplasty, bypass, etc) and follow- gain of 0.03
up costs. The cost of complications QALYs
associated with CA was also compared to
included. Adverse events related to 2D TOF MRA
other imaging procedures were not Below the knee
considered relevant for inclusion. 2D TOF MRA is 2D TOF MRA 2D TOF MRA There was a 70%
 Perspective: UK NHS £362 more was the most costs £43, 272 probability that DUS is
costly than DUS effective per QALY and the most cost-effective
strategy, is therefore strategy at a threshold
resulting in a not of £20, 000.
gain of 0.008 considered
QALYs cost-effective.
compared to DUS is the
DUS most cost
effective
strategy.
Above the knee
CE MRA is £155 CE MRA was CE MRA costs There was a 75%
more costly the most £143, 389 per probability that 2D
than 2D MRA effective QALY and is MRA was the most cost
strategy, therefore not effective strategy at a
resulting in a considered threshold of £20, 000.
0.001 QALY cost effective.
gain 2D MRA is the
Final version
86
PAD
compared to most cost

2D MRA effective
strategy.
CE MRA vs. DUS
de Vries 200750 Potentially Partially  RCT-based cost-utility analysis CE MRA was CE MRA CE MRA costs If the investment costs
serious applicable(d)  Population: People with IC and CLI £275 more resulted in a £13, 750 per for the MR imager
limitations(c) costly than 0.02 QALY QALY gain(e) were reduced by 50%,
 Time horizon: 6 months
DUS(e) gain the incremental
 Costs: Initial and additional imaging, compared to difference in total
procedural and outpatient costs over DUS(e) costs was reduced to
6 months. £48, resulting in an
 Perspective: Netherlands, hospital ICER of £2, 400.
 Other: Approximately equal
proportions of patients underwent
exercise therapy, angioplasty and
bypass in each arm.
DSA vs. CTA
Kock 200747 Potentially Partially  RCT-based cost-utility analysis CTA is £547 less CTA results in CTA is the If the reported
serious applicable(g)  Population: People with IC and CLI costly than a 0.07 QALY dominant unadjusted values are
limitations(f) DSA(h) gain(h) treatment used to calculate the
 Time horizon: 6 months
option(h) incremental costs and
 Costs: Initial and additional imaging, QALYs, CTA is £1, 788
vascular interventions out to 6 more costly than DSA
months from initial imaging. and results in a 0.04
 Outcomes: Costs over 4 months, QALY grain. The ICER is
change in quality of life at 6 months, therefore £44, 450.
additional imaging, patient comfort. And CTA is not
 Perspective: Netherlands, hospital considered cost-
 Other: Approximately equal effective.
proportions of patients underwent
exercise therapy, angioplasty and
bypass in each group.
CE MRA vs. CTA
Ouwendijk 2005 Potentially Partially  Cost-utility analysis based on RCT CTA is £2,425 CTA results in CTA is the If the reported
48
serious applicable(j) less costly than a 0.02 QALY domain unadjusted values are
Final version
87
PAD
limitation(i)  Population: people with PAD CE MRA(k) gain(k) treatment used to calculate
(symptom group unclear) option(k) incremental costs and
 Time horizon: 6 months QALYs, CE MRA is £2,
710 more costly than
 Costs: Direct costs over 6 months;
CTA and results in a
details not given.
gain of 0.03 QALYs. The
 Outcomes: Costs at 6 months, ICER is therefore £9,
change in quality of life at 6 months, 033 and CE MRA would
therapeutic confidence. be considered cost
 Perspective: Netherlands, hospital effective.
 Other: Approximately one third of Altering the initial
patients underwent angioplasty, investment costs of
bypass and conservative treatment imaging equipment
in each group. does not change the
conclusion of the
analysis.
Costs were converted to GBP using OECD Purchasing Power Parity Index(OECD), 2010 16360 /id} and inflated to 2008/09 GBP using PRSSU Pay and Prices Index51.
(a) Probability of intervention differs according to imaging modality as reported by the studies included in the clinical review; no lifetime analysis of cost and QALY gain (1 year time horizon);
intervention outcomes differ from those identified in the literature included in the current clinical review; source of health state descriptions is unclear; resource use and unit cost estimates
for downstream interventions differ from those included as part of the economic review; inadequate sensitivity analysis.
(b) Analysis did not include all relevant comparators; downstream consequences differ from those considered appropriate by the GDG.
(c) Cut-off criteria and factors used to determine intervention treatment pathway not reported; sensitivity and specificity not reported (difficult to compare to results of current clinical evidence
review); included investment costs for imaging equipment.
(d) Dutch healthcare setting; did not include all possible comparators.
(e) Unadjusted for predictive variables at baseline (disease severity (IC vs. CLI), renal disease, cardiac diseases, cerebrovascular disease, and diabetes mellitus, hospital setting, and study
group).
(f) Cut-off criteria and factors used to determine intervention treatment pathway not reported; sensitivity and specificity not reported (difficult to compare to results of current clinical evidence
review).
(g) Dutch healthcare setting; did not include all possible comparators; downstream consequences differ from those considered appropriate by GDG.
(h) Adjusted for predictive variables at baseline (disease severity (IC vs. CLI), renal insufficiency, cerebrovascular disease, and diabetes mellitus).
(i) Cut-off criteria and factors used to determine intervention treatment pathway not reported; sensitivity and specificity not reported (difficult to compare to results of current clinical evidence
review); included investment costs for imaging equipment.
(j) Dutch healthcare setting; did not include all possible comparators; downstream consequences differ from those considered appropriate by GDG.
(k) Adjusted for predictive variables at baseline (disease severity (IC vs. CLI), renal insufficiency, cardiac diseases, cerebrovascular disease, and diabetes mellitus).
Final version
88
PAD
8.2.2.1 Clinical
Two-dimensional phase-contrast magnetic resonance angiography (2D PC MRA)
In comparison to the reference standard of DSA, the review showed that 2D PC MRA:
 Had a sensitivity of 97.6% (95% CI 95.1 to 99.1) and specificity of 73.7% (95% CI 51.2 to 88.2) for
assessment of whole leg, 50-100% stenosis [1 systematic review based on 1 study, 253 segments,
Two-dimensional time-of-flight magnetic resonance angiography (2D TOF MRA)
In comparison to the reference standard of DSA, the review showed that 2D TOF MRA:
 Had a sensitivity of 88.07% and specificity of 85.38% assessment of whole leg, 50-100% stenosis [1
systematic review based on 5 studies, 2668 segments, moderate quality evidence]37
assessment of whole leg, ≥70% stenosis [1 systematic review based on 1 study, 206 segments,
 Had a sensitivity range of 76.9 to 100% and specificity range of 85.1 to 98.3% for assessment of
whole leg, occlusion [1 systematic review based on 4 studies, 2290 segments, moderate quality
evidence]37
 Had a sensitivity range of 71.4 to 97.3% and specificity range of 84.4 to 100% for assessment of
above knee [1 systematic review based on 3 studies, 800 segments, moderate quality evidence]37
 Had a sensitivity of 92.53% and specificity of 94.73% for assessment of below knee [1 systematic
review based on 3 studies, 1823 segments, high quality evidence]37
assessment of the foot [1 systematic review based on 1 study, 33 segments, low quality
evidence]37
Contrast-enhanced magnetic resonance angiography (CE-MRA)
In comparison to the reference standard of DSA, the review showed CE MRA:

 Had a sensitivity of 94.96% and specificity of 96.37% assessment of whole leg, ≥50% stenosis [3
studies and 1 systematic review based on 7 studies, 7710 segments, high quality evidence]37,40,42,43
whole leg, ≥70% stenosis [1 systematic review based on 4 studies, 2773 segments, moderate
quality evidence]37
 Had a sensitivity of 91.83% and specificity of 98.71% for assessment of whole leg, occlusion [2
studies and 1 systematic review based on 6 studies, 6403 segments, high quality evidence]37,39,40
 Had a sensitivity range of 81.4 to 100% and specificity range of 91.9 to 95.9 for assessment of
above knee, ≥50% stenosis [1 systematic review based on 4 studies, 742 segments, moderate
quality evidence]37
 Had a sensitivity of 90.5% (95% CI 83 to 94.9) and specificity of 97.3% (95% CI 95.4 to 98.4) for
assessment of above knee, ≥70% stenosis [1 systematic review based on 1 study, 576 segments,
high quality evidence]37
 Had a sensitivity of range 86.7 to 100% and specificity of range 99.5 to 100% for assessment of
above knee, occlusion [1 systematic review based on 4 studies, 742 segments, moderate quality
evidence]37
Final version
89
PAD
 Had a sensitivity range of 71.1 to 96.5% and specificity range of 87 to 95.8% for assessment of
below knee, ≥50% stenosis [1 systematic review based on 3 studies, 721 segments, moderate
quality evidence]37
assessment of below knee, ≥70% stenosis [1 systematic review based on 1 study, 298 segments,
 Had a sensitivity range of 86.2 to 95.2% and specificity range of 92.9 to 96.8% for assessment of
below knee, occlusion [1 systematic review based on 2 studies, 627 segments, moderate quality
evidence]37
the foot [1 systematic review based on 1 study, 286 segments, moderate quality evidence]37
Computed tomography angiography (CTA)
In comparison to the reference standard DSA, the review showed CTA:

 Had a sensitivity of 93.5% and specificity of 91.13% for assessment of whole leg, ≥50% stenosis [1
systematic review based on 6 studies, 4270 segments, high quality evidence]37
 Had a sensitivity range of 97 to 98.5% and specificity range of 97 to 98.5% for assessment of
whole leg, ≥70% stenosis [1 study and 1 systematic review based on 3 studies, 9599 segments,
moderate quality evidence]37,44
quality evidence]37
 Had a sensitivity range of 99 to 100% and specificity range of 99.4 to 100% for assessment of
above knee, ≥70% stenosis [1 study and 1 systematic review based on 2 studies study, 1150
segments, moderate quality evidence]37,38
 Had a sensitivity range of 95.1 to 96% and specificity range of 99.2 to 100% for assessment of
above knee, occlusion [1 systematic review based on 2 studies, 338 segments, moderate quality
evidence]37
assessment of below knee, ≥50% stenosis [1 systematic review based on 1 study, 390 segments,
 Had a sensitivity of 98.2% and specificity of 99.7% for assessment of below knee, ≥70% stenosis [1
study, 539 segments, moderate quality evidence]38
Duplex ultrasound scanning (DUS)
In comparison to the reference standard DSA, the review showed that DUS:
 Had a sensitivity of 89.7% and specificity of 95.64% for assessment of whole leg, ≥50% stenosis [2
studies and 1 systematic review based on 9 studies, 8335 segments, high quality evidence]37,40,42
study and 1 systematic review based on 8 studies, 7396 segments, high quality evidence]37,40
 Had a sensitivity range of 60.9 to 88% and specificity range of 79 to 99.7% for assessment of
whole leg, other stenosis thresholds [1 study and 1 systematic review based on 3 studies, 3021
segments, moderate quality evidence]37,41
 Had a sensitivity of 91.54% and specificity of 92.58% for assessment of above knee, ≥50% stenosis
[1 systematic review based on 9 studies, 1970 segments, high quality evidence]37
Final version
90
PAD
 Had a sensitivity range of 65.4 to 100% and specificity range of 95.2 to 98% for assessment of
quality evidence]37
 Had a sensitivity of 92.58% and specificity of 97.84% for assessment of above knee, occlusion [1
above knee, other stenosis thresholds [1 systematic review based on 3 studies, 682 segments,
below knee, ≥50% stenosis [1 systematic review based on 4 studies, 767 segments, moderate
quality evidence]37
 Had a sensitivity of 79.52% and specificity of 90.57% for assessment of below knee, occlusion [1
systematic review based on 6 studies, 2562 segments, moderate quality evidence]37
below knee, other stenosis thresholds [1 systematic review based on 2 studies, 1772 segments,
assessment of the foot [1 systematic review based on 1 study, 140 segments, low quality
evidence]37
8.2.2.2 Economic
 No cost-effectiveness studies were identified that included all relevant comparators.
 For whole leg and below the knee imaging, one economic decision model concluded that DUS was
more cost effective than 2D TOF MRA, CE MRA and CA. For above the knee imaging, the same
economic decision model found that 2D TOF MRA was more cost effective than DUS, CE MRA and
CA. [Partially applicable with potentially serious limitations]45
 It was difficult to draw conclusions from studies of pair-wise comparisons
o One RCT determined that CE MRA was more cost effective than DUS [Partially applicable with
potentially serious limitations]46
o One RCT determined that CTA was more cost effective than DSA [Partially applicable with
o One RCT determined that CTA was more cost effective than DSA [Partially applicable with
7. Offer duplex ultrasound as first-line imaging to all people with

peripheral arterial disease for whom revascularisation is being
considered.
8. Offer contrast-enhanced magnetic resonance angiography to

people with peripheral arterial disease who need further
imaging (after duplex ultrasound) before considering
revascularisation.
9. Offer computed tomography angiography to people with

peripheral arterial disease who need further imaging (after
duplex ultrasound) if contrast-enhanced magnetic resonance
Recommendations angiography is contraindicated or not tolerated.
Final version
91
PAD
Relative value of different It is difficult to make a definitive comparison between the accuracy of these
outcomes imaging techniques because none of the studies included compared all the
techniques to each other. In addition, the studies combined results from
different disease sites. Furthermore, the reference standard was taken as
DSA, but this can occasionally miss vessels which are picked up by other
techniques. The HTA (Collins 2007)37 concluded that CE MRA was superior in
diagnostic accuracy than DUS and CTA and would be a suitable alternative to
the reference standard of DSA, and having noted the difficulties in
interpreting the data the GDG found that the newer evidence did not
substantially alter this conclusion, although they noted that differences in
sensitivity and specificity were not large.
Trade off between clinical The GDG noted that all imaging techniques are relatively safe. The avoidance
benefit and harms of intravascular contrast media (not required for DUS) and of exposure to
ionising radiation (not required for DUS or for CE-MRA) are important
considerations. Allergic reactions to contrast medium are rare, but the
potential nephrotoxic effects of iodinated contrast media are of concern.
There are non-contrast techniques other than DUS e.g. TOF MRA but these
are agreed to be less accurate (see below). In addition, the avoidance of
unnecessary reduplication of imaging is important (time and cost).
Whilst DSA is considered the gold standard, it is much less commonly used in
routine practice. It involves both administration of a contrast medium and
ionising radiation. In addition, discomfort is experienced by some patients.
DUS was not perceived as having any major risks. DUS may be technically
more difficult in large or obese patients and/or in the presence of
calcification (particularly in diabetic patients) or where there are ulcers and
bandaging near the sites of the vessels. Stenosis and occlusion are important
with regard to sensitivity of DUS for below knee lesions.
CE MRA offers better spatial resolution, is faster to perform and is less

dependent on blood flow than DUS. However, it is contraindicated in people
with intra-cranial clips and pacemakers. In addition, some people are unable
to tolerate MRA due to claustrophobia.
CTA is not recommended for people with an eGFR of <30ml/min. The latter is
not an absolute contra-indication but would also be considered a relative
contra-indication to CE MRA. If the creatinine is <200 CTA could be
performed with safeguards.
Economic considerations The GDG discussed the methods, results and limitations of each study
included in the economic evidence review and agreed that it was very
difficult to draw a robust conclusion from the current cost-effectiveness
evidence base.
The group discussed the current costs associated with each imaging modality
from an NHS perspective and the costs and consequences of the pathways
that they expect patients to follow based on the results of imaging. They
agreed that for patients in whom revascularisation may be beneficial, DUS
represents the least costly and least invasive method of determining the
location and extent of the lesion, and may well provide sufficient
information. If the results of DUS are not suitable for planning an
intervention, the GDG agreed that CE MRA and CTA represent useful
modalities for gathering more detailed information.
Final version
92
PAD
Quality of evidence The quality of the evidence was rated from high quality to low by the GRADE
criteria.
There was concern about using the degree of stenosis as part of this
assessment. In practice, treatment is based on severity of symptoms.
Although uncommon, sometimes people with <50% stenosis are treated and
many people with >50% stenosis will remain asymptomatic. Furthermore,
the degree of stenosis is unknown until the imaging has taken place, and it
therefore cannot be used as a means of judging which test to do in advance.
It was noted that the accuracy of the techniques could be affected by the use
of different imaging protocols.
The GDG expressed caution when interpreting the results of the HTA.
Although the HTA is relatively recent, clinical practice has changed
significantly in that short time.
The GDG did not consider 2D TOF MRA within the recommendations as it
was not thought to be a relevant comparator. 2D TOF MRA is much more
prone to artefacts such as movement and is susceptible to non-uniformity of
blood flow. Imaging times are longer than for CE-MRA.
The recommendations were made based on the clinical and cost

effectiveness evidence, and expert opinion.
Other considerations Based on the clinical and cost effectiveness data, and expert opinion, the
GDG agreed that DUS should be used as a first option for people being
considered for revascularisation. However, they noted that it might not
provide sufficient information, and that ultrasound is easier to perform in
some people than others. They therefore felt that other imaging modalities
should be available and that the recommendation should reflect this.
Most units will have access to MRA in the UK. However the amount of time
available on the scanner will vary from site to site and generally it is used
much less frequently than DUS; CTA might also utilised more widely. The
local expertise may therefore be limited in some units.

The GDG identified the recommendation on offering CE-MRA as a key priority
for implementation. There is considerable variability in the investigations
used and whilst MRA is considered preferable to CTA, the widespread
adoption may have significant implications for training and the availability of
expertise and equipment.
Final version
93
PAD
9 Management of intermittent claudication

9.1 Introduction
Intermittent claudication (IC) is a tight, cramp like pain in the muscles of the calf, thigh or buttock
which comes on only after walking and is relieved by resting. The pain is caused by diminished
circulation.
The aim of treatment for intermittent claudication is two-fold. First, as people with PAD are at high
risk of other cardiovascular events, the aim is to reduce this risk. The first basic intervention for PAD,
is to offer information including general information about cardiovascular risk and potential
interventions to reduce this (cardiovascular exercise, quitting smoking, healthy eating, medicines –
see chapter 6 for the recommendations relating to this) as well as specific information about risks to
the limb. This could be termed best medical treatment for PAD.
The second aim of treatment is to improve walking distance. The decision to directly attempt to
improve walking distance should be decided by the patient, balancing the impact their symptoms
have on their day to day life, and the chance of success versus the risks of treatment. People with
claudication are a heterogeneous group. Many will only be mildly troubled by their symptoms or
have other significant co-morbidity that reduces their mobility. Others however may be severely
restricted by their claudication which can significantly alter their lifestyle. It is the role of the clinician
to help the patient decide on the best therapeutic option for them based on the impact of their
symptoms on their quality of life.
The purpose of this chapter is to set out the GDG’s consideration of the evidence comparing the
various management options for IC. It is assumed that the diagnosis has been correctly established
(see chapter 7).
9.1.1 Role of exercise

Physical exercise has been shown to be of benefit to people with established cardiovascular disease
(NICE guideline on lipid modification 52) and increased exercise in people with IC can result in
improvements in walking distance. People with IC should be encouraged to walk to near maximal
pain. A variety of methods have been employed to support people with IC in exercising, including
treadmill walking, exercise classes and gym membership (supervised exercise).
9.1.2 Role of naftidrofuryl oxalate

There are a number of vasoactive drugs currently licensed for treating the symptoms of IC. There is
some evidence that vasoactive drugs can increase walking distance compared to placebo.53 The NICE
technology appraisal (TA 223) on “Cilostazol, naftidrofuryl oxalate, pentoxifylline and inositol
nicotinate for the treatment of intermittent claudication in people with peripheral arterial disease”53
recommended naftidrofuryl oxalate as the preferred treatment.
From a clinical viewpoint, although there is a small benefit identified in drug treatment, the question
remains as to whether drug treatment is preferred to other treatments such as supervised exercise
therapy, angioplasty or stents, when patients are suitable for more than one of these options.
9.1.3 Endovascular techniques

A proportion of people with IC will achieve reasonable symptom control after cardiovascular risk
prevention measures have been taken and a regular exercise regimen has been established. IC can
also be treated using endovascular procedures (angioplasty +/- stent placement) or bypass surgery,
Final version
94
PAD
both of which constitute a more direct means of addressing the problem since they are directed at
the arterial lesions causing claudication.
9.1.3.1 Angioplasty
In recent years there has been rapid development of endovascular techniques for the management
of PAD. These are minimally invasive procedures in which catheters and guide wires are introduced
through small punctures in the artery, carried out under local anaesthetic. These techniques are used
to introduce devices that can be used to unblock or dilate areas where there are obstructions to
blood flow. The most common technique is the use of an inflatable balloon to dilate an area of artery
(angioplasty). This has some limitations in that it may not be possible to open up the artery
sufficiently or the procedure may lead to complications, such as the development of a flap of the
lining of the artery (dissection) or dislodging material that passes further down the artery and causes
another blockage (embolisation).
9.1.3.2 Stenting
A treatment that can be used to improve the results of angioplasty is the insertion of a stent. Stents
are small spring like structures that are usually made of metal (known as bare metal stents) and can
be placed within the artery in order to try and hold it open. The potential benefits of the use of
stents are that they may improve the diameter of the treated artery, where angioplasty alone is
inadequate. They may also help to prevent or treat complications by pinning down a flap of lining
that has developed or preventing embolisation and may alter the risks of long term re-stenosis or re-
occlusion of the treated section of artery.
There are two different approaches to the use of stents. One is to use them as an adjunct to
angioplasty only in those cases where the result of the initial angioplasty is thought to be sub-
optimal. The alternative is to insert a stent as part of an angioplasty procedure, which is termed
primary stenting.
Over recent years new drug eluting stents have been developed which have a coating of material
containing drugs that are gradually released over a long period of time and are intended to reduce
the risk of narrowing of the artery after treatment.
9.1.4 Bypass surgery

The most invasive treatments for people with PAD, who have not been suitable for or responded to
other treatments, are open surgical procedures to improve the circulation to the limb.
The most common operations are bypass grafts in which a new blood vessel is created by joining a
conduit to above and below the blocked artery. In treating blocked arteries in the leg below the groin
there are a number of options for bypass material. The patient’s own vein (autologous) can be used
in the bypass procedure. This usually involves taking the long saphenous vein from the same leg as
the blockage. Autologous grafting has the advantage of being less likely to become infected or cause
a serious reaction. However there are not always suitable veins available and because of the valves in
the vein it either needs to be completely removed and reversed, resulting in the need for long
incision down the leg, or needs to have a procedure to destroy the valves, which may damage the
interior of the vein leading to a risk of complications or subsequent narrowing. The other option is to
use an artificial artery made out of a prosthetic material, often PTFE or Dacron.
Final version
95
PAD
Chapter overview
The GDG wished to know whether the results of angioplasty or bypass surgery were superior to
exercise or, since the mechanism of benefit is different to exercise, whether they add anything to the
benefit obtained from exercise.
In formulating their review of the evidence, the GDG considered types of treatment that could be
used in addition to best medical treatment (i.e. management of secondary cardiovascular risk
factors) for PAD. Literature searches were performed to answer a series of questions in which
treatment options were compared head-to-head, the options being some form of exercise, surgery,
and endovascular therapy. The GDG felt that the first question they needed to answer was how best
to help people with PAD achieve an optimal level of exercise - exercise in some form was accepted as
beneficial on a priori grounds. For people with PAD the possibilities for exercise therapy range from
simple advice on exercise, through individualised plans, to participation in formal supervised exercise
sessions. Additional treatment might then consist of nothing more (i.e. exercise alone has
successfully controlled symptoms), an endovascular procedure or surgery; the assessment of these
measures needed to allow that they might be added to either unsupervised or supervised exercise
(depending on the outcome of that first head-to-head comparison). In order to assess all the
possibilities using both clinical and health-economic data, papers covering all potential treatment
comparisons under the umbrella heading of exercise versus endovascular therapy versus surgery
were sought and assessed, and an original health economic model was developed covering all 3
forms of intervention.
The situation is further complicated by questions within each separate general treatment modality.
Within endovascular therapy, the GDG wished to know whether angioplasty alone is sufficient or
whether stents should also be placed; and if stenting is employed, whether it should be with a bare
metal stent or a drug-eluting stent. For the surgical question, it was felt appropriate to compare
autologous vein grafts with prosthetic grafts. These different possibilities were also accounted for in
the large health-economic model, but to minimise complexity they will be presented separately from
the over-arching exercise vs endovascular therapy vs surgery questions.
Finally on intermittent claudication, the GDG needed to incorporate the NICE Technology Appraisal
(TA) of vasoactive drugs. It was felt that these drugs are generally used in current practice when
other treatment is not possible or when turned down by the person with PAD, and that they do not
confer any prognostic advantage nor offer a likely cure for symptoms. Moreover the TA had already
considered their cost-effectiveness. They are therefore slightly separate from the other forms of
treatment covered by this guideline, and are not included in the direct comparison of the other
forms of treatment for intermittent claudication.
9.2 Supervised exercise compared to unsupervised exercise

What is the clinical and cost effectiveness of supervised exercise therapy compared to unsupervised
exercise therapy for the treatment of PAD in adults with intermittent claudication?
For the purpose of this review, unsupervised exercise was defined as advice to exercise for
approximately 30 minutes three to five times per week, walking until the onset of symptoms, then
resting to recover. Supervised exercise was defined as a community-based exercise including hospital
or gym based programme supervised by healthcare professionals (typically two physiotherapists with
approximately ten patients per group). A programme typically consists of approximately two hours of
classes per week for a period of up to three months during which patients exercise until the onset of
symptoms, and then rest and repeat.
Final version
96
PAD
Two Cochrane reviews were identified (Bendermacher 200654 and Watson 200855) comparing
unsupervised exercise to supervised exercise for the treatment of intermittent claudication. These
studies were not included or updated in the current review as they did not meet the review question
protocol defined by the GDG, which had a wider definition of the exercise interventions compared to
the Cochrane reviews. However they were used as a source to ensure that studies identified in the
Cochrane review which matched the current review protocol had been considered for inclusion.
Twelve RCTs comparing supervised and unsupervised exercise56-67 were included in the clinical
review. Table 27 describes the duration and content of the supervised exercise programmes in each
included study. These are summarised in the clinical evidence profiles below (Table 28 and Table 29).
See also the full clinical evidence tables in Appendix H and forest plots in Appendix J. The reason for
withdrawal from the exercise interventions are summarised in Table 30.
One study reported quality of life as measured by the EQ-5D58; five papers (representing an
additional four studies) included the SF-36 questionnaire56-60; and one study included the SF-20 as a
measure of health related quality of life.61 Methods for mapping SF-36 health state descriptions to
health state valuations based on the EQ-5D have been developed and reported by Ara and Brazier
2008.68 Where the results of each dimension score were not reported in full, authors reporting the
use of this measure were contacted for additional data and all replied. Cheetham 200456 and Nicolai
201058 provided average values for each of the 8 dimensions; Pinto 199759 replied that although
these data were collected they were not available. All available values were mapped to preference
based values using Equation 1 as reported by Ara and Brazier68 and probabilistic simulation methods.
A summary of mapped values is presented in Table 31 and Table 32; additional data are reported in
the economic modelling report in Appendix K.
Final version
97
PAD
Table 27: Study characteristics: Summary of exercise interventions

Study Unsupervised exercise Supervised exercise programme
Method Content Duration Setting Method Content
Cheetham 200456 Advice only  Written and verbal advice to 1 x 45 min/week Hospital gym Weekly  5-10 minutes talk about benefits of
exercise for half an hour at least 3 for 6 months education, exercise
times per week to near maximal circuit  30 minutes exercise. Alternating
pain. training walking for 2 min and exercise
 Additional exercise such as stair stations for 2 minutes: stair
climbing and toe walking also climbing, high-step climbing, tip toe
advised. walking, calf raises, and power
 Progress reviewed every 3 months. walking/jogging.
Gardner 201167 Home based  12 week programme of walking to 3 x per week for 3 Research Treadmill  Intermittent treadmill walking for 3
exercise near maximal claudication pain months centre walking days/week at speed of 2mph.
 3 days per week at self-selected  Walking began at 15 mins for the
pace. first two weeks and increased by 5
minutes biweekly until a total of 40
minutes of walking during final 2
weeks of the programme.
Kakkos 200557 Advice only  Advised to exercise for at least 45 3 x 60 min/week Physiotherapy Treadmill  Each class consisted of a 5 minute
minutes per day, walking to near for 6 months department walking warm up, 50 minutes of exercise
maximal pain. and a 5 minute cool down.
 Patients started walking at 2mph
and 0% until pain became severe,
then rested. Increased speed by
0.5mph or grade by 1%-2% every 10
minutes.
Nicolai 201058 Advice only  Verbal and written advice to 3 x 30 min/week Local Treadmill  Interval training; encouraged to
exercise three times per day. for 12 months physiotherapy walking perform at least three walking
 During each session, near maximal practice sessions per day, walking to sub-
pain level should be reached three maximal pain with short intervals.
times.  Also included walking pattern
improvement and endurance and
strength exercises.
Final version
98
PAD
Pinto 199759 Education,  Attended weekly lecture and 3 x 30 min/week Not specified Weekly  20 minutes stationary or arm
advice, verbally advised to walk for 20-40 for 3 months education, cycling followed by walking. Initially,
exercise minutes at least 3 times per week treadmill speed was set to produce maximum
journal and to near maximal pain. walking, pain at 3-5min, then asked to rest.
weekly in-  Asked to record pause durations in cycling  Exercise log was kept.
person home log.  Patients also attended a weekly
support
 Vascular nurses provided feedback lecture as per the control group.
prior to weekly lecture.
Regensteiner Detailed  Detailed, personalised written 3 x 60 min/week Hospital gym Treadmill  Walked until a moderate level of
199761 advice and advice to walk for between 35 to for 3 months walking pain developed, then asked to rest.
weekly 50 minutes at least three times per Began at 35 minutes per class,
telephone week. increasing by 5 minutes to a total of
support  Advised to walk to near maximal 50 minutes while also gradually
pain. increasing speed and grade.
 Patients contacted weekly by
telephone to provide feedback and
encouragement.
Savage 200160 Advice and  Verbal advice to exercise 3 times 3 x 40 min/week Hospital gym Treadmill  Walked at 2mph at 60% of max
monthly per week, walking to the point of for 3 months walking intensity achieved in initial test.
telephone maximal pain, then resting, Walked to the point of intense pain,
support gradually increasing to 40 minutes then rest. Began at 15minutes per
each time. class, increasing to a total of 40
 Contacted by telephone every minutes.
month to provide feedback and
encouragement.
Stewart 200862 Advice only Not specified 2 x 60 min/week Hospital gym Circuit  Five different exercises per class
for 3 months training with 8 minutes spent at each. Ten
minutes warm up & cool down.
Patients asked to rest when pain
became intolerable.
 Exercises could all be performed at
home; no treadmills used.
 Asked to continue to exercise at
home following programme
completion.
Final version
99
PAD
Tew 200963 Advice only  Verbal advice on benefits of an 2 x 20-40 Exercise Arm-crank  Trained in cycles of 2 minutes of
active lifestyle. min/week for 3 laboratory exercise crank exercise at a rate of 50 rev.
months /min, followed by two minutes of
rest.
 Intensity was increased to maintain
a ‘somewhat hard’ rating of
perceived exhaustion.
Tisi 199764 Advice only  Verbally encouraged to exercise at 1 x 60 min/week Not specified Leg  Series of active and passive leg
least 45 minutes each day in the for 1 month exercises exercises performed to the limit of
home and walk 1 mile daily. claudication pain.
 Also encouraged to exercise for at
least 45 minutes per day and walk 1
mile daily.
Treat-Jacobson Advice,  Provided with IC-specific, 3 x 70 min/week Exercise Treadmill  Began walking at 2 mph at 0%
200965 exercise standardised written exercise for 3 months laboratory walking grade. Walked until onset of
journal, instructions and exercise log. moderately severe pain, then asked
weekly in-  Advised to exercise at moderate to rest. Once able to walk for 8
person intensity exercise (walking), for a minutes, grade increased by
support minimum of 30 minutes a session, increments of 0.5% until 8-10%
at least three times per week. grade achieved , then speed
increased in increments of 0.1 – 0.2
mph.
Zwierska 200566 Advice only  Verbally encouraged to exercise 2 x 40 min/week Exercise Arm- and  Cycles of 2 minutes of exercise at a
regularly. for 6 months laboratory leg- cranking rate of 50 rev. /min, followed by
exercises two minutes of rest. Initial intensity
was 9W with power output
increased by 7W and 14W per
increment in the arm-cranking test
and leg-cranking test, respectively.
Final version
100
PAD
Table 28: Clinical evidence profile: Supervised exercise compared to unsupervised exercise for people with intermittent claudication due to femoro-
popliteal disease
Quality assessment No of patients Effect

Quality
No of Other Supervised Unsupervised Relative
Design Risk of bias Inconsistency Indirectness Imprecision Absolute
studies considerations exercise exercise (95% CI)
Quality of life at 6 months
157 RCT Serious(a) No serious No serious Serious(b) None 12 9 See Table 32 LOW
inconsistency indirectness
Quality of life at 1 year
157 RCT Serious(a) No serious No serious Serious(b) None 12 9 See Table 32 LOW
Withdrawal at 3 months
162 RCT Serious(a) No serious No serious Very None 2/30 2/30 RR 1 (0.15 to 0 fewer per VERY LOW
inconsistency indirectness serious(c) (6.7%) (6.7%) 6.64) 1000 (from 57
fewer to 376
more)
Withdrawal at 6 months (random effects)
257,62 RCT Serious(a) Serious(d) No serious Very None 7/42 7/39 RR 1.03 (0.18 to 5 more per 1000 VERY LOW
indirectness serious(c) (16.7%) (17.9%) 5.81) (from 147 fewer
to 863 more)
Withdrawal at 1 year
157 RCT Serious(a) No serious No serious Very None 6/12 2/9 RR 2.25 (0.59 to 278 more per VERY LOW
inconsistency indirectness serious(c) (50%) (22.2%) 8.65) 1000 (from 91
fewer to 1000
more)
ABPI at 6 months
166 RCT Serious(e) No serious No serious No serious None 71 33 - MD 0.01 lower MODERATE
Final version
101
PAD
inconsistency indirectness imprecision (0.09 lower to

0.01 higher)
(a) Unclear allocation concealment and blinding.
(b) No information on variability was given in the study, therefore the calculation of the standard deviation was not possible and the mean difference and CI were not estimable.
(c) 95% CI crosses both MIDs.
(d) Unexplained heterogeneity.
(e) Unclear randomisation process, allocation concealment and blinding.
Table 29: Clinical evidence profile: Supervised compared to unsupervised exercise for people with intermittent claudication due to unknown disease.

Quality
No of Other Supervised Unsupervised Relative
Design Risk of bias Inconsistency Indirectness Imprecision Absolute
studies considerations exercise exercise (95% CI)
356,58,60 RCT Very No serious No serious Serious(b) None 149 142 See Table 31 and Table 32 VERY LOW
serious(a) inconsistency indirectness
356,58,60 RCT Very No serious No serious Serious(b) None 149 142 See Table 31 and Table 32 VERY LOW
serious(a) inconsistency indirectness
256,58 RCT Serious(c) No serious No serious Serious(b) None 138 132 See Table 31 and Table 32 LOW
Quality of life at 1 year
256,58 RCT Serious(c) No serious No serious Serious(b) None 138 132 See Table 31 and Table 32 LOW
Maximum walking distance at 3 months (combined end and change results)
360,63,65 RCT Very No serious No serious Serious(e) None 71 42 - MD 154.49 higher VERY LOW
serious(d) inconsistency indirectness (85.73 to 223.26
higher)
Maximum walking distance at 6 months (combined end and change results)
Final version
102
PAD
260,65 RCT Very No serious No serious Serious(e) None 36 16 - MD 136.74 higher VERY LOW
serious(f) inconsistency indirectness (51.94 to 221.54
higher)
Pain free walking distance at 3 months (combined end and change results)
360,63,65 RCT Very No serious No serious Serious(e) None 71 42 - MD 74.71 higher VERY LOW
serious(d) inconsistency indirectness (30.48 to 118.95
higher)
Pain free walking distance at 6 months (combined end and change results)
260,65 RCT Very No serious No serious Serious(e) None 36 16 - MD 76.32 higher VERY LOW
serious(f) inconsistency indirectness (18.37 to 134.26
higher)
Adverse events at 3 months
167 RCT Very No serious No serious Very None 3/33 4/29 RR 0.66 47 fewer per 1000 VERY LOW
serious(f) inconsistency indirectness serious(g) (9.1%) (13.8%) (0.16 to (from 116 fewer to
2.7) 234 more)
Percentage of sessions attended in 3 months of treatment
167 RCT Very No serious No serious Serious(b) None Unsupervised exercise - completed 82.5% of sessions (33 VERY LOW
serious(f) inconsistency indirectness people)
Supervised exercise - completed 84.8% of sessions (29 people)
Withdrawal at 3 months
259,65 RCT Very No serious No serious Very None 7/60 5/36 RR 0.87 18 fewer per 1000 VERY LOW
2.79) 249 more)
259,65 RCT Very No serious No serious Very None 20/60 10/36 RR 1.16 44 more per 1000 VERY LOW
2.32) 367 more)
Withdrawal at 1 year
158 RCT No serious No serious No serious Very None 16/109 18/102 RR 0.83 30 fewer per 1000 LOW
risk of bias inconsistency indirectness serious(g) (14.7%) (17.6%) (0.45 to (from 97 fewer to
1.54) 95 more)
Final version
103
PAD
ABPI at 3 months
460,61,63,64 RCT Very No serious No serious No serious None 70 61 - MD 0.02 lower (0.06 LOW
serious(h) inconsistency indirectness imprecision lower to 0.02
higher)
ABPI at 6 months (random effects)
260,64 RCT Very Serious(i) No serious Serious(e) None 33 27 - MD 0 lower (0.16 VERY LOW
serious(f) indirectness lower to 0.17
higher)
ABPI at 1 year
164 RCT Very No serious No serious Serious(e) None 22 17 - MD 0.1 lower (0.27 VERY LOW
serious(f) inconsistency indirectness lower to 0.07
higher)
(a) 1 of 3 studies had unclear methodology; 1 of 3 studies had unclear allocation concealment and blinding; 1 of 3 studies had low risk of bias.
(b) No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
(c) 1 of 2 studies had unclear allocation concealment and blinding and baseline differences; 1 of 2 studies had low risk of bias
(d) 2 of 3 studies had unclear methodology; 1 of 3 studies had unclear allocation concealment and blinding.
(e) 95% CI crosses one MID.
(f) Unclear methodology.
(g) 95% CI crosses both MIDs.
(h) 1 of 4 studies had unclear allocation concealment and blinding; 3 of 4 studies had unclear methodology.
(i) Unexplained heterogeneity.
Table 30: Study characteristics: Reason for withdrawal from exercise programmes
Study Unsupervised exercise programme (n) Supervised exercise programme (n)
3 months
Treat-Jacobson 2009 2/8 4/33
Lost to follow-up (1), Study unrelated health problem (1) Family crisis (3), Unrelated injury (1)
Stewart 2008 2/30 2/30
Withdrawal given without a reason (1), Aggravated back injury (1) Fatal stroke (1), Aggravated back injury (1)
6 months
Treat-Jacobson 2009 2/8 12/33
Lost to follow-up (1), Unrelated health problem (1) Family crises (3), Unrelated injury (1), Lost to follow-up (2),
Unrelated health problem (6)
Final version
104
PAD
Stewart 2008 4/30 3/30

Fatal stroke (1), Withdrew without giving a reason (2), Aggravated Developed leg ulcer (1), Fatal stroke (1), Aggravated back
back injury (1) injury (1)
Kakkos 2005 1/9 4/12(a)
Due to development of rest pain and had a bypass Due to: Fatigue, Bladder cancer, GI bleeding, Leg injuries,
Personal choice
12 months
Kakkos 2005 2/9 6/12
Withdrew consent (1), Developed rest pain and had bypass (1) Fatigue, Bladder cancer, GI bleeding, Leg injuries, Personal
choice withdrew consent
Nicolai 2010 18/102 16/109
Lack of motivation (7), CHD (1), CVA (1), Orthopaedic disease (2) Lack of motivation (3), Progression of PAD (2), CHD (2),
Other concomitant disease (4), Death (3) Orthopaedic disease (2), Diabetic foot (1), Other
concomitant disease (3), Death (4)
(a) Numbers attributed to each reason not reported within the study.
Table 31: EQ-5D: Unsupervised compared to supervised exercise

Unsupervised exercise Supervised exercise
Baseline 3 months 6 months 9 months 12 months Baseline 3 months 6 months 9 months 12 months
58,69
Nicolai 2010 & van Asselt 2011 – Mean (SD)
0.62 ± 0.23 0.68 ± 0.23 0.69 ± 0.19 0.68 ± 0.23 0.66 ± 0.26 0.66 ± 0.2 0.69 ± 0.21 0.72 ± 0.17 0.73 ± 0.21 0.74 ± 0.2
Table 32: SF 36 individual domain results and mapped EQ-5D values – unsupervised compared to supervised exercise
Cheetham 200456 - Median (IQR)
PF 50 (20) 55 (NR) 55 (NR) 55 (NR) 55 (NR) 60 (20) 65 (NR) 70 (NR) 70 (NR) 70 (NR)
RP 56 (19) 53 (NR) 56 (NR) 56 (NR) 56 (NR) 75 (44) 75 (NR) 84 (NR) 81 (NR) 88 (NR)
BP 70 (36) 71 (NR) 70 (NR) 77 (NR) 71 (NR) 59 (29) 72 (NR) 71 (NR) 72 (NR) 72 (NR)
GH 59 (27) 56 (NR) 59 (NR) 63 (NR) 59 (NR) 67 (22) 65 (NR) 67 (NR) 70 (NR) 62 (NR)
Final version
105
PAD
V 53 (12) 53 (NR) 59 (NR) 56 (NR) 53 (NR) 56 (37) 56 (NR) 62 (NR) 65 (NR) 62 (NR)
SF 81 (37) 81 (NR) 81 (NR) 81 (NR) 81 (NR) 88 (50) 88 (NR) 88 (NR) 88 (NR) 88 (NR)
RE 67 (42) 71 (NR) 75 (NR) 67 (NR) 67 (NR) 67 (50) 67 (NR) 67 (NR) 67 (NR) 67 (NR)
MH 70 (40) 70 (NR) 70 (NR) 73 (NR) 70 (NR) 75 (35) 75 (NR) 80 (NR) 80 (NR) 75 (NR)
(a)
EQ-5D 0.65 (0.02) 0.71 (0.02) 0.70 (0.02) 0.73 (0.02) 0.71 (0.02) 0.71 (0.02) 0.76 (0.02) 0.79 (0.02) 0.79 (0.02) 0.78 (0.02)
57
Kakkos 2005 – Median (IQR)
PF 50 (30) NR 60 (23) NR 45 (25) 65 (14) NR 65 (23) NR 50 (30)
RP 100 (50) NR 75 (38) NR 50 (75) 50 (44) NR 50 (12) NR 0 (100)
BP 60 (45) NR 62 (27) NR 51 (43) 60 (27) NR 70 (42) NR 62 (43)
GH 35 (31) NR 40 (14) NR 40 (10) 35 (19) NR 35 (13) NR 50 (30)
V 60 (22) NR 65 (24) NR 50 (15) 70 (10) NR 60 (25) NR 50 (30)
SF 78 (11) NR 72 (20) NR 89 (78) 78 (20) NR 78 (11) NR 89 (22)
RE 33 (33) NR 33 (0) NR 67 (100) 0 (25) NR 0 (33) NR 0 (33)
MH 52 (28) NR 44 (27) NR 88 (36) 44 (20) NR 56 (20) NR 76 (20)
(b)
EQ-5D NR NR NR NR NR NR NR NR NR NR
58
Nicolai 2010 – Mean (SD)
PF 52.4 (15.0) 59.4 (16.6) 61.3 (15.8) 55.4 (18.0) 59.0 (19.0) 52.8 (14.3) 61.7 (16.4) 65.9 (16.7) 62.3 (16.9) 65.1 (16.8)
RP 51.0 (40.8) 56.8 (38.0) 55.2 (39.0) 51.8 (40.8) 55.8 (39.8) 45.8 (39.1) 53.5 (40.7) 58.5 (38.9) 57.9 (39.0) 65.3 (36.2)
BP 52.0 (18.0) 54.5 (19.8) 56.1 (21.7) 51.9 (24.3) 55.8 (22.7) 51.1 (16.6) 57.4 (20.9) 61.2 (22.6) 60.9 (23.6) 64.8 (22.5)
GH 54.9 (13.0) 48.4 (21.5) 55.7 (12.1) 55.6 (12.2) 54.2 (12.8) 53.7 (12.6) 55.6 (12.8) 56.1 (12.1) 55.0 (12.6) 53.6 (14.3)
V 63.0 (20.3) 62.6 (21.1) 60.3 (18.3) 57.9 (21.2) 59.2 (19.8) 61.6 (18.7) 62.2 (18.3) 62.5 (19.2) 60.4 (19.6) 62.0 (18.9)
SF 79.9 (19.6) 79.5 (24.2) 78.6 (24.3) 72.4 (27.3) 75.4 (25.3) 77.1 (22.8) 80.6 (21.6) 79.0 (21.7) 76.7 (23.6) 81.7 (22.8)
RE 85.1 (29.0) 82.5 (34.8) 85.5 (29.4) 82.0 (32.4) 82.4 (34.9) 85.2 (32.6) 87.9 (29.0) 85.2 (30.5) 85.8 (29.6) 86.1 (29.1)
MH 76.4 (17.2) 75.2 (17.8) 72.8 (24.3) 73.5 (17.8) 74.6 (19.1) 75.5 (17.8) 76.4 (18.4) 76.4 (17.6) 74.4 (18.8) 74.9 (20.3)
(a)
EQ-5D 0.66 (0.01) 0.68 (0.01) 0.69 (0.01) 0.65 (0.01) 0.68 (0.01) 0.65 (0.01) 0.71 (0.01) 0.73 (0.01) 0.71 (0.01) 0.74 (0.01)
60
Savage 2001 – Mean (SD)
PF 45 (17) 61 (10) 54 (27) NR NR 54 (14) 60 (16) 56 (14) NR NR
RP 47 (47) 68 (43) 47 (46) NR NR 84 (30) 77 (34) 84 (19) NR NR
BP 50 (13) 72 (23) 64 (14) NR NR 59 (20) 70 (18) 65 (19) NR NR
Final version
106
PAD
GH 67 (9) 65 (17) 65 (19) NR NR 71 (17) 64 (14) 66 (18) NR NR

V 49 (22) 47 (6) 52 (19) NR NR 66 (17) 68 (17) 63 (16) NR NR
SF 85 (19) 90 (15) 85 (20) NR NR 91 (11) 92 (10) 91 (10) NR NR
RE 75 (46) 81 (38) 74 (43) NR NR 97 (10) 82 (35) 71 (45) NR NR
MH 83 (13) 74 (17) 65 (31) NR NR 79 (16) 82 (12) 73 (17) NR NR
(a)
EQ-5D 0.66 (0.03)* 0.76 (0.03)* 0.68 (0.04)* NA NA 0.68 (0.03)* 0.74 (0.03)* 0.69 (0.03)* NA NA
(a) Mapped based on algorithm (Equation1) reported by Ara and Brazier 200868
(b) Not estimable based on median values of 0.
Abbreviations: PF = physical function; RP = role physical; BP = bodily pain; GH = general health; V = vitality; SF = social functioning; RE = role emotional; MH = mental health; SD= standard
deviation; IQR = interquartile range; NA = not applicable; NR = not reported.
Final version
107
PAD
Two cost-utility analyses were included from the economic literature search. Both studies were
based on clinical trials with a time horizon of one year. The analysis by van Asselt 201169 was based
on an RCT included in the current clinical review. Using bootstrap analysis, this study reported there
was only a 35% probability that supervised exercise was cost effective at a threshold of £20, 000. The
study by Lee 200770 concluded that supervised exercise is cost effective compared to unsupervised
exercise in a UK NHS setting. However, the evidence used to inform this analysis was taken from a
non-randomised trial with a non-preference based method of QALY calculation. Study characteristics
and a summary of results are presented in Table 33. Detailed economic evidence tables can be found
in Appendix I.
The GDG considered compliance to the prescribed exercise programme (i.e. the proportion of people
continue to exercise long-term following either intervention) to be a key factor for determining the
long term cost-effectiveness of supervised exercise. Neither of the included studies was thought to
sufficiently capture the long-term compliance to each type of treatment nor were they designed to
evaluate the benefit to cardiovascular health that is associated with exercise. Therefore, an original
economic model was developed using data collected from the clinical review and supplementary
evidence where required.
9.2.1.3 Original economic model
NB: A detailed report of the methods and results of this model can be found in Appendix K.
Methods
A cost-utility analysis was undertaken to evaluate the cost-effectiveness of unsupervised compared

to supervised exercise for the treatment of IC. A Markov model (see Figure 2) was used to estimate
the lifetime quality-adjusted life years (QALYs) and costs from a 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 guidance. The model was built probabilistically to take into account uncertainty
surrounding each of the model input parameters.
Approach to modelling
Intermittent claudication (IC) is associated with an increased mortality and risk of cardiovascular
morbidity, and a decreased quality of life. Participation in regular physical activity is associated with a
reduction in the risk of cardiovascular events, greater life expectancy, and an improvement in quality
of life.
However, the benefits of exercise therapy are lost if the person ceases to be active. Improvements in
cardiovascular function that occur with exercise rapidly deteriorate with inactivity or a reduction in
the volume of exercise training71 and there is evidence that the quality of life gain reported by people
who have completed an exercise programme is only maintained if individuals continue be active.72
The model therefore contains two primary health states: active and sedentary. The ‘active’ state was
used to describe people who maintain a similar level of activity to that reported in the clinical trials.
The level of activity described by the trials closely matches the definition of an ‘active’ lifestyle used
by several other sources included in the model, including the 2006 Health Survey for England.b
‘Sedentary’ was used to describe people who are less active or inactive.
b The HSE defines an active lifestyle as undertaking 30 minutes or more of moderate of vigorous physical activity on one
to four days per week.
Final version
108
PAD
Table 33: Economic evidence profile: Unsupervised exercise vs. supervised exercise
Incremental Incremental Cost Uncertainty
Study Limitations Applicability Other comments cost effects effectiveness
Lee 200770 Serious Directly  Cost utility analysis based on Supervised Supervised Supervised  Not evaluated by
limitations(a) applicable(b) a non-randomised trial by Lee exercise was exercise exercise cost authors.
200770 £52 more resulted in a £1, 935 per
 Population: People with IC costly than gain of 0.027 QALY gained
unsupervised QALYs
 Time horizon: 1 year
exercise compared to
 Costs: Supervised exercise unsupervised
programme exercise
 Perspective: UK NHS
Exercise Therapy Minor Partially  Cost utility analysis based on Supervised Supervised Supervised There was a 20%
in Peripheral limitations(c) applicable(d) RCT by Nicolai 201058 exercise was exercise exercise cost probability that supervised
Arterial Disease  Population: People with IC £874 more resulted in a £22, 997 per exercise is cost-effective
(EXITPAD) study69 costly than gain of 0.038 QALY gained based on bootstrap
unsupervised QALYs analyses.
 Costs: All healthcare and non- exercise compared to
healthcare costs based on unsupervised
retrospective patient exercise
questionnaire
 Perspective: Netherlands,
societal
(a) Non-randomised sources of clinical effectiveness included studies evaluated exercise programmes of different durations.
(b) No preference weighting assigned to SF-36 scores (invalid QALY valuation); short time horizon.
(c) Societal perspective; short time horizon.
(d) Dutch healthcare setting.
Final version
109
PAD
The main assumption of the model was therefore that compliance to the recommended level of
physical activity is needed to provide the benefits associated with these programmes. People who
revert to a sedentary state were assigned baseline cardiovascular risk, mortality and quality of life
estimates. As a necessary simplification, it was assumed that those who stop exercising remain
sedentary. Please see Appendix L for the model evaluating sequential exercise and endovascular
interventions.
In order to explore the impact that different levels of compliance have on the cost and effects of
each type of programme, two different scenarios were modelled: in Scenario 1, supervised exercise
leads to greater short and long term compliance; and in Scenario 2, supervised exercise leads to
greater short term compliance and no difference in long term compliance.
As a necessary simplification, people who experience a cardiovascular event enter a semi-absorbing

health state from which the only available transition is death. Average costs and quality of life
associated with post-cardiovascular event states were applied to this health state, and the same
mortality rate as sedentary people was assumed.
The GDG decided to use the quality of life data from the RCTs included in the clinical review as the
primary measure of clinical effectiveness. The group were aware that other models, such as the TA
developed by Squires 201053, used maximum walking distance (MWD) as a proxy for calculating QALY
values. However, the GDG agreed that this was an inferior measure of effectiveness when quality of
life outcomes were directly available from the included RCTs.
Figure 2: Markov model structure
Active
Death
(all cause)
Nonfatal
Sedentary cardiovascular
Post - nonfatal event
cardiovascular
event
Schematic diagram of the Markov model designed to compare the cost-effectiveness of supervised to unsupervised exercise
programmes for the treatment of people with IC. The Markov modelling approach involves a transition between different
health states over time. The model is divided into three month cycles. At the end of each cycle a time-dependant transition
to another health state is possible, unless people enter into an ‘absorbing state’ from which they do not recover. In this
model, the absorbing state is death.
Baseline mortality and relative risk associated with exercise
Age- and sex-specific all cause mortality was based on the most recent available life tables of the
general population in England and Wales. Rates were adjusted for people with IC by multiplying by
the standardised risk of all cause mortality observed over 10 years in people with IC.73
Final version
110
PAD
No randomised evidence of exercise-associated risk of mortality in people with IC was identified. The
GDG agreed that evidence from people with cardiovascular disease would represent a reasonable
proxy. A recent Cochrane review of randomised controlled trials was therefore used to inform the
risk of total mortality among people participating in exercise rehabilitation compared to non-active
controls.74 A summary of the values used to inform this parameter is provided in Table 34. The GDG
discussed the limitations of using an indirect population to inform this parameter and the effect of
this value on the model result was further explored in sensitivity analysis.
Table 34: Total mortality

10-year total mortality for Relative risk of total Relative risk of total mortality
the general population mortality in people with IC in people who exercise
based on Life Tables for compared to those without compared to those who do not
England and Wales(a) IC exercise
Mortality 25.0% 3.1 (95% CI, 1.9 to 4.9)73 0.87 (95% CI , 0.75 to 0.99)74
(a) Assuming that the average age of the baseline population is 67 years and 66% are male.
Baseline risk of cardiovascular events and relative risk associated with exercise
The average baseline probability of stroke and MI was calculated by age and gender using the
Framingham risk equations and risk calculator spreadsheet developed by Rupert Payne at the
University of Edinburgh.75,76 Risk factor inputs for each sex were obtained from the 2006 Health
Survey for England (HSE; Table 35).77 Average age- and sex- specific blood pressure values were
obtained from the 2011 NICE Hypertension update guideline78, which used individual patient level
data from the 2006 HSE. A recent study by the Ankle Brachial Index Collaboration found that when
combined with Framingham risk scores, an ABPI of between 0.61 and 0.70 approximately triples the
risk of major cardiovascular events for men and women.79
The risk of myocardial infarction (MI) in patients who exercise compared to those who are not active
in an exercise programme was obtained from the Cochrane review by Heran et al (2011).74 A meta-
analysis of the effect of physical activity on the incidence of stroke was used to inform the risk of
stroke for active compared to sedentary people in the model.80 A summary of the values used to
inform these parameters is provided in Table 35. As with estimates of the relative risk of total
mortality, these data sources are subject to several limitations and the effect of these values on the
model were explored in sensitivity analysis.
Table 35: Major cardiovascular events

10 year risk of MI and
stroke for general Relative risk of major Relative risk of MI and stroke
population according to cardiovascular events in in people who exercise
the Framingham people with IC compared to compared to those who do not
equations(a) those without IC(b) exercise
MI 7.2% Men: 2.71 (95% CI, 2.01 to 0.97 (95% CI, 0.82 to 1.15)74
3.64)
Women: 3.82 (95% CI, 2.86 to
5.11)
Stroke 4.4% Men: 2.71 (95% CI, 2.01 to 0.80 (95% CI, 0.74 to 0.86)80
3.64)
Women: 3.82 (95% CI, 2.86 to
5.11)
(a) Calculated using Framingham MI and stroke risk equations75,76 and risk factor inputs derived from the 2006 Health
Survey for England77, assuming that the average age of the baseline population is 67 years and 66% are male.
(b) Based on a risk of cardiovascular events for mean and women with an ABPI of 0.61 to 0.7 compared to men and women
with normal ABPI.79
Final version
111
PAD
Quality of life
In cost-utility analyses, measures of health benefit are valued in terms of quality adjusted life years
(QALYs). The QALY is a measure of a person’s length of life weighted by a valuation of their health
related quality of life (HRQoL) over that period. The quality of life weighting comprises two elements:
the description of changes in HRQoL and an overall valuation of that description. Questionnaires such
as the SF-36 and SF-12 provide generic methods of describing HRQoL while the EQ-5D, HUI, and SF-
6D also include preference-based valuations of each health state.
Quality of life data was collected from all RCTs included in the clinical review (see Appendix H). One
study included the EQ-5D as a measure of HRQoL.58 Five papers (representing an additional four
trials) reported SF-36 data.56-60 According to the NICE reference case10, EQ 5D data is the preferred
measure of quality of life for use in cost utility analyses. Because Nicolai 2010 and van Asselt 2011
report different quality of life outcomes from the same study (EXITPAD), in the base case analysis,
the EQ-5D values reported by van Asselt 2011 were used in preference to SF-36 scores reported by
Nicolai 2010.
Recently, several algorithms have been developed which can be used to map generic descriptions of
HRQoL to preference-based utility indexes. In 2008, Ara and Brazier68 published a method of
predicting mean EQ-5D preference based index score using published mean cohort statistics from the
eight dimensions of the SF-36 health profile. In order to use these algorithms, values for each of the
eight dimensions of the questionnaire are required. Two57,60 provided all the necessary values and
the authors of the remaining three studies56,58,59 were contacted to request the required data.
Nicolai 2010 and Cheetham 2004 granted access to mean SF-36 scores and permission to include it in
the current analysis. The authors of the study by Pinto 2001 were unable to provide similar data as it
was no longer available. The data reported by Kakkos and colleagues 2005 was found to produce
invalid values for mapping and was excluded. Therefore, of the eleven RCTs identified in the clinical
review, those by Cheetham 200456, Nicolai 2010/van Asselt 201158,69 and Savage 200160 were used to
calculate quality of life following supervised and unsupervised exercise programmes.
Mapping SF-36 to EQ-5D using published algorithms and probabilistic simulation
For each trial, it is the change in quality of life over time and the difference in this change between
interventions (i.e. mean difference in change) that is the key to determining the relative
effectiveness of each intervention. In order to calculate the mean difference in change between each
three month time interval while taking into account the uncertainty surrounding each estimate, the
mean and standard error of each dimension of the SF-36 were assigned a beta distribution according
to the method of moments described by Briggs 2006.81 Probabilistic mapped values were then
calculated using Equation 4 from the paper by Ara and Brazier68, who specify that ‘when comparing
incremental differences between study arms or changes over time, Equation 4 is the preferred
choice’. A simulation was run 10, 000 times in order to calculate a mean, standard error and
confidence interval surrounding each mapped estimate. For the purposes of clinical validation,
absolute mean mapped values were calculated using Equation 1 according to the same method.
Note that mean difference in change calculated using Equation 4 is not expected to equal the
incremental difference between the mean mapped values from Equation 1 as they are derived using
different models. Alternative methods of calculating relative differences in quality of life between
treatment arms were explored in sensitivity analysis. Note also that because the covariance matrices
for the regression coefficients were not available it was not possible to account for uncertainty in the
mapping algorithm in the probabilistic analysis.
Final version
112
PAD
Inputs and assumptions used to inform model utilities
In the base case analysis, an average utility value was weighted according to the total number of
people in the study at each time point and entered into the probabilistic model using a beta
distribution. In order to preserve within-study randomisation, the weighted average incremental
change in quality of life associated with supervised exercise as calculated by the probabilistic
simulation described above was added to the baseline quality of life across the two trials. Quality of
life gains achieved after exercise intervention were maintained for people who continued to exercise.
Those who stopped exercising were assigned the baseline quality of life.
The weighted average absolute values and weighted mean difference in change are reported in Table
36. Please see Appendix K for further details.
Table 36: Quality of life

Weighted mean (SE)
Weighted average (SE) quality of life(a) Weighted average
difference in change
(SE) baseline quality
between each follow-
of life
Unsupervised Supervised up interval(b)
Baseline 0.636 (0.017) 0.672 (0.014) 0.654 (0.011)
3 months 0.692 (0.017) 0.709 (0.015) -0.021 (0.033)
6 months 0.692 (0.014) 0.732 (0.013) 0.026 (0.032)
9 months 0.692 (0.018) 0.744 (0.016) 0.010 (0.034)
12 months 0.671 (0.023) 0.748 (0.017) 0.029 (0.039)
(a) Calculated based on Equation 1 from Ara and Brazier 200868 and weighted according to the number of patients in each
trial.
(b) Calculated based on Equation 4 from Ara and Brazier 200868 and weighted according to the number of patients in each
trial. Positive values indicate a net benefit of supervised exercise. Note that these values do not equal the mean
difference in change between absolute weighted mean values because they are calculated using different mapping
equations.
Abbreviation: SE = standard error of the mean.
Costs
The cost of a supervised programme was based on estimates of resource use informed by expert
opinion and unit costs obtained from the 2010 PSSRU. A breakdown of the assumptions and unit
costs used to calculate per-patient cost of a supervised exercise programme are provided in Table 37.
Because the cost of the initial GP consultation is common to both supervised and unsupervised
exercise, it is not included in the cost of either intervention arm (i.e. it ‘cancels out’). The cost of
unsupervised exercise was therefore set at £0. This was varied in sensitivity analysis to account for
different levels of support provided by different types of unsupervised programmes.
Table 37: Cost of a 3 month supervised exercise programme

Programme duration and intensity
Two hours of class per week for three months (13 weeks)(a)
Ten people per class(b)
Resource use Unit cost
(b)
Two physiotherapists £37 (x2) per hour(c)
One physiotherapist technician(b) £22 per hour(c)
Room hire and equipment rental(b) £15 per hour(b)
Associated cost of supervised exercise programme
Total programme cost (per 10-person group) £2, 886
Final version
113
PAD
Programme duration and intensity

Total programme cost per patient £288
(a) Average length and duration of exercise programmes evaluated by RCTs included in clinical review (see Table 27).
(b) Based on expert opinion (with thanks to Lysa Downing, Ricky Mullis and Martin Fox): several GDG members sent
requests for information to their clinical colleagues and commissioning managers and responses were received from
around the country. A number of different models were described and discussed by the GDG. The resource use described
in the table was thought to represent the typical pattern for outpatient care for people with IC.
(c) Obtained from the 2010 PSSRU51
The approach to modelling cardiovascular events was based on the model developed for the NICE
hypertension guideline update (CG 127 http://www.nice.org.uk/Guidance/cg127).78 As in the
hypertension model, when people with IC experienced a cardiovascular event they were assigned an
initial cost representing the acute management and/or diagnosis cost (MI = £4, 792; stroke = £9,
630). In subsequent cycles they were assigned an ongoing cost representing the average costs
following an event (MI = £141; stoke = £559).
Compliance to supervised and unsupervised exercise
Several studies identified in the clinical review reported either total dropout rates or dropouts
associated with each study arm (Table 30). However, the GDG did not consider compliance within a
trial setting to be representative of real world behaviour. The literature was reviewed for estimates
of short and long-term compliance to supervised and unsupervised exercise programmes in people
with PAD, cardiovascular disease or older adults in the community; no relevant evidence was
identified. Therefore, based on input from the GDG, two different scenarios were modelled: in
Scenario 1, supervised exercise leads to greater short and long-term compliance (Figure 3); and in
Scenario 2, supervised exercise leads to greater short term compliance and no difference in long-
term compliance (Figure 4).
Figure 3: Scenario 1 – Greater long term compliance to supervised exercise
100%
90%
80%
70%
Exercise compliance
60%
50%
Supervised exercise
40%
Unsupervised exercise
30%
20%
10%
0%
0 1 2 3 4 5 6 7 8 9
Cycle (1 cycle = 3 months)
Time period Cycle Supervised Unsupervised

Lowest Most likely Highest Lowest Most likely Highest
3 months 1 40% 68% 83% 17% 43% 56%
6 months 2 25% 50% 66% 10% 33% 45%
1 year 4 14% 37% 54% 7% 22% 37%
2 years 8 12% 31% 47% 5% 16% 31%
Final version
114
PAD
Figure 4: Scenario 2 – Equal long term compliance to supervised and unsupervised
100%
90%
80%
70%
Exercise compliance
60%
50%
Supervised exercise
40%
30%
20%
10%
0%
0 1 2 3 4 5 6 7 8 9
Time point Cycle Supervised Unsupervised

Lower Most Upper Lower Most Likely Upper
likely
3 months 1 40% 68% 80% 17% 43% 56%
6 months 2 15% 40% 57% 10% 33% 45%
1 year 4 4% 22% 40% 7% 22% 37%
2 years 8 5% 16% 32% 5% 16% 31%
Results
This analysis found that supervised exercise is more cost effective than unsupervised exercise. By
taking into account the standard error of each model input, probabilistic analysis revealed that if
supervised exercise leads to greater compliance over both the short and long term, it is cost effective
in 79% of model iterations at an average cost of £711 per QALY gained. If supervised exercise does
not lead to an increase in activity levels over the long term, it remains cost effective in 75% of model
iterations at an average cost of £1, 608 per QALY gained (Table 38).
Final version
115
PAD
Figure 5: Distribution of incremental costs and effects
ICER scatterplot (500 model iterations)

350
£20, 000 per QALY thresholda
300
Incremental costs (£)
250
200
150 Scenario 1
Scenario 2
100
50
0
-1 -0.5 0 0.5 1 1.5 2
Incremental effectiveness (QALYs)

aPoints lying to the right of the £20, 000 threshold are considered cost effective.
Table 38: Mean probabilistic results of cost effectiveness model

Incremental Total Incremental Incremental Probability
Strategy Total Cost Cost QALYs QALYs cost per QALY of being CE
Scenario 1 – Greater long term compliance to supervised exercise
Unsupervised £2, 499 Baseline 5.082 Baseline Baseline 21%
Supervised £2, 690 £191 5.350 0.268 £711 79%
Scenario 2 – Equal long term compliance
Unsupervised £2, 499 Baseline 5.078 Baseline Baseline 25%
Supervised £2, 714 £215 5.212 0.134 £1, 608 75%
Disaggregating the results of the analysis by cost and QALYs allows us to examine the impact of key
components of the model on the overall result. Table 39 illustrates that the cost of the supervised
exercise programme is the major driver in cost differences between the two interventions. As would
be expected, the cost associated with the prevention of cardiovascular events is greater in the
scenario with greater difference compliance between interventions (Scenario 1), but in both
scenarios the incremental cost associated with cardiovascular morbidity is relatively small. Table 40
shows the impact of the reduction in mortality attributed to people who continue to be active in
terms of the difference in baseline QALY gain between the two interventions. Although the reduction
in mortality associated with exercise plays a role in driving the results of the model, this table
illustrates that the main driver in the difference in quality of life between the two exercise strategies
is the difference in quality of life associated with the intervention itself. The effect of cardiovascular
morbidity on the results of the model is negligible.
Final version
116
PAD
Table 39: Breakdown of total costs (probabilistic)

Unsupervised Supervised Incremental cost of
exercise exercise supervised exercise
Scenario 1- Greater long term compliance to supervised exercise
Supervised exercise programme £0 £219 £219
Initial CV events £1, 186 £1, 176 £-10
Follow-up CV event £1, 259 £1, 241 £-18
Scenario 2– Equal long term compliance
Supervised exercise programme £0 £219 £219
Initial CV events £1, 186 £1, 184 £-2
Follow-up CV event £1, 259 £1, 256 £-3
Table 40: Breakdown of total QALYs (probabilistic)

Unsupervised Supervised Difference
exercise exercise (Supervised – Unsupervised)
Scenario 1- Greater long term compliance to supervised exercise
Baseline quality of life 5.191 5.230 0.039
Supervised exercise programme 0.000 0.250 0.250
CV events (initial and follow-up) -0.010 -0.010 0.000
Scenario 2– Equal long term compliance
Baseline quality of life 5.185 5.189 0.004
Supervised exercise programme 0.000 0.132 0.132
CV events (initial and follow-up) -0.010 -0.010 0.000
Sensitivity analysis
A wide range of sensitivity analyses were undertaken to explore the effect of different parameter
inputs and assumptions on the results of the model. The results of these sensitivity analyses showed
that supervised exercise is the most cost effective strategy under the majority of data sources and
assumptions tested. The exception to this was if all key assumptions about the benefits of exercise
were removed from the model. If we do not extrapolate quality of life beyond the trial end dates and
do not include any measure of mortality or cardiovascular benefit in people who are active,
supervised exercise programmes are unlikely to be cost effective compared to unsupervised exercise.
The full results of all sensitivity analyses are presented in Appendix K.
Interpretation and limitations
The clinical review was not designed to distinguish between trials of varying length, duration or
exercise intensity. As such, it is not possible to determine whether certain types of supervised
programmes are more cost effective than others. For this guideline, the definition of each type of
exercise programme was based on a simple average of studies included in the clinical review. The
supervised exercise programme described by this method was also found to match programmes
familiar to the GDG.
Currently, no published RCT data exist to inform the relative risk of cardiovascular events and
mortality in people who exercise compared to those who do not in people with IC. The data used in
this model was obtained from two meta-analyses of trials conducted in two different populations:
people with CHD who had experienced MI or coronary revascularisation and a mixed population of
people who had and had not had a stroke.
Final version
117
PAD
Limited published data was available to inform the impact of each type of exercise programme on
quality of life beyond one year. Although this data was not comparative, it suggested that quality of
life is maintained in those who continue to exercise; this was a key assumption of the analysis. If this
assumption is removed from the model, there is still a high probability that supervised exercise is
cost effective under the level of compliance suggested by Scenario 1, but there is a higher level of
uncertainty under Scenario 2.
The effectiveness of supervised and unsupervised exercise programmes is directly related to the
ability of each intervention to produce a lasting change on the activity levels of participating
individuals. Currently, data about the short and long term compliance to these regimens is not
available in the public domain. In the absence of this evidence, the GDG and their colleagues were
surveyed in order to elicit an expert opinion on which to base this parameter. The resulting estimates
that were used to inform the model represent the group’s most plausible scenarios for a population
of people with IC based on their clinical experience. However, long term data from real clinical
practices is needed to better inform future modelling in this area.
9.2.2.1 Clinical
Intermittent claudication due to aorto-iliac disease:
No clinical evidence was reported for people with IC due to aorto-iliac disease.
Intermittent claudication due to femoro-popliteal disease:
There was no statistically significant difference between supervised exercise and unsupervised
exercise for:
 Withdrawal at 3 months [1 study, 60 participants, very low quality evidence]62
 Withdrawal at 6 months [2 studies, 81 participants, very low quality evidence]57,62
 Withdrawal at 1 year [1 study, 21 participants, very low quality evidence]57
 ABPI at 6 months [41 study, 104 participants, moderate quality evidence]66
Evidence statement for outcomes where meta-analysis was not possible – no statistical analysis
performed
 Quality of life increased in most SF-36 domains for both supervised exercise and unsupervised
exercise at 6 months [1 study, 21 participants, low quality evidence]57
 Quality of life decreased in most SF-36 domains for both supervised exercise and unsupervised
exercise at 1 year [1 study, 21 participants, low quality evidence]57
Intermittent claudication - unknown disease location:
Supervised exercise was significantly better than unsupervised exercise for:

 Maximum walking distance at 3 months [3 studies, 113 participants, very low quality
evidence]60,63,65
 Maximum walking distance at 6 months [2 studies, 52 participants, very low quality evidence]60,65
 Pain free walking distance at 3 months [3 studies, 113 participants, very low quality
evidence]60,63,65
 Pain free walking distance at 6 months [2 studies, 52 participants, very low quality evidence]60,65
Final version
118
PAD
There was no statistically significant difference between supervised exercise and unsupervised
exercise for:
 Adverse events at 3 months [1 study, 62 participants, very low quality evidence]67
 Withdrawal at 3 months and 6 months [2 studies, 96 participants, very low quality evidence]59,65
 Withdrawal at 1 year [1 study, 211 participants, low quality evidence]58
 ABPI at 3 months [4 studies, 131 participants, low quality evidence]60,61,63,64
 ABPI at 6 months [2 studies, 60 participants, very low quality evidence]60,64
 ABPI at 1 year [1 study, 39 participants, very low quality evidence]64
performed:
 Quality of life increased for both supervised exercise and unsupervised exercise at 3 months [3
studies, 291 participants, very low quality evidence]56,58,60
 Quality of life mostly increased for supervised exercise and mostly decreased for unsupervised
exercise at 6 months [3 studies, 291 participants, very low quality evidence]56,58 60
exercise at 9 months [2 studies, 270 participants, low quality evidence]56,58
exercise at 1 year [2 studies, 270 participants, low quality evidence]56,58
 One study showed people treated with supervised exercise completed 84.8% of sessions during 3
months of treatment compared to people treated with unsupervised exercise completed 82.5% of
sessions during 3 months of treatment [1 study, 62 participants, very low quality evidence]67
9.2.2.2 Economic
 One trial-based study concluded that unsupervised exercise was more cost effective than
supervised exercise in 65% of patients [partially applicable with minor limitations]69
 One trial-based cost-utility evaluation concluded that supervised exercise is more cost effective
than unsupervised exercise [directly applicable with potentially serious limitations]70
 According to the results of an original economic model based on the current clinical evidence
review and GDG input, it is highly likely that supervised exercise represents a cost effective
treatment for people with IC [directly applicable with minor limitations]
10. Offer a supervised exercise programme to all people with

intermittent claudication.
11.Consider providing a supervised exercise programme for people

with intermittent claudication which involves:
 2 hours of supervised exercise a week for a 3-month period
Recommendations  encouraging people to exercise to the point of maximal pain.
Relative values of The GDG were interested in whether supervised exercise programmes would
different outcomes influence mortality as well as quality of life, but the available studies did not
address this issue.
The absolute change in maximum walking distance (MWD) and quality of life
were considered to be the most important outcomes in measuring success of
exercise interventions. MWD is the most widely reported outcome in studies for
Final version
119
PAD


intermittent claudication. Improvement can be reported as absolute or
percentage change in MWD; there is value in knowing both, although the GDG
agreed that the absolute change was more important. It was also recognised
that exercise is likely to have additional benefits such as improvements in
psychological or emotional well-being that should be captured by changes in
quality of life measures. Overall the studies suggested that participation in a
supervised exercise programme was associated with a greater improvement in
MWD.
The GDG placed less importance on changes in ABPI and pain free walking
distance (PFWD). This was because they did not expect ABPI to be greatly
affected by the different exercise programmes (as clinical benefit is more likely
to be due to improved muscle metabolism rather than blood flow) and because
PFWD was considered too subjective a measure of improvement to allow
meaningful comparisons between individuals and studies.
None of the studies reported data on cardiovascular events or limb loss,

although these outcomes were felt to be of less importance in IC than CLI.
Trade off between Based on their collective clinical experience, the GDG agreed that the risks
clinical benefits and associated with a supervised exercise programme are minimal, while the
harms benefits may include an increase in walking distance, quality of life, and
decreased risk of cardiovascular events.
Both exercise interventions require a time commitment from the patient.

Supervised exercise may also be associated with transportation costs. These
considerations should be discussed with each patient on an individual basis.
Economic considerations An original economic model was developed to combine best available evidence
on the efficacy of supervised compared to unsupervised exercise for the
treatment of IC. The primary outcome of the model was quality of life as
reported by the RCTs included in the clinical review. The cost of a supervised
exercise programme was calculated from an NHS and social services perspective.
Quality of life and costs associated with cardiovascular events were also
included, as was the decreased risk of mortality and cardiovascular events
experienced by people who are physically active. Compliance to exercise was a
key component of the model; two theoretical compliance scenarios were
included in the base case analysis.
Based on the results of the model, supervised exercise is a cost effective

treatment choice in over 75% of model simulations. Although supervised
exercise is more expensive than unsupervised, it is also more effective. If we
assume that supervised exercise leads to greater compliance over both the short
and long term, these programmes cost approximately £711 per QALY gained. If
we assume that there is no difference in exercise levels over the long term
compared to unsupervised exercise, then supervised exercise programmes cost
Final version
120
PAD


£1, 608 per QALY gained.
The model was robust to the majority of sensitivity analyses surrounding key
assumptions and data used to inform the model. However, the results were
sensitive to the assumption that those who continue to exercise maintain the
improvement in quality of life demonstrated at the end of one year. If the results
of the intervention are not sustained beyond the end of each trial, the
probability that supervised exercise is the most cost-effective option is much
more uncertain. The results were also dependant upon assumptions about
compliance to exercise over the short and long term.
Two published papers reported health economic analyses and were also
considered by the GDG.69,70 They had a very short time horizon and did not take
into account the expected beneficial effect of exercise on mortality and
cardiovascular morbidity. These studies presented conflicting results and were
not thought to be as relevant as the model developed for this guideline.
Supervised exercise programmes for PAD are not widely available and the GDG
recognised that this recommendation would likely have a significant
implementation cost. However, the GDG considered that the basic infrastructure
required may already exist within cardiology and respiratory services.
Quality of evidence The following quality issues were highlighted by the GDG, relating both to the
studies themselves and to difficulties in synthesising their results:
 The effect size tended to be small
 The included studies were rated moderate to very low quality by GRADE
criteria
 Trials differed in terms of types of exercise (upper versus lower body)
 With interventions of this type, it is possible that improvements could be
related to increased contact and attention from healthcare providers rather
than a true effect of exercise
 There is also a documented training effect of treadmill walking, which could
have lead to greater walking distances in the supervised exercise group82
 Limited data were available about withdrawals, but the GDG felt that the
reported rates were lower than they would have expected based on their
experience of real world behaviour, and that those who are prepared to enter
randomised trials involving supervised exercise were already pre-selected. In
clinical practice the overall proportion prepared to participate in and continue
with exercise programmes may be significantly lower
 The definitions of intermittent claudication varied between trials
 The long-term benefits of supervised exercise programmes are not clear in this
population.
Other considerations There is potential for confusion when considering exercise for IC. At one level,
Final version
121
PAD


simple advice to exercise should be seen as part of the lifestyle changes that the
patient should be advised about when the diagnosis of PAD is first made. It is
more formal exercise intervention which is being considered in the clinical
studies reviewed here.
The GDG discussed access issues and noted that at present patients tend to be
offered other interventional treatments ahead of supervised exercise. It was
noted that some patients may lack motivation to undertake a programme and
others may experience anxiety particularly if they have other co-morbidities
such as angina. It is therefore considered important to discuss the choices
available to the patient and recognise that some may prefer advice and
instruction about unsupervised exercise.
With patients undertaking exercise, there could be less need for secondary
interventions and patients may have better cardiovascular outcomes. However,
it was recognised that further research is required to assess the long term
benefits of supervised exercise programmes for IC. It was agreed by the GDG
that the benefits are likely to decrease with reduced compliance with exercise
following completion of a programme. The GDG therefore made a research
recommendation about monitoring long term effects.
Based on the available evidence the GDG concluded that, in the absence of
significant comorbidity where exercise would be contra-indicated, they could
recommend that people with intermittent claudication should be offered a
supervised exercise programme.
There was considerable variation in the nature of supervised exercise

programmes identified in the literature review – most commonly these were
carried out twice a week for 3 months and this was the frequency considered in
the cost effectiveness modelling. The GDG formed a consensus on the features
of a suitable exercise programme, which they felt should include the following
features:
 Although there is uncertainty about the best type of exercise for people with
PAD, most of the programmes described in the evidence review involved
walking to near maximal pain. The GDG agreed that patients should be
encouraged to walk to the point of maximal pain
 The frequency of the exercise programme should be approximately 2 hours
per week for 3 months
 The programme should be goal orientated and have a defined educational
component i.e. discussions about lifestyle change, benefits of exercise for PAD
patients and attitudes to the disease
 Supervised exercise programmes should be managed by an experienced and
suitably qualified healthcare professional
 The location of the exercise programme should be as close to the person’s
home as possible.
Final version
122
PAD



The GDG identified this recommendation as a key priority for implementation.
Supervised exercise programmes appear to be a cost-effective intervention yet
the GDG are aware that availability and access to such programmes varies
geographically, which results in inequality. In addition, supervised exercise may
be preferred by patients rather than undergoing revascularisation. Exercise can
have a positive impact on patient outcomes, such as walking distance.
2. What is the clinical and cost effectiveness of supervised exercise programmes compared with
unsupervised exercise for treating people with intermittent claudication, taking into account
the effects on long-term outcomes and continuing levels of exercise?
Research has shown that taking part in exercise and physical activity can lead to improvements in
symptoms in the short term for people with intermittent claudication. However, the benefits of
exercise are quickly lost if it is not frequent and regular. Supervised exercise programmes have been
shown to produce superior results when compared with advice to exercise (unsupervised) in the
short term, but they are more expensive, and there is a lack of robust evidence on long-term
effectiveness. A community-based randomised controlled trial is required to compare the long-term
clinical and cost effectiveness of a supervised exercise programme and unsupervised exercise. The
trial should enrol people with peripheral arterial disease-related claudication, but exclude those with
previous endovascular or surgical interventions. The primary outcome measure should be maximal
walking distance, with secondary outcome measures including quality of life, function, level of
uptake of exercise programmes and long-term engagement in physical activity.
9.3 Naftidrofuryl oxalate

What is the clinical and cost effectiveness of naftidrofuryl oxalate compared to exercise therapy,
angioplasty or stents for the treatment of intermittent claudication in adults with PAD?
NICE recently published a technology appraisal (TA 223) on “Cilostazol, naftidrofuryl oxalate,
pentoxifylline and inositol nicotinate for the treatment of intermittent claudication in people with
peripheral arterial disease”.53 The TA reviewed the evidence for the four named vasoactive drugs in
treating IC not controlled by best medical treatment, which was the TA term for what is referred to
as best medical treatment for PAD in section 9.1 above. Naftidofuryl oxalate was recommended as
the preferred treatment. The technology appraisal did not examine evidence comparing the
vasoactive drugs to exercise therapy, angioplasty or stents.
Final version
123
PAD
A literature search was conducted for RCTs that compared the effectiveness of naftidrofuryl oxalate
to exercise therapy, angioplasty or stents. No limits were set on time, sample size or duration of
follow-up. Indirect populations and emergency settings were excluded.
No relevant RCTs were identified.
NICE TA exploratory analysis
The GDG were presented with a summary of the methods and results of the exploratory economic
analysis conducted as part of the NICE TA comparing naftidrofuryl oxalate to angioplasty.53 The GDG
considered that this analysis was not based on comparative evidence; it represents an exploration of
several theoretically possible outcomes of angioplasty compared to naftidrofuryl oxalate. The costs
associated with angioplasty in the NICE TA analysis were based on the literature (procedural cost of
£92583) and were very different from those estimated according to NHS Reference Costs [procedural
cost of £3, 661 to £9, 367 (see Appendix L)]. The sensitivity analysis developed for the NICE TA was
also limited in that only angioplasty was included as an alternative to vasoactive drugs (those in the
‘no drug treatment’ arm of the exploratory model all underwent angioplasty). The GDG considered
exercise a more appropriate alternative for people with IC.
Original economic model
Methods
Without comparative clinical evidence it was not possible to evaluate the relative cost-effectiveness
of vasoactive drugs compared to exercise programmes in the base case analysis. Instead, the GDG
decided to incorporate the use of naftidrofuryl oxalate as a sensitivity analysis in the original
economic model developed to compare unsupervised to supervised exercise. Costs and
discontinuation rates associated with naftidrofuryl oxalate were obtained from the NICE TA and
incorporated into the current model. As in the NICE TA, it was assumed that naftidrofuryl oxalate
does not have any effect on the risk of mortality or cardiovascular events. Evidence of comparative
efficacy (as measured by quality of life was left blank and a threshold analysis was run to determine
the incremental gain in quality of life that would be necessary for naftidrofuryl to be considered cost-
effective compared to supervised and unsupervised exercise. For a full discussion of the methods and
results of this model please refer to Appendix K. Parameter inputs used to inform threshold analysis
of naftidrofuryl oxalate are reported in Table 41.
Table 41: Parameter inputs used to inform threshold analysis of naftidrofuryl oxalate
Probability Distribution
Parameter Point estimate Value range distribution parameters Source
3 month cost of £30.49 NA Fixed NA NHS Drug
naftidrofuryl Tariff84
oxalate
Discontinuation 11% NA Fixed NA Squires 201053
at 6 months
Discontinuation 68% NA Fixed NA Hiatt 2008 in
at 36 months Squires 201053
Relative effect 1 NA Fixed NA Squires 201053
on mortality(a)
Final version
124
PAD
Relative effect 1 NA Fixed NA Squires 201053

on stroke &
MI(a)
(a) As in the NICE TA 223, it was assumed that naftidrofuryl oxalate does not have any effect on mortality or CV risk.
Results
Compared to exercise, the threshold at which naftidrofuryl oxalate becomes the most cost effective
treatment strategy depends on the assumed level of compliance to each exercise programme. Where
there is a higher level of compliance to supervised exercise over both the short and long term,
naftidrofuryl oxalate becomes more cost effective when people achieve a gain of 0.029 QALYs per
cycle compared to unsupervised exercise. If compliance is equal over the long term, a QALY gain of
0.017 per cycle is needed (Table 42).
According to the utility calculations undertaken by the NICE TA53, people taking naftidrofuryl oxalate
had a mean utility of 0.5088 after 24 weeks of treatment. Compared to the baseline utility of 0.4873
for people not taking vasoactive drugs, this represents a utility gain of 0.021. According to this
estimate naftidrofuryl oxalate would be dominated by supervised exercise in both scenarios and is
therefore not likely to be cost effective compared to supervised exercise. However, it is difficult to
make comparisons due to differences in the methods used to estimate utility values.
Table 42: Threshold at which naftidrofuryl oxalate is more cost effective than supervised exercise
Additional utility with naftidrofuryl compared to unsupervised exercise
Scenario 1 0.029 x 4
Scenario 2 0.017 x 4
9.3.2.1 Clinical
No clinical evidence was identified for this question.
9.3.2.2 Economic
Based on the results of a threshold analysis undertaken as part of the original cost effectiveness
model developed for this guideline, naftidrofuryl oxalate is unlikely to be more cost effective than
supervised exercise for the treatment of IC under the base case assumptions of the model. However,
the GDG did not identify any clinical evidence to support a strong conclusion in this area.
Naftidrofuryl oxalate may also be considered an option when people do not wish to undertake an
exercise programme; in this case, the question is not one of choice between different treatments and
the scenario represented by the economic model is not relevant.
18.Consider naftidrofuryl oxalate for treating people with

intermittent claudication, starting with the least costly
preparation, only when:
 supervised exercise has not led to satisfactory improvement
and
 the person prefers not to be referred for consideration of
Recommendations angioplasty or bypass surgery.
Final version
125
PAD
Review progress after 3-6 months and discontinue

naftidrofuryl oxalate if there has been no symptomatic
benefit.
Relative value of different In line with the NICE TA 223 recommendation, the GDG focussed on the
outcomes clinical and cost effectiveness for naftidrofuryl and decided to compare it to
exercise therapy, angioplasty or stents for the treatment IC in adults. No
evidence could be identified to allow comparisons of clinical efficacy to be
made.
Trade off between clinical Naftidrofuryl oxalate is contraindicated in people with a history of
benefits and harms hyperoxaluria or recurrent calcium-containing stones. The summary of
product characteristics should be consulted for a full list of side effects and
contraindications.
The GDG were of the opinion that, because it may be more convenient to
prescribe a drug than to refer for further assessment for an invasive
intervention, there is a risk that naftidrofuryl may sometimes be used when
other treatment modalities (e.g. revascularisation) are likely to be superior
in terms of outcomes.
Economic considerations The GDG considered the cost of naftidrofuryl oxalate discontinuation rates
as reported by the NICE TA, and the gain quality of life needed to make it a
more cost effective strategy than supervised exercise according to the
results of the economic model. They noted that naftidrofuryl is unlikely to
be cost-effective given that the gain in quality of life needed for
naftidrofuryl to be a cost effective option is greater than that reported in
the NICE TA.
The GDG also considered that there may be situations in which best
medical treatment has been unsuccessful and people do not wish to
undertake an exercise programme or interventional treatment. In these
situations, the GDG considered that the use of naftidrofuryl oxalate would
be cost-effective.
Quality of evidence Whilst no evidence was identified, the GDG observed that the effect sizes
reported in NICE TA 223 for walking distance were considerably lower than
the minimally important differences the GDG had identified for this
guideline.
Other considerations There was no evidence to identify those sub-groups of people with IC who
may benefit from naftidrofuryl and where in the care pathway this should
be offered. The GDG agreed by consensus that naftidrofuryl should not be
given as first line treatment for IC.
The GDG discussed at length the importance of referral to secondary care

when a person’s symptoms have not resolved or have worsened, and their
quality of life is affected. It is important that people with IC are offered the
most appropriate treatment option within the care pathway. In addition,
the use of naftidrofuryl should be reviewed to ensure that patients do not
remain on the therapy when there is no beneficial effect.
The discussion around treatment options must take account of patient

choice. It must be recognised that some patients may not wish to undergo
Final version
126
PAD
referral or other treatments and therefore, wish to trial naftidrofuryl.
9.4 Comparisons between treatment options: exercise, best medical

treatment, angioplasty and bypass surgery
What is the clinical and cost effectiveness of endovascular or surgical techniques compared to or in
combination with exercise or best medical treatment for the treatment of people with intermittent
claudication?
A literature search was conducted for RCTs that compared the effectiveness of endovascular or
surgical techniques to or in combination with exercise or best medical treatment. No time limit was
placed on the literature search, and there were no limitations on sample size. Indirect populations
and emergency settings were excluded.
Twelve studies of eight RCTs85-87 88-96 were included in the review. The trials did not report outcome
data for people with diabetes.
The interventions evaluated in these trials could be divided into five pair wise comparisons:
1. Best medical treatment compared to best medical treatment with angioplasty (see section 9.4.2)
2. Supervised exercise with best medical treatment compared to supervised exercise, best medical
treatment plus angioplasty (see section 9.4.3)
3. Best medical treatment with angioplasty compared to best medical treatment with angioplasty
and supervised exercise (see section 9.4.4)
4. Angioplasty compared to supervised exercise (see section 9.4.5)
5. Bypass surgery compared to supervised exercise (see section 9.4.6)
9.4.2 Best medical treatment compared to best medical treatment with angioplasty
Clinical evidence
For this comparison, four studies of two RCTs were included that compared best medical treatment
alone to best medical treatment with angioplasty.85-88 One Cochrane review was identified97 which
considered angioplasty compared to non surgical management for intermittent claudication. The
Cochrane review was not included or updated as it did not meet the review question protocol
defined by the GDG, which also included the comparison of best medical treatment to surgery.
However it was used as a source to ensure that studies identified in the Cochrane review which
matched the current review protocol had been considered for inclusion.
The study characteristics are reported in Table 43. The quality and results of included studies are
reported in the clinical evidence profiles (
Table 44 and Table 45). The forest plots for each clinical outcome are reported in Appendix J.
Final version
127
PAD
Table 43: Study characteristics: Best medical treatment compared to best medical treatment with angioplasty for intermittent claudication
Study Disease location BMT BMT with Angioplasty
The Oslo Balloon Combined lesions in aorto-iliac and BMT BMT
Angioplasty versus femoro-popliteal arteries  Smoking cessation  Smoking cessation
Conservative Treatment
 Home-based exercise advice  Home-based exercise advice
Study (OBACT)85,86
 Nutritional advice and individualised optimal  Nutritional advice and individualised optimal
Mediterranean-type diet Mediterranean-type diet
 Aspirin 160 mg daily or clopidogrel 75 mg daily  Aspirin 160 mg daily or clopidogrel 75 mg daily
for peptic ulcer history for peptic ulcer history
 Statins for untreated hypercholesterolaemia.  Statins for untreated hypercholesterolaemia
 High blood pressure treatment  High blood pressure treatment
 Angioplasty
 Iliac occlusions treated with primary stenting;
iliac stenoses were selectively stented
 Stents were not used infra-inguinally
Whyman, 1996; Whyman, Femoro-popliteal arteries BMT BMT
199787,88  Low dose aspirin  Low dose aspirin
 Smoking advice  Smoking advice
 Exercise advice  Exercise advice
Angioplasty
 Angioplasty by balloon dilation
 Arterial stenting not routinely used
Final version
128
PAD
Table 44: Clinical evidence profile: Best medical treatment compared to best medical treatment with angioplasty for intermittent claudication due to
femoro-popliteal and aorto-iliac disease

Quality
No of Other BMT + Relative
Design Risk of bias Inconsistency Indirectness Imprecision BMT Absolute
studies considerations angioplasty (95% CI)
Maximum walking distance at 3 months
186 RCT Serious(a) No serious No serious Serious(b) None 28 28 - MD 123.9 higher LOW
inconsistency indirectness (16.69 to 231.11
higher)
Maximum walking distance at 1 year
higher)
Maximum walking distance at 2 years
inconsistency indirectness (122.12 to
317.28 higher)
Pain free walking distance at 3 months
higher)
Pain free walking distance at 1 year
inconsistency indirectness (172.61 to
378.39 higher)
Pain free walking distance at 2 years
Final version
129
PAD

higher)
Major complications at 1 year
186 Observational No serious No serious No serious No serious None 0 out of 28 people in the best medical treatment LOW
studies risk of bias(c) inconsistency indirectness imprecision plus angioplasty group had major complications
Re-intervention at 1 year
186 Observational No serious No serious No serious No serious None 0 out of 28 people in the best medical treatment LOW
studies risk of bias(c) inconsistency indirectness imprecision plus angioplasty group had re-intervention
ABPI at 3 months
186 RCT Serious(a) No serious No serious No serious None 28 28 - MD 0.24 higher MODERA
inconsistency indirectness imprecision (0.23 to 0.25 TE
higher)
ABPI at 1 year
higher)
ABPI at 2 years
higher)
(b) 95% CI crosses one MID.
(c) Data taken from a RCT, non-comparative outcome.
Table 45: Clinical evidence profile: Best medical treatment compared to best medical treatment with angioplasty for intermittent claudication due to
femoro-popliteal disease

Quality
No of Other BMT + Relative
Design Risk of bias Inconsistency Indirectness Imprecision BMT Absolute
studies considerations angioplasty (95% CI)
Final version
130
PAD
Mortality at 2 years
187 RCT Serious(a) No serious No serious Very None 0/29 2/30 RR 0.21 (0.01 53 fewer per VERY LOW
inconsistency indirectness serious(b) (0%) (6.7%) to 4.13) 1000 (from 66
fewer to 209
more)
Major complications at 6 months
188 Observational No serious No serious No serious No serious None 0 out of 29 people in the BMT plus angioplasty LOW
studies risk of bias(c) inconsistency indirectness imprecision group had major complications
Re-intervention at 6 months
studies risk of bias(c) inconsistency indirectness imprecision group had re-intervention
Re-intervention at 2 years
studies risk of bias(c) inconsistency indirectness imprecision group had re-intervention
ABPI at 6 months
188 RCT Serious(a) No serious No serious Serious(d) None 29 30 - MD 0.14 higher LOW
higher)
ABPI at 2 years
187 RCT Serious(a) No serious No serious No serious None 29 30 - MD 0.06 higher MODERAT
inconsistency indirectness imprecision (0.04 to 0.08 E
higher)
(a) Unclear blinding.
(b) 95% CI crosses both MIDs.
(c) Data from a RCT, non-comparative outcome.
(d) 95% CI crosses one MID.
Final version
131
PAD
9.4.3 Supervised exercise with best medical treatment compared to supervised exercise, best
medical treatment plus angioplasty
Clinical evidence
Three RCTs89,90,95 were found which addressed the question and were included in the review.
The study characteristics are reported in Table 46.The quality and results of included studies are
reported in Table 47 and Table 48. The mapped EQ-5D are reported in Table 49. The forest plots for
each clinical outcome are reported in Appendix J.
Final version
132
PAD
Table 46: Study characteristics: Best medical treatment with supervised exercise compared to best medical treatment with angioplasty and supervised
exercise
Intervention 1 Intervention 2
Study Disease location BMT + Supervised exercise BMT + Angioplasty + Supervised exercise
Greenhalgh, Aorto-iliac and BMT BMT
200889 femoro-popliteal  Aspirin 75 mg or clopidogrel if intolerant to aspirin  Aspirin 75 mg or clopidogrel if intolerant to aspirin
arteries  Blood pressure, total and high-density lipoprotein serum  Blood pressure, total and high-density lipoprotein serum
cholesterol and serum glucose were assessed and drug cholesterol and serum glucose were assessed and drug
therapy commenced where necessary therapy commenced where necessary
 Smoking cessation advice and support  Smoking cessation advice and support
Supervised exercise Angioplasty

 ≥1 session per week for 6 months  Balloon angioplasty with selective stent placement (number of
 Each session consisted of 30 minutes continuous exercise stent placed = x/y)
to a maximum pain threshold using a walking circuit
interspersed with lower-limb training stations. Supervised exercise
 ≥1 session per week for 6 months
 Each session consisted of 30 minutes continuous exercise to a
maximum pain threshold using a walking circuit interspersed
with lower-limb training stations.
Mazari, 201090 Femoro-popliteal BMT BMT
arteries  Antiplatelet therapy (aspirin and/or clopidogrel).  Antiplatelet therapy (aspirin and/or clopidogrel).
 Risk factor modification (target orientated management of  Risk factor modification (target orientated management of
hypertension, hypercholesterolemia and diabetes). hypertension, hypercholesterolemia and diabetes).
 Advice leaflet regarding exercise.  Advice leaflet regarding exercise.

 3 sessions per week for 12 weeks  Balloon angioplasty with selective stent placement (number of
 Classes consisted of a circuit of exercise stations. stents placed = 0/y)
Supervised exercise
Final version
133
PAD
 3 sessions per week for 12 weeks (beginning one week

following angioplasty).
 Classes consisted of a circuit of exercise stations.
Mazari, 201295 Femoropopliteal BMT BMT
arteries  Antiplatelet therapy (aspirin and/or clopidogrel)  Antiplatelet therapy (aspirin and/or clopidogrel)
 Risk factor management  Risk factor management
 Advice leaflets of physical activity and exercise  Advice leaflets of physical activity and exercise

 3 times a week for 12 weeks under supervision of  Angioplasty was performed by a consultant vascular
physiotherapist or doctor. radiologist in accordance with the units standard procedure
 Closed circuit training on six stations each for 2 minutes
with 2 minutes brisk walking between each station. Supervised exercise
 Patients completed one full circuit for the first 6 weeks  3 times a week for 12 weeks under supervision of
followed by an additional increment of 1 station per week physiotherapist or doctor.
for the next 6 weeks ending with completing 2 full circuits.
 Closed circuit training on six stations each for 2 minutes with 2
minutes brisk walking between each station.
 Patients completed one full circuit for the first 6 weeks
followed by an additional increment of 1 station per week for
the next 6 weeks ending with completing 2 full circuits.
Table 47: Clinical evidence profile: Supervised exercise with best medical treatment compared to supervised exercise with best medical treatment and
angioplasty for intermittent claudication due to aorto-iliac disease

Quality
Other BMT/SE/angio Relative
No of studies Design Risk of bias Inconsistency Indirectness Imprecision BMT/SE Absolute
considerations plasty (95% CI)
Final version
134
PAD

Maximum walking distance (no sd) at 2 years 
189 RCT Serious(a) No serious No serious Serious(b) None 11 12 - not LOW
inconsistency indirectness pooled
Pain free walking distance (% attaining 200 m without pain) at 2 years
189 RCT Serious(a) No serious No serious Very None 7/11 3/12 RR 2.55 387 more VERY LOW
inconsistency indirectness serious(c) (63.6%) (25%) (0.87 to per 1000
7.47) (from 32
fewer to
1000
more)
Complications following procedure
189 Observational No serious No serious No serious No serious None 4 out of 19 people in the BMT with supervised LOW
studies risk of bias(d) inconsistency indirectness imprecision exercise and angioplasty group has complications
Compliance with exercise programme
inconsistency indirectness serious(c) (52.6%) (46.7%) (0.57 to per 1000
2.25) (from 201
fewer to
583
more)
(d) Data taken from a RCT, non-comparative outcome.
Final version
135
PAD
Table 48: Clinical evidence profile: Supervised exercise with best medical treatment compared to supervised exercise with best medical treatment and
angioplasty and for intermittent claudication due to femoro-popliteal disease

Quality
BMT/SE / Relative
No of studies Design Risk of bias Inconsistency Indirectness Imprecision Other considerations BMT/SE Absolute
angioplasty (95% CI)
289 90 RCT Serious(a) No serious No serious Serious(b) None 95 94 See Table 49 LOW
289 90 RCT Serious(a) No serious No serious Serious(b) None 95 94 See Table 49 LOW
Maximum walking distance (no sd) at 2 years
189 RCT Serious(a) No serious No serious Serious(b) None 37 34 - not pooled LOW
Pain free walking distance (% patients attaining 200 m without pain) at 2 years
189 RCT Serious(a) No serious No serious No serious None 23/37 7/34 RR 3.02 416 more MODERATE
inconsistency indirectness imprecision (62.2%) (20.6%) (1.49 to per 1000
6.12) (from 101
more to
1000 more)
Complications following procedure
studies risk of bias(c) inconsistency indirectness imprecision exercise and angioplasty group has
complications
Complications at 3 months
studies risk of bias(c) inconsistency indirectness imprecision exercise and angioplasty group has
complications
Final version
136
PAD
195 Observational No serious No serious No serious No serious None 0 out of 58 people in the BMT with supervised LOW
(c)
studies risk of bias inconsistency indirectness imprecision exercise and angioplasty group had re-
interventions
Compliance with exercise programme
inconsistency indirectness serious(d) (62.5%) (60%) (0.75 to per 1000
1.44) (from 150
fewer to
264 more)
Withdrawal rate at 3 months
190 RCT Serious(a) No serious No serious Very None 10/58 8/60 RR 1.29 39 more per VERY LOW
inconsistency indirectness serious(d) (17.2%) (13.3%) (0.55 to 1000 (from
3.05) 60 fewer to
273 more)
(a) Unclear allocation concealment.
(b) No information on the variability was given in the study, therefore the calculation of the standard deviation was not possible and the mean difference and CI were not estimable.
(c) Data taken from an RCT, non-comparative outcome.
(d) 95% CI crosses both MIDs.
Table 49: SF-36 individual domain results and mapped EQ-5D values – Supervised exercise compared to angioplasty with supervised exercise
Supervised exercise Angioplasty + supervised exercise
Greenhalgh 200898 – Femoro-popliteal arteries – Mean (SD)
PF 35.8 (8.0) NR 37.9 (8.9) NR 37.2 (9.3) 37.2 (8.0) NR 40.0 (9.9) NR 40.2 (9.1)
RP 41.2 (10.9) NR 41.5 (10.1) NR 41.1 (11.4) 42.3 (10.3) NR 42.8 (11.1) NR 43.6 (10.1)
BP 42.7 (9.4) NR 41.8 (9.1) NR 43.1 (8.7) 42.9 (8.7) NR 44.3 (10.4) NR 44.3 (9.9)
GH 45.9 (9.4) NR 44.1 (9.8) NR 44.2 (8.4) 43.8 (8.8) NR 41.9 (10.0) NR 42.6 (9.4)
V 45.8 (9.7) NR 44.3 (12.0) NR 43.6 (11.1) 46.6 (11.2) NR 47.7 (11.8) NR 47.2 (12.1)
SF 44.4 (11.9) NR 43.2 (12.0) NR 44.7 (12.0) 45.0 (10.6) NR 45.3 (10.3) NR 45.1 (10.3)
RE 42.1 (13.6) NR 41.2 (13.7) NR 40.4 (15.7) 44.8 (13.2) NR 46.5 (12.3) NR 46.2 (9.9)
MH 46.7 (12.4) NR 46.6 (12.2) NR 46.4 (12.5) 50.3 (10.4) NR 50.4 (10.3) NR 49.9 (10.4)
EQ-5D(a) 0.45 (0.01) NA 0.45 (0.01) NA 0.46 (0.01) 0.47 (0.01) NA 0.48 (0.01) NA 0.48 (0.01)
Final version
137
PAD
Greenhalgh 200898 – Aorto-iliac arteries – Mean (SD)

PF 35.4 (8.3) NR 36.2 (6.2) NR 34.7 (9.2) 33.2 (8.5) NR 43.6 (10.1) NR 44.4 (10.5)
RP 40.0 (12.0) NR 41.0 (10.5) NR 38.4 (11.4) 41.1 (8.8) NR 47.1 (11.7) NR 46.6 (10.1)
BP 40.1 (7.4) NR 37.4 (5.1) NR 38.8 (10.1) 40.7 (9.6) NR 48.0 (10.7) NR 49.5 (11.5)
GH 39.2 (7.4) NR 36.6 (9.8) NR 36.1 (7.4) 42.2 (13.1) NR 45.5 (10.5) NR 45.6 (9.6)
V 43.8 (8.5) NR 43.2 (8.2) NR 40.6 (11.1) 42.4 (9.6) NR 47.2 (10.9) NR 45.0 (11.3)
SF 39.4 (13.6) NR 38.0 (10.8) NR 39.2 (11.8) 38.2 (10.2) NR 50.0 (9.7) NR 45.9 (9.9)
RE 40.3 (14.6) NR 43.9 (11.6) NR 41.8 (13.1) 39.9 (14.4) NR 44.0 (12.7) NR 43.4 (11.8)
MH 43.4 (9.7) NR 42.6 (9.9) NR 42.6 (11.7) 42.7 (10.1) NR 48.1 (8.9) NR 44.6 (12.7)
(a)
EQ-5D 0.45 (0.01) NA 0.42 (0.01) NA 0.42 (0.02) 0.42 (0.01) NA 0.50 (0.01) NA 0.50 (0.02)
90
Mazari 2010 – Median (range)
PF 30 (35) 55 (48) NA NA NA 40 (30) 60 (43) NA NA NA
RP 20 (30) 25 (100) NA NA NA 25 (75) 75 (75) NA NA NA
BP 41 (42) 55 (43) NA NA NA 41 (31) 62 (32) NA NA NA
GH 55 (37) 60 (30) NA NA NA 55 (25) 62 (20) NA NA NA
V 45 (20) 50 (35) NA NA NA 45 (21) 55 (30) NA NA NA
SF 62 (50) 75 (50) NA NA NA 62 (35) 75 (50) NA NA NA
RE 33 (100) 83 (100) NA NA NA 66 (67) 83 (67) NA NA NA
MH 68 (28) 72 (30) NA NA NA 70 (25) 82 (27) NA NA NA
(b)
EQ-5D NE NE NA NA NA NE NE NA NA NA
(b) Only the range was reported; probabilistic mapped values not estimable.
deviation; NA = not applicable; NR = not reported; NE = not estimable.
Final version
138
PAD
9.4.4 Best medical treatment with angioplasty compared to best medical treatment with
angioplasty and supervised exercise for intermittent claudication
Clinical evidence
Two RCTs95,96 were found which addressed the question and were included in the review.
reported in Table 51. The reasons for withdrawal are reported in Table 52. The forest plots for each
clinical outcome are reported in Appendix J.
Final version
139
PAD
Table 50: Study characteristics: Best medical treatment with angioplasty compared to best medical treatment with angioplasty and supervised exercise
Intervention 1 Intervention 2
Study Disease location BMT + Angioplasty BMT + Angioplasty + Supervised exercise
Kruidenier, 201196 Aorto-iliac BMT BMT
arteries  Cardiovascular risk factor modification (inc.  Cardiovascular risk factor modification (inc. antiplatelet
antiplatelet inhibitor and a statin and treatment for inhibitor and a statin and treatment for hypertension
hypertension and/or diabetes as required and/or diabetes as required
 Advice to quit smoking if required and offer of a  Advice to quit smoking if required and offer of a smoking
smoking cessation programme cessation programme
 Lifestyle changes (e.g. physical activity, weight, diet)  Lifestyle changes (e.g. physical activity, weight, diet)
Angioplasty Angioplasty
 Performed by experienced interventional radiologist  Performed by experienced interventional radiologist
 Iliac angioplasty with selective stent placement for  Iliac angioplasty with selective stent placement for iliac
iliac stenosis; angioplasty with primary stent stenosis; angioplasty with primary stent placement for
placement for superficial femoral artery stenosis or superficial femoral artery stenosis or recanalisation with
recanalisation with primary stent placement for iliac primary stent placement for iliac and femoral occlusions
and femoral occlusions
Supervised exercise
 Began with 3 weeks of rest following angioplasty
 Community based setting, supervised by a trained
physiotherapist in proximity to their homes
 Generally started with a frequency of 2-3 sessions of 30
minutes a week, frequency reduced according to patients
progress
 Patients encouraged to walk on a daily basis in addition to
physiotherapy sessions.
Mazari, 201295 Femoropopliteal BMT BMT
arteries  Antiplatelet therapy (aspirin and/or clopidogrel)  Antiplatelet therapy (aspirin and/or clopidogrel)
 Advice leaflets of physical activity and exercise  Advice leaflets of physical activity and exercise
Final version
140
PAD
Angioplasty Angioplasty
 Angioplasty was performed by a consultant vascular  Angioplasty was performed by a consultant vascular
radiologist in accordance with the units standard radiologist in accordance with the units standard
procedure. procedure.
Supervised exercise
 3 times a week for 12 weeks under supervision of
physiotherapist or doctor.
 Closed circuit training on six stations each for 2 minutes
with 2 minutes brisk walking between each station.
 Patients completed one full circuit for the first 6 weeks
followed by an additional increment of 1 station per week
for the next 6 weeks ending with completing 2 full circuits.
Table 51: Clinical evidence profile: Best medical treatment with angioplasty compared to best medical treatment with angioplasty and supervised
exercise for intermittent claudication due to aorto-iliac disease

Quality
Risk of Other BMT/SE/angio BMT/angio Relative
No of studies Design Inconsistency Indirectness Imprecision Absolute
bias considerations plasty plasty (95% CI)
196 RCT Serious(a) No serious No serious Serious(b) None 33 29 See Table 53 and Table LOW
inconsistency indirectness 54
196 RCT Serious(c) No serious No serious Serious(d) None 32 29 - MD 191.1 LOW
inconsistency indirectness higher
(35.1 lower
to 417.3
higher)
Final version
141
PAD

(68.43 to
474.17
higher)
(15.77 to
455.43
higher)
(106.19 to
484.21
higher)
Major adverse events at 6 months
196 RCT Serious(c) No serious No serious Very None 3/35 0/35 RR 7 (0.37 to - VERY LOW
inconsistency indirectness serious(e) (8.6%) (0%) 130.69)
195 RCT Very No serious No serious Serious(d) None 0/58 9/60 RR 0.05 (0 to 142 fewer VERY LOW
serious(f) inconsistency indirectness (0%) (15%) 0.91) per 1000
(from 13
fewer to
150 fewer)
196 RCT Serious(c) No serious No serious Very None 7/35 1/35 RR 7 (0.91 to 171 more VERY LOW
(20%)
inconsistency indirectness serious(e) (2.9%) 53.95) per 1000
(from 3
fewer to
1000 more)
(a) Unclear allocation concealment and blinding, baseline characteristic differences.
Final version
142
PAD
(c) Unclear allocation concealment and blinding.
(e) 95% CI crosses both MIDs.
(f) Unclear methodology.
Table 52: Study characteristics: Reason for withdrawal from treatment

Study BMT/SE / angioplasty (n) BMT/SE (n)
6 months
Kruidenier, 201196 7/35: 1/35:
Not motivated (1); too busy (2); insurance related (1); orthopaedic co- Requested supervised exercise (1)
morbidity (1); unknown (2)
Table 53: EQ-5D: Angioplasty compared to angioplasty with supervised exercise

96
Kruidenier 2011 – Mean (SD)
0.63 ± 0.19 NR 0.77 ± 0.20 NR NR 0.55 ± 0.27 NR 0.79 ± 0.19 NR NR
(a) NR = not reported
Table 54: SF-36 individual domain results and mapped EQ-5D values – Angioplasty compared to angioplasty with supervised exercise
Angioplasty Angioplasty + supervised exercise
96
Kruidenier 2011 – Mean (SD)
PF 41.6 (17.5) NR 72.2 (18.0) NR NR 43.6 (19.4) NR 72.7 (22.3) NR NR
RP 39.1 (43.5) NR 71.6 (37.0) NR NR 33.3 (39.9) NR 56.3 (40.2) NR NR
BP 43.0 (16.4) NR 64.7 (26.0) NR NR 41.4 (19.9) NR 70.0 (22.8) NR NR
GH 52.2 (13.2) NR 53.7 (12.5) NR NR 51.5 (11.3) NR 56.9 (12.6) NR NR
V 51.2 (18.8) NR 57.1 (20.0) NR NR 57.4 (20.2) NR 67.3 (17.7) NR NR
SF 69.1 (28.0) NR 77.2 (31.0) NR NR 64.0 (22.8) NR 80.7 (19.8) NR NR
RE 83.9 (35.4) NR 77.0 (40.9) NR NR 80.8 (38.2) NR 82.3 (35.9) NR NR
MH 72.8 (18.3) NR 68.0 (19.5) NR NR 72.2 (20.8) NR 79.4 (17.5) NR NR
Final version
143
PAD
Angioplasty Angioplasty + supervised exercise

(a)
EQ-5D 0.58 (0.02) NA 0.74 (0.02) NA NA 0.57 (0.01) NA 0.79 (0.02) NA NA
deviation; NA = not applicable; NR = not reported; NE = not estimable.
Final version
144
PAD
9.4.5 Angioplasty compared to supervised exercise
Clinical evidence
Five RCTs91-93,99 95 were found which addressed the question and were included in the review.
reported in Table 57, Table 58 and Table 59. The mapped EQ-5D values are reported in Table 60.The
forest plots for each clinical outcome are reported in Appendix J.
Spronk 200991 reported values for the four physical domains of the SF-36 (physical functioning,
physical role, bodily pain and general health). The authors were contacted to request the remaining
domains and they replied that these domains were not collected. Baseline and mean score
improvement in EQ-5D were reported in the cost-effectiveness paper based on this randomised
control trial.100 By assigning a distribution to each reported EQ-5D value, the mean score
improvement at 6 and 12 months was added to the baseline value to calculate mean quality of life at
each time point. This simulation was run 20,000 times and the results are reported in Table 55.
Table 55: Simulated mean EQ-5D values from Spronk 2008 based on mean score improvement
Mean Mean score Simulated Mean Mean score Simulated
value (SE) improvement mean value value (SE) improvement mean value
(SE) (SE) (SE) (SE)
Baseline 0.69 (0.02) 0.66 (0.02)
6 months 0.09 (0.03) 0.780 (0.034) 0.16 (0.02) 0.820 (0.31)
12 months 0.07 (0.02) 0.076 (0.032) 0.11 (0.03) 0.770 (0.036)
Final version
145
PAD
Table 56: Study characteristics: Angioplasty compared to supervised exercise for intermittent claudication
Study Disease location Supervised exercise Angioplasty
Spronk, 200991 Aorto-iliac arteries BMT BMT
 Atherosclerotic risk factor treatment that included  Atherosclerotic risk factor treatment that included
hypertension, serum glucose, cholesterol, lipid profile, hypertension, serum glucose, cholesterol, lipid profile,
and homocysteinemia (in patients <50 years of age) and homocysteinemia (in patients <50 years of age)
management and, all patients were prescribed aspirin management, and all patients were prescribed aspirin
therapy (100 mg/d). therapy (100 mg/d).
 All smokers were strongly and repeatedly advised to  All smokers were strongly and repeatedly advised to quit
quit smoking, and were offered a smoking-cessation smoking, and were offered a smoking-cessation
programme. programme.
 Risk factor management continued during follow-up  Risk factor management continued during follow-up

Twice weekly 30 minute sessions on treadmills for 24 Balloon angioplasty. Self-expanding stent placed if
weeks angioplasty was considered unsuccessful.
Perkins, 1996; Aorto-iliac and femoro- BMT BMT
Creasy, 199092,93 popliteal arteries No details No details

Supervised exercise programme of twice weekly 30 Angioplasty using conventional guide-wire and catheter
minute sessions for 6 months. Each session consisted of technique
dynamic leg exercises
Mazari, 201099 Femoro-popliteal BMT BMT
arteries All patients received: All patients received:
 Antiplatelet therapy (aspirin and/or clopidogrel)  Antiplatelet therapy (aspirin and/or clopidogrel)
 Smoking cessation advice and support (including  Smoking cessation advice and support (including nicotine
nicotine replacement therapy and NHS smoking replacement therapy and NHS smoking cessation
cessation programme) programme)
 Risk factor modification (target orientated  Risk factor modification (target orientated management
management of hypertension, hypercholesterolemia of hypertension, hypercholesterolemia and diabetes).
and diabetes).  Advice leaflet regarding exercise.
Final version
146
PAD
 Advice leaflet regarding exercise.

Angioplasty
Supervised exercise Balloon angioplasty. Primary stenting or adjunctive
Supervised exercise sessions 3 times a week for 12 procedures were not performed in any case
weeks, classes involved a circuit of exercise stations
Mazari, 201295 Femoropopliteal arteries BMT BMT
 Antiplatelet therapy (aspirin and/or clopidogrel)  Antiplatelet therapy (aspirin and/or clopidogrel)
 Advice leaflets of physical activity and exercise.  Advice leaflets of physical activity and exercise.

 3 times a week for 12 weeks under supervision of  Angioplasty was performed by a consultant vascular
physiotherapist or doctor radiologist in accordance with the units standard
 Closed circuit training on six stations each for 2 procedure
minutes with 2 minutes brisk walking between each
station
 Patients completed one full circuit for the first 6
weeks followed by an additional increment of 1
station per week for the next 6 weeks.
Table 57: Clinical evidence profile: Angioplasty compared to supervised exercise for intermittent claudication due to aorto-iliac disease

Quality
No of Risk of Other Supervised Relative
Design Inconsistency Indirectness Imprecision Angioplasty Absolute
studies bias considerations Exercise (95% CI)
Quality of life at 6 months
Final version
147
PAD
Maximum walking distance from baseline at 6 months
191 RCT Serious(a) No serious No serious No serious None 75 75 - MD 383 MODERATE
inconsistency indirectness imprecision lower (537.62
to 228.38
lower)
Maximum walking distance from baseline at 1 year
191 RCT Serious(a) No serious No serious Serious(c) None 75 75 - MD 208 LOW
inconsistency indirectness lower (359.79
to 56.21
lower)
Pain free walking distance from baseline at 6 months
to 48.38
lower)
Pain free walking distance from baseline at 1 year
lower to
31.66 higher)
Number of patients who doubled their maximum walking distance at 3 months
193 RCT Very No serious No serious Very serious(e) None 4/16 7/15 RR 0.54 (0.2 215 fewer per VERY LOW
serious(d) inconsistency indirectness (25%) (46.7%) to 1.47) 1000 (from
373 fewer to
219 more)
Final version
148
PAD
193 RCT Very No serious No serious Serious(c) None 5/14 9/12 RR 0.48 (0.22 390 fewer per VERY LOW
(c)
serious inconsistency indirectness (35.7%) (75%) to 1.03) 1000 (from
585 fewer to
22 more)
193 RCT Very No serious No serious Very serious(e) None 4/11 9/12 RR 0.48 (0.21 390 fewer per VERY LOW
serious(d) inconsistency indirectness (36.4%) (75%) to 1.13) 1000 (from
593 fewer to
97 more)
Number of patients who doubled their maximum walking distance at 1 year
193 RCT Very No serious No serious Very serious(e) None 2/5 6/7 RR 0.47 (0.15 454 fewer VERY LOW
serious(d) inconsistency indirectness (40%) (85.7%) to 1.42) per 1000
(from 729
fewer to 360
more)
Complications at 1 year
291,93 Observat No No serious No serious No serious None 11 out of 95 people in the angioplasty group had LOW
ional serious inconsistency indirectness imprecision complications
studies risk of
bias(f)
Re-interventions at 6 months
191 Observat No No serious No serious No serious None 5 out of 75 people in the angioplasty group had re- LOW
ional serious inconsistency indirectness imprecision intervention
studies risk of
bias(f)
Re-interventions at 1 year
291,93 Observa No No serious No serious No serious None 8 out of 95 people in the angioplasty group had re- LOW
tional serious inconsistency indirectness imprecision intervention
studies risk of
bias(f)
Number of good attenders to exercise (on average > 1 session per week) at 6 months
193 Observa No No serious No serious No serious None 8 out of 16 people in the exercise group were good LOW
Final version
149
PAD
tional serious inconsistency indirectness imprecision attenders

studies risk of
bias(f)
Number of poor attenders to exercise (on average < 1 session per week) at 6 months
193 Observa No No serious No serious No serious None 8 out of 16 people in the exercise group were poor LOW
tional serious inconsistency indirectness imprecision attenders
studies risk of
bias(f)
199 RCT Serious(a) No serious No serious Very serious(e) None 3/60 8/60 RR 0.38 (0.1 83 fewer per VERY LOW
inconsistency indirectness (5%) (13.3%) to 1.35) 1000 (from
120 fewer to
47 more)
ABPI at rest from baseline at 6 months
191 RCT Serious(a) No serious No serious Serious(c) None 75 75 - MD 0.11 LOW
inconsistency indirectness higher (0.06 to
0.16 higher)
ABPI at rest from baseline at 1 year
191 RCT Serious(a) No serious No serious Serious(e) None 75 75 - MD 0.12 LOW
0.17 higher)
ABPI after exercise from baseline at 6 months
191 RCT Serious(a) No serious No serious Serious(e) None 75 75 - MD 0.13 LOW
0.2 higher)
ABPI after exercise from baseline at 1 year
191 RCT Serious(a) No serious No serious No serious None 75 75 - MD 0.07 MODERATE
inconsistency indirectness imprecision higher (0.02 to
0.12 higher)
(c) 95% CI crosses one MID.
(d) Unclear methodology.
Final version
150
PAD
(e) 95% crosses both MIDs.

(f) Data taken from an RCT, non-comparative outcome.
Table 58: Clinical evidence profile: Angioplasty compared to supervised exercise for intermittent claudication due to aorto-iliac and femoro-popliteal
disease

Quality
No of Other Supervised Relative
Design Risk of bias Inconsistency Indirectness Imprecision Angioplasty Absolute
studies considerations Exercise (95% CI)
192 Observational No serious No serious No serious No serious None 3 out of 30 people in the angioplasty group had LOW
studies risk of bias(a) inconsistency indirectness imprecision re-intervention
Number of patients exercising daily at 5-6 years
192 Observational No serious No serious No serious No serious None 2 out of 26 people in the exercise group were LOW
studies risk of bias(a) inconsistency indirectness imprecision exercising daily
Number of patients exercising more than twice a week at 5-6 years
192 Observational No serious No serious No serious No serious None 3 out of 26 people in the exercise group were LOW
studies risk of bias(a) inconsistency indirectness imprecision exercising more than twice a week
(a) Data taken from a RCT, non-comparative outcome.
Table 59: Clinical evidence profile: Angioplasty compared to supervised exercise for intermittent claudication due to femoro-popliteal disease

Quality
No of Other Supervised Relative
Design Risk of bias Inconsistency Indirectness Imprecision Angioplasty Absolute
studies considerations Exercise (95% CI)
195 Observational No serious No serious No serious No serious None 9 out of 60 people in the angioplasty group had LOW
studies risk of bias(a) inconsistency indirectness imprecision re-intervention
(a) Data taken from a RCT, non-comparative outcome.
Final version
151
PAD
Table 60: SF-36 individual domain results and mapped EQ-5D values – Angioplasty compared to supervised exercise
90
Mazari 2010 – Median (range)
PF 30 (35) 55 (48) NA NA NA 35 (30) 52 (40) NA NA NA
RP 20 (30) 25 (100) NA NA NA 25 (65) 25 (75) NA NA NA
BP 41 (42) 55 (43) NA NA NA 41 (40) 61 (46) NA NA NA
GH 55 (37) 60 (30) NA NA NA 57 (37) 54 (41) NA NA NA
V 45 (20) 50 (35) NA NA NA 50 (35) 55 (35) NA NA NA
SF 62 (50) 75 (50) NA NA NA 75 (50) 88 (50) NA NA NA
RE 33 (100) 83 (100) NA NA NA 66 (100) 100 (100) NA NA NA
MH 68 (28) 72 (30) NA NA NA 72 (28) 82 (25) NA NA NA
(a)
EQ-5D NE NE NA NA NA NE NE NA NA NA
(a) Only the range was reported; probabilistic mapped values not estimable.
Abbreviations: PF = physical function; RP = role physical; BP = bodily pain; GH = general health; V = vitality; SF = social functioning; RE = role emotional; MH = mental health; NA = not
applicable; NE = not estimable.
Final version
152
PAD
9.4.6 Bypass surgery compared to supervised exercise
Clinical evidence
One RCT94 was found which addressed the question and was included in the review.
The study characteristics are reported in Table 61. The quality and results of the included study are
reported in Table 62. The forest plots for each clinical outcome are reported in Appendix J.
Final version
153
PAD
Table 61: Study characteristics: Bypass surgery compared to supervised exercise

Study Disease location Combination Supervised exercise Bypass surgery
Lundgren, 198994 Aorto-iliac and Supervised exercise 6 weeks Supervised exercise programme; 3 sessions per week Bypass with saphenous vein or
femoro-popliteal following bypass operation for a minimum of 6 months. Each session consisted expanded PTFE graft.
disease of 30 minutes of dynamic leg exercises beyond the
appearance of leg pain due to arterial insufficiency
Table 62: Clinical evidence profile: Bypass surgery compared to supervised exercise for intermittent claudication due to aorto-iliac and femoro-popliteal
disease

Quality
No of Other Relative
Design Risk of bias Inconsistency Indirectness Imprecision Bypass Exercise Absolute
studies considerations (95% CI)
Mortality at 1 year
194 RCT Serious(a) No serious No serious Very serious(b) None 2/25 0/25 RR 5 (0.25 - VERY LOW
inconsistency indirectness (8%) (0%) to 99.16)
Maximum walking distance from baseline at 1 year
194 RCT Serious(a) No serious No serious Serious(c) None 25 25 - MD 85 higher LOW
inconsistency indirectness (107.88 lower to
277.88 higher)
Pain free walking distance at 1 year
194 RCT Serious(a) No serious No serious Serious(c) None 25 25 - MD 200 higher LOW
higher)
Complication at 30 days
194 Observational No serious risk No serious No serious No serious None 6 out of 25 people in the surgery group had LOW
studies of bias(d) inconsistency indirectness imprecision complications
Re-intervention at 30 days
194 Observational No serious risk No serious No serious No serious None 3 out of 25 people in the surgery group had LOW
Final version
154
PAD
studies of bias(d) inconsistency indirectness imprecision re-intervention

Withdrawal from exercise programme
194 Observational No serious risk No serious No serious No serious None 4 out of 25 in the exercise group withdrew LOW
studies of bias(d) inconsistency indirectness imprecision from exercise
(a) Study had unclear allocation concealment; unclear blinding.

(b) 95% crosses both MIDs.
(c) 95% CI crosses MID.
(d) Data taken from an RCT, non-comparative outcome.
Final version
155
PAD
9.4.7 Angioplasty compared to endovascular techniques
9.4.7.1 Review question
A literature search was conducted for RCTs that compared the effectiveness of angioplasty versus
bypass surgery. No time limit was placed on the literature search, and there were no limitations on
sample size. Indirect populations and emergency settings were excluded.
Seven studies of four RCTs101-104 105 106,107 were identified which addressed the question and were
included in the review. The trials did not report outcome data for people with diabetes.
The quality and results of included studies are reported in Table 63 and Table 64. The forest plots for
each clinical outcome are reported in appendix J.
For the clinical evidence statements, see section 9.4.9.
Final version
156
PAD
Table 63: Clinical evidence profile: Angioplasty compared to bypass surgery for people with intermittent claudication due to aorto-iliac disease

Quality
No of Risk of Other Relative
Design Inconsistency Indirectness Imprecision Angioplasty Bypass Absolute
studies bias considerations (95% CI)
Mortality at 30 days
1104 RCT Serious(a) No serious No serious Very serious(b) None 0/130 1/133 RR 0.34 (0.01 5 fewer per 1000 (from 7 VERY LOW
inconsistency indirectness (0%) (0.75%) to 8.29) fewer to 55 more)
Mortality at 3 months
1104 RCT Serious(a) No serious No serious Very serious(b) None 0/130 2/133 RR 0.2 (0.01 12 fewer per 1000 (from VERY LOW
inconsistency indirectness (0%) (1.5%) to 4.22) 15 fewer to 48 more)
Mortality at 1 year
inconsistency indirectness (0%) (2.3%) to 2.8) 22 fewer to 41 more)
inconsistency indirectness (15.4%) (19.5%) to 1.34) 106 fewer to 66 more)
Amputation post procedure
1104 RCT Serious(a) No serious No serious Very serious(b) None 2/130 2/133 RR 1.02 (0.15 0 more per 1000 (from 13 VERY LOW
inconsistency indirectness (1.5%) (1.5%) to 7.16) fewer to 93 more)
Amputation at 2 years
inconsistency indirectness (6.2%) (9.8%) to 1.47) 71 fewer to 46 more)
1105 RCT Serious(a) No serious No serious Very serious(b) None 6/59 3/59 RR 2 (0.52 to 51 more per 1000 (from VERY LOW
inconsistency indirectness (10.2%) (5.1%) 7.62) 24 fewer to 337 more)
Complications post procedure
Final version
157
PAD
1104 RCT Serious(a) No serious No serious Very serious(b) None 27/130 10/133 RR 2.76 (1.39 132 more per 1000 (from VERY LOW
inconsistency indirectness (20.8%) (7.5%) to 5.47) 29 more to 336 more)
1104 RCT Serious(a) No serious No serious Very serious(b) None 26/130 20/133 RR 1.33 (0.78 50 more per 1000 (from VERY LOW
inconsistency indirectness (20%) (15%) to 2.26) 33 fewer to 189 more)
ABPI after treatment (no specific time point)
1104 RCT Serious(a) No serious No serious No serious None 130 133 - MD 0.04 lower (0.04 to MODERATE
inconsistency indirectness imprecision 0.04 lower)
ABPI at 3 years
1104 RCT Serious(a) No serious No serious No serious None 130 133 - MD 0.02 higher (0.01 to MODERATE
inconsistency indirectness imprecision 0.03 higher)
Table 64: Clinical evidence profile: Angioplasty compared to bypass surgery for intermittent claudication due to femoro-popliteal disease

Quality
Risk of Other Relative
No of studies Design Inconsistency Indirectness Imprecision Angioplasty Bypass Absolute
bias considerations (95% CI)
2 106,107 RCT Serious(a) No serious No serious No serious None 0/53 0/48 not not pooled MODERATE
inconsistency indirectness imprecision(b) (0%) (0%) pooled
Mortality at 1 year
2101,106 RCT Very No serious No serious Very serious(d) None 5/63 4/64 RR 1.31 19 more per VERY LOW
serious(c) inconsistency indirectness (7.9%) (6.3%) (0.39 to 1000 (from 38
4.39) fewer to 212
more)
1103 RCT Very No serious No serious Very serious(d) None 6/40 5/46 RR 1.38 41 more per VERY LOW
serious(e) inconsistency indirectness (15%) (10.9%) (0.46 to 1000 (from 59
Final version
158
PAD
4.18) fewer to 346

more)
serious(e) inconsistency indirectness (22.5%) (17.4%) (0.55 to 1000 (from 78
3.04) fewer to 355
more)
Amputation at 1 year
3101,106,107 RCT Very No serious No serious Very serious(d) None 3/103 5/93 RR 0.61 21 fewer per VERY LOW
serious(f) inconsistency indirectness (2.9%) (5.4%) (0.17 to 1000 (from 45
2.18) fewer to 63
more)
1103 RCT Very No serious No serious Very serious(d) None 1/50 5/50 RR 0.2 80 fewer per VERY LOW
serious(e) inconsistency indirectness (2%) (10%) (0.02 to 1000 (from 98
1.65) fewer to 65
more)
2102,105 RCT Very No serious No serious Very serious(d) None 4/88 11/85 RR 0.35 84 fewer per VERY LOW
serious(g) inconsistency indirectness (4.5%) (12.9%) (0.11 to 1000 (from
1.05) 115 fewer to 6
more)
Minor complications post procedure
5.32) fewer to 254
more)
Major adverse event at 1 year
1107 RCT Serious(a) No serious No serious Very serious(d) None 0/30 2/25 RR 0.17 66 fewer per VERY LOW
inconsistency indirectness (0%) (8%) (0.01 to 1000 (from 79
3.34) fewer to 187
more)
Final version
159
PAD
Minor adverse event at 1 year

1107 RCT Serious(a) No serious No serious Very serious(d) None 0/30 2/25 RR 0.17 66 fewer per VERY LOW
inconsistency indirectness (0%) (8%) (0.01 to 1000 (from 79
3.34) fewer to 187
more)
2.06) fewer to 184
more)
1103 RCT Very No serious No serious Very serious(d) None 17/50 17/50 RR 1 (0.58 0 fewer per VERY LOW
serious(e) inconsistency indirectness (34%) (34%) to 1.73) 1000 (from
143 fewer to
248 more)
serious(e) inconsistency indirectness (36%) (30%) (0.68 to 1000 (from 96
2.11) fewer to 333
more)
ABPI at 1 year
1106 RCT Serious(a) No serious No serious Serious(h) None 23 18 - MD 0.12 LOW
0.17 higher)
ABPI at 1 year (no sd)
1101 RCT Very No serious No serious Serious(i) None 50 50 - not pooled VERY LOW
serious(e) inconsistency indirectness
ABPI at 2 years (no sd)
1103 RCT Very No serious No serious Serious(i) None 50 50 - not pooled VERY LOW
serious(e) inconsistency indirectness
(b) No events in either group.
Final version
160
PAD
(c) 1 of 2 studies had unclear methodology;1 of 2 studies had unclear allocation concealment and blinding.
(d) 95% CI crosses both MIDs.
(e) Unclear methodology.
(f) 1 of 3 studies had unclear methodology; 2 of 3 studies had unclear allocation concealment and blinding.
(g) 1 of 2 studies had unclear methodology; 1 of 2 studies had unclear blinding.
(h) 95% CI crosses one MID.
(i) No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
Final version
161
PAD

Five cost-utility analyses were identified that compared exercise and endovascular interventions for
the treatment of IC. One was a pair-wise comparison based on an RCT100 and the remaining four
were decision analytic models evaluating different intervention sequences.108-111 None of the studies
included all interventions under consideration by the GDG. Study characteristics and results are
summarised in Table 65, subdivided according to included intervention strategies.
Spronk 2008 analysed cost and outcome data from a RCT comparing supervised exercise (performed
twice weekly for 30 minutes per session over 24 weeks on a treadmill) to angioplasty with selective
stent placement. This RCT was included in the clinical review. At 12 month follow-up and after
adjustment for baseline variables, the treatment groups did not differ significantly in functional
capacity or quality of life. However, there was a large difference in cost favouring supervised
exercise. This analysis found that supervised exercise is highly likely to be cost effective compared to
angioplasty with selective stent placement, however it was limited by a short time horizon and did
not include all relevant comparators.
Visser 2003111 developed a model to compare supervised exercise to angioplasty followed by exercise
and angioplasty followed by exercise. Each endovascular strategy was preceded by a different
imaging modality. The results of this analysis suggest that angioplasty preceded by MRA is the most
cost-effective initial intervention, with supervised exercise for those who are not suitable for
angioplasty. However, this study did not exercise as a primary treatment for IC and therefore is
missing an important comparator.
De Vries 2002109 developed a model to compare five combinations of supervised exercise (advised to
walk 2-6km every day for six months with four check-in periods) and revascularisation (angioplasty
with selective stent placement for supra-inguinal disease) and angioplasty or bypass surgery for infra-
inguinal disease) with clinical outcomes based on a retrospective database analysis. The results of
this model indicate that none of the evaluated strategies fall within the £20, 000 per QALY threshold
compared to a baseline strategy of unsupervised exercise.
Based on data from the Dutch Iliac Stent Trial and several meta-analyses, Bosch 1998108 developed a
decision model to evaluate the cost effectiveness of treating claudication due to iliac arterial stenosis
with primary stent placement, selective stent placement or angioplasty without stent placement.
This model assumes that 40% of patients undergoing angioplasty require selective stent placement
and that compared to angioplasty alone, the relative risk of failure associated with stent placement is
0.61. The results of this model suggest that angioplasty with selective stent placement for both
primary and secondary treatment is more cost effective than both selective stent placement
followed by conservative management and primary stent placement followed by selective stent
placement. This conclusion was robust to a wide range of sensitivity analyses.
The same model (with American costs) was used in a later analysis by Bosch 2000.112 Based on the
results of their previous study (Bosch 1999113), which concluded that primary stent placement was
not cost-effective, the authors did not include angioplasty with primary stent placement as a
comparison in this analysis. Because this comparison was not relevant to the study question it was
excluded from the review.
Hunink 1995110 evaluated the cost-effectiveness of revascularisation for femoro-popliteal disease

using angioplasty, bypass surgery and combinations of the two treatments in people with disabling
claudication. Only patients requiring revascularisation were included and strategies such as exercise,
medical therapy or amputation were not considered. The results of bypass surgery were sub-grouped
according to graft material (autologous vein vs. prosthetic bypass) and lesion type. Although the
results of the analysis are different for each subgroup, the conclusions are broadly the same.
Final version
162
PAD
Table 65: Economic evidence profiles

Incremental cost Incremental Cost Uncertainty
Study Limitations Applicability Other comments effect effectiveness
Supervised exercise compared to angioplasty with selective stent placement
Spronk 2008100 Minor Partially  Cost utility analysis based on RCT Angioplasty with Angioplasty Angioplasty  At a threshold
limitations(a) applicable(b) by Spronk 200991 selective stent with selective with selective of
 Population: People with IC placement is £3, stent placement stent placement approximately
867 more costly results in a 0.02 costs £193, 374 £60k, there was
than supervised QALY gain. per QALY a 5% probability
 Costs: All healthcare costs exercise. gained. that angioplasty
 Perspective: Netherlands, is more cost
hospital effective than
supervised
exercise.
Supervised exercise vs. MRA and angioplasty or exercise vs. MRA and angioplasty or bypass surgery vs. DUS and angioplasty or exercise vs. DUS and angioplasty or
bypass vs. DSA and angioplasty or exercise vs. DSA and angioplasty or bypass.
Visser 2003111 Potentially Partially  Decision analytic model MRA + Primary MRA + Primary MRA +Primary  The results
serious applicable(d)  Population: People with IC angioplasty was angioplasty angioplasty cost were robust to
limitations(c) £1, 821 more resulted in £20, 670 per most sensitivity
 Time horizon: Lifetime
costly than 0.0881 QALYs QALY gained analyses.
 Costs: All healthcare costs supervised gain compared compared to
 Perspective: Netherlands, exercise. DSA to supervised supervised
societal followed by exercise alone. exercise alone.
angioplasty or DSA followed by DSA followed by
bypass was £10, angioplasty or angioplasty or
287 more costly bypass = 0.0767 bypass cost
than MRA QALYs gained £134, 120, 074
followed by compared to per QALY gained
angioplasty or MRA followed compared to
exercise. by angioplasty MRA followed
or exercise. by angioplasty
or exercise.
Unsupervised exercise only vs. unsupervised exercise followed by angioplasty for treatment failure vs. Unsupervised exercise followed by angioplasty or bypass for
treatment failure vs. angioplasty or unsupervised exercise followed by angioplasty for treatment failure vs. angioplasty, bypass or unsupervised exercise followed
by angioplasty or bypass for treatment failure.
Final version
163
PAD
de Vries 2002109 Potentially Partially  Decision analytic model Angioplasty or Angioplasty or Angioplasty or  ICER for
serious applicable(f)  Population: People with IC exercise followed exercise followed exercise interventional
limitations(e) by angioplasty = by angioplasty = followed by strategies
 Horizon: Lifetime
£3, 838 more 0.10 QALYs angioplasty = increased with
 Costs: Diagnostic and costly compared gained compared £38, 376 age or a
interventional procedures, short- to unsupervised to unsupervised QALYs gained positive history
and long-term follow-up costs exercise alone. exercise alone. compared to of CAD, due to
 Perspective: USA/Netherlands Exercise followed Exercise followed unsupervised increased
societal perspective by angioplasty or by angioplasty or exercise. procedural risk
bypass = £21, 985 bypass = 0.07 Exercise and reduced
more costly QALY gain followed by life expectancy
compared to compared angioplasty or in older
angioplasty or angioplasty or bypass = £314, patients with
exercise followed exercise followed 079 per QALY cardiac
by angioplasty. by angioplasty. gained. ischaemia.
Unsupervised exercise only vs. selective stent placement followed by unsupervised exercise vs. selective stent placement followed by selective stent placement vs.
Bosch 1998108 Minor Partially  Decision analytic model Selective stent Selective stent Selective stent Robust to
limitations applicable(g)  Population: People with IC placement placement placement changes in the
followed by followed by followed by risk of long term
 Horizon: Lifetime
selective stent selective stent selective stent failure following
 Costs: Diagnostic costs, placement was placement placement stent placement,
interventional procedures, and £3, 960 more resulted in 0.32 cost £12, 376 proportion of
patient costs. costly than QALYs gained per QALY patients requiring
 Perspective: Netherlands societal selective stent compared to gained a stent, and stent
perspective placement selective stent cost.
followed by no placement
revascularisation. followed by no
revascularisation.
No treatment vs. angioplasty followed by no treatment vs. angioplasty followed by angioplasty vs. angioplasty followed by bypass vs. bypass followed by no
treatment vs. bypass followed by bypass revision
Hunink 1995110 Potentially Partially  Decision analytic model Vein graft for IC stenosis
serious applicable(i)  Population: People with IC Angioplasty Angioplasty Angioplasty In people with IC
limitations(h)
 Horizon: Lifetime followed by followed by followed by due to occlusion,
 Costs: Costs of angioplasty and bypass surgery is bypass surgery bypass surgery the conclusion of
bypass for patients with the least costly is the most was the the model was
strategy effective dominant unchanged
Final version
164
PAD
claudication and critical limb strategy treatment

ischaemia; annual follow-up strategy
costs for angioplasty and bypass PTFE-AK for IC stenosis
patients; cost of amputation and
rehabilitation; annual cost of Angioplasty Angioplasty Angioplasty In people with IC
post amputation care; annual followed by followed by followed by due to occlusion,
cost of care with major angioplasty was angioplasty was angioplasty was the conclusion of
morbidity. the least costly the most the dominant the model was
strategy effective treatment unchanged
 Perspective: UK NHS
strategy strategy
PTFE-BK for IC stenosis
Angioplasty Angioplasty Angioplasty In people with IC
followed by followed by followed by due to occlusion,
angioplasty was angioplasty was angioplasty was the conclusion of
the least costly the most the dominant the model was
strategy effective treatment unchanged
strategy strategy
(a) This analysis took a societal perspective but reported disaggregated costs – societal costs have been subtracted for the purposes of reporting this study. Short time horizon.
(b) Data derived from US and Dutch databases; patency not reported, making between study comparisons difficult.
(c) Societal perspective; assumed that symptom progression necessitated reintervention.
(d) Did not include exercise as a primary treatment strategy.
(e) Costs and QALY results read off graph and imputed from reported ICERs.
(f) Assumed angioplasty preceded by catheter angiography.
(g) Dutch healthcare setting; societal perspective.
(h) Quality of life estimated using Torrence Multi Attribute Scale by healthcare workers; patency failure assumed to be equivalent to symptom progression & re-intervention (according to the
GDG, not all patients who experience failure or symptom progression following angioplasty will undergo reintervention. Only those who return to their healthcare provider will be
considered, and of those, the probability of treatment will depend on the location and extent of the lesion); progression of symptoms not modelled due to lack of data.
(i) Resource use based on American hospital records.
Final version
165
PAD
9.4.8.1 Original economic model
None of the cost-utility studies identified in the literature included all relevant comparators for the
treatment of people with IC. Therefore, the GDG decided to prioritise this area for original cost
effectiveness modelling. The aim of this analysis was to determine the most cost-effective treatment
pathway for people with IC in England and Wales who are suitable for both exercise and angioplasty
as first-line treatment options.
The analysis was undertaken from the perspective of the NHS and personal social services, in
accordance with NICE guidelines methodology. Relevant costs consisted of the cost of a supervised
exercise programme and treatment for stroke and MI. All costs are reported in 2009/10 British
pounds. The primary measure of outcome is the quality-adjusted life-year (QALY). The model was
evaluated over a lifetime horizon with both costs and QALYs discounted at a rate of 3.5% per year.
Alternative discount rates of 1.5% for QALYs and 3.5% for costs were explored in sensitivity analysis.
9.4.8.2 Methods
Comparators
The model was designed to compare 13 alternative treatment strategies for people with intermittent
claudication (four primary interventions followed by three secondary interventions, plus one
additional combined intervention). A treatment strategy was defined as the initial therapy combined
with secondary intervention options if the initial treatment should fail (Table 66).
The model did not consider bypass surgery as a primary strategy because the GDG did not consider
bypass to be an appropriate first-line therapy for people with claudication; bypass was included as a
secondary procedure following unsatisfactory results from supervised exercise or angioplasty. Stent
placement was included as a planned (‘primary stent placement’) and bail-out (‘selective stent
placement’) procedure for angioplasty. In both primary and selective stent strategies, only bare
metal stents were considered as the GDG decided not to recommend the routine use of drug eluting
stents following a review of the clinical evidence (see section 9.6). Angioplasty with primary stent
was not considered as a secondary intervention as the GDG did not think that there was anything to
recommend it over selective stent placement.
Table 66: Evaluated treatment strategies

Strategy Initial treatment Secondary treatment
1 Unsupervised exercise Supervised exercise
2 Unsupervised exercise Angioplasty with selective stent
3 Unsupervised exercise Bypass surgery
4 Supervised exercise Supervised exercise
5 Supervised exercise Angioplasty with selective stent
6 Supervised exercise Bypass surgery
7 Angioplasty with selective stent Supervised exercise
8 Angioplasty with selective stent Angioplasty with selective stent
9 Angioplasty with selective stent Bypass surgery
10 Angioplasty with primary stent Supervised exercise
11 Angioplasty with primary stent Angioplasty with selective stent
12 Angioplasty with primary stent Bypass surgery
13 Angioplasty with selective stent + supervised exercise
Final version
166
PAD
Population
The hypothetical population included in the analysis was people with IC who are suitable for and
willing to undergo either exercise or angioplasty. Based on the baseline characteristics of people in
the included RCTs, a starting age of 67 years was used to represent the average age of people with
IC. The hypothetical cohort was 70% male and had an average ABPI of 0.64. Twenty four percent of
people were diabetic and 43% were current smokers. The prevalence of diabetes and smokers was
used to inform the baseline risk of stroke and MI in the model.
Not included were people with co-morbidities which prevent participation in an exercise program;
people who are either not interested in undergoing angioplasty or not considered anatomically
suitable for an endovascular procedure; people who have recently undergone an endovascular
procedure; or people with CLI. People who drop out after beginning an exercise programme are
included in the model.
According to the methods used in the clinical review, patients with IC due to stenosis in the aorto-
iliac and femoro-popliteal arteries were considered as separate subgroups. All were assumed to be
receiving best medical therapy (antiplatelet therapy, anti-hypertensive therapy, cholesterol-lowering
agents, diabetes control and smoking cessation advice) at baseline, consistent with the included
RCTs.
Approach to modelling
Intermittent claudication is associated with high mortality, increased risk of cardiovascular morbidity
and a decreased quality of life. Primary treatment options for IC include exercise and angioplasty.
Exercise may take the form of either a supervised or unsupervised programme and angioplasty may
be performed with either primary or selective stent placement. If symptoms do not improve, patients
may be offered a supervised exercise programme or referred for assessment for angioplasty or
bypass surgery. In order to determine which interventions represent the most cost effective pathway
for people with IC, the model included 13 different treatment sequences: four primary alternatives,
three secondary interventions and one combination treatment. As a necessary simplification, no
more than two treatment options were considered. If patients’ symptoms deteriorate following
secondary intervention, they were assumed to revert to their baseline quality of life.
As for the model comparing supervised to unsupervised exercise (Appendix K), compliance to the
recommended level of physical activity was associated with a decreased risk of mortality and
cardiovascular events. The most conservative estimate of compliance to exercise (scenario 2;
Appendix K) was used in the base case analysis with other scenarios explored in sensitivity analysis.
Treatment failure following exercise was defined as a worsening of symptoms. Epidemiological
studies suggest that approximately a quarter of patients with intermittent claudication experience
deterioration in their symptoms over a five year period.5 Currently, there is no evidence to suggest
that exercise has any impact on the rate of disease progression. It was assumed that patients who
undertake supervised and unsupervised exercise programmes experience the same rate of
symptomatic progression as observed in the epidemiological literature.
There is no evidence to suggest that angioplasty has any impact on long term mortality or
cardiovascular risk factors. Therefore, people who underwent angioplasty were assumed to have the
same mortality and cardiovascular risk as those who were inactive (i.e. baseline risk). Failure
following angioplasty was defined as patency failure plus symptom deterioration requiring secondary
intervention. Relative risk of re-intervention for people who had undergone selective and primary
stent placement were obtained from the systematic clinical review. In the absence of evidence of the
effectiveness of secondary interventions, it was assumed that they were associated with the same
relative risk of mortality and morbidity as those observed in primary procedures. People who failed
secondary intervention and were left with persistent claudication had no further intervention, unless
they subsequently progressed to CLI.
Final version
167
PAD
The GDG noted that currently there is no evidence to suggest a relationship between treatment for
claudication and progression to critical limb ischaemia (CLI). In the base case analysis, the risk of
progression to CLI was included as a constant background rate irrespective of treatment pathway,
effectively ‘cancelling out’ of the model. The treatment for critical limb ischaemia was the same for
all strategies: 25% underwent amputation. The potential impact of different treatments on the rate
of progression to CLI (and therefore to amputation) was explored in sensitivity analysis.
People who experience a cardiovascular event enter a semi-absorbing health state from which the
only available transition is death. Average costs and quality of life associated with post-
cardiovascular event states were applied to this health state, and the same mortality rate as
sedentary people was assumed. It was also assumed that all patients would undergo a general
examination and treatment for cardiovascular risk factors.
The treatment goal for people with IC is to improve health related quality of life. As in the previous
model comparing supervised to unsupervised exercise (Appendix K), the GDG decided to use the
quality of life data from the RCTs included in the clinical review as the primary measure of clinical
effectiveness. Symptomatic progression, cardiovascular events, and lower limb amputation resulted
in a reduced quality of life according to published estimates.
Based on clinical experience, it was assumed that patients who drop out of supervised exercise
programmes do so within the first few weeks. They were assigned a quarter of the cost of a course of
supervised exercise and assumed not to accrue any health benefit from their time spent in the
programme.
The model was built probabilistically to take account of the uncertainty surrounding each input
parameter. In order to characterise uncertainty, a probability distribution was defined for each
parameter based on error estimates from the data sources (e.g. standard errors or confidence
intervals). The way in which distributions are defined reflects the nature of the data. When the
model was run, a value for each input was randomly selected from its respective distribution. The
model was run repeatedly (10,000 times) to obtain mean cost and QALY values.
Final version
168
PAD
Figure 6: Schematic Markov model structure
Final version
169
PAD
Unsupervised Supervised Angioplasty with selective Angioplasty with Angioplasty with

exercise exercise stent + supervised exercise selective stent primary stent
Active
Successful
(symptoms stable)
Nonfatal (symptoms stable)
Cardiovascular event
Sedentary
(symptoms stable) Symptom
deterioration
Post-nonfatal
cardiovascular event
Amputation Critical limb

ischaemia
Symptom
Re-intervention
deterioration
Death
Supervised Angioplasty with Bypass

exercise selective stent
As above As above As above

Schematic diagram of the Markov model designed to compare the cost-effectiveness of different exercise and endovascular treatment strategies for people with IC. The Markov modelling approach involves
a transition between different health states over time, represented by arrows. The model is divided into three month cycles. At the end of each cycle a time-dependant transition to another health state is
possible, unless people enter into an ‘absorbing state’ from which they do not recover. In this model, the absorbing state is death. In the base case model, transition to CLI (and therefore amputation) occurs
at a constant rate, represented by dashed grey arrows.
Final version
170
PAD
Baseline mortality and relative risk associated with exercise
Age- and sex-specific all cause mortality was based on the most recent available life tables of the
general population in England and Wales. Rates were adjusted for people with IC by multiplying by
the standardised risk of all cause mortality observed over 10 years in people with IC.73
No randomised evidence of exercise-associated risk of mortality in people with IC was identified. The
GDG agreed that evidence from people with cardiovascular disease would represent a reasonable
proxy. A recent Cochrane review of randomised controlled trials was therefore used to inform the
risk of total mortality among people participating in exercise rehabilitation compared to non-active
controls.74 A summary of the values used to inform this parameter is provided in Table 67. The GDG
discussed the limitations of using an indirect population to inform this parameter and the effect of
this value on the model result was further explored in sensitivity analysis.
Table 67: Total mortality

10-year total mortality for Relative risk of total Relative risk of total mortality
the general population mortality in people with IC in people who exercise
based on Life Tables for compared to those without compared to those who do not
England and Wales(a) IC exercise
Mortality 25.0% 3.1 (95% CI, 1.9 to 4.9)73 0.87 (95% CI , 0.75 to 0.99)74
(a) Assuming that the average age of the baseline population is 67 years and 66% are male.
Baseline risk of cardiovascular events and relative risk associated with exercise
The average baseline probability of stroke and MI was calculated by age and gender using the
Framingham risk equations and risk calculator spreadsheet developed by Rupert Payne at the
University of Edinburgh.75,76 Risk factor inputs for each sex were obtained from the 2006 Health
Survey for England (HSE; Table 35).77 Average age- and sex- specific blood pressure values were
obtained from the 2011 NICE Hypertension update guideline78, which used individual patient level
data from the 2006 HSE. A recent study by the Ankle Brachial Index Collaboration found that when
combined with Framingham risk scores, an ABPI of between 0.61 and 0.70 approximately triples the
risk of major cardiovascular events for men and women.79
The risk of MI in patients who exercise compared to those who are not active in an exercise
programme was obtained from the Cochrane review by Heran et al (2011).74 A meta-analysis of the
effect of physical activity on the incidence of stroke was used to inform the risk of stroke for active
compared to sedentary people in the model.80 A summary of the values used to inform these
parameters is provided in Table 68. As with estimates of the relative risk of total mortality, these
data sources are subject to several limitations and the effect of these values on the model were
explored in sensitivity analysis.
Table 68: Major cardiovascular events

10 year risk of MI and
stroke for general Relative risk of MI and stroke
population according Relative risk of major cardiovascular in people who exercise
to the Framingham events in people with IC compared compared to those who do not
equations(a) to those without IC(b) exercise
MI 7.2% Men: 2.71 (95% CI, 2.01 to 3.64) 0.97 (95% CI, 0.82 to 1.15)74
Women: 3.82 (95% CI, 2.86 to 5.11)
Stroke 4.4% Men: 2.71 (95% CI, 2.01 to 3.64) 0.80 (95% CI, 0.74 to 0.86)80
Women: 3.82 (95% CI, 2.86 to 5.11)
(a) Calculated using Framingham MI and stroke risk equations75,76 and risk factor inputs derived from the 2006 Health
Survey for England77, assuming that the average age of the baseline population is 67 years and 66% are male.
Final version
171
PAD
(b) Based on a risk of cardiovascular events for mean and women with an ABPI of 0.61 to 0.7 compared to men and women
with normal ABP.I79
Compliance to supervised and unsupervised exercise
Levels of short- and long-term compliance to supervised and unsupervised exercise programs among
people with IC is an area of great uncertainty. Following a review of the literature and survey of GDG
members and their colleagues across the country (Appendix K), two scenarios were developed to
represent different theoretical rates of compliance each exercise programme. In order to simplify
reporting for this model, the more conservative of the two scenarios was used to inform the base
case analysis. Under this assumption, compliance to supervised exercise is greater than unsupervised
exercise over the short term and equal over the long term (Figure 7). The impact of different levels of
compliance on the outcome of the model was explored in sensitivity analysis.
Figure 7: Equal long term compliance to unsupervised and supervised over the long term
100%
90%
80%
70%
Exercise compliance
60%
50%
Supervised exercise
40%
30%
20%
10%
0%
0 1 2 3 4 5 6 7 8 9
Time point Cycle Supervised Unsupervised

Lower Most likely Upper Lower Most Likely Upper
3 months 1 40% 68% 80% 17% 43% 56%
6 months 2 15% 40% 57% 10% 33% 45%
1 year 4 4% 22% 40% 7% 22% 37%
2 years 8 5% 16% 32% 5% 16% 31%
Symptom deterioration after a period of exercise
Few studies have measured disease progression among patients with intermittent claudication. Most
articles on the natural history of the disease report that claudication remains stable in 70% to 80% of
patients over a five year period Hirsch 20065, Rosenbloom 1988, Edi study 1996). In the remainder of
patients, it may progress to disabling claudication or critical limb ischaemia requiring
revascularisation. Based on these estimates, it was assumed that claudication symptoms worsen to
the point of requiring revascularisation in 25% (range = 20% to 30%) of people with IC over 5 years.
This is equivalent to a one year probability of 5.6% and a three month probability of 1.4%.
Final version
172
PAD
Currently, there is no evidence to suggest that the probability of symptom deterioration differs
between patients who exercise and those who do not. The probability of requiring revascularisation
was assumed to be equal regardless of activity status and therefore did not differ according to
whether patients had undertaken a supervised or unsupervised exercise programme.
CLI and amputation
Amputation is a relatively rare outcome of claudication and is usually a result of the patient
developing CLI. It was assumed that 2% of people with claudication progress to CLI over a 5 years
and that 25% of those with CLI 25% undergo amputation as a primary intervention.5 In the base case
analysis, progression to CLI was applied at a constant rate regardless of a person’s position in the
treatment pathway. It was assumed that the development of CLI is a function of the disease process
and does not differ by intervention. This assumption was further explored in sensitivity analysis.
The one year mortality rate in people with CLI is approximately 25%.114 For those who undergo
amputation, this is considerably higher with a 35% probability of mortality in the first year following
amputation and 19% probability every year thereafter.115
Baseline and relative treatment effects
Baseline rates of mortality, major complications and amputation associated with angioplasty with
selective stent placement were obtained from a prospective audit by the Royal College of Surgeons
of England.116 Because the results of the audit were not reported by lesion location, the reported
outcomes were assumed to represent an average value across both vessels. The audit found that 33
(2.4%) of total angioplasties were complicated by major medical morbidity which was unrelated to
the technique of angioplasty. This was used as the baseline probability of major complication
following angioplasty with selective stent placement. None of the patients undergoing angioplasty
for claudication died within 30 days of the procedure. Although the GDG agreed that the risk of
death as a result of angioplasty was small, they thought that there was still a risk associated with the
procedure. It was assumed that 0.5 (out of 841) people with IC undergoing angioplasty die due to the
procedure. Similarly, none of the patients experienced limb loss as a result of acute ischaemia
following angioplasty.116 However, the GDG indicated that although small, there is a risk of
amputation as a result of angioplasty. Therefore, as for mortality, it was assumed that 0.5 of 841
angioplasty procedures for claudication could be expected to result in amputation.
People who undergo endovascular procedure may experience a reoccurrence of symptoms over the
following months or years. Based on primary patency results reported in the TASC II guideline and
the clinical experience of the GDG, it was assumed that each year after angioplasty, a certain
percentage of people with aorto-iliac and femoro-popliteal disease experience patency failure. Not
all of those who experience patency failure will undergo reintervention. Of those who return to their
healthcare provider, the GDG noted that people with aorto-iliac disease are more likely to undergo
secondary intervention compared to those with stenoses or occlusions of the femoro-popliteal
artery. The weighted average probability of reintervention for each artery is presented in Table 69.
Evidence of relative clinical effectiveness between different interventions was collected from the
pooled results of the clinical systematic review. For each outcome, angioplasty with selective stent
placement was used as the baseline comparator. Relative risks were entered into the model
probabilistically to reflect the uncertainty surrounding each point estimate. For two outcomes (30-
day mortality and post-operative amputation) there was no data reported for one of the two
arteries. Where the GDG considered that there was no a priori reason to assume a difference in
treatment efficacy based on location, and if the 95% CI in one anatomical area included one, a
default value of 1 was used to inform the missing risk ratio. Where the GDG considered there was an
a priori reason for considering that there would be a difference, the results for one anatomical area
Final version
173
PAD
were used as the basis for estimating the other. See Table 69 for a summary of all clinical
effectiveness outcomes applied in the model.
Table 69: Baseline probabilities and relative treatment effects

Point Value range
Parameter estimate Source
Probability of 30-day mortality for angioplasty with selective stent
Baseline probability of 30-day 0.06% 0.0% - 0.9% Expert opinion informed by Royal College
mortality of Surgeons 2002116
Relative risk of 30-day mortality for angioplasty with primary stent (compared to selective stent)
Aorto-iliac Not reported. Assumed no difference between interventions (RR = 1)
Femoro-popliteal 0.20 0.01 – 4.17 Cejna 2001117
Probability of major complications for angioplasty with selective stent
Baseline probability of major 2.4% 1.7% - 3.3% Royal College of Surgeons 2002116
complications
Relative risk of major complications for angioplasty with primary stent (compared to selective stent)
Aorto-iliac 0.57 0.21 – 1.54 Tetteroo 1998118
Femoro-popliteal 1.26 0.33 – 1.93 Dick 2009119, Krankenberg 2007120,
Schillinger 2006121, Vroegindewij 1997122
Baseline probability of post operative amputation following angioplasty with selective stent
Baseline probability of post 0.06% 0.0% - 0.9% Expert opinion informed by Royal College
operative amputation of Surgeons 2002116
Relative risk of post operative amputation following angioplasty with primary stent (compared to
selective stent)
Aorto-iliac Not reported. Assumed no difference between interventions (RR = 1)
Femoro-popliteal 0.50 0.09 – 2.63 Cejna 2001117
Probability of IC symptom worsening following angioplasty (selective stent & primary stent)
Aorto-iliac 7.5% 5% - 10% Expert opinion
Femoro-popliteal 34% 28% - 40% Expert opinion
Baseline probability of reintervention following symptom worsening (selective stent only)
Aorto-iliac 71% 66% - 76% Expert opinion
Femoro-popliteal 28% 18% - 38% Expert opinion
Relative risk of re-intervention following angioplasty with primary stent (compared to selective stent)
Aorto-iliac 1.63 0.58 – 4.61 Tetteroo 1998118
Femoro-popliteal 0.50 0.22 – 1.13 Schillinger 2007123,124
Relative risk of 30-day mortality following bypass (compared to selective stent)
Aorto-iliac 2.94 0.12 – 73.19 Wilson 1989104
Femoro-popliteal 2.94 0.12 – 73.19 Expert opinion (see text)
Relative risk of perioperative major complications following bypass (compared to selective stent)
Femoro-popliteal 0.60 0.17 – 2.17 McQuade 2009103
Relative risk of amputation within 30-days of bypass (compared to selective stent)
Femoro-popliteal Not reported. Assumed no difference between interventions (RR = 1)
Final version
174
PAD
Utilities
In cost-utility analyses, measures of health benefit are valued in terms of quality adjusted life years
(QALYs). The QALY is a measure of a person’s length of life weighted by a valuation of their health
related quality of life (HRQoL) over that period. The quality of life weighting comprises two elements:
the description of changes in HRQoL and an overall valuation of that description. Questionnaires such
as the SF-36 and SF-12 provide generic methods of describing HRQoL while the EQ-5D, HUI, and SF-
6D also include preference-based valuations of each health state.
Quality of life data were collected from all RCTs included in the clinical review. Four studies included
the EQ-5D as a measure of HRQoL. Thirteen papers (representing an additional nine trials) reported
SF-36 data. According to the NICE reference case, EQ 5D data are the preferred measure of quality of
life for use in cost utility analyses. Therefore, of the four trials that reported both measures, EQ-5D
was used in preference to SF-36.
Recently, several algorithms have been developed which can be used to map generic descriptions of
HRQoL to preference-based utility indexes. In 2008, Ara and Brazier68 published a method of
predicting mean EQ-5D preference based index score using published mean cohort statistics from the
eight dimensions of the SF-36 health profile. In order to use these algorithms, values for each of the
eight dimensions of the questionnaire are required. Four provided all the necessary values and the
authors of the remaining nine studies were contacted to request the required data (Appendix L).
Mapping SF-36 to EQ-5D using published algorithms and probabilistic simulation
For each trial, it is the change in quality of life over time and the difference in this change between
interventions (i.e. mean difference in change) that is the key to determining the relative
effectiveness of each intervention. In order to calculate the mean difference in change between each
three month time interval while taking into account the uncertainty surrounding each estimate, the
mean and standard error of each dimension of the SF-36 were assigned a beta distribution according
to the method of moments described by Briggs 2006.81 Probabilistic mapped values were then
calculated using Equation 4 from the paper by Ara and Brazier68, who specify that ‘when comparing
incremental differences between study arms or changes over time, Equation 4 is the preferred
choice’. A simulation was run 20, 000 times in order to calculate a mean, standard error and
confidence interval surrounding each mapped estimate. For the purposes of clinical validation,
absolute mean mapped values were calculated using Equation 1 according to the same method. The
results of these simulations are reported in Table 72.
Equation 1: 0.03256 + 0.0037 x PF + 0.0011 x SF – 0.00024 x RP + 0.00024 x RE + 0.00256 x MH –

0.00063 x VT + 0.00286 x BP + 0.00052 x GH
Equation 4: -0.18105 + 0.00781 x PF +0.00213 x SF + 0.00022 x RE + 0.00472 x BP + 0.00064 x GH
Note that mean difference in change calculated using Equation 4 is not expected to equal the
incremental difference between the mean mapped values from Equation 1 as they are derived using
different models. Alternative methods of calculating relative differences in quality of life between
treatment arms were explored in sensitivity analysis. Note also that because the covariance matrices
for the regression coefficients were not available it was not possible to account for uncertainty in the
mapping algorithm in the probabilistic analysis.
Inputs and assumptions used to inform model utilities
In the base case analysis, an average utility value was weighted according to the total number of
people in the study at each time point. In order to preserve within-study randomisation, the
weighted average incremental change in quality of life associated with each intervention (as
calculated by the probabilistic simulation; Table 72) was applied in an additive method. For example,
Final version
175
PAD
at 3 months, the mean difference in change from baseline between selective stent placement and
supervised exercise is 0.035 QALYs. And at the same time point, the mean difference in change
between supervised exercise and unsupervised exercise is -0.021 QALYs. Adding these values results
in a mean difference in change between selective stent placement and unsupervised exercise of
0.014 QALYs between baseline and three months.
None of the studies that included bypass surgery as an intervention measured quality of life as an
outcome. The exclusion list of the clinical evidence review was searched for non-randomised data
from which to draw utility data, however none reported this information. Based on discussions with
the GDG and observational studies in the literature125, it was assumed that the utility gain associated
with angioplasty with primary stent is equal to that associated with bypass.
The duration of supervised exercise programmes differed between each trial (Savage = 3 months;
Cheetham = 6 months; Nicolai = 12 months). The GDG agreed that in order to make use of all
available evidence the data from all trials should be combined using a weighted average. Quality of
life gains achieved after exercise intervention were maintained for people who continued to exercise.
Those who stopped exercising were assigned the baseline quality of life.
Quality of life associated with cardiovascular events
Quality of life associated with cardiovascular events was derived from the most recent NICE
Hypertension guideline update, which in turn was obtained from a comprehensive review of the
literature undertaken by the authors of the NICE statins HTA (Table 70).
Table 70: Quality of life following cardiovascular events

Event Mean utility SE Source
MI 0.760 0.018 Goodacre 2004126
Stroke 0.629 0.040 Tengs 2003127
In line with the methods used by the hypertension guideline, it was assumed that full health was
equal to a utility of one. The utility value for each cardiovascular event was then multiplied by the
baseline quality of life experienced by people with IC for each artery (e.g. 0.76 x baseline). The
difference between this value and the baseline quality of life was used to inform the decrease in
quality of life associated with each event. It was assumed that the quality of life decrement in the
years following a cardiovascular event is half that experienced in the first year. Each calculation was
performed using a probabilistic simulation (n= 20, 000). Simulated absolute mean values and mean
utility decrements are summarised in Table 71. In the model, each utility decrease was divided by
four to account for the three month cycle length.
Quality of life following amputation
The quality of life associated with amputation was obtained from a cost-utility analysis by Sculpher et
al 1996.128 This analysis estimated that the utility for someone with an amputation above the knee is
0.20 (0.00 – 0.40) and 0.61 (0.41 – 0.81) for below the knee. It has previously been estimated that
52% of amputations are above the knee. An overall utility value for people who have had an
amputation was estimated by assigning a distribution to each above- and below- the knee utility
value, applying this proportional estimate, and running a probabilistic simulation. The resulting value
of 0.396 (0.264 – 0.546) was used to represent the average quality of life of people who have had an
amputation.
Table 71: Simulated mean utility and mean utility decrements compared to baseline
Utility associated with each health state Corresponding utility decrease from baseline
Health state Mean SE 95% CI Mean SE 95% CI
Final version
176
PAD
Aorto-iliac arteries
IC (baseline) 0.580 0.048 0.490 – 0.674
MI 0.441 0.038 0.370 – 0.515 -0.139 0.016 -0.171 to -0.111
Post MI 0.510 0.42 0.430 – 0.593 -0.070 0.008 -0.086 to -0.055
Stroke 0.365 0.038 0.293 – 0.442 -0.215 0.029 -0.276 to -0.162
Post stroke 0.472 0.041 0.396 – 0.553 -0.108 0.015 -0.138 to -0.081
CLI 0.350 0.051 0.253 – 0.454 -0.231 0.070 -0.367 to -0.094
Amputation 0.396 0.072 0.264 - 0.546 -0.185 0.086 -0.349 to -0.009
Femoro-popliteal arteries
IC (baseline) 0.573 0.044 0.489 – 0.659
MI 0.435 0.35 0.369 – 0.505 -0.138 0.015 -0.168 to 0.110
Post MI 0.504 0.039 0.430 – 0.581 -0.069 0.007 -0.084 to -0.055
Stroke 0.360 0.036 0.292 – 0.434 -0.213 0.028 -0.271 to -0.162
Post stroke 0.467 0.038 0.395 – 0.542 -0.106 0.014 -0.136 to -0.081
CLI 0.350 0.051 0.253 – 0.454 -0.223 0.068 -0.356 to -0.092
Amputation 0.396 0.072 0.264 - 0.546 -0.177 0.084 -0.546 to -0.264
Final version
177
PAD
Table 72: Mean quality of life and mean difference in change between time points
Angioplasty with
Unsupervised Supervised selective stent + Angioplasty with Angioplasty with
Mean difference in change
exercise exercise supervised selective stent primary stent
exercise
Mean SE Mean SE Mean SE Mean SE Mean SE Interval Mean SE
Weighted average of Nicolai 2010, Cheetham 2004, Savage 2001
Baseline 0.636 0.017 0.672 0.014
3 months 0.691 0.017 0.709 0.015 Baseline to 3 months -0.021 0.033
6 months 0.692 0.015 0.732 0.016 3 months to 6 months 0.026 0.032
Greenhalgh 2008 (Aorto-iliac)
Baseline 0.426 0.012 0.419 0.012
3 months 0.422 0.008 0.461 0.009 Baseline to 3 months 0.077 0.020
24 months 0.451 0.017 0.507 0.014 12 month to 24 months -0.059 0.051
Greenhalgh 2008 (Femoro-popliteal)
Baseline 0.451 0.008 0.466 0.007
9 months 0.456 0.006 0.479 0.006 6 months to 9 months -0.001 0.013
24 months 0.458 0.009 0.486 0.009 12 month to 24 months 0.014 0.028
Spronk 2009 (Aorto-iliac & Femoro-popliteal)
Baseline 0.690 0.024 0.660 0.023
Final version
178
PAD

Bosch 1999 (Aorto-iliac)
Baseline 0.461 0.154 0.459 0.204
Mean difference in change = change in utility between time points within one trial arm subtracted from the change in the same time interval in the other trial arm. A positive value indicates an
improvement in quality of life in the trial arm in the right column of each intervention pair.
Final version
179
PAD
Costs
As in the model comparing unsupervised to supervised exercise, the cost of supervised exercise was
based on estimates of resource use informed by expert opinion and unit costs obtained from the
2010 PSSRU. A breakdown of the assumptions and unit costs used to calculate per-patient cost of a
supervised exercise programme are provided in Table 37 (section 9.4.8.2).
Endovascular intervention costs were obtained from the most recent 2009/2010 NHS Reference
Costs. The GDG estimated that approximately 5% of angioplasty procedures performed as a primary
strategy for people with intermittent claudication are non elective and that 10% of angioplasty
procedures performed as a secondary strategy are unplanned, and that 55% of amputations
preformed for people with CLI would be performed as emergency non elective procedures.
In the absence of recent relevant estimates of the cost of post-amputation care in the literature, the
GDG provided estimates of resource use based on their experience and the expertise of colleagues
around the country. These resources were grouped according to those that occur in the first year
after amputation and those occurring in subsequent years.
Table 73: Costs and cost-related variables

Point
Parameter estimate Value range Source
Cost of CV events
Initial MI (first 3 months) £4, 792 £3, 853 – £5, 731 Hypertension guideline 201178
Post nonfatal MI £141 £113 – £169 Hypertension guideline 201178
(subsequent 3 month
cycles)
Initial stroke (first 3 £9, 630 £7, 743 – £11, 517 Hypertension guideline 201178
months)
Post nonfatal stroke £559 £449 – £669 Hypertension guideline 201178
(subsequent 3 month
cycles)
Unsupervised and supervised exercise intervention cost
Unsupervised exercise £0 NA Expert opinion
Supervised exercise £288 £232 – £345 Expert opinion
Angioplasty with primary and selective stent intervention cost
Diagnostic imaging £90 £53 - £102 NHS Reference Costs 2009/1049
Stent (bare metal) £550 £450 - £650 Expert opinion
Primary angioplasty with £3, 661 £2, 204 - £4, 480 NHS Reference Costs 2009/1049
no complications
Primary angioplasty with £9, 367 £2, 200 - £14, 270 NHS Reference Costs 2009/1049
major complications
Secondary angioplasty £3, 695 £2, 206 - £4, 524 NHS Reference Costs 2009/1049
with no complications
Secondary angioplasty £9, 385 £2, 329 - £14, 154 NHS Reference costs 2009/1049
with major complications
Proportion of patients receiving stents (selective stent)
Aorto-iliac 35.2% 28.5% - 42.9% Based on included RCTs108,113,118
Femoro popliteal 16.2% 10.5% - 24.4% Based on included RCTs101,102,120,121,124
Average number of stents used where stents are placed
Final version
180
PAD
Aorto-iliac 2 NA Expert opinion

Femoro-popliteal 2 NA Expert opinion
Bypass intervention cost
Bypass with no/major £5, 988 £4, 417 - £7, 025 NHS Reference Costs 2009/1049
complications
Bypass with major £7, 139 £5, 185 - £8, 641 NHS Reference Costs 2009/1049
complications
Amputation procedural cost
Cost of amputation £9, 224 £6, 862 - £10, 481 NHS Reference Costs 2009/1049
without major
complications
Cost of amputation with £15, 001 £7, 862 - £18, 600 NHS Reference Costs 2009/1049
major complications
Probability of major 14.3% 12.2% - 16.6% Aulivola 2004115
complications
Amputation cost of care in first year following amputation and each subsequent year
Cost of care during first £28, 270 £25, 499 - £31, 040 Expert opinion
year
Annual cost of care in £23, 502 £21, 199 - £25, 806 Expert opinion
subsequent years
9.4.8.3 Results
Aorto-iliac artery
After excluding strategies which are dominated or extendedly dominated (Figure 8), the results of
the analysis show that supervised exercise followed by angioplasty with selective stent placement
(strategy 5) is the most cost-effective treatment strategy for people with IC at a cost of £16, 289 per
QALY. Although angioplasty with selective stent followed by angioplasty with selective stent (strategy
8) results in the greatest QALY gain, the incremental cost per QALY is greater than that which is
considered cost-effective by NICE (Table 76). The cost effectiveness acceptability frontier shows that
at a threshold of between £20, 000 and £30, 000, strategy 5 is the option with the greatest
probability of being cost effective (Figure 9).
Final version
181
PAD
Figure 8: Cost effectiveness plane: Aorto-iliac artery
Aorto-iliac artery Unsupervised exercise followed by supervised

12000
exercise
Supervised exercise followed by supervised
exercise
Supervised exercise followed by selective stent
10000
Selective stent followed by selective stent
Selective stent followed by supervised exercise

8000
Unsupervised exercise followed by selective

Costs (£)
stent
Unsupervised exercise followed by bypass
6000
Supervised exercise followed by bypass
4000
Primary stent followed by supervised exercise
Selective stent + supervised exercise
2000 Primary stent followed by selective stent
Selective stent followed by bypass
0 Primary stent followed by bypass

4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8
Not Dominated
Effectiveness (QALYs)
Figure 9: Cost effectiveness acceptability frontier: Aorto-iliac-artery
Aorto-iliac artery
1
0.9
Prob. Cost-Effective (Optimal Strategy Only)
0.8
Unsupervised exercise followed by
0.7 supervised exercise
0.6
Supervised exercise followed by
0.4
Supervised exercise followed by selective
0.3 stent
0.2
0.1
0
0 5000 10000 15000 20000 25000 30000
Cost effectiveness threshold (£)
Table 74: Probabilistic base case results without dominated options: Aorto-iliac artery
Total Incremental
Strategy Total Cost Incremental Cost QALYs QALYs Cost effectiveness
1 £4, 046 Baseline 4.415 Baseline Baseline
4 £4, 263 £217 4.506 0.091 2, 387
Final version
182
PAD
5 £5, 411 £1, 147 4.576 0.070 £16, 289

8 £9, 661 £4, 250 4.716 0.140 £30, 408
Femoro-popliteal artery
The results of the analysis in the femoro-popliteal artery show that supervised exercise followed by
angioplasty with selective stent placement (strategy 5) is also the most cost-effective treatment
strategy at a cost of £16, 024 per QALY (Figure 10). In this artery, angioplasty with selective stent
followed by angioplasty with selective stent (strategy 8) also results in the greatest QALY gain, but
the incremental cost per QALY is greater than that which is considered cost-effective by NICE (Table
75). The cost effectiveness acceptability frontier shows that at a threshold of between £20, 000 and
£30, 000, strategy 5 is the option with the greatest probability of being cost effective (Figure 11).
Final version
183
PAD
Figure 10: Cost effectiveness plane: Femoro-popliteal artery
9000
Unsupervised exercise followed by supervised exercise
8000
Supervised exercise followed by supervised exercise
Supervised exercise followed by selective stent

7000
Selective stent followed by selective stent

6000
Unsupervised exercise followed by selective stent
5000 Selective stent followed by supervised exercise

Cost (£)
Unsupervised exercise followed by bypass

4000
Supervised exercise followed by bypass
3000 Primary stent followed by supervised exercise
Primary stent followed by selective stent

2000
Primary stent followed by bypass
1000 Selective stent followed by bypass
Selective stent + supervised exercise

0
4.35 4.4 4.45 4.5 4.55 4.6
Not Dominated
Effectiveness (QALYs)
Figure 11: Cost effectiveness acceptability frontier: Femoro-popliteal artery
1
0.9
Prob. Cost-Effective (Optimal Strategy Only)
0.8
Unsupervised exercise followed by
0.6
Supervised exercise followed by

0.5
supervised exercise
0.4
0.3 Supervised exercise followed by

selective stent
0.2
0.1
0
0 5000 10000 15000 20000 25000 30000
Cost effectiveness threshold (£)
Table 75: Probabilistic base case results without dominated options: Femoro-popliteal artery
Incremental Total Incremental Cost
Strategy Total Cost Cost QALYs QALYs effectiveness
Final version
184
PAD
1 £4, 059 Baseline 4.374 Baseline Baseline

4 £4, 276 £217 4.466 0.092 £2, 362
5 £5, 378 £1, 102 4.534 0.069 £16, 024
8 £6, 603 £1, 225 4.572 0.037 £32, 898
Sensitivity analyses
A wide range of sensitivity analyses were undertaken to test the effect of different inputs and
assumptions on the results of the model. These analyses showed that the results were subject to a
high degree of uncertainty and the conclusion was sensitive to many of the key assumptions used to
inform the model. In particular, the results were sensitive to the assumption that exercise reduces
the risk of mortality in people who are active. By reducing the assumed increase life expectancy
associated with activity, a primary selective stent strategy becomes more effective in comparison.
Under this sensitivity analysis, selective stent followed by selective stent is the most cost effective
option in both arteries. The results of the model are also sensitive to the assumption that the change
in quality of life observed at the end of the trial period persists over a person’s lifetime so long as
they do not experience a recurrence of symptoms, and in those undertaking exercise intervention,
they remain active.
Limitations and interpretation
This model was developed based on a combination of best available clinical evidence and expert
opinion. It is directly relevant to the treatment of people with IC in England and Wales. It was built
probabilistically to account for the uncertainty surrounding each parameter. The results of the
analysis reflect the overall uncertainty in the treatment decision for an average population who are
suitable for all of the evaluated interventions.
The model was developed on the assumption that secondary interventions are associated with the
same relative risk of mortality and morbidity as those observed in primary procedures. In practice,
the GDG indicated that there are many risk factors or clinical features which may differentially affect
the outcome of secondary interventions. For example, a patient who did not benefit from or dropped
out of a supervised exercise programme is unlikely to benefit from a secondary course in the same
way as someone who has had a positive outcome or no previous experience of the same programme.
Similarly, secondary procedures at the same site may have an increased risk of failure. Many factors
including anatomic disease extent and clinical presentation, patient preference, and patient co-
morbidities will influence treatment options which are most appropriate for individual patients. This
model is not intended as a substitute to expert clinical judgement; patients must be considered on an
individual basis where there are factors which may affect the expected outcome.
The model was designed to address questions set by the guideline scope. Different methods of post
operative management were not included in the scope of the guideline and were therefore not
included in the model. Similarly, specific pre-operative characteristics were not accounted for. With
respect to exercise interventions, the clinical review was not designed to distinguish between trials of
varying length, duration or exercise intensity. As such, it is not possible to determine whether
certain types of supervised programmes are more cost effective than others. For this guideline, the
definition of each type of exercise programme was based on a simple average of studies included in
the clinical review. The supervised exercise programme described by this method was also found to
match programmes familiar to the GDG.
Currently, no published RCT data exist to inform the relative risk of cardiovascular events and
mortality in people who exercise compared to those who do not in people with IC. The data used in
this model was obtained from two meta-analyses of trials conducted in two different populations:
Final version
185
PAD
people with CHD who had experienced MI or coronary revascularisation and a mixed population of
people who had and had not had a stroke.
Limited published data was available to inform the impact of each type of exercise programme on
quality of life beyond one year. Although this data was not comparative, it suggested that quality of
life is maintained in those who continue to exercise. It was also assumed that changes in quality of
life observed in people undergoing endovascular treatment is maintained so long as symptom
progression (either to claudication of CLI) does not occur. This was a key assumption of the analysis.
If this assumption is removed from the model, none of the evaluated interventions are effective
enough to justify their cost in the aorto-iliac artery and the baseline intervention should be
prescribed. In the femoro-popliteal artery, removing this assumption results in selective stent
followed by supervised exercise is the most cost effective. Because the long-term effect of these
interventions is not known, it is not possible to know which scenario represents the most likely long
term outcome. More long-term research into the effects of these treatments is needed.
9.4.9 Clinical evidence statements
9.4.9.1 Best medical treatment compared to best medical treatment plus angioplasty (see section 9.4.2 for
clinical evidence)
Intermittent claudication due to femoro-popliteal or aorto-iliac disease:
Best medical treatment with angioplasty was significantly better than best medical treatment alone
for:
 Maximum walking distance at 3 months and 1 year [1 study, 56 participants, low quality
evidence]86
 Maximum walking distance at 2 years [1 study, 56 participants, low quality evidence]85
 Pain free walking distance at 3 months and 1 year [1 study, 56 participants, low quality
evidence]86
 Pain free walking distance at 2 years [1 study, 56 participants, low quality evidence]85
 ABPI at 3 months and 1 year [1 study, 56 participants, moderate quality evidence]86
 ABPI at 2 years [1 study, 56 participants, moderate quality evidence]85
performed:
 No complications at 1 year [1 study, 28 participants, low quality evidence]86
 No re-interventions at 1 year [1 study, 28 participants, low quality evidence]86
Best medical treatment with angioplasty was significantly better than best medical treatment alone
for:
 ABPI at 6 months [1 study, 59 participants, low quality evidence]88
There was no statistically significant difference between best medical treatment with angioplasty and
best medical treatment for:
 Mortality at 2 years [1 study, 59 participants, very low quality evidence]87
Final version
186
PAD
performed:
 No major complications at 6 months [1 study, 29 participants, low quality evidence]88
 No re-interventions at 6 months [1 study, 29 participants, low quality evidence]88
 One re-intervention in 29 people at 2 years [1 study, 29 participants, low quality evidence]87
9.4.9.2 Best medical treatment with supervised exercise and angioplasty compared to best medical
treatment with supervised exercise (for clinical evidence see 9.4.3)
There was no statistically significant difference between best medical treatment with supervised
exercise and angioplasty compared to best medical treatment and supervised exercise for:
 Pain free walking distance at 2 years [1 study, 23 participants, very low quality evidence]89
 Compliance with the exercise programme [1 study, 34 participants, very low quality evidence]89
performed
 Quality of life increased for both best medical treatment with supervised exercise and angioplasty
compared to best medical treatment and supervised exercise at 6 months and 1 year [1 study, 23
participants, low quality evidence]89
 People who had best medical therapy with supervised exercise and angioplasty had a high
maximum walking distance at 2 years [1 study, 23 participants, low quality evidence]89
 Four of 19 people had complications following the angioplasty [1 study, 19 participants, low
quality evidence]89
Best medical treatment with supervised exercise and angioplasty was significantly better than best
medical treatment and supervised exercise for:
 Pain free walking distance at 2 years [1 study, 71 participants, moderate quality evidence]89
exercise and angioplasty compared to best medical treatment and supervised exercise for:
 Compliance with the exercise programme [1study, 93 participants, very low quality evidence]89
 Withdrawal rates at 3 months [1 study, 118 participants, very low quality evidence]90
performed
 Quality of life increased for both best medical treatment with supervised exercise and angioplasty
compared to best medical treatment and supervised exercise at 6 months and 1 year [2 studies,
189 participants, low quality evidence]89,90
 People who had best medical therapy with supervised exercise and angioplasty had a high
maximum walking distance at 2 years [1 study, 71 participants, low quality evidence]89
 Six of 48 people had complications following the angioplasty [1 study, 48 participants, low quality
evidence]89
 No complications were reported at 3 months [1 study, 58 participants, low quality evidence]90
 0 out of 58 people had re-interventions at 1 year in the angioplasty group [1 study, 58
Final version
187
PAD
9.4.9.3 Best medical treatment with supervised exercise and angioplasty compared to best medical
treatment with angioplasty (for clinical evidence see section 9.4.4)
Best medical treatment with supervised exercise and angioplasty was statistically significantly better
than best medical treatment and angioplasty for:
 Maximum walking distance at 6 months [1 study, 61 participants, low quality evidence]96
 Pain free walking distance at 3 months [1 study, 60 participants, low quality evidence]96
 Pain free walking distance at 6 months [1 study, 61 participants, low quality evidence]96
 Re-intervention at 12 months [1 study, 118 participants, very low quality evidence]95
exercise and angioplasty compared to best medical treatment and angioplasty for:
 Major adverse events at 6 months [1 study, 70 participants, very low quality evidence]96
 Withdrawal from treatment at 6 months [1 study, 70 participants, very low quality evidence]96
No clinical evidence was found for people with IC due to femoro-popliteal disease.
9.4.9.4 Angioplasty compared to supervised exercise (for clinical evidence see section 9.4.5)
Supervised exercise was significantly better than angioplasty for:

 Maximum walking distance at 6 months [1 study, 150 participants, moderate quality evidence]91
 Maximum walking distance at 1 year [1 study, 150 participants, low quality evidence]91
 Pain free walking distance at 6 months and at 1 year [1 study, 150 participants, low quality
evidence]91
Angioplasty was significantly better than supervised exercise for:

 ABPI at rest at 6 months [1 study, 150 participants, moderate quality evidence]91
 ABPI at rest at 1 year [1 study, 150 participants, low quality evidence]91
 ABPI after exercise at 6 months [1 study, 150 participants, low quality evidence]91
 ABPI after exercise at 1 year [1 study, 150 participants, moderate quality evidence]91
There was no statistically significant difference between angioplasty and supervised exercise for:
 Number of people who doubled their maximum walking distance at 3 months [1 study, 31
participants, very low quality evidence]93
 Number of people who doubled their maximum walking distance at 1 year [1 study, 12
 Withdrawal at 3 months [1 study, 120 participants, very low quality evidence]90
Final version
188
PAD
performed
 Quality of life increased for both angioplasty and supervised exercise at 3 months [1 study, 120
 Quality of life increased for both angioplasty and supervised exercise at 6 months and 1 year [1
study, 150 participants, low quality evidence]91
performed:
 Eleven of 95 people had complications following angioplasty at 1 year [2 studies, 95 participants,
low quality evidence]91,93
 Five of 75 people had re-intervention following angioplasty at 6 months [1 study, 75 participants,
low quality evidence]91
 Eight of 95 people had re-intervention following angioplasty at 1 year [2 studies, 95 participants,
low quality evidence]91,93
 Eight of 16 people were good attenders for exercise (attended an average of > 1 session per
week) to exercise at 6 months [1 study, 16 participants, low quality evidence]93
Intermittent claudication due to aorto-iliac and femoro-popliteal disease:
performed:
 Three of 30 people had re-intervention following angioplasty at 15 months [1 study, 30
 Three of 26 people were exercising daily at 5-6 years [1 study, 26 participants, low quality
evidence]92
 Three of 26 people were more exercising than twice a week at 5-6 years [1 study, 26 participants,
performed:
 9 out of 60 people had re-interventions at 1 year in the angioplasty group [1 study, 60
9.4.9.5 Bypass surgery compared to supervised exercise
Intermittent claudication due to aorto-iliac and femoro-popliteal disease (for clinical evidence see
section 9.4.6):
Bypass surgery was significantly better than supervised exercise for:

 Pain free walking distance at 1 year [1 study, 50 participants, low quality evidence]94
There was no statistically significant difference between bypass surgery and exercise for:
 Mortality at 1 year [1 study, 50 participants, very low quality evidence]94
performed
Final version
189
PAD
 Six of 25 people had complications following bypass surgery at 30 days [1 study, 25 participants,
 Three of 25 people had re-interventions following bypass surgery at 30 days [1 study, 25
 Four of 25 people withdrew from the exercise programme [1 study, 25 participants, low quality
evidence]94
9.4.9.6 Angioplasty compared to bypass surgery (for clinical evidence see section 9.4.7)
Bypass surgery was significantly better than angioplasty for:

 ABPI after treatment (time point not specified) [1 study, 263 participants, moderate quality
evidence]104
Angioplasty was significantly better than bypass surgery for:

There was no statistically significant difference between angioplasty and bypass surgery for:
 Mortality at 30 days, 3 months, 1 year and 2 years [1 study, 263 participants, very low quality
evidence]104
 Amputation at post procedure and 2 years [1 study, 263 participants, very low quality evidence]104
 Amputation at 4 years [1 study, 118 participants, very low quality evidence]105
 Complications post procedure [1 study, 263 participants, very low quality evidence]104
 Re-intervention at 2 years [1 study, 263 participants, very low quality evidence]104

 ABPI at 1 year [1 study, 41 participants, low quality evidence]106
There was no difference between angioplasty and bypass surgery for:

 Mortality at 30 days [2 studies, 101 participants, moderate quality evidence]106,107
 Mortality at 1 year [2 studies, 127 participants, very low quality evidence]101 106
 Mortality at 2 years and 4 years [1 study, 86 participants, very low quality evidence]103
 Amputation at 1 year [3 studies, 196 participants, very low quality evidence]106 101 107
 Amputation at 4 years [2 studies, 173 participants, very low quality evidence]102,105
 Minor complications post procedure [2 studies, 141 participants, very low quality evidence]106 103
 Major adverse events at 1 year [1 study, 55 participants, very low quality evidence]107
 Minor adverse events at 1 year [1 study, 55 participants, very low quality evidence]107
 Re-intervention at 1 year [2 studies, 155 participants, very low quality evidence]101 107
 Re-intervention at 2 years and 4 years [1 study, 100 participants, very low quality evidence]103
Evidence statement for individual studies where meta-analysis was not possible – no statistical
analysis performed:
Final version
190
PAD
 One study found that in people with IC due to femoro-popliteal disease, ABPI at 1 year was higher
in those that had bypass surgery compared to those that had angioplasty [1 study, 100
 One study found that in people with IC due to femoro-popliteal disease, ABPI at 2 years was
higher in those that had bypass surgery compared to those that had angioplasty [1 study, 100
9.4.10 Economic evidence statements

None of the studies identified in the economic literature search included all comparators:
 One RCT-based analysis suggested that supervised exercise is more cost effective than angioplasty
with selective stent placement [partially applicable with minor limitations]100
 One decision analytic model suggested that primary angioplasty for those who are suitable is
more cost-effective than supervised exercise alone [partially applicable with potentially serious
limitations]111
 One decision analytic model suggested that unsupervised exercise is more cost-effective than
both exercise followed by angioplasty and angioplasty followed by angioplasty [partially
applicable with potentially serious limitations]109
 Two decision analytic models suggested that angioplasty with selective stent placement followed
by angioplasty with selective stent placement for long term treatment failure is more cost
effective than no revascularisation and revascularisation with primary stent placement [partially
applicable with minor limitations]108,112
 One decision analytic model suggested that depending on lesion type, graft type and indication,
either angioplasty or bypass were cost effective secondary treatments [partially applicable with
According to the results of the original economic model based on the current clinical evidence review
and GDG input, there is a high degree of uncertainty regarding the most cost-effective sequence of
interventions for the treatment of intermittent claudication. The results of the model suggest that
supervised exercise followed by angioplasty with selective stent placement has the highest
probability of being cost effective in both the aorto-iliac and femoro-popliteal artery [directly
applicable with minor limitations]. Please see Appendix L for a full description of the methods and
results of the original economic model.
12.Offer angioplasty for treating people with intermittent

claudication only when:
 advice on the benefits of modifying risk factors has been
reinforced (see recommendation 3) and
 a supervised exercise programme has not led to a satisfactory
improvement in symptoms and
 imaging has confirmed that angioplasty is suitable for the
person.
16.Offer bypass surgery for treating people with severe lifestyle-

limiting intermittent claudication only when:
 angioplasty has been unsuccessful or is unsuitable and
 imaging has confirmed that bypass surgery is appropriate for
Final version
191
PAD
the person.
Recommendations
Relative values of different The GDG agreed that population considered in these comparisons are a
outcomes relatively select group. These patients will have severe claudication that is not
responding to other measures i.e. secondary prevention, and exercise.
Mortality is always an important outcome, but death rates from intermittent

claudication would not be expected to be high and the patient numbers in
these studies were small or modest. Moreover, the follow-up period generally
extended to no more than a year. Where it was reported in these studies,
there was no significant difference in mortality between any of the
interventions.
The GDG were particularly interested in maximum walking distance and any
available quality of life data. Pain free walking distance was not thought to be
as clinically important an outcome since it is subject to more individual
variation and because practically speaking most patients will walk through
their pain for some distance. Measurement of improvement in ABPI is of
interest in that it is totally objective unlike either index of walking distance, but
the GDG were unanimous in regarding it as the least important of these
outcomes since it is not patient-centred.
For angioplasty, there was clear evidence of an improvement in both maximal

and pain-free walking distance when the endovascular intervention was
compared to no intervention (although people in both arms of the studies also
received best medical treatment). The evidence of benefit from angioplasty
was less clear when subjects in both arms also underwent supervised exercise.
Here the additional value of angioplasty was only apparent for the outcome
measure of pain-free walking distance, and only in the group of patients with
femoro-popliteal disease. However, the GDG noted that this difference was
seen at 2 years post-intervention whereas most of the studies did not follow
up patients for this length of time. This evidence indicates that angioplasty is
effective, but when directly compared to supervised exercise it produced less
improvement in both maximal and pain-free walking distance measured up to
one year. Improvements in ABPI favoured angioplasty over supervised
exercise, but as already noted the GDG regarded this as of lesser importance.
The comparison of bypass surgery with exercise was based on a single study
performed over 20 years ago94. The group of patients undergoing bypass
achieved a better maximal walking distance and a better pain free walking
distance at a 1 year time-point. A third group in this study took part in a
supervised exercise programme after bypass, and this combined intervention
produced a greater improvement than either alone.
Although bypass surgery and angioplasty were compared directly in a number

of studies, none of these reported maximal walking distance or pain free
walking distance. Some differences in ABPI were found but these were
inconsistent in that surgery appeared to produce more improvement at one
year whereas the measurement favoured the angioplasty group at 3 years.
Trade off between clinical Comparison of adverse effects in these studies was hard to synthesize, and
benefits and harms indeed the 3 interventions all have very different potential risks. Exercise
therapy is non-invasive, but carries the risk of exacerbating problems such as
those caused by chronic musculo-skeletal disease.
Final version
192
PAD
Angioplasty can produce local haematomas and these were reported in the
studies evaluated. Bypass surgery is associated with significant risks including
those of an anaesthetic, haematoma and wound infection, and these should be
discussed fully with the patient. The complication rates in the studies directly
comparing angioplasty to surgery were not significantly different, and nor were
re-intervention rates at the time points reported.
There is a problem with compliance to supervised exercise programmes, which

may limit their usefulness, partly related to the willingness and ability of
people to attend them. The studies reported that withdrawal rates were
related to distance from home and lack of transport.
Economic considerations An original economic model was developed to compare the cost-effectiveness
of several different intervention strategies for the treatment of people with IC.
The analysis combined evidence of effectiveness and quality of life collected as
part of the clinical review with current cost data. See Appendix L for a full
report of the methods used to inform this analysis.
According to the results of the model, supervised exercise followed by

angioplasty with selective stent placement for people with worsening
claudication is the most cost effective intervention pathway at a cost of
approximately £16, 000 per QALY gained.
If angioplasty does not represent a treatment option for people with IC,
supervised exercise followed by bypass surgery is the next most cost-effective
option.
The results are sensitive to several key assumptions of the model, such as the
assumption that exercise results in a reduced risk of mortality among those
who are active. For a full description of results and sensitivity analyses please
refer to Appendix L.
The GDG were satisfied with the robustness of the economic modelling, its
assumptions and sensitivity analysis.
Quality of evidence The quality of the evidence generally ranged from very low to low by GRADE
criteria, although occasional outcome measures were rated of better quality.
The evidence was downgraded for a variety of reasons, but typically on unclear
blinding, risk of bias and imprecision.
The GDG found some difficulty in comparing studies because the definitions of
best medical treatment differed, and it was not always clear what was included
in the background treatment applied to both study arms. The use of
medication such as statins has increased over the past 2 decades and patients
in studies performed at different time points cannot be assumed to have
similar treatment beyond the study interventions. They also noted that the
only available comparison of surgery against supervised exercise was over 20
years old, and that techniques for surgery, and for supporting care, have
changed in that time.
Other considerations Whilst the trials with arms that included combined surgery and supervised
exercise showed some benefit from combined treatment the economic
modelling suggests that simultaneous use is likely to be less cost effective than
sequential use. The GDG also took the view that supervised exercise would
Final version
193
PAD
normally be offered before considering endovascular or surgical interventional

and there were no trials that had specifically examined sequenced compared
to simultaneous treatments.
The GDG agreed that patients who required further intervention after
attempting supervised exercise should be considered first for an endovascular
procedure, based on the greater potential hazards of surgery and on the
health-economic analysis. However, they felt that their recommendations
should reflect the fact that in some people the nature of their arterial disease
will make them unsuitable for an endovascular procedure, and that in these
instances surgery could be considered.
9.5 Angioplasty with selective stent placement compared to

angioplasty with primary stent placement
What is the clinical and cost effectiveness of angioplasty with selective stent placement compared to
angioplasty with primary stent placement for the treatment of PAD in adults with intermittent
claudication?
A literature search was conducted for RCTs that compared the effectiveness of angioplasty with
selective stent placement compared to angioplasty with primary stent placement. No time limit was
placed on the literature search, and there were no limitations on sample size. Indirect populations
and emergency settings were excluded.
Two Cochrane reviews were identified129,130 which considered angioplasty without stents compared
to angioplasty with stents for the superficial femoral artery or for intermittent claudication. The
Cochrane reviews were not included or updated as they did not meet the review question protocol
defined by the GDG, which included all arteries of the leg and accepted papers with mixed
populations not only pure intermittent claudication populations. However they were used as a
source to ensure that studies identified in the Cochrane reviews which matched the current review
protocol had been considered for inclusion.
Fifteen studies of ten RCTs113,117-122,124,131-137 were identified which addressed the question and were
included in the review. The trials did not report outcome data for people with diabetes. There were
unit of analysis issues in some of the trials where data were analysed by limb or lesion rather than
per person randomised. These trials have been dealt with separately.
The quality and results of included studies are reported in the clinical evidence profiles (Table 76,
Table 77, Table 78 and Table 79. The forest plots for each clinical outcome are reported in Appendix
J.
Final version
194
PAD
Table 76: Clinical evidence profile: Angioplasty with selective stent placement compared to angioplasty with primary stent placement for intermittent
claudication due to aorto-iliac disease (person randomised data)
Angioplasty Quality
Angioplasty with
No of Risk of Other with primary Relative
Design Inconsistency Indirectness Imprecision selective stent Absolute
studies bias considerations stent (95% CI)
placement
placement
1118 RCT Serious(a) No serious No serious No serious None 0/136 0/136 not pooled not pooled MODERATE
inconsistency indirectness imprecision(b) (0%) (0%)
Mortality at 1 year
1118 RCT Serious(a) No serious No serious Very serious(c) None 2/136 1/143 RR 2.1 8 more per 1000 VERY LOW
inconsistency indirectness (1.5%) (0.7%) (0.19 to (from 6 fewer to
22.93) 153 more)
1118 RCT Serious(a) No serious No serious Very serious(c) None 2/136 1/136 RR 2 (0.18 7 more per 1000 VERY LOW
inconsistency indirectness (1.5%) (0.74%) to 21.8) (from 6 fewer to
153 more)
Mortality at 5 years
1133 RCT Serious(a) No serious No serious Very serious(c) None 22/136 21/143 RR 1.1 15 more per 1000 VERY LOW
1.91) 134 more)
Amputation at 5 years
1133 RCT Serious(a) No serious No serious Very serious(c) None 8/136 3/143 RR 2.8 38 more per 1000 VERY LOW
10.35) 196 more)
1113 RCT Serious(a) No serious No serious Serious(d) None 136 143 See Table 80 and Table 81 LOW
Final version
195
PAD
1113 RCT Serious(a) No serious No serious Serious(d) None 136 143 See Table 80 and Table 81 LOW
Quality of life at 2 years
1113 RCT Serious(a) No serious No serious Serious(d) None 136 143 See Table 80 LOW
1113 RCT Serious(a) No serious No serious No serious None 136 143 - MD 8 lower (22.25 MODERATE
inconsistency indirectness imprecision lower to 6.25
higher)
1113 RCT Serious(a) No serious No serious No serious None 136 143 - MD 2 higher MODERATE
inconsistency indirectness imprecision (12.48 lower to
16.48 higher)
Maximum walking distance at 2 years
1113 RCT Serious(a) No serious No serious No serious None 136 143 - MD 3 lower (18.96 MODERATE
inconsistency indirectness imprecision lower to 12.96
higher)
Adverse events at 30 days
4.69) 155 more)
7.36) 89 more)
1118 RCT Serious(a) No serious No serious Very serious(c) None 4/136 6/143 RR 0.7 (0.2 13 fewer per 1000 VERY LOW
inconsistency indirectness (2.9%) (4.2%) to 2.43) (from 34 fewer to
60 more)
Final version
196
PAD
1118 RCT Serious(a) No serious No serious Very serious(c) None 6/136 10/143 RR 0.63 26 fewer per 1000 VERY LOW
inconsistency indirectness (4.4%) (7%) (0.24 to (from 53 fewer to
1.69) 48 more)
ABPI at 3 months
1113 RCT Serious(a) No serious No serious No serious None 136 143 - MD 0.01 higher MODERATE
inconsistency indirectness imprecision (0.05 lower to 0.07
higher)
ABPI at 1 year
inconsistency indirectness imprecision (0.03 lower to 0.07
higher)
ABPI at 2 years
inconsistency indirectness imprecision (0.03 to 0.13
higher)
(d) No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
claudication due to aorto-iliac disease (limb/lesion randomised data)
Quality
Angioplasty with Angioplasty with
Design Inconsistency Indirectness Imprecision selective stent primary stent Absolute
placement placement
1133 RCT Serious(a) No serious No serious Very serious(b) None 33/169 33/187 RR 1.11 19 more per 1000 VERY LOW
Final version
197
PAD
1.71) 125 more)

Re-intervention at 6 – 8 years
1137 RCT Serious(a) No serious No serious Very serious(b) None 21/118 12/118 RR 1.75 76 more per 1000 VERY LOW
3.39) 243 more)
ABPI at 6 – 8 years
inconsistency indirectness imprecision (0.01 to 0.11
higher)
claudication due to femoro-popliteal disease (person randomised data)
Angioplasty with Angioplasty with Quality

placement placement
2132,134 RCT Very No serious No serious No serious None 0/95 0/164 not pooled not pooled LOW
serious(a) inconsistency indirectness imprecision(b) (0%) (0%)
1121 RCT Serious(c) No serious No serious No serious None 0/53 0/51 not pooled not pooled MODERATE
Mortality at 1 year (random effects)
4120,121,13 RCT Very Serious(d) No serious Very None 18/543 19/550 RR 0.73 (0.2 to 9 fewer per
5,136
serious(c) indirectness serious(e) (3.3%) (3.5%) 2.61) 1000 (from VERY LOW
28 fewer to
56 more)
Procedure related mortality at 1 year
Final version
198
PAD
1136 RCT Serious(c) No serious No serious No serious None 0/238 0/241 not pooled not pooled MODERATE
Amputation at 6 months
5120,121,13 RCT Very No serious No serious Very None 1/615 3/684 RR 0.57 (0.12 2 fewer per VERY LOW
4-136
serious(f) inconsistency indirectness serious(e) (0.16%) (0.44%) to 2.64) 1000 (from
4 fewer to
7 more)
1124 RCT Serious(c) No serious No serious Very None 1/52 0/46 RR 2.66 (0.11 - VERY LOW
inconsistency indirectness serious(e) (1.9%) (0%) to 63.75)
1131 RCT Serious(c) no serious no serious Serious(g) None 53 51 See Table 81 LOW
1131 RCT Serious(c) No serious No serious Serious(g) None 53 51 See Table 81 LOW
1121 RCT Serious(c) No serious No serious Serious(h) None 53 51 - MD 93 LOW
inconsistency indirectness lower
(214.24
lower to
28.24
higher)
Maximum walking distance at 6 months (no sd)
1119 RCT Serious(c) No serious No serious Serious(g) None 39 34 - not pooled LOW
Final version
199
PAD
1121 RCT Serious(c) No serious No serious Serious(h) None 53 51 - MD 120 LOW

inconsistency indirectness lower
(237.36 to
2.64 lower)
Maximum walking distance at 1 year (no sd)
2119,120 RCT Serious(c) No serious No serious Serious(g) None 61 50 - not pooled LOW
Maximum walking distance at 2 years (no sd)
Pain free walking distance at 30 days
1132 RCT Very No serious No serious Serious(h) None 23 30 - MD 83.2 VERY LOW
serious(i) inconsistency indirectness higher
(123.82
lower to
290.22
higher)
Major adverse events at 30 days
4119-122 RCT Serious(c) No serious No serious Very None 8/237 10/230 RR 0.79 (0.33 9 fewer per VERY LOW
inconsistency indirectness serious(e) (3.4%) (4.3%) to 1.93) 1000 (from
29 fewer to
40 more)
Minor adverse events at 30 days
Major adverse events at 1 year
1135 RCT Very No serious No serious Very None 11/131 6/135 RR 1.89 (0.72 40 more VERY LOW
serious(i) inconsistency indirectness serious(e) (8.4%) (4.4%) to 4.96) per 1000
(from 12
fewer to
176 more)
Final version
200
PAD
3120,121,13 RCT Serious(j) No serious No serious Very None 44/196 39/202 RR 1.17 (0.8 to 33 more VERY LOW
2
inconsistency indirectness serious(e) (22.4%) (19.3%) 1.71) per 1000
(from 39
fewer to
137 more)
1124 RCT Serious(c) No serious No serious Serious(h) None 28/52 17/46 RR 1.46 (0.93 170 more LOW
inconsistency indirectness (53.8%) (37%) to 2.29) per 1000
(from 26
fewer to
477 more)
Target lesion revascularisation at 6 months
1134 RCT Serious(c) No serious No serious No serious None 34/72 2/134 RR 31.64 (7.83 457 more MODERAT
inconsistency indirectness imprecision (47.2%) (1.5%) to 127.92) per 1000 E
(from 102
more to
1000 more)
Target lesion revascularisation at 1 year (random effects)
2134,136 RCT Serious(c) Very serious(d) No serious No serious None 79/310 38/375 RR 2.87 (1.25 189 more VERY LOW
indirectness imprecision (25.5%) (10.1%) to 6.6) per 1000
(from 25
more to
567 more)
ABPI at 30 days
1132 RCT Very No serious No serious Serious(h) None 23 30 - MD 0.06 VERY LOW
serious(i) inconsistency indirectness lower (0.17
lower to
0.05 higher)
ABPI at 6 months
1121 RCT Serious(c) No serious No serious Serious(h) None 53 51 - MD 0.08 LOW
lower to
0.01 higher)
Final version
201
PAD
ABPI at 6 months (no sd)

ABPI at 9 months
1135 RCT Very No serious No serious Serious(h) None 64 83 - MD 0.11 VERY LOW
serious(i) inconsistency indirectness lower (0.17
to 0.05
lower)
ABPI at 1 year random effects
3121,122,13 RCT Serious(c) Very serious(d) No serious No serious None 318 316 - MD 0.04 VERY LOW
6
indirectness imprecision lower (0.12
lower to
0.04 higher)
ABPI at 1 year (no sd)
2119,120 RCT Serious(c) No serious No serious Serious(g) None 61 50 - not pooled LOW
ABPI at 2 years
1124 RCT Serious(c) No serious No serious Serious(h) None 52 46 - MD 0.1 LOW
to 0.03
lower)
(a) 1 of 2 studies had unclear allocation concealment and blinding; 1 of 2 studies had unclear methodology.
(d) Unexplained heterogeneity.
(e) 95% CI crosses both MIDs.
(f) 4 of 5 studies had unclear allocation concealment and blinding; 1 of 5 studies had unclear methodology.
(g) No information on variability was given in the study, therefore calculation of standard deviation was not possible and the mean difference and CI were not estimable.
(h) 95% CI crosses one MID.
(i) Unclear methodology.
(j) 2of 3 studies had unclear allocation concealment and blinding; 1 of 3 studies had unclear methodology.
Final version
202
PAD
claudication due to femoro-popliteal disease (limb/lesion randomised data)
Quality
placement placement
1117 RCT Very No serious No serious Very serious(b) None 2/77 0/77 RR 5 (0.24 to - VERY LOW
(a)
serious inconsistency indirectness (2.6%) (0%) 102.47)
Mortality at 1 year
1117 RCT Very No serious No serious Very serious(b) None 7/77 12/77 RR 0.58 65 fewer per 1000 VERY LOW
serious(a) inconsistency indirectness (9.1%) (15.6%) (0.24 to 1.4) (from 118 fewer
to 62 more)
Amputation at 30 days
1117 RCT Very No serious No serious Very serious(b) None 4/77 2/77 RR 2 (0.38 to 26 more per 1000 VERY LOW
serious(a) inconsistency indirectness (5.2%) (2.6%) 10.6) (from 16 fewer to
249 more)
1117 RCT Very No serious No serious Very serious(b) None 16/77 28/77 RR 0.57 156 fewer per VERY LOW
serious(a) inconsistency indirectness (20.8%) (36.4%) (0.34 to 1000 (from 11
0.97) fewer to 240
fewer)
Major complications at 30 days
1117 RCT Very No serious No serious Very serious(b) None 6/77 7/77 RR 0.86 (0.3 13 fewer per 1000 VERY LOW
serious(a) inconsistency indirectness (7.8%) (9.1%) to 2.43) (from 64 fewer to
130 more)
ABPI time point not specified
1117 RCT Very No serious No serious No serious None 77 77 - MD 0.02 lower LOW
serious(a) inconsistency indirectness imprecision (0.08 lower to
Final version
203
PAD
0.04 higher)
(a) Unclear methodology.
Table 80: EQ-5D – Angioplasty with selective stent placement compared to angioplasty with primary stent placement
Angioplasty with selective stent placement Angioplasty with primary stent placement
Baseline 3 months 12 months 24 months Baseline 3 months 12 months 24 months
Bosch 1999 – Median (95% CI)
0.46 (0.15-0.75) 0.70 (0.20-1.00) 0.70 (0.15-1.0) 0.66 (0.15-1.0) 0.46 (0.20-0.75) 0.75 (0.15-1.00) 0.59 (0.19-1.0) 0.70 (0.09 -1.0)
Table 81: SF-36 individual domain results and mapped EQ-5D values – Angioplasty with selective stent placement compared to angioplasty with primary
stent placement
Bosch 1999 – Median (95% CI)
PF 45 (10 – 85) 80 (15-100) NR NR 85 (20-100) 40 (5-79) 85 (10-100) NR NR 70 (7-100)
RP 0 (0-100) 100 (0-100) NR NR 100 (0-100) 0 (0-100) 100 (0-100) NR NR 100 (0-100)
BP 45 (0-99) 78 (10-100) NR NR 80 (22-100) 45 (3-100) 90 (20-100) NR NR 78 (4-100)
GH 55 (10-90) 60 (10-95) NR NR 65 (15-95) 55 (15-94) 65 (15 -100) NR NR 63 (15-100)
V 50 (5-90) 70 (20-100) NR NR 65 (16-100) 50 (6-95) 70 (15-100) NR NR 65 (12-100)
SF 75 (13-100) 88 (13-100) NR NR 88 (25-100) 60 (0-100) 100 (14-100) NR NR 100 (0-100)
RE 67 (0-100) 100 (0-100) NR NR 100 (0-100) 100 (0-100) 100 (0-100) NR NR 100 (0-100)
MH 74 (20-100) 80 (28-100) NR NR 76 (30-100) 76 (13-100) 84 (28-100) NR NR 80 (6-100)
EQ-5D± NA NA NR NR NA NA NA NR NR NA
Sabeti 1983 – Median (IQR)
PF 45 (25) NR 62 (50) NR 67 (7) 50 (28) NR 60 (50) NR 65 (37)
RP 0 (50) NR 0 (100) NR 0 (100) 0 (75) NR 0 (100) NR 25 (75)
BP 22 (30) NR 52 (44) NR 46 (54) 30 (29) NR 51 (78) NR 52 (78)
GH 45 (28) NR 47 (38) NR 50 (41) 52 (27) NR 47 (38) NR 52 (29)
V 40 (23) NR 47 (33) NR 45 (36) 45 (25) NR 50 (24) NR 50 (30)
Final version
204
PAD
SF 75 (50) NR 88 (41) NR 88 (41) 88 (50) NR 88 (37) NR 100 (25)
RE 100 (100) NR 100 (100) NR 67 (100) 67 (100) NR 100 (67) NR 100 (33)
MH 64 (26) NR 66 (32) NR 60 (36) 64 (34) NR 72 (36) NR 72 (26)
±
EQ-5D NA NA NA NA NA NA NA NA NA NA
deviation; NR = not reported.
±Mapped based on algorithm (Equation1) reported by Ara and Brazier 2008 68
° Only the range was reported; probabilistic mapped values not estimable.
Final version
205
PAD
One published cost-effectiveness analyses were identified for this question. Bosch 1998108 developed
a decision model to evaluate the cost effectiveness of treating claudication due to iliac arterial
stenosis with primary stent placement, selective stent placement or angioplasty without stent
placement. This model assumes that 40% of patients undergoing angioplasty require selective stent
placement and that compared to angioplasty alone, the relative risk of failure associated with stent
placement is 0.61. The results of this model suggest that angioplasty with selective stent placement
for both primary and secondary treatment is more cost effective than both selective stent placement
followed by conservative management and primary stent placement followed by selective stent
placement. This conclusion was robust to a wide range of sensitivity analyses. The characteristics and
results of this study are summarised in Table 65 and Appendix I.
The same model (with American costs) was used in a later analysis by Bosch 2000.112 Based on the
results of their previous study (Bosch 1999113), which concluded that primary stent placement was
not cost-effective, the authors did not include angioplasty with primary stent placement as a
comparison in this analysis. Because this comparison was not relevant to the study question it was
excluded from the review. A full list of excluded studies is included in Appendix F.
Angioplasty with primary stent placement and angioplasty with selective stent placement were
included in the original model designed to assess the cost-effectiveness of different methods of
treatment for people with IC. Based on the results of this model, primary stent placement is not a
cost-effective option for the treatment of people with IC. It is both less effective and more expensive
than the majority of other treatment alternatives. Please refer to section 9.4.8.1 (page 166) for a
summary of the methods and results of this model and Appendix L for the full model write-up.
9.5.2.1 Clinical
Intermittent claudication due to aorto-iliac disease (person randomised data):
There was no difference between angioplasty with selective stent placement and angioplasty with
primary stents placement for:
 Mortality at 3 months [1 study, 272 participants, moderate quality evidence]118
There was no statistically significant difference between angioplasty with selective stent placement
and angioplasty with primary stents placement for:
 Maximum walking distance at 3 months, 1 year and 2 years [1 study, 279 participants, moderate
quality evidence]113
 Adverse events at 30 days [1 study, 279 participants, very low quality evidence]118
 Re-intervention at 3 months, 1 year and 2 years [1 study, 279 participants, very low quality
evidence]118
 ABPI at 3 months and 1 year [1 study, 279 participants, moderate quality evidence]113
Angioplasty with selective stent placement was significantly better than angioplasty with primary
stent placements for:
Final version
206
PAD
performed
 Quality of life increased for both angioplasty with selective stent placement and angioplasty with
primary stent placement at 3 months [1 study, 279 participants, low quality evidence]113
 Quality of life decreased for angioplasty with selective stent placement and remained the same
for angioplasty with primary stent placement quality of life at 1 year [1 study, 279 participants,
 Quality of life increased for angioplasty with selective stent placement and decreased for
angioplasty with primary stent placement at 2 years [1 study, 279 participants, low quality
evidence]113
Intermittent claudication due to femoro-popliteal disease (person randomised data):
Angioplasty with primary stent placements was significantly better than angioplasty with selective
stent placement for:
 Target lesion revascularisation at 6 months 1 study, 206 participants, moderate quality
evidence]134
 Target lesion revascularisation at 1 year [2 studies, 685 participants, very low quality
evidence]134,136
 ABPI at 9 months [1 study, 147 participants, very low quality evidence]135
 ABPI at 2 years [1 study, 98 participants, low quality evidence]124
There was no difference between angioplasty with selective stent placement and angioplasty with
primary stent placement for:
 Mortality at 30 days [2 studies, 259 participants, low quality evidence]132,134
 Mortality at 6 months [1 study, 104 participants, moderate quality evidence]121
 Procedure related mortality at 1 year [1 study, 479 participants, moderate quality evidence]136
 Amputation at 6 months [1 study, 104 participants, moderate quality evidence]121
 Minor adverse events at 30 days [1 study, 104 participants, moderate quality evidence]121
and angioplasty with primary stent placement for:
 Mortality at 1 year [4 studies, 1093 participants, very low quality evidence]120,121,135,136
 Amputation at 1 year [5 studies, 1299 participants, very low quality evidence]120,121,134-136
 Pain free walking distance at 30 days [1 study, 53 participants, very low quality evidence]132
 Major adverse events at 30 days [4 studies, 467 participants, very low quality evidence]120 121
119,122
 Major adverse event at 1 year [1 study, 266 participants, very low quality evidence]135
 Re-intervention at 1 year [3 studies, 398 participants, very low quality evidence]120,121,132
 Re-intervention at 2 years [1 study, 98 participants, low quality evidence]124
 ABPI at 30 days [1 study, 53 participants, very low quality evidence]132
 ABPI at 6 months [1 study, 104 participants, low quality evidence]121
Final version
207
PAD
 ABPI at 1 year [3 studies, 634 participants, very low quality evidence]122,124,136
analysis performed:
 Quality of life increased in most domains of SF-36 for both angioplasty with selective stent
placement and angioplasty with primary stent placement at 6 months and 1 year [1 study, 104
 Mean maximum walking distance at 6 months was higher in those that had angioplasty with
primary stent placements compared to those that had angioplasty with selective stent
placements [1 study, 73 participants, low quality evidence]119
 Mean maximum walking distance at 1 year was higher in those that had angioplasty with primary
stent placements compared to those that had angioplasty with selective stent placements [2
studies, 111 participants, low quality evidence]119,120
 The mean maximum walking distance at 2 years was higher in those that had angioplasty with
primary stent placement compared to those that had angioplasty with selective stent placements
[1 study, 98 participants, low quality evidence]124
 Mean ABPI at 6 months was higher in those that had angioplasty with primary stent placement
compared to those that had angioplasty with selective stent placements [1 study, 73 participants,
 Mean ABPI at 1 year was higher in those that had angioplasty with primary stent placement
compared to those that had angioplasty with selective stent placements [2 studies, 111
participants, low quality evidence]119,120
Intermittent claudication due to aorto-iliac disease (Limb/lesion randomised data):
stent placements for:
 ABPI at 6 to 8 years [1 study, 228 limbs, moderate quality evidence]137
 Re-intervention at 5 years [1 study, 356 limbs, very low quality evidence]133
 Re-intervention at 6 to 8 years [1 study, 236 limbs, very low quality evidence]137
Intermittent claudication due to femoro-popliteal disease (limb/lesion randomised data):
 Mortality at 30 days and 1 year [1 study,154 participants, very low quality evidence]117
 Amputation at 30 days [1 study, 154 participants, very low quality evidence]117
 Major complications at 30 days [1 study, 154 participants, very low quality evidence]117
 ABPI (time point not specified) [1 study, 154 participants, low quality evidence]117
stent placement for:
 Re-intervention at 1 year [1 study, 154 participants, very low quality evidence]117
9.5.2.2 Economic evidence statements

 One decision analytic model found that that in people with claudication due to iliac artery
stenosis, angioplasty with selective stent placement for both primary and secondary treatment is
Final version
208
PAD
more cost effective than both selective stent placement followed by conservative management
and primary stent placement followed by selective stent placement [partially applicable with
minor limitations]108
According to the results of the original economic model based on the current clinical evidence review
and GDG input, there is a high degree of uncertainty regarding the most cost-effective sequence of
interventions for the treatment of intermittent claudication. However, primary stent placement does
not appear to be a cost-effective option for the treatment of people with IC either as a first line or
second line therapy. It is both less effective and more expensive than the majority of other treatment
alternatives [directly applicable with minor limitations]. Please refer to section 9.4.8.1 (page 166) of
this guideline for a summary of the methods and results of this model and Appendix L for a full
description of the methods and results of the original economic model.
13.Do not offer primary stent placement for treating people with
intermittent claudication caused by aorto-iliac disease (except
complete occlusion) or femoro-popliteal disease.
14.Consider primary stent placement for treating people with

intermittent claudication caused by complete aorto-iliac
Recommendations occlusion (rather than stenosis).
Relative value of different Mortality and amputation are not frequent outcomes in the population
outcomes with intermittent claudication. The GDG felt that walking distance was the
most important outcome of those for which sufficient data was available.
However, the majority of the evidence showed no significant differences
between angioplasty with or without stents and stents alone for walking
distance, APBI, mortality, re-intervention, amputation and adverse effects.
Walking distance was improved at one time point and one intervention
site, but not in any other sub-analysis. There were also some differences in
ABPI results, but these did not consistently favour angioplasty with
selective or primary stenting.
Patency was not considered as a relevant outcome for the reasons detailed
in section 3.1.1 in methodology chapter.
Trade off between clinical The GDG were concerned that stents may give the operator the impression
benefits and harms that a procedure has been technically successful at the time the procedure
is performed, but noted that no consistent later benefit was demonstrated
in comparison with angioplasty.
The GDG considered that the routine use of stents as opposed to selective
use in conjunction with angioplasty carried the disadvantages of additional
cost, increased procedure time, and potential risks of additional
instrumentation.
Endovascular procedures carry a potential risk of causing embolisation of

material from the diseased artery which can cause blockage of smaller
arteries further down the leg. This is thought to be a greater risk with
complete occlusion of the aorto-iliac arteries than with stenosis or
occlusion in smaller vessels. There is also a risk of restenosis following
endovascular treatment and having foreign material such as a stent in the
artery may increase this risk, particularly in smaller vessels.
Final version
209
PAD
The GDG considered that it is generally accepted that stenting is

advantageous in terms of embolisation rates although the evidence
reviewed in these studies did not reflect this.
Economic considerations Although the GDG noted that there was little difference in outcomes
between selective and primary stent placement, for completeness primary
stent placement was included as a primary intervention in the original
economic model developed for this guideline. It was not included as a
secondary comparator.
The results of the model show that strategies which include primary stent
placement as a first-line intervention are both more expensive and less
effective than most other options (see Figure 8 and Figure 10). Primary
stent placement is therefore not a cost-effective strategy for the treatment
of IC in either the aorto-iliac or femoro-popliteal arteries. For a full report
of methods and results of the analysis please refer to Appendix L.
Quality of evidence The evidence was rated as low or moderate by GRADE criteria. The GDG
also highlighted that the data on ABPI and walking distance were short-
term and that evidence on the long term benefits would have been
extremely useful.
The GDG noted that the trial (Schillenger, 2006 121) which showed most of
the statistically significant differences was performed in a selective
population with intermittent claudication secondary to short arterial
lesions. The results may well not reflect the likely outcomes in longer, more
complex lesions.
Other considerations This comparison is about whether to place stent in all patients undergoing
an endovascular intervention for PAD, or only those in whom the operator
deems it necessary. Although the latter seems more open to error, the
former may be wasteful, and in this group of studies no clear evidence in
favour of primary stenting emerged, and the health economic data
suggests that this would not be a cost-effective strategy.
The GDG considered the difference between balloon expandable and self
expanding stents. They recognised that most current treatment is with self
expanding stents and that the evidence of benefit relates to SES alone,
which were the majority of trials used to derive the treatment effects in the
economic model. Exclusion of BES trials from the analysis did not change
the level of significance for any of the outcomes and resulted in increased
heterogeneity for some comparisons. It was therefore considered that
separating the trails was unnecessary and would not lead to a change in
conclusions or recommendations.
Primary stenting for femoro-popliteal disease or stenotic disease of the

aorto-iliac arteries is not standard UK practice and the GDG felt that there
was insufficient evidence to recommend a change to this situation.
Primary stents are currently used in aorto-iliac occlusion in the UK because

of concern about the risk of embolisation. The GDG recognised that they
had identified no evidence to justify this as routine, but also noted that
embolisation was not an endpoint specifically sought in these studies.
Final version
210
PAD
9.6 Bare metal compared to drug eluting stents

What is the clinical and cost effectiveness of bare metal stents compared to drug eluting stents for
the treatment of PAD in adults with intermittent claudication?
A literature search was conducted for RCTs that compared the effectiveness of bare metal stents to
drug eluting stents. No time limit was placed on the literature search, and there were no limitations
on sample size. Indirect populations and emergency settings were excluded.
One RCT136 was submitted during a call for evidence which addressed the question and were
included in the review. The trials did not report separate outcome data for people with diabetes.
NOTE: a second RCT (Rastan 2011)138 was also identified in the search, but was not included in the
evidence review because the patient population was not applicable to the UK setting (the GDG
considered that it is unusual to treat infra-geniculate disease for IC in the UK). However the study
was felt to be more relevant for the CLI population and was included in the analysis (see section
10.4).
The quality and results of included studies are reported in Table 82. Forest plots for each clinical
outcome are reported in Appendix J. NOTE: the outcomes of patency and target lesion
revascularisation were reported as percentages in the study and therefore data were unsuitable for
forest plots and could not be pooled in GRADE.
Final version
211
PAD
Table 82: Clinical evidence profile: Bare metal stents (BMS) compared to drug eluting stents (DES) for people with intermittent claudication due to
femoro-popliteal disease after angioplasty failure

Quality
Design Inconsistency Indirectness Imprecision BMS DES Absolute
All cause mortality at 12 months
1136 RCT Serious(a) No serious No serious Very serious(b) None 1/59 0/61 RR 3.10 (0.13 not pooled
inconsistency indirectness (1.7%) (0%) to 74.61) VERY LOW
Procedure / device related deaths at 12 months
1136 RCT Very No serious No serious No serious None 0/59 0/61 not pooled not pooled
serious(a) inconsistency indirectness imprecision(c) (0%) (0%) LOW
Patency at 1 year
1136 RCT Very No serious No serious Serious(d) None 73.0%(e) 89.9%(e) not pooled(e) not pooled
(a)
serious inconsistency indirectness VERY LOW
Patency at 2 years
1136 RCT Very No serious No serious Serious(d) None 62.7%(f) 81.2%(f) not pooled(f) not pooled
serious(a) inconsistency indirectness VERY LOW
Target lesion revascularisation (TLR) at 2 years
1136 RCT Very No serious No serious Serious(d) None 10.8%(g) 23.1%(g) not pooled(g) not pooled
serious(a) inconsistency indirectness VERY LOW
(a) Inadequate randomisation method, unclear blinding and allocation concealment.
(c) There were no events in either group.
(d) No numerator or denominator was given in the study, and therefore the calculation of the relative risk was not possible and the CI was not estimable.
(e) p=0.01 reported in the paper for BMS vs DES
(f) p<0.01 reported in conference presentation for BMS vs DES.
(g) p=0.05 reported in conference presentation for 'freedom from TLR' for BMS vs DES. No p-value reported for TLR.
Final version
212
PAD
Final version
213
PAD
No published cost-effectiveness analyses were identified for this question. In the absence of
published evidence, the GDG were presented with the current cost of bare metal and drug eluting
stents to aid decision making.
Vascular stents are excluded from the NHS reference cost for angioplasty and incur an additional cost
according to the number and type used per procedure. The unit cost of vascular stents was not
available from the NHS Supply Catalogue. A buyer for cardiology and radiology products at the NHS
Supply chain was asked to provide a list of prices for all vascular stents currently in use in England
and Wales. However, the GDG found that this list included many stents which are not used in
peripheral vascular surgery (such as cerebrovascular and cardiovascular stents) and did not include
many stents which are designed for use in peripheral vessels. As a result, this list was not used to
calculate average stent costs. Members of the GDG were then asked to provide prices from their
hospitals. Based on prices obtained by GDG members, the group estimated bare metal stents cost
approximately £550 and drug eluting stents approximately £900. The estimated cost of bare metal
stents was found to be consistent with that identified in an audit of NHS radiology departments
undertaken by Cox and Koutroumanos 2010.139
9.6.2.1 Clinical
Drug-eluting stents were significantly better than bare metal stents for:
 Patency at 1 year and two years [1 studyc, very low quality evidence]136
There was no statistically significant difference between bare metal stents and drug eluting stents
for:
 All cause mortality at 1 year [1 study, 120 participants, very low quality evidence]136
 Target lesion revascularisation at 2 years [1 studyd, 120 participants, very low quality evidence]136
There was no difference between bare metal stents and drug eluting stents for:
 Procedure / device related mortality at 12 months [1 study, 120 participants, low quality
evidence]136
9.6.2.2 Economic
No cost-effectiveness evidence was identified for this question.
c Exact number of participants used in the analyses for patency was not reported in these studies.
d p=0.05 reported in conference presentation for ‘freedom from TLR’ for BMS vs DES. No p-value reported for TLR.
Final version
214
PAD
15.Use bare metal stents when stenting is used for treating people
Recommendation with intermittent claudication.
Relative values of different The trial comparing bare metal versus drug eluting stents in the femoro-
outcomes popliteal circulation of people with IC was received through a call for evidence.
Data were reported on all cause mortality and procedure related mortality at
12 months, and also on the need for revascularisation. No significant
differences were noted for these clinical outcomes.
Target lesion revascularisation (TLR) at 24 months was also reported. The

difference favoured drug-eluting stents (DES) but this result was of borderline
statistical significance. Moreover, this was a single small study in which TLR
was not a pre-specified outcome.
The GDG discussed the issue of patency at length and the GDG recognised the
value of this measure was limited in the absence of associated evidence to link
patency to more clinically relevant outcomes. The major concern was that the
usefulness of patency as an outcome depended upon clear evidence to make
the link between patency and clinical outcomes of relevance to people with
PAD. The GDG noted that some treatments that are known to have an effect
upon symptoms in people with PAD have no effect upon patency. Their clinical
experience and knowledge of the literature suggests that it is common for
people to develop recurrent symptoms despite a patent segment of vessel or
to develop re-stenosis or re-occlusion without having recurrent symptoms.
They therefore considered that the results of treatment were far better
measured by outcomes of relevance to patients such as symptoms, quality of
life and the need for further interventions.
The only situation where the GDG considered that patency would be a
potentially useful outcome was where the two treatments being compared
were expected to have identical mechanical effects, such as in comparing
similar stents with and without drug elution. Even in this situation clinical
outcomes would be preferred where available and the usefulness of the
surrogate outcome would depend upon the availability of evidence to link this
to clinical outcomes which, for the reasons above, would need to be related to
the specific treatment. Therefore, the GDG considered patency for the bare
metal compared to drug eluting stents.
The data showed that drug-eluting stents offered better patency at the 12
month and 24 months time-point. The GDG did not feel that these results
provided robust evidence of clinical benefit of drug eluting stents on which to
make a positive recommendation.
Trade off between clinical The method of placement of the two forms of stents is identical, and therefore
benefits and harms the main potential adverse effects are also the same. No unexpected
difference emerged in the trial evidence.
Economic considerations There was no cost effectiveness evidence identified for this question. Drug
eluting stents are more expensive than bare metal stents.
If drug eluting stents were to reduce the need for reintervention then this may
potentially offset the additional cost of the stents. However the GDG did not
consider that there was sufficient evidence on which to base estimates of such
Final version
215
PAD
an effect and in the absence of robust evidence to suggest that clinical

outcomes are improved with the use of drug eluting stents, the GDG agreed
that the increased cost does not represent a cost effective use of NHS
resources.
Quality of evidence The evidence presented was categorised as low or very low by GRADE score.
The GDG noted the absence of data on walking distance. This would have been
of interest, although maximum walking distance measured within trials e.g.
treadmill test, is not necessarily a realistic measure of a patient’s walking
distance in real life circumstances.
Other considerations The GDG considered whether evidence of the use of different types of stent for
coronary artery disease offered any useful information for the treatment of IC
in the PAD population. However, they felt that the anatomical differences
between the two sites did not allow extrapolation from one to the other.
The GDG noted that drug eluting stents are amongst a number of new
technologies, such as drug coated balloons, cutting balloons, cryotherapy and
brachytherapy, which aim to reduce re-stenosis following endovascular
treatment. These other technologies have not been considered in the current
scope, but may prove to be of value in the future.
There was no evidence identified relating to aorto-iliac disease. The GDG were
concerned as to whether it was appropriate to generalise the results to all
anatomical sites and all types of drug eluting stent. However in the absence of
additional evidence to differentiate between these they considered that their
conclusions should apply to all sites and all kinds of DES.
In summary, the evidence shows a difference in patency favouring DES, a

borderline result for target vessel revascularisation at one time point, but no
data on clinical outcomes such as walking distance. As there is no strong
evidence of a difference in clinical outcomes between the two stent types and
drug eluting stents are more costly, the GDG formed a consensus judgement
that use of bare metal stents is the preferred option.
9.7 Autologous vein compared to prosthetic bypass

What is the clinical effectiveness of autologous vein versus prosthetic bypass for the treatment of
intermittent claudication in adults?
A literature search was conducted for RCTs that compared the effectiveness of autologous vein
versus prosthetic bypass grafting. No time limit was placed on the literature search, and there were
no limitations on sample size. Indirect populations and emergency settings were excluded. One
Cochrane review was identifieD140 which considered graft type in bypass surgery for femoro-popliteal
disease in both intermittent claudication and critical limb ischemia. The Cochrane review was not
included or updated as it did not meet the review question protocol defined by the GDG, which
included all arteries of the leg. However it was used as a source to ensure that studies identified in
the Cochrane review which matched the current review protocol had been considered for inclusion.
Final version
216
PAD
Two reports of one RCT141,142 were found which addressed the question and were included in the
review. None of the trials reported on subgroups for patients with diabetes as the main outcome.
The quality and results of included studies are reported in Table 83. Forest plots for each clinical
outcome are reported in Appendix J. No forest plot was available for perioperative mortality
(≤30days).
Final version
217
PAD
1 Table 83: Clinical evidence profile: Autologous vein compared to prosthetic bypass for intermittent claudication due to femoro-popliteal disease

Quality
No of Risk of Other Autologous Prosthetic Relative
Design Inconsistency Indirectness Imprecision Absolute
studies bias considerations vein bypass (95% CI)
1141 RCT Serious(a) No serious No serious Very None 24/75 18/76 RR 1.35 (0.8 to 83 more per 1000 VERY LOW
inconsistency indirectness serious(c) (32%) (23.7%) 2.28) (from 47 fewer to
303 more)
inconsistency indirectness serious(c) (2.7%) (2.6%) 7.01) (from 22 fewer to
158 more)
Perioperative minor adverse event
inconsistency indirectness serious(c) (5.3%) (3.9%) 5.83) (from 27 fewer to
191 more)
1142 RCT Serious(a) No serious No serious Very serious3 None 1/75 4/76 RR 0.25 (0.03 to 39 fewer per 1000 VERY LOW
inconsistency indirectness (1.3%) (5.3%) 2.21) (from 51 fewer to
64 more)
1141 RCT Serious(a) No serious No serious No serious None 5/75 16/76 RR 0.32 (0.12 to 143 fewer per MODERATE
inconsistency indirectness imprecision (6.7%) (21.1%) 0.82) 1000 (from 38
fewer to 185
fewer)
Final version
218
PAD
1 (a) Unclear randomisation procedure and no participant blinding.

2 (b) No events in either intervention.
3 (c) 95% CI crosses both MIDs.
Final version
219
PAD
No cost effectiveness studies were identified for this question.
The cost-utility analysis by Hunick et al 1995 (reported in Table 65) subgrouped the results of their
clinical analysis by graft material. Although the study was not designed to directly compare the cost-
effectiveness of one type of material to another, according to the results of the model, bypass
surgery using autologous vein grafts results in higher quality of life and lower cost than bypass
surgery using synthetic grafts.
The GDG also discussed the cost of autologous and prosthetic grafts in an NHS context. The group
considered that although the same NHS Reference Cost applies to patients undergoing both
procedures, prosthetic veins cost several hundred pounds, varying widely depending on graft length
and material (official cost estimates were not available from standard sources). However, the
procedure associated with prosthetic vein bypass is slightly shorter than that for autologous vein as
there is no need to harvest the vein. In addition, the average hospital stay is slightly less for
prosthetic vein bypass operations. However, autologous vein bypass is associated with a reduced
rate of infection and fewer complications. Based on the clinical evidence and clinical experience, the
GDG agreed that autologous vein bypass was likely to represent the least costly of the two
procedures. A formal cost estimation was not undertaken as it was thought that this was
unnecessary (as the most effective option was also thought to be the least costly) and time
consuming.
9.7.2.1 Clinical
Autologous vein was significantly better than prosthetic bypass for:

 Re-intervention at five year follow up [1 study, 151 participants, moderate quality evidence]141
There was no statistically significant difference between autologous vein and prosthetic bypass for:
 Mortality at 5 years, [1 study, 151 participants, very low quality evidence]141
 Re-intervention at 2 years [1 study, 151 participants, very low quality evidence]142
 Amputation rates at five years [1 study, 151 participants, very low quality evidence]141
 Perioperative minor adverse event [1 study, 151 participants, very low quality evidence]141
There was no difference between autologous vein and prosthetic bypass for:
 Mortality at 30 days [1 study, 151 participants, moderate quality evidence]141
9.7.2.2 Economic
No cost effectiveness studies were identified for this question.
Final version
220
PAD
17.Use an autologous vein whenever possible for people with

Recommendation intermittent claudication having infra-inguinal bypass surgery.
Relative values of different Re-intervention, complications and mortality were considered the important
outcomes outcomes for decision making for this question. Quality of life was also
considered important but no data was identified for this outcome. The GDG
did not expect amputation rates to be high within an IC population but
looked specifically at this outcome as a marker of success or failure of the
intervention.
There was discussion around use of other measures of patency, but the GDG
did not feel that these were as important as clinical success of an
intervention.
Although there was no difference between the graft types for most
outcomes, re-intervention rates tended to favour autologous grafts and this
difference was significant at the longest (5-year) time-point reported.
Trade off between benefits The GDG noted that the formal evidence suggested benefit from autologous
and harms vein grafts in terms of the need for re-intervention but did not show any
noteworthy difference in complication rates. There were slightly more peri-
operative complications with autologous grafts but the difference was not
statistically significant.
Current clinical practice within the UK has moved away from use of
prosthetic grafts because of a perception, with some support from
observational studies (not reviewed here), that prosthetic material is
associated with more infection. The risk of MRSA infection in prosthetic graft
has been linked with higher mortality rate. There was some concern that the
RCT evidence may not accurately reflect infection rates.
Economic considerations Although autologous vein bypass is associated with a slightly higher rate of
perioperative adverse events, which might have cost implications, conversely
prosthetic vein grafts are associated with a significantly higher re-
intervention rate. Indirectly, the economic model published by Hunick 1995
suggested that autologous grafts were more cost-effective. The GDG agreed
that prosthetic vein bypass grafts do not represent a cost effective use of
NHS resources for people undergoing infra-inguinal bypass surgery.
Quality of the evidence The GDG discussed some issues around the quality of the trials. It was noted
that the evidence presented was not recent and that no trials beyond 2003
were available. They also noted that the studies were underpowered for
some outcomes. However, they thought it unlikely that there would be any
support for a new randomised trial.
The GDG noted that the technology has advanced considerably since 2003
and that the bypass surgery is done less frequently because other
endovascular procedures can now be used successfully.
Other considerations The GDG recognise that by focussing on RCTs there is a risk of losing some
important data in terms of morbidity and mortality.
Although, there was no clear advantage for either autologous vein or
Final version
221
PAD
prosthetic graft, the GDG felt that where there were differences these
favoured autologous grafts, and this is supported by their clinical experience.
Final version
222
PAD
10 Management of critical limb ischaemia

10.1 Introduction: chapter overview
People with critical limb ischaemia (CLI) face an enormous cardiovascular risk and there is a 50%
mortality rate within 1 year of diagnosis. These patients also tend to be older and have significant co-
morbidities, which need to be optimised. People with CLI require prompt referral to specialist
services to be assessed for revascularisation. Delays in referral and treatment can result in poorer
outcomes for people with CLI including major amputation. People with critical limb ischaemia should
be encouraged to manage cardiovascular disease through the secondary prevention measures as
described in chapter 9.1.
Options for revascularisation include angioplasty or bypass surgery. These have been compared in
the previous chapter in the context of intermittent claudication, but require separate consideration
for people with CLI, in whom mortality and the risk of limb amputation are considerably greater.
There will be patients in whom revascularisation has not been possible or has been unsuccessful. In
such cases, patients may proceed to amputation. The extent to which effort should be made to avoid
amputation is open to some debate. Although it can be regarded as a failure of treatment it may be
in a patient’s best interest, if clinical assessment and supporting investigation suggest that attempts
at angioplasty or bypass are unlikely to succeed, to proceed straight to amputation. Trying to save
the limb in these circumstances may prolong the patient’s discomfort, delay eventual recovery, and
also entail unnecessary expense for the Health Service. It was originally intended that this chapter
would include a comparison of amputation with bypass and endovascular treatment, but in the
absence of data (see section 10.2.1 below) it was decided to consider amputation separately. This is
dealt with in chapter 12.
10.2 Angioplasty compared to bypass surgery

What is the clinical and cost effectiveness of angioplasty compared to bypass surgery or amputation
for the treatment of critical limb ischaemia in adults with PAD?
A literature search was conducted for RCTs that compared the effectiveness of angioplasty to bypass
surgery, and for RCTs and observational studies comparing angioplasty or bypass compared to
amputation. No time limit was placed on the literature search, and there were no limitations on
sample size. Indirect populations and emergency settings were excluded. One Cochrane review was
identified Fowkes, 200897 which considered bypass compared to other treatment for critical limb
ischaemia. The Cochrane review was not included or updated as it did not meet the protocol defined
by the GDG, which only compared bypass to angioplasty, where as the Cochrane compared bypass to
angioplasty and other interventions. However it was cross checked for included studies which
matched the review protocol.
Four relevant RCTs143 105,106 were included in the review. The trials did not report outcome data for
people with diabetes.
No RCTs or observational studies comparing angioplasty or bypass surgery to amputation were

identified.
Final version
223
PAD
The quality and results of included studies are reported in Table 84, Table 85 and Table 86. Quality of
life and mapped EQ-5D values are reported in Table 87 and Table 88. The forest plots for each clinical
outcome are reported in Appendix J.
Final version
224
PAD
Table 84: Clinical evidence profile: Angioplasty compared to bypass surgery for critical limb ischaemia due to aorto-iliac disease

Quality
No of Other Relative
Design Risk of bias Inconsistency Indirectness Imprecision Angioplasty Bypass Absolute
studies considerations (95% CI)
Limb salvage at 4 years
1105 RCT Serious(a) No serious No serious Very None 16/22 17/23 RR 0.98 15 fewer per VERY LOW
inconsistency indirectness serious(b) (72.7%) (73.9%) (0.69 to 1000 (from 229
1.4) fewer to 296
more)
Table 85: Clinical evidence profile: Angioplasty compared to bypass surgery for critical limb ischaemia due to femoro-popliteal disease

Quality
2106,143 RCT Serious(a) No serious No serious Very None 7/254 11/259 RR 0.65 15 fewer per VERY LOW
inconsistency indirectness serious(b) (2.8%) (4.2%) (0.26 to 1000 (from
1.64) 31 fewer to
27 more)
Mortality at 1 year
1106 RCT Serious(c) No serious No serious Very None 5/30 4/31 RR 1.29 37 more per VERY LOW
4.35) 80 fewer to
432 more)
Final version
225
PAD
1143 RCT Serious(d) No serious No serious Very None 131/224 119/228 RR 1.12 63 more per VERY LOW
1.32) 26 fewer to
167 more)
Amputation rate at 1 year
1106 RCT Serious(c) No serious No serious Very None 2/30 8/31 RR 0.26 191 fewer VERY LOW
inconsistency indirectness serious(b) (6.7%) (25.8%) (0.06 to per 1000
1.12) (from 243
fewer to 31
more)
Amputation free survival rate at 3 years
1143 RCT Serious(d) No serious No serious Very None 82/224 86/228 RR 0.97 11 fewer per VERY LOW
1.23) 91 fewer to
87 more)
Limb salvage rate at 4 years
1105 RCT Serious(e) No serious No serious Very None 10/11 10/16 RR 1.45 281 more VERY LOW
inconsistency indirectness serious(b) (90.9%) (62.5%) (0.95 to per 1000
2.22) (from 31
fewer to 763
more)
1143 RCT Serious(d) No serious No serious Serious(e) None 164 152 See Table 87 and Table LOW
Quality of life at 2 years
Final version
226
PAD
Quality of life at 3 years
1143 RCT Serious(d) No serious No serious Serious(g) None 36/224 51/228 RR 0.72 63 fewer per LOW
inconsistency indirectness (16.1%) (22.4%) (0.49 to 1000 (from
1.06) 114 fewer to
13 more)
1143 RCT Serious(d) No serious No serious No serious None 70/224 109/228 RR 0.65 167 fewer MODERATE
inconsistency indirectness imprecision (31.3%) (47.8%) (0.52 to per 1000
0.83) (from 81
fewer to 229
fewer)
Minor adverse events at 1 year
1143 RCT Serious(c) No serious No serious No serious None 3/30 12/31 RR 0.26 286 fewer MODERATE
inconsistency indirectness imprecision (10%) (38.7%) (0.08 to per 1000
0.83) (from 66
fewer to 356
fewer)
Re-intervention at 30 days
1143 RCT Serious(d) No serious No serious Very serious(b) None 67/224 41/228 RR 1.66 119 more VERY LOW
inconsistency indirectness (29.9%) (18%) (1.18 to per 1000
2.34) (from 32
more to 241
more)
1106 RCT Serious(d) No serious No serious Very serious(b) None 10/53 4/49 RR 2.31 107 more VERY LOW
inconsistency indirectness (18.9%) (8.2%) (0.78 to per 1000
6.89) (from 18
Final version
227
PAD
fewer to 481
more)
ABPI at 1 year
1106 RCT Serious(d) No serious No serious Very serious(b) None 30 31 - MD 0.01 VERY LOW
inconsistency indirectness higher (0.2
lower to 0.22
higher)
(a) 1 of 2 studies had unclear allocation concealment; 1 of 2 studies had unclear allocation concealment and blinding.
(d) Unclear allocation concealment.
(e) Unclear blinding.
(f) No information on variability was given in the study, therefore the calculation of the standard deviation was not possible and the mean difference and CI were not estimable.
(g) 95% CI crosses one MID.
Table 86: Clinical evidence profile: Angioplasty compared to bypass surgery for critical limb ischaemia due to femoro-popliteal disease – Adjusted
hazard ratios

Quality
Overall survival before 2 years
1143 RCT serious(a) no serious no serious very serious(b) None 224 228 HR 1.19 (0.84 to - VERY LOW
inconsistency indirectness 1.68)
Overall survival after 2 years
1143 RCT serious(a) no serious no serious no serious none 224 228 HR 0.61 (0.5 to - MODERATE
inconsistency indirectness imprecision 0.75)
Amputation free survival before 2 years
1143 RCT serious(a) no serious no serious very serious(b) None 224 228 HR 1.03 (0.76 to - VERY LOW
Amputation free survival after 2 years
Final version
228
PAD
1143 RCT serious(a) no serious no serious serious(c) None 224 228 HR 0.85 (0.5 to - LOW
(a) Unclear allocation concealment; hazard ratio taken from data reported in study.
(c) 95% CI crosses one MID.
Table 87: EQ-5D – Angioplasty compared to bypass surgery for critical limb ischaemia
Angioplasty Bypass
Baseline Change 0-3 Change Change Change Change Baseline Change 0-3 Change Change Change Change
months 3-6 6-12 12-24 24-36 months 3-6 6-12 12-24 24-36
months months months months months months months months
Bradbury, 2010 – Mean (sd)
0.26 (0.32) 0.53 (0.31) 0.52 0.55 0.56 0.61 0.29 (0.34) 0.57 (0.28) 0.56 0.62 0.59 0.54
(0.34) (0.31) (0.32) (0.25) (0.31) (0.29) (0.34) (0.35)
Table 88: SF- 36 summary component score – Angioplasty compared to bypass for critical limb ischaemia
Angioplasty Bypass
Baseline Change 0-3 Change 3-6 Change 6-12 Baseline Change 0-3 Change 3-6 Change 6-12 months
months months 12 months months months
Bradbury 2010 – Mean (sd)
Physical 17.50 (7.97) 23.80 (11.68) 24.62 (11.58) 24.58 (11.70) 17.80 (9.06) 24.37 (12.45) 24.88 (13.51) 26.13 (13.54)
Mental 43.47 (11.64) 47.69 (11.28) 46.67 (12.19) 48.26 (11.76) 45.17 (11.96) 48.68 (11.13) 48.60 (10.75) 50.16 (10.60)
Final version
229
PAD
Three relevant cost-effectiveness studies were identified for this question by Bradbury, 2010143,
Hunick, 1995110, and Brothers, 1999144). One study was a cost-utility analysis based on the BASIL trial
which compared the costs and effects of angioplasty to bypass surgery143; the decision analytic
model by Hunick, 1995110 compared the costs and QALYs of several different intervention sequences
involving angioplasty and bypass surgery; and the model by Brothers, 1999144 compared the costs
and QALYs expected from treating patients with either bypass or amputation. These studies are
summarised in the economic evidence profile below (Table 89). Full evidence tables can be found in
Appendix I and a list of excluded studies in Appendix F.
This question was originally prioritised by the GDG for original economic modelling. It was as to be
structured around the results of a network meta-analysis with amputation free survival as the main
outcome, and health state utility values from published sources used to determine QALYs. However,
the studies in the clinical review did not report sufficient data to allow us to complete this analysis in
a way that would add to what was already available in the literature.
Final version
230
PAD
Table 89: Economic study characteristics: Angioplasty compared to bypass surgery compared to amputation in people with critical limb ischaemia
Study Applicability Limitations Other Comments costs QALYs effectiveness
Comparators: Primary angioplasty vs Primary bypass surgery
Bradbury Directly Potentially  Economic evaluation based on Bypass was Bypass resulted Bypass cost Uncertainty around the
2010143 applicable(a) serious BASIL RCT £3,795 more in 0.028 QALYs £135,517 per primary outcome (cost
limitations(b)  Population: patients with severe costly than gained QALY gained per QALY) was reported
limb ischaemia angioplasty compared to in one cost effectiveness
angioplasty acceptability curve.
 Time horizon: 3 years
There was a 20%
 Costs: All hospital costs over 3 probability that bypass
years. surgery was cost-
 Country of analysis: UK effective at a threshold
of £20k.
Comparators: No treatment vs. Primary angioplasty followed by angioplasty for treatment failure vs. Primary angioplasty followed by bypass for treatment failure vs.
Primary bypass surgery
Hunick 1995110 Partially Potentially  Decision analytic model based on Vein graft for rest pain stenosis
applicable (c) serious a variety of published sources. Primary Angioplasty Angioplasty For patients with rest
limitations(d)  Population: patients with angioplasty followed by followed by pain occlusion, the
femoropopliteal disease followed by bypass surgery bypass conclusion was
 Time horizon: Lifetime bypass surgery was the most surgery is the unchanged
 Costs: All hospital costs were was the least effective dominant
obtained from hospital records. costly strategy strategy strategy.
The cost of care for patients PTFE-AK for rest pain stenosis
immobilised and dependant Angioplasty Angioplasty Angioplasty For patients with rest
following amputation was based followed by followed by followed by pain occlusion,
on published literature. angioplasty was angioplasty was angioplasty angioplasty followed by
 Country of analysis: USA the least costly the most costly was the bypass was the
strategy strategy dominant dominant strategy
strategy
PTFE-BK for rest pain stenosis
Angioplasty Angioplasty Angioplasty For patients with rest
followed by followed by followed by pain occlusion,
Final version
231
PAD
angioplasty was angioplasty was angioplasty angioplasty followed by

the least costly the most costly was the bypass was the
strategy strategy dominant dominant strategy
strategy
Comparators: Primary bypass surgery vs. Primary amputation vs. Primary medical management
Brothers Potentially Partially  Decision analytic model Primary bypass Primary bypass Primary One- and two-way
1999144 serious applicable(f)  Population: people with first was £5, 466 resulted in a bypass costs sensitivity analyses were
limitations(e) presentation of limb-threatening more expensive gain of 1.16 £4, 712 per performed to evaluate
ischaemia caused by tibial- than non- QALYs QALY gained the effect of varying
peroneal artery occlusive disease operative compared to compared to expected utility,
expectant non-operative non-operative incremental costs, early
 Outcomes: QALYs
management(g) expectant expectant patency, late patency
 Costs: Hospital, outpatient and management(h) management(i and peri-operative
physician charges obtained from mortality rates. The
patient records authors reported the
 Perspective: USA hospital results of these analyses
in graphical form only
and did not excluded
dominated options,
therefore, it is not
possible to analyse the
results of these analyses.
Based on threshold
analysis, the authors
concluded that primary
amputation becomes the
most cost-effective
strategy when primary
bypass patency is less
than 11%. Expectant
management is the most
cost-effective treatment
when operative
mortality for
revascularisation or
amputation exceeds
55%.
Final version
232
PAD
(a) UK NHS setting (English and Scottish centres).

(b) Three year time horizon; resource use and unit costs not reported; analysis of uncertainty based on undiscounted costs and discounted QALYs; cost of amputation not accounted for.
(c) Resource use based on American hospital records.
(d) Quality of life estimated using Torrence Multi Attribute Scale by healthcare workers; patency failure assumed to be equivalent to symptom progression & re-intervention; progression of
symptoms not modelled due to lack of data.
(e) Long-term patient survival, limb salvage rate, and primary and cumulative secondary patency rates were obtained from the results of retrospective analyses previously conducted by the
authors with no evidence of a systematic search; utility values were obtained from people with CLI rather than patients who had experienced each health state QALY gain was considered
only over a 5-year horizon, therefore, this study will underestimate the long-term effect of reduced operative mortality expected from both the expectant management and primary
amputation strategies; unclear method of QALY elicitation and valuation.
(f) USA hospital perspective.
(g) Primary amputation was £2, 186 more costly than non-operative expectant management.
(h) Primary amputation resulted in a gain of 0.06 QALYs compared to non-operative management.
(i) Primary amputation is excluded by extended dominance.
Final version
233
PAD
10.2.2.1 Clinical
Critical limb ischaemia due to aorto-iliac disease:
 Limb salvage at 4 years [1 study, 45 participants, very low quality evidence]105
Critical limb ischaemia due to femoro-popliteal disease:
Bypass surgery was significantly better than angioplasty for:

 Overall survival after 2 years (adjusted HR) [1 study, 452 participants, moderate quality
evidence]143

 Minor adverse events at 30 days [1 study, 452 participants, moderate quality evidence]143
 Minor adverse events at 1 year [1 study, 61 participants, moderate quality evidence]106
 Mortality at 30 days [2 studies, 513 participants, very low quality evidence]106,143
 Overall survival before 2 years (adjusted HR) [1 study, 452 participants, very low quality
evidence]143
 Amputation at 1 year [1 study, 61 participants, very low quality evidence]106
 Amputation free survival before 2 years (adjusted HR) [1 study, 452 participants, very low quality
evidence]143
 Amputation free survival after 2 years (adjusted HR) [1 study, 452 participants, low quality
evidence]143
 Amputation free survival at 3 years [1 study, 452 participants, very low quality evidence]143
 Limb salvage rate at 4 years [1 study, 27 participants, very low quality evidence]105
 Major adverse events at 30 days [1 study, 452 participants, low quality evidence]143
 Re-intervention at 30 days [1 study, 452 participants, very low quality evidence]143
 ABPI at 1 year [1 study, 61 participants, very low quality evidence]106
analysis performed:
 Quality of life increased for both angioplasty and bypass at 3 months [1 study, 316 participants,
 Quality of life decreased for both angioplasty and bypass at 6 months [1 study, 275 participants,
 Quality of life increased for both angioplasty and bypass at 1 year [1 study, 252 participants, low
 Quality of life increased for angioplasty and decreased bypass at 2 years [1 study, 139
Final version
234
PAD
 Quality of life increased for angioplasty and decreased bypass at 3 years [1 study, 97 participants,
10.2.2.2 Economic
 One study found that angioplasty is more cost effective than bypass surgery for the treatment of
people with SLI [directly applicable with potentially serious limitations]143
 One study found that angioplasty followed by (autologous vein) bypass surgery is the most cost
effective treatment option in people with CLI due to stenoses and occlusions [partially applicable
with potentially serious limitations]110
 One study found that primary bypass surgery may be more cost-effective than primary
amputation in people with CLI [partially applicable with potentially serious limitations]144
19.Ensure that all people with critical limb ischaemia are assessed
by a vascular multidisciplinary team before treatment
decisions are made.
20.Offer angioplasty or bypass surgery for treating people with

critical limb ischaemia who require revascularisation, taking
into account factors including:
 comorbidities
 pattern of disease
 availability of a vein
Recommendations  patient preference.
Relative value of different The GDG considered mortality as an outcome of major importance, but
outcomes were also concerned to consider quality of life. Both amputation and the
need for further intervention, irrespective of whether angioplasty or bypass
surgery is done first, will impact on quality of life and these outcome
measures were also considered carefully.
A difference in mortality was observed in the form of an adjusted hazard

ration in favour of surgery at the 2-year time point. There was no significant
difference in unadjusted figures, nor was there any mortality difference at
any other time-point whether shorter or longer than 2 years. The GDG also
discussed a post-hoc analysis of the BASIL trial which suggested that there
is a mortality benefit for patients undergoing bypass who live beyond 2
years. However, this analysis has not been validated, and it is not clear how
to predict >2-year survival before the intervention actually takes place.
(This evidence did not meet the primary literature search criteria and was
therefore not part of the formal evidence review set out above).
Although fairly large differences were seen in amputation rates at some

time points there were no statistically significant differences in this
outcome measure, nor in re-intervention rates.
Trade off between clinical Adverse events were more frequently observed with bypass surgery than
benefit and harms with angioplasty, although this difference was significant only for minor
events.
There was debate around the technical failure rate with angioplasty. Having
Final version
235
PAD
bypass surgery after angioplasty resulted in poorer outcomes than going

straight to bypass in the BASIL study, which may indicate that angioplasty
had changed the bypass opportunity. However, it is also possible that this
group of people, who required two procedures were those with a poorer
natural prognosis and that they would not have had good results with
either procedure. This is difficult to tease out of the study data and the
GDG were not unanimous in their view of the implied risk of attempting
angioplasty first in people suitable for bypass.
Economic considerations The GDG considered the results and the limitations of the cost-
effectiveness analyses by Bradbury 2010 and Hunick 1995. On balance,
they agreed that angioplasty is most likely to be the most cost effective
primary treatment strategy for people with CLI. However, due to the
limitations of the evidence base and the considerable uncertainty reported
in the analyses, the GDG did not feel that either form of intervention could
be unequivocally recommended as preferable on health economic grounds.
A patient’s likely benefit from either angioplasty or surgery needs to be
judged on an individual basis and therefore referral to a specialist centre
where a multi-disciplinary assessment can take place should form a key
part of determining the most cost-effective pathway for each patient.
Quality of evidence The GDG noted that the evidence reviewed was moderate to very low
quality by the GRADE criteria. The evidence was downgraded on a number
of issues including allocation concealment and blinding. This led to a
discussion on the trial methodology for these interventions. It is not
possible to blind those performing the relevant procedure or to blind the
participants to the interventions received. Therefore, under GRADE criteria,
the evidence would never receive a high quality scoring. The GDG
concluded that the RCTs presented were the most robust available for a
comparison of angioplasty with surgery.
The patient population with CLI have few clinical options available. The
GDG felt that many of patients included within these trials were likely to
have been more suited to either the angioplasty or bypass intervention,
because of differences, for example, in anatomy or co-morbidity. The
number of potential subjects with genuinely equal suitability for either
intervention is, in their experience, fairly small.
No evidence was reviewed for the benefits of multi-disciplinary review. The

recommendation was based on the GDG experience and consensus.
Other considerations It is difficult to make a blanket recommendation for all patients with CLI as
many of them have features which make them unsuitable for either
angioplasty or bypass. The GDG advocated that all patients are considered
on an individual basis by a multi-disciplinary team. basis as the following
factors determine which intervention is considered optimal:
 Age of the patients
 Fitness for surgery
 Severity of disease
 Size and shape of patient
 Co-morbidities involved
 Presence or absence of a suitable vein
 Technical ability to undertake angioplasty
 Balance of benefit versus harms.
Final version
236
PAD
Most units now have multi-disciplinary teams and they are considered
standard practice. An MDT review will ensure that patients have access to
all treatment options and the decisions are made based on individual
needs. The GDG did not review evidence relating to multi-disciplinary
review in people with CLI eligible for revascularisation or bypass surgery
but agreed by consensus that such a recommendation was important.
In practice, angioplasty tends to be undertaken as the first line option,

although there is geographical variation around the UK. Of the two
procedures, there are likely to be fewer patients unsuitable for angioplasty
than unsuitable for surgery. The clinical studies did not show any clear
advantage either way, but the health economic evidence favoured
angioplasty as the first procedure. The GDG therefore agreed that in the
small number of cases in which the two procedures look equally likely to
succeed, angioplasty should be tried first.
Patient choice must be part of the decision making process. It was

recognised that some patients may even prefer to undergo amputation
instead of repeated interventions, which are associated with longer
hospital stay and healing times.
No evidence was found on management of patients with diabetes. There is

a recognition that the prevalence of diabetes is increasing. However, the
GDG felt that the data could not be extrapolated to make a separate
recommendation for the diabetic population.

The GDG highlighted recommendation 17 as a key priority for
implementation. The reason for selecting this recommendation as a priority
was that the GDG considered it important for CLI patients to be reviewed
by a MDT in order that all possible options for treatment to be considered.
10.2.4 Research recommendations
3. What is the clinical and cost effectiveness of a 'bypass surgery first' strategy compared with an
'angioplasty first' strategy for treating people with critical limb ischaemia caused by disease of
the infra-geniculate (below the knee) arteries?
Many people with critical limb ischaemia, especially those with diabetic vascular disease, also have
disease of the infra-geniculate (below the knee) arteries in the calf. For many years, the standard of
care has been bypass surgery. Although such surgery may be associated with significant morbidity,
the resulting long-term amputation-free survival rates are generally good. In recent years there has
been a trend towards treating infra-geniculate disease with angioplasty, on the grounds that it is
associated with less morbidity than surgery. However, this change in practice is not evidence-based,
and serious concerns remain about the durability of angioplasty in this anatomical area. A
multicentre, randomised controlled trial with a full health economic analysis is required to address
this. The primary endpoint should be amputation-free survival, with secondary endpoints including
overall survival, health-related quality of life, healing of tissue loss, and relief of ischaemic pain.
Final version
237
PAD
4. What is the clinical and cost effectiveness of primary amputation compared to an attempt at
revascularisation (either angioplasty or bypass surgery) for selected people presenting with
critical limb ischaemia who are thought to be at high risk of failure following revascularisation?
About 50% of people presenting with critical limb ischaemia (CLI) are offered revascularisation either
by means of angioplasty or bypass surgery. However, in those undergoing revascularisation it is
possible to recognise a subgroup in which the success of intervention is so low that primary
amputation might be a better strategy. Conversely, in the 50% of people with CLI who are treated
conservatively or with primary amputation there may be a subgroup in which revascularisation
would be appropriate. A multicentre, hospital-based, randomised controlled trial is required to
define the most clinically and cost-effective strategy for the highest-risk people with CLI in whom
there is equipoise between revascularisation, either via angioplasty or bypass surgery, and primary
amputation. The primary endpoint should be amputation free survival with secondary endpoints
including overall survival, health-related quality of life, healing of tissue loss, and relief of ischaemic
pain. A full health economic analysis should also be undertaken.
10.3 Angioplasty with selective stent placement compared with

angioplasty with primary stent placement
What is the clinical and cost effectiveness of angioplasty with selective stent placement compared to
angioplasty with primary stent placement for the treatment of critical limb ischemia in adults with
PAD?
A literature search was conducted for RCTs that compared the effectiveness of angioplasty with
selective stent placement to primary stent placement. No time limit was placed on the literature
search, there were no limitations on sample size, and outcomes were subgrouped according to lesion
location (femoro-popliteal and aorto-iliac). Indirect populations and emergency settings were
excluded. One Cochrane review was identified129 which considered angioplasty without stents
compared to angioplasty with stents for the superficial femoral artery. The Cochrane review was not
included or updated as it did not meet the protocol defined by the GDG, which included all arteries
of the leg. However it was cross checked for included studies which matched the review protocol.
Five relevant RCTs145-149 were included in the review. The trials did not report outcome data for
people with diabetes and no data was identified for people with CLI due to aorto-iliac disease.
There were unit of analysis issues in some of the trials where data were analysed by the limb or
lesion rather than by person randomised. These trials have been analysed separately.
The quality and results of included studies are reported in Table 90 and Table 91. The forest plots for
each clinical outcome are reported in Appendix J.
Final version
238
PAD
Table 90: Clinical evidence profile: Angioplasty with selective stent placement compared to angioplasty with primary stent placement for critical limb
ischaemia due to femoro-popliteal disease (person randomised data)
Quality
placement placement
1148 RCT Very No serious No serious No serious None 0/17 0/15 not pooled not pooled LOW
serious(a) inconsistency indirectness imprecision (0%) (0%)
Mortality at 3 months
1146 RCT Very No serious No serious Very serious(b) None 3/32 5/33 RR 0.62 58 fewer per 1000 VERY
serious(a) inconsistency indirectness (9.4%) (15.2%) (0.16 to (from 127 fewer to LOW
2.38) 209 more)
Mortality at 9 months
2.34) 353 more)
2.21) 220 more)
(a)
serious inconsistency indirectness (3.7%) (8.3%) (0.04 to 4.6) (from 80 fewer to LOW
300 more)
1146 RCT Very No serious No serious Very serious(b) None 7/24 10/19 RR 0.55 237 fewer per VERY
serious(a) inconsistency indirectness (29.2%) (52.6%) (0.26 to 1000 (from 389 LOW
Final version
239
PAD
1.18) fewer to 95 more)

1148 RCT Very No serious No serious No serious None 0/17 0/15 not pooled not pooled LOW
serious(a) inconsistency indirectness imprecision (0%) (0%)
Major adverse events at 1 year
1148 RCT Very No serious No serious Very serious(b) None 4/17 1/15 RR 3.53 169 more per 1000 VERY
(a)
serious inconsistency indirectness (23.5%) (6.7%) (0.44 to (from 37 fewer to LOW
28.21) 1000 more)
Minor adverse events at 1 year
1149 RCT Serious(c) No serious No serious Very serious(b) None 4/33 0/21 RR 5.82 - VERY
inconsistency indirectness (12.1%) (0%) (0.33 to LOW
102.93)
1145 RCT Very No serious No serious Very serious(b) None 0/27 1/24 RR 0.3 (0.01 29 fewer per 1000 VERY
serious(a) inconsistency indirectness (0%) (4.2%) to 6.98) (from 41 fewer to LOW
249 more)
serious(a) inconsistency indirectness (11.8%) (13.3%) (0.14 to (from 115 fewer to LOW
5.52) 603 more)
Target lesion revascularisation at 3 months
serious(a) inconsistency indirectness (0%) (3%) (0.01 to (from 30 fewer to LOW
8.13) 216 more)
1146 RCT Very No serious No serious Very serious(b) None 3/24 7/19 RR 0.34 (0.1 243 fewer per VERY
serious(a) inconsistency indirectness (12.5%) (36.8%) to 1.14) 1000 (from 332 LOW
fewer to 52 more)
ABPI at 3 months
1146 RCT Very No serious No serious Serious(d) None 32 33 - MD 0.2 lower VERY
serious(a) inconsistency indirectness (0.31 to 0.09 LOW
Final version
240
PAD
lower)
ABPI at 9 months
1146 RCT Very No serious No serious Very serious(b) None 32 33 - MD 0 higher (0.11 VERY
serious(a) inconsistency indirectness lower to 0.11 LOW
higher)
(a) Unclear methodology.
Table 91: Clinical evidence profile: Angioplasty with selective stent placement compared to angioplasty with primary stent placement for critical limb
ischaemia due to femoro-popliteal disease (limb / lesion randomised data)
Quality
placement placement
inconsistency indirectness serious(b) (4.5%) (6.3%) to 10.78) 1000 (from 59
fewer to 611
more)
1147 RCT Serious(a) No serious No serious Very None 7/22 3/16 RR 1.7 (0.52 131 more per VERY LOW
fewer to 857
more)
inconsistency indirectness serious(b) (13.6%) (25%) to 2.11) 1000 (from 215
fewer to 277
more)
Final version
241
PAD

fewer to 611
more)
inconsistency indirectness serious(b) (9.1%) (25%) to 1.75) 1000 (from 230
fewer to 188
more)
Major adverse event at 2 years
fewer to 454
more)
1147 RCT Serious(a) No serious No serious Very None 5/22 2/16 RR 1.82 (0.4 103 more per VERY LOW
fewer to 901
more)
(b) 95% CI crosses both MIDS.
Final version
242
PAD
In the absence of published evidence, the GDG were presented with the cost of (bare metal) stents.
Vascular stents are excluded from the NHS reference cost for angioplasty and incur an additional cost
according to the number and type used per procedure. The unit cost of vascular stents was not
available from the NHS Supply Catalogue. A buyer for cardiology and radiology products at the NHS
Supply chain was asked to provide a list of prices for all vascular stents currently in use in England
and Wales, however the GDG concluded that this list was not inclusive. Members of the GDG were
then asked to provide prices from their hospitals. Based on prices obtained by GDG members, the
group estimated bare metal stents cost approximately £550. The GDG also indicated that on average
two stents are used per procedure.
The clinical studies included in this review did not provide details of the number of patients requiring
selective stent placement. Assuming that the proportion is similar to those in IC, approximately 40%
of patients require stent placement. 108 According to the evidence included in the clinical review,
4.5% selective stent placement procedures resulted in major adverse events at 30 days compared to
6.3% of primary stent placement procedures.147 Applying this data to the NHS reference costs
presented in Table 92, the average cost of angioplasty with selective stent placement is £4, 171 and
the cost of angioplasty with primary stent placement is £4, 603. Therefore, the incremental cost of
angioplasty with primary stent placement is approximately £432.
Table 92: Costs of angioplasty procedure – Elective and non-elective

National Lower Upper
Currency average unit quartile unit quartile unit
code Currency description Activity cost cost cost
Elective inpatient (long stay) HRG data
QZ15A Therapeutic endovascular 114 £9, 200 £1, 940 £14, 255
procedure with major
complications
QZ15C Therapeutic endovascular 7, 991 £1, 888 £940 £2, 248
procedure without complications
Elective inpatient (long stay) excess bed day HRG data
QZ15A Therapeutic endovascular 132 £173 £152 £152
complications
QZ15C Therapeutic endovascular 1, 580 £344 £250 £433
Average cost
Elective angioplasty with major complications £9, 349 (£2, 071 - £14, 386)
Elective angioplasty without complications £3, 627 (£2, 204 - £4, 435)
Non elective inpatient (long stay) HRG data
QZ15A Therapeutic endovascular 611 £9, 518 £4, 547 £11, 821
complications
QZ15C Therapeutic endovascular 1, 820 £4, 206 £2, 148 £5, 200
Non elective inpatient (long stay) excess bed day HRG data
QZ15A Therapeutic endovascular 850 £255 £140 £338
Final version
243
PAD

complications
QZ15C Therapeutic endovascular 7, 054 £357 £229 £454
Average cost
Non elective angioplasty with major complications £9, 702 (£4, 647 - £12, 064)
Non elective angioplasty without complications £4, 298 (£2, 206 - £5, 317)
Weighted average cost of angioplasty (assuming 10% of procedures are non elective)
Angioplasty with major complications £9, 385 (£2, 329 to £14, 154)
Angioplasty without complications £3, 695 (£2, 204 to £4, 524)
Source/Note: All costs obtained from 2009/10 NHS Reference Costs49
10.3.2.1 Clinical
No clinical evidence was reported for people with CLI due to aorto-iliac disease.
Critical limb ischaemia due to femoro-popliteal disease (person randomised data):
Angioplasty with primary stent placement was statistically significantly better than angioplasty with
selective stent placement for:
 Mortality at 3 months [1 study, 65 participants, very low quality evidence]146
 Mortality at 30 days [1 study, 32 participants, low quality evidence]148
 Amputation at 3 months [1 study, 65 participants, very low quality evidence]146
 Amputation at 1 year [1 study, 32 participants, low quality evidence]148
 Major adverse events at 1 year [1 study, 32 participants, very low quality evidence]148
 Minor adverse events at 1 year [1 study, 54 participants, very low quality evidence]149
 Re-intervention at 6 months [1 study, 51 participants, very low quality evidence]145
 Target lesion revascularisation at 3 months [1 study, 65 participants, very low quality evidence]146
 Target lesion revascularisation at 9 months [1 study, 43 participants, very low quality evidence]146
Critical limb ischaemia due to femoro-popliteal disease (limb/lesion randomised data):
 Mortality at 30 days and 2 years [1 study, 38 limbs, very low quality evidence]147
Final version
244
PAD
 Amputation at 2 years [1 study, 38 limbs, very low quality evidence]147

 Major adverse events at 30 days and 2 years [1 study, 38 limbs, very low quality evidence]147
 Minor adverse events at 30 days [1 study, 38 limbs, very low quality evidence]147
 Re-intervention at 2 years [1 study, 38 limbs, very low quality evidence]147
10.3.2.2 Economic
21.Do not offer primary stent placement for treating people with
critical limb ischaemia caused by aorto-iliac disease (except
complete occlusion) or femoro-popliteal disease.
22.Consider primary stent placement for treating people with critical

limb ischaemia caused by complete aorto-iliac occlusion (rather
Recommendations than stenosis).
Relative values of different There were no differences in any of the reported outcome measures.
outcomes
Patency was not considered as a relevant outcome for the reasons detailed in
section 3.1.1 in the methodology chapter.
Trade off between clinical The GDG were concerned that stents may give the operator the impression
benefits and harms that a procedure has been technically successful at the time the procedure is
performed, but noted that no consistent later benefit was demonstrated in
comparison with angioplasty.
The GDG considered that the routine use of stents as opposed to selective use
in conjunction with angioplasty carried the disadvantages of additional cost,
increased procedure time, and potential risks of additional instrumentation.
Endovascular procedures carry a potential risk of causing embolisation of

material from the diseased artery which can cause blockage of smaller arteries
further down the leg. This is thought to be a greater risk with complete
occlusion of the aorto-iliac arteries than with stenosis or occlusion in smaller
vessels. There is also a risk of restenosis following endovascular treatment and
having foreign material such as a stent in the artery may increase this risk,
particularly in smaller vessels.
The GDG considered that it is generally accepted that stenting is advantageous

in terms of embolisation rates although the evidence reviewed in these studies
did not reflect this.
Economic considerations No cost effectiveness evidence was identified for this question. The GDG
considered the increased cost associated with primary stent placement
compared to selective stent placement. They agreed that in light of clinical
evidence suggesting that there is no clear benefit associated with primary
stent placement, it does not represent value for money and should not be
recommended for routine use.
Quality of evidence The evidence was rated as low to very low by GRADE criteria. The GDG noted
Final version
245
PAD
that the included studies were small.
Other considerations This comparison is about whether to place stent in all patients undergoing an
endovascular intervention for PAD, or only those in whom the operator deems
it necessary. Although the latter seems more open to error, the former may
be wasteful, and in this group of studies no clear evidence in favour of primary
stenting emerged.
Primary stenting for femoro-popliteal disease or stenotic disease of the aorto-

iliac arteries is not standard UK practice and the GDG felt that there was
insufficient evidence to recommend a change to this situation.
Primary stents are currently used in aorto-iliac disease in the UK because of

concern about the risk of embolisation. The GDG recognised that they had
identified no evidence to justify this as routine, but also noted that
embolisation was not an endpoint specifically sought in these studies.
In the absence of any clear evidence for or against primary stent placement,
the GDG made their decision based on the extra cost of routinely employing
stents and developed recommendations which would discourage primary
stenting, but acknowledge the possible value for aorto-iliac occlusive disease.
5. What is the clinical and cost effectiveness of selective stent placement compared with
angioplasty plus primary stent placement for treating people with critical limb ischaemia caused
by disease of the infra-geniculate arteries?
Studies comparing angioplasty plus selective stent placement with primary stent placement have
been limited to the aorto-iliac and femoro-popliteal segment. There is also a significant group of
people with critical ischaemia caused by disease of the infra-geniculate vessels in which there is a
potential for endovascular treatment. Infra-geniculate disease is more complex to treat by
endovascular means, and the risks and benefits of different treatment options may differ from those
for the more proximal vessels. A multicentre, randomised controlled trial with a full health economic
analysis is required to address the optimum policy as regards the choice of method for angioplasty
and stent placement for the infra-geniculate arteries. The primary endpoint should be amputation-
free survival, with secondary endpoints including overall survival, re-intervention rates, health-
related quality of life, healing of tissue loss, and relief of ischaemic pain. What is the clinical and cost
effectiveness of selective stent placement compared with angioplasty plus primary stent placement
for treating people with critical limb ischaemia caused by disease in the infra-geniculate arteries?
10.4 Bare metal compared to drug eluting stents

What is the clinical and cost effectiveness of bare metal stents compared to drug eluting stents for
the treatment of critical limb ischemia in adults with PAD?
A literature search was conducted for RCTs that compared the effectiveness of bare metal stents to
drug eluting stents. No time limit was placed on the literature search, there were no limitations on
Final version
246
PAD
sample size, and outcomes were sub-grouped according to lesion location (femoro-popliteal and
aorto-iliac). Indirect populations and emergency settings were excluded.
Four relevant RCTs of two trials138,150-152 were included in the review. The trials did not report
outcome data for people with diabetes and no data was identified for people with CLI due to aorto-
iliac disease.
The quality and results of included studies are reported in Table 93. The forest plots for each clinical
Final version
247
PAD
Table 93: Clinical evidence profile: Bare metal compared to drug eluting stents for critical limb ischaemia due to femoro-popliteal disease
Quality of evidence No of patients Effect
Risk of Other Relative Quality
No of studies Design Inconsistency Indirectness Imprecision BMS DES Absolute
bias considerations (95% CI)
1152 RCT Very No serious No serious Very serious(b) None RR 0.52
1/28 2/29 33 fewer per 1000 (from 66 VERY
serious(a) inconsistency indirectness (0.05 to
(3.6%) (6.9%) fewer to 303 more) LOW
5.4)
2/46 7/47 106 fewer per 1000 (from VERY
(4.3%) (14.9%) 140 fewer to 49 more) LOW
1.33)
1151 RCT Very No serious No serious No serious None 0/46 0/47 not
not pooled LOW
serious(a) inconsistency indirectness imprecision(c) (0%) (0%) pooled
Major adverse events at 6 months
(0%) (3.6%) fewer to 235 more) LOW
7.59)
Minor adverse events intra-operative
(6.9%) (7.1%) fewer to 385 more) LOW
6.39)
Minor adverse events at 6 months
2150,151 RCT Very No serious No serious Very serious(b) None RR 0.98
(4.3%) (4.3%) fewer to 247 more) LOW
6.67)
Minor adverse events at 2 years
1151 RCT Very No serious No serious Very serious(b) None 3/46 RR 7.15 0 more per 1000 (from 0 VERY
0/47
serious(a) inconsistency indirectness (6.5%) (0.38 to fewer to 0 more) LOW
Final version
248
PAD
0% 134.66)
Revascularisation procedure on contralateral leg before discharge at 6 months
2/28 2/29 3 more per 1000 (from 58 VERY
(7.1%) (6.9%) fewer to 404 more) LOW
6.86)
Revascularisation procedure on contralateral leg after discharge at 6 months
(7.1%) (10.3%) fewer to 293 more) LOW
3.83)
Target vessel revascularisation at 6 months
1152 RCT Very No serious No serious Very serious(b) None 3/28 RR 3.11
1/29 73 more per 1000 (from 23 VERY
serious(a) inconsistency indirectness (10.7% (0.34 to
(3.4%) fewer to 935 more) LOW
) 28.12)
Target vessel revascularisation at 2 years
1151 RCT Very No serious No serious Very serious(b) None 10/46 RR 1.7
(a) 6/47 89 more per 1000 (from 42 VERY
serious inconsistency indirectness (21.7% (0.67 to
(12.8%) fewer to 421 more) LOW
) 4.3)
Target vessel revascularisation at 2 years (Hazard Ratio)
1151 RCT Very No serious No serious No serious None 10/46 HR 7.27
6/47 502 more per 1000 (from
serious(a) inconsistency indirectness imprecision (21.7% (1.75 to LOW
(12.8%) 85 more to 856 more)
) 30.26)
1152 RCT Very No serious No serious No serious None 0/28 0/29 not
not pooled LOW
serious(a) inconsistency indirectness imprecision(c) (0%) (0%) pooled
Target lesion revascularisation at 2 years
6/46 3/47 66 more per 1000 (from 29 VERY
(13%) (6.4%) fewer to 427 more) LOW
7.69)
Target lesion revascularisation at 2 years (Hazard Ratio)
Final version
249
PAD
1151 RCT Very No serious No serious Very serious(b) None HR 3.9

(a)
serious inconsistency indirectness 6/46 3/47 163 more per 1000 (from VERY
(0.77 to
(13%) (6.4%) 14 fewer to 662 more) LOW
19.63)
ABPI at 6 months
1152 RCT Very No serious No serious No serious None MD 0.04 lower (0.13 lower
24 23 - LOW
serious(a) inconsistency indirectness imprecision to 0.05 higher)
ABPI at 2 years
1151 RCT Very No serious No serious Serious(d) None MD 0.06 lower (0.15 lower VERY
37 35 -
serious(a) inconsistency indirectness to 0.03 higher) LOW
Patency at 6 months
1152 RCT Very No serious No serious Serious(d) None HR 1.05 47 more per 1000 (from 76
22/22 19/20 VERY
serious(a) inconsistency indirectness (0.92 to fewer to 190 more)
(100%) (95%) LOW
1.20)
(a) Unclear allocation concealment and randomisation.
(c) There were no events in either group.
Table 94: Clinical evidence profile: Bare metal compared to drug eluting stents for critical limb ischaemia due to infra-geniculate disease
No of
Quality of evidence Effect
patients
Relative Quality
No of Other
Design Risk of bias Inconsistency Indirectness Imprecision BMS DES (95% Absolute
studies considerations
CI)
Mortality at 1 year
1138 RCT No serious risk No serious No serious Very serious(a) None 8/33 9/42 RR 1.13 28 more per 1000 LOW
of bias inconsistency indirectness (24.2 (21.4 (0.49 to (from 109 fewer to 345
%) %) 2.61) more)
1138 RCT No serious risk No serious No serious Very serious(a) None 2/33 2/42 RR 1.27 13 more per 1000 LOW
of bias inconsistency indirectness (6.1%) (4.8%) (0.19 to (from 39 fewer to 360
Final version
250
PAD
8.56) more)
Target lesion revascularisation at 1 year
1138 RCT No serious risk No serious No serious Very serious(a) None 3/33 4/42 RR 0.95 5 fewer per 1000 (from LOW
of bias inconsistency indirectness (9.1%) (9.5%) (0.23 to 73 fewer to 283 more)
3.97)
ABPI at 1 year
1138 RCT No serious risk No serious No serious No serious None 33 42 - MD 0.07 lower (0.13 to HIGH
of bias inconsistency indirectness imprecision 0.01 lower)
Patency at 1 year
1138 RCT No serious risk No serious No serious Serious(b) None 56.5% 75% not not pooled MODERATE
of bias inconsistency indirectness pooled
p=0.23(c
)
(a) 95% CI crosses both MIDs

(b) No denominator was given in the study, and therefore the calculation of the relative risk was not possible and the CI was not estimable
(c) p=0.23 reported in the paper for BMS vs DES
Final version
251
PAD
In the absence of any published cost effectiveness evidence, the GDG were presented with the
average cost of bare metal and drug eluting stents and the cost of angioplasty with and without
complications (Table 95 and Table 96).
Table 95: Peripheral vascular stent cost

Vascular stent type Approximate average cost Source
Bare metal £550 GDG opinion based on hospital records
Drug eluting £950 GDG opinion based on hospital records
Table 96: Angioplasty procedural cost

HRG Average Lower and upper
code Description unit cost quartile unit cost Source
Elective inpatient (long stay) HRG and excess bed day data
QZ15A Therapeutic endovascular £9, 349 £2, 071 to £14, 386 NHS Reference
procedures with major complications Costs
QZ15B Therapeutic endovascular procedure £3, 397 £1, 850 to £4, 104 NHS Reference
with intermediate complications Costs
QZ15C Therapeutic endovascular procedure £3, 627 £2, 204 to £4, 435 NHS Reference
without complications Costs
Non elective inpatient (long stay) HRG and excess bed day data
QZ15A Therapeutic endovascular £9, 701 £4, 647 to £12, 064 NHS Reference
procedures with major complications Costs
QZ15B Therapeutic endovascular procedure £5, 197 £3, 369 to £6, 353 NHS Reference
with intermediate complications Costs
QZ15C Therapeutic endovascular procedure £4, 298 £2, 206 to £5, 317 NHS Reference
without complications Costs
Source/Note: 2009/10 NHS Reference costs49.
10.4.2.1 Clinical
Drug eluting stents were significantly better than bare metal stents for:
 Target vessel revascularisation at 2 years (using hazard ratio) [1 study, 93 participants, low quality
evidence]151
for:
Final version
252
PAD
 Major adverse events at 6 months [1 study, 57 participants, very low quality evidence]152
 Minor adverse events intra-operatively [1 study, 57 participants, very low quality evidence]152
 Minor adverse events at 6 months and 2 years [2 studies, 93 participants, very low quality
evidence]150,151
 Revascularisation on contralateral leg before at 6 months, and after discharge at 6 months [1
study, 57 participants, very low quality evidence]152
 Target vessel revascularisation at 6 months [1 study, 75 participants, very low quality evidence]152
 Target vessel revascularisation at 2 years (using relative risk) [1 study, 93 participants, very low
 Target lesion revascularisation at 2 years (using relative risk and hazard ratio) [1 study, 93
 ABPI at 6 months [1 study, 47 participants, low quality evidence}152
 ABPI at 2 years [1 study, 72 participants, very low quality evidence]151
 Patency at 6 months [1 study, 42 participants, very low quality evidence]152
There was no difference between bare metal stents and drug eluting stents for:
 Amputation at 2 years [1 study, 93 participants, low quality evidence]151
 Target lesion revascularisation at 6 months [1 study, 57 participants, low quality evidence]152
Critical limb ischaemia due to infra-geniculate disease:
Drug eluting stents were significantly better than bare metal stents for:
 ABPI at 1 year [1 study, 75 participants, high quality evidence]138
for:
 Mortality at 1 year [1 study, 75 participants, low quality evidence]138
 Amputation at 1 year [1 study, 75 participants, low quality evidence]138
 Target lesion revascularisation at 1 year [1 study, 75 participants, low quality evidence]138
 Patency at 1 year [1 studye, moderate quality evidence]138
10.4.2.2 Economic
23.Use bare metal stents when stenting is used for treating people
Recommendation with critical limb ischaemia.
Relative values of different The GDG considered amputation free survival and re-intervention rates to be
outcomes the key clinical outcomes for this question. They also wished to know whether
there were any differences in quality of life, mortality and ABPI. Although data
were available at several different time points, few differences were found in
the reported outcomes. No data was reported on quality of life or walking
distance.
Drug-eluting stents were shown to be superior in terms of target vessel re-

vascularisation rates and ABPI, both at one time-point only. The GDG did not
e Exact number of participants used in the analyses for patency was not reported in this study.
Final version
253
PAD
feel that great importance could be attached to these when set against the
large number of comparisons showing no difference, and bearing in mind that
these were not felt to be the key outcome measures.
Patency was considered as a potential outcome measure for this comparison,

although the GDG recognised the value of this measure was limited in the
absence of associated evidence to link patency to more clinically relevant
outcomes.
Trade off between clinical According to the results of the clinical review, there was no significant
benefits and harms difference in mortality, adverse events and amputation between bare metal
and drug eluting stents.
The GDG did not feel there was any difference between the two types of stent
in terms of technical difficulty in placement. The method of placement of the
two forms of stents is identical, and therefore the main potential adverse
effects are also the same.
Economic considerations There was no cost effectiveness evidence identified for this question. Drug
eluting stents are more expensive than bare metal stents.
In the absence of evidence to suggest that clinical outcomes are improved with
the use of drug eluting stents, the GDG agreed that the increased cost does not
represent a cost effective use of NHS resources.
Quality of evidence The quality of evidence comparing bare metal and drug eluting stents was of
low quality by GRADE criteria. The GDG noted that the SIROCCO study was
halted as no differences were found between bare metal and drug eluting
stents.152
Other considerations The GDG were aware of studies evaluating the effectiveness of bare metal
versus drug eluting stents in coronary arteries. However, they did not think
that these results could be extrapolated to the peripheral arteries because of
the considerable differences in anatomy.
The evidence identified related to femoro-popliteal and infra-geniculate

disease with no evidence relevant to aorto-iliac disease. The GDG discussed
this and considered that there was no evidence to suggest that drug eluting
stents would be more cost effective at any particular site, and that their
recommendation should therefore apply to all sites.
There is no clinically relevant difference in benefit between the two stent types
and drug eluting stents are more costly. The GDG therefore formed a
consensus judgement that bare metal stent placement is the preferred option.
10.5 Autologous vein compared to prosthetic bypass

What is the clinical effectiveness of autologous vein versus prosthetic bypass graft for the treatment
of CLI in adults with PAD?
Final version
254
PAD
The review question sought to examine evidence for the type of graft to be used when bypass is
indicated in the patient. It was also necessary to consider the importance of the anatomical extent
and distribution of disease and co-morbidities that are likely to affect outcome such as diabetes.
A literature search was conducted for RCTs that compared the effectiveness of autologous vein
versus prosthetic bypass grafting. No time limit was placed on the literature search, and there were
no limitations on sample size. Indirect populations and emergency settings were excluded. One
Cochrane review was identified140 which considered graft type in bypass surgery for the femoro-
popliteal disease. The Cochrane review was not included or updated as it did not meet the protocol
defined by the GDG, which included all arteries of the leg. However it was cross checked for included
studies which matched the review protocol.
Two relevant RCTs153,154 were included in the review. The trials did not report outcome data for
people with diabetes.
The quality and results of included studies are reported in Table 97. The forest plots for each clinical
Final version
255
PAD
Table 97: Clinical evidence profile: Autologous vein versus prosthetic bypass for critical limb ischaemia due to femoro-popliteal disease

Quality
No of Risk of Other Autologous Prosthetic Relative
Design Inconsistency Indirectness Imprecision Absolute
studies bias considerations vein bypass (95% CI)
Perioperative mortality at 30 days
2153,154 RCT Serious(a) No serious No serious No serious None 0/76 0/75 not pooled not pooled MODERATE
Peri-operative amputation at 30 days
2153,154 RCT Serious(a) No serious No serious No serious None 0/76 8/75 RR 0.06 (0 100 fewer per 1000 MODERATE
inconsistency indirectness imprecision (0%) (10.7%) to 0.93) (from 7 fewer to 107
fewer)
Perioperative minor adverse event at 30 days
1154 RCT Serious(a) No serious No serious Very serious(c) None 5/25 4/24 RR 1.2 (0.37 33 more per 1000 VERY LOW
inconsistency indirectness (20%) (16.7%) to 3.94) (from 105 fewer to
490 more)
2153,154 RCT Serious(a) No serious No serious Very serious(c) None 1/76 7/75 RR 0.2 (0.04 75 fewer per 1000 VERY LOW
inconsistency indirectness (1.3%) (9.3%) to 1.11) (from 90 fewer to 10
more)
ABPI following surgery (no time point given)
1154 RCT Serious(a) No serious No serious Serious(d) None 25 24 - MD 0.07 lower (0.16 LOW
Final version
256
PAD
inconsistency indirectness lower to 0.02 higher)

(a) Unclear allocation concealment and participant blinding.
(b) No events in either intervention.
Final version
257
PAD
The cost-utility analysis by Hunick et al 1995 (reported in Table 89) subgrouped the results of their
clinical analysis by graft material. Although the study was not designed to directly compare the cost-
effectiveness of one type of material to another, according to the results of the model, bypass
surgery using autologous vein grafts results in higher quality of life and lower cost than bypass
surgery using synthetic grafts.
The GDG also discussed the cost of autologous and prosthetic grafts in an NHS context. The group
considered that although the same NHS Reference Cost applies to patients undergoing both
procedures, prosthetic veins cost several hundred pounds, varying widely depending on graft length
and material (official cost estimates were not available from standard sources). However, the
procedure associated with prosthetic vein bypass is slightly shorter than that for autologous vein as
there is no need to harvest the vein. In addition, the average hospital stay is slightly less for
prosthetic vein bypass operations. However, autologous vein bypass is associated with a reduced
rate of infection and fewer complications. Based on the clinical evidence and clinical experience, the
GDG agreed that autologous vein bypass was likely to represent the least costly of the two
procedures. A formal cost estimation was not undertaken as it was thought that this was
unnecessary (as the most effective option was also thought to be the least costly) and time
consuming.
10.5.2.1 Clinical
Autologous vein was significantly better than prosthetic bypass for:

 Amputation at 5 years [2 studies, 151 participants, moderate quality evidence]153,154
There were no statistically significant difference between autologous vein and prosthetic bypass for:
 Peri-operative minor adverse event at 30 days [1 study, 49 participants, very low quality
evidence]154
 Reintervention at 5 years [2 studies, 151 participants, very low quality evidence]153 154
 ABPI after surgery (no time point given by surgery) [1 study, 49 participants, low quality
evidence]154
There was no difference between autologous vein and prosthetic bypass for:
 Peri-operative mortality at 30 days, [2 studies, 151 participants, moderate quality evidence]153,154
 Mortality at 5 years, [1 study, 49 participants, moderate quality evidence]154
 Peri-operative amputation at 30 days, [1 study, 102 participants, moderate quality evidence]153
10.5.2.2 Economic
Final version
258
PAD
24.Use an autologous vein whenever possible for people with

Recommendation critical limb ischaemia having infra-inguinal bypass surgery.
Relative values of different The GDG were particularly interested in a comparison of amputation
outcomes rates between the 2 types of graft. The evidence suggested a better
outcome in this regard with autologous grafts although this was only
significant at one time point. The need for re-intervention was also better
(i.e. less re-intervention) with autologous grafts, but this difference fell
just short of statistical significance.
Information on quality of life, or any reflection of symptomatic well-

being, would have been useful in informing decisions, but these data
were not available from the retrieved papers.
Trade off between benefits and Observational studies (not reviewed as part of this question), and clinical
harms experience of the GDG suggest that prosthetic material is associated with
more infection and poorer limb salvage rates. As a result, there has been
a change in UK clinical practice away from use of prosthetic grafts. The
risk of MRSA infection in prosthetic graft has been linked with a higher
mortality rate than in patients undergoing autologous bypass. The GDG
felt that RCT evidence does not accurately reflect these important issues.
Economic considerations The GDG noted that prosthetic bypass is associated with a greater cost,
higher infection rate and higher 5-year rate of amputation compared to
autologous vein bypass. Indirectly, the economic model published by
Hunick 1995 suggested that autologous grafts were more cost-effective.
The GDG agreed that prosthetic vein bypass grafts do not represent a cost
effective use of NHS resources for people undergoing infra-inguinal
bypass surgery.
Quality of the evidence The GDG were disappointed that no recent evidence was identified as
part of the review. They also felt that the available papers did not identify
outcome data that they considered important, particularly regarding
infection rates. In the GDG experience, there is higher likelihood of
serious infection and death through use of prosthetics.
The mortality rate at 5 years was zero with both autologous and
prosthetic grafts. This is surprising in a cohort of patients with critical limb
ischaemia. Therefore the population in the study may not be
representative of all those with CLI.
One significant difference (in amputation rates) was noted and this was
graded as of moderate quality by GRADE criteria.
Other considerations Although, the was no clear benefit between autologous and prosthetic
bypass, the GDG felt that the recommendation should be made in favour
of autologous bypass. This was based on (a) their consensus view (b)
supported by the superiority in terms of amputation rate, the one
significant difference within the available data(c) supported by a non-
significant trend in re-intervention rates (d) and finally, based on their
assessment of the likely economic advantage of using autologous grafts..
Final version
259
PAD
Management of ischaemic pain in critical limb ischaemia
11 Management of ischaemic pain in critical limb

ischaemia
11.1 Introduction
Critical limb ischaemia (CLI) is characterised by persistent and severe ischaemic rest pain associated
with poor tissue perfusion, tissue loss and ulceration. The preferred option is to improve tissue
perfusion through endovascular or surgical treatment, therefore reducing the pain. In some cases,
however, such treatment is not possible. This may be due to un-reconstructable disease or degree of
tissue loss. Treatment to reperfuse the limb may have been attempted but have been unsuccessful,
or the patient’s preferences may be towards conservative treatment. This results in continued pain.
Whilst amputation is sometimes required, this outcome may be prevented or delayed if it is possible
to adequately control pain.
The impact of pain can vary between patients as pain is a very personal experience. Pain is typically
worse at night in bed because when the limb is elevated perfusion does not have gravity to assist it.
This results in sleep deprivation. It is common for patients to attempt sleep with their leg hanging out
of the bed or to choose to sleep in a chair. Ischaemic pain is often described by patients as a
relentless, unbearable, deep burning pain. It impacts on all aspects of their life as they are unable to
function properly. They are unlikely to pursue their normal activities and may well need help with
daily tasks. They often become irritable with strains placed on their relationships. Appetite is
compromised so they suffer nutritionally. Studies have highlighted that people with PAD have a fear
about increasing pain.17,19,20
Appropriate pain management is dependent on the accurate diagnosis of the cause of foot pain (see
chapter 7). This chapter deals with the management of ischaemia pain. Neuropathic pain, although
sometimes associated with CLI, will not be dealt with in this guideline and is covered in Neuropathic
pain: Pharmacological management, NICE clinical guideline CG96.155
11.2 Management options for pain in critical limb ischaemia

What is the clinical and cost effectiveness of chemical sympathectomy, opioids, gabapentin,
pregabalin or tricyclic antidepressants compared to each other in any combination for the
management of ischaemic pain in adults with critical limb ischemia?
To improve patient outcomes and quality of life, the GDG sought to identify RCT and observational
evidence for interventions to manage ongoing or escalating ischaemic pain. Spinal cord stimulation
was not included in the evidence review as the NICE technology appraisal 159156 does not
recommend its use for ischaemic pain outwith the context of a clinical trial. The treatments
considered in the review were: chemical sympathectomy, opioids, gabapentin, pregabalin or tricyclic
antidepressants (amitriptyline, nortiptyline and imipramine). The literature search was limited to
studies with a follow-up duration of more than one week and indirect populations were excluded.
No RCTs or observational studies which compared chemical sympathectomy, opioids, gabapentin,

pregabalin or tricyclic antidepressants to each other in any combination were identified.
Final version
260
PAD
No cost-effectiveness evidence comparing chemical sympathectomy, opioids, gabapentin, pregabalin

or tricyclic antidepressants to each other or in combination was identified in the literature. In the
absence of relevant published evidence, the GDG were presented with current UK costs to inform
decision making (Table 98 and Table 99).
Table 98: Cost of drugs for the treatment of ischaemic pain in critical limb ischaemia
Drugs commonly Total
Dose Cost per used to treat side Cost per cost per
Drug regimen 28 days Common side effects effects 28 days 28 days
Paracetamol 1 gram £3.23 Rare NA NA £3.23
(Generic) per day
Tramadol 50 – 100 £3.48 Dizziness, nausea, Cyclizine (50mg 3 £7.41 £10.89
(Generic) grams vomiting (esp. in acute times a day when
phase when treatment required-
starts and subsides assumed one 100-
over time –not a long tablet pack)
term side effect)
Co-codamol 2 tablets £8.18 Constipation, nausea, Laxatives (Senna – £4.27 £12.45
(30/500) four vomiting, dizziness, 2 tablets at night)
times a light-headedness,
day confusion, drowsiness
and urinary retention
Oxycodone 5mg £22.72 Constipation, nausea, Laxatives (Senna, £4.27 £26.99
(OXYNORM four vomiting, drowsiness, 2 tablets at night;
immediate times a pruritus, somnolence, Lactulose, 10ml
release day confusion twice a day)
capsules or when
liquid) required
Oxycodone 20mg £49.91 Constipation, nausea, Laxatives (Senna, £4.27 £54.18
(OXYCONTIN twice a vomiting, drowsiness, 2 tablets at night;
slow release day pruritus, somnolence, Lactulose, 10ml
tablets) confusion twice a day)
Morphine 10mg £10.56 Constipation, nausea, Laxatives (Senna, £4.27 £14.83
(Oramorph four vomiting, drowsiness, 2 tablets at night;
liquid or times a pruritus, somnolence, Lactulose, 10ml
Sevredol day confusion twice a day)
immediate when
release required
tablets)
Morphine 30mg £11.75 Constipation, nausea, Laxatives (Senna, £4.27 £16.02
(MST slow twice a vomiting, drowsiness, 2 tablets at night;
release day‡ pruritus, somnolence, Lactulose, 10ml
tablets) confusion twice a day)
Pregabalin 150mg £64.40 Dizziness, somnolence Discontinue use --- £64.40
(Lyrica twice a
capsules) day
Amtryptyline 50mg at £1.00 Constipation, dry Laxatives (Senna, £4.27 £5.27
(Generic) night mouth, sedation, 2 tablets at night;
cardiotoxicity, postural Lactulose, 10ml
hypotension, bladder twice a day)
problems
Gabapentin 300mg £7.42 Viral infection, Discontinue use --- £7.42
Final version
261
PAD
(Generic) three somnolence, dizziness,

times a ataxia, fatigue, fever
day
Source/Note: Dosing and side effect data based on expert opinion (Ammy Lam; personal communication). All cost data
is from the March 2011 British National Formulary {BNF 2011 /id}. Where there are multiple brands listed, the lowest cost
dose is reported; ‡ Where morphine is used in greater doses it will be associated with greater associated side effects.
Table 99: Cost of chemical sympathectomy

Average
National Average Average cost excess Total
average length of per excess bed average
Intervention Reference cost HRG unit cost stay bed day days cost
‡
Day cases (65%)
Chemical Complex pain £687 NA NA NA £687
sympathectomy procedures (AB03Z
Chemical destruction of
lumbar sympathetic
nerve)
Inpatient cases (35%)‡
Chemical Complex pain £1, 866 1.51 days £260 0.5 £1, 996
sympathectomy procedures (AB03Z
Chemical destruction of
lumbar sympathetic
nerve)
Total average cost of chemical sympathectomy = £1, 145
‡Based on the most recent Hospital Episode Statistics, there were a total of 752 admissions for ‘Chemical destruction of
lumbar sympathetic nerve’; 490 of these were day cases. It was assumed that the remaining 262 were inpatient procedures.
11.2.2.1 Clinical
No clinical evidence was identified.
11.2.2.2 Economic
No cost-effectiveness evidence was identified.
Final version
262
PAD
25.Offer paracetamol, and either weak or strong opioids

depending on the severity of pain, to people with critical
limb ischaemic pain.
26.Offer drugs such as laxatives and anti-emetics to manage the

adverse effects of strong opioids, in line with the person's
needs and preferences.
27.Refer people with critical limb ischaemic pain to a specialist

pain management service if any of the following apply:
 their pain is not adequately controlled or revascularisation
is inappropriate or impossible
 ongoing high doses of opioids are required for pain control
 pain persists after revascularisation or amputation.
28.Do not offer chemical sympathectomy to people with critical

Recommendations limb ischaemic pain, except in the context of a clinical trial.
Relative values of different For patients with pain associated with critical limb ischaemia, the GDG
outcomes considered pain relief and quality of life as the most important outcomes.
Improved quality of life which would include the ability to sleep, maintain
normal activities of daily living, maintaining a level of independence is of
high importance to the patient. No data on these outcomes were found.
In addition to considering various drugs with analgesic properties, the

GDG were particularly interested to look at evidence relating to chemical
sympathectomy, an old established operation which is still performed in
some centres. No satisfactory evidence was found.
Trade off between benefits and The GDG considered the side effects associated with each type of
harms analgesia (such as constipation, nausea and drowsiness). The group
agreed that a tiered approach to pain management would minimise
adverse events associated with stronger preparations while ensuring that
adequate pain relief was provided. The adequate management of pain
can improve a patient’s quality of life.
The GDG noted that prolonged use of pain medication is often associated
with side-effects, and that tolerance and dependence to pain relief need
to be considered. Patients should therefore be reviewed on a regular
basis. Particular note was taken of the potential risks of prolonged strong
opioid use, and the GDG felt that this situation should be one in which
advice from, and monitoring by, a pain specialist should be sought.
Economic considerations The GDG considered the cost of each analgesic treatment and the cost of
treating their associated side effects (e.g. laxatives for constipation). They
thought that a tiered approach to pain management would likely be the
most cost effective treatment strategy as mild preparations are generally
the least costly, have the fewest side effects and are often effective in
providing adequate pain relief.
The GDG considered the potential for improved pain management and
Final version
263
PAD
the cost of ineffective analgesics alongside the cost associated with

referral to specialist pain management services. The group agreed that
for people who require strong analgesic preparations, are taking a
maximum dosage and/or have poorly managed pain, the potential for
improved quality of life would be likely to justify the cost of specialist
treatment.
Quality of the evidence No RCT or observational evidence was identified to allow comparison of
chemical sympathectomy, opioids, gabapentin, pregabalin or tricyclic
anti-depressants in any combination for managing CLI pain. The
recommendations were based on GDG consensus and expert opinion.
Other considerations The GDG recognised that the management of CLI pain is often poor,
sometimes due to ischaemic pain being misdiagnosed but also because
the root cause is difficult to treat. Pain occurs through out the disease
process. Patients require increasing pain management towards the end of
the care pathway. Such patients may have had a failed revascularisation
procedure or are not considered suitable for revascularisation. Pain
management should be considered before referral for amputation
The GDG agreed that the principles of pain management for CLI are not
intrinsically different from managing pain in other chronic conditions.
Other NICE guidance such as Low back pain 157 and Osteoarthritis158 have
used a stepped approach in pain management. The WHO pain ladder is
widely used in the management of pain for various conditions. Pain relief
is escalated in cases of persistent or increasing pain. The GDG were of the
opinion, particularly in the absence of strong evidence, that a stepped
approach would also be appropriate for ischaemic leg pain.
The GDG agreed by consensus that pain management should begin with
paracetamol. Where this is insufficient, weak opioids such as Tramadol or
codeine should be given alongside paracetamol. Strong opioids such as
morphine or oxycodone are recommended for short term use only. Other
medications such as laxatives and anti-emetics should be offered
alongside strong opioids.
Patients should be commenced on a dose that suits their individual

needs. Dosing can be escalated where pain is not controlled. The optimal
dosing required for pain relief can differ between individuals and this
must be taken into account during the decision-making process.
These patients will have been managed in secondary care but are often
referred back to primary care for the management of their pain. It is
important, therefore, that there is clear guidance for pain management
for primary care representatives including when to refer to a pain
specialist.
The GDG suggested that the following good practice principles of pain
management:
 The person should be regularly reviewed when on pain medication to
ensure that it is giving adequate pain relief and no serious side effects.
Patient preference must also be considered.
 Any pain relief used without marketing authorisation for the treatment
of pain associated with must be documented and informed consent
taken.
Final version
264
PAD
 All pain relief measures must be monitored as per local protocols and in
accordance with the BNF.
Chemical sympathectomy
The GDG noted that observational evidence is available on chemical
sympathectomy, but this does not compare the procedure to the other
interventions of interest for this review question.
In current practice, chemical sympathectomy is undertaken where

revascularisation has not succeeded or is not an option and after other
pain relief options have failed, but usually prior to amputation. The
patients concerned tend to be those with non-healing ulcers, severe rest
pain or not responding to strong opioids. The GDG debated the concern
that there may be a minority of such patients who would benefit from
chemical sympathectomy and that any recommendation against offering
this may be harmful to these patients. They also acknowledged that the
technique for undertaking a chemical sympathectomy has changed; the
treatment is now performed using imaging to guide the needle. This new
technique has not been fully explored and may be more effective.
However, the GDG were also aware that the availability of chemical
sympathectomy varies around the country. It is not performed in some
areas because it is thought to be ineffective, and those who hold this view
regard sympathectomy as a procedure which delays the initiation of more
effective pain relief. The placebo effect was also noted to be common in
pain management techniques and without randomised controlled trials,
the true effect of a pain treatment can not be known.
The issue of chemical sympathectomy was discussed at length by the

GDG because of difficulty in obtaining consensus. It was agreed that
practice is variable, there is a reduction in overall usage and many centres
do not provide this treatment. It was also agreed that there was no
convincing formal evidence of benefit, but disagreement on the
consequences of this lack of evidence, some holding that use of
sympathectomy should not be discouraged until lack of benefit is proven,
while others felt that it would be wrong to continue the practice without
obtaining proper outcome evidence. The majority view was that
sympathectomy should only be offered in the context of a clinical trial,
and that this was the best way to ensure that practice was consistent and
that there would be the development of evidence to support future
guidance in this area.

What is the clinical and cost effectiveness of chemical sympathectomy in comparison with other
methods of pain control for managing critical limb ischaemic pain?
Approximately 1 in 5 people with critical limb ischaemia cannot be offered procedures to improve
the blood supply to their leg because of either the pattern of their disease or other comorbidities. In
this group the therapeutic options are pain control or primary amputation. Chemical lumbar
sympathectomy, which involves the destruction of the lumbar sympathetic chain (usually the L2 and
L3 ganglia), has been suggested to reduce pain and improve wound healing, and may prevent
amputation in some patients. Initially achieved surgically, it is now most commonly performed using
Final version
265
PAD
chemical agents such as phenol to destroy the lumbar sympathetic chain. Despite having been used
for over 60 years, the role of chemical lumbar sympathectomy remains unclear. Improvement in skin
blood flow and modification of pain perception control have been demonstrated, and this has
prompted the use of chemical lumbar sympathectomy for treating a range of conditions such as
regional pain syndrome, vasospastic conditions and critical limb ischaemia. However, in critical limb
ischaemia the use of chemical lumbar sympathectomy varies widely between units in England, the
mode of action and indications are unclear, and there is currently no randomised controlled trial
evidence demonstrating its clinical value. Therefore a randomised control trial comparing chemical
lumbar sympathectomy with other methods of pain relief is recommended.
Final version
266
PAD
Major amputation for critical limb ischaemia
12 Major amputation for critical limb ischaemia

12.1 Introduction
People with peripheral arterial disease (PAD) are usually offered amputation for ‘unreconstructable’
critical limb ischaemia (CLI). In other words, amputation is offered when ischaemic rest pain and/or
tissue loss (ulceration, gangrene), and any associated infection cannot be controlled by medical
therapy and when a multidisciplinary team (MDT) of vascular specialists has deemed that the blood
supply to the leg cannot be restored by means of angioplasty or bypass surgery. In a minority of
people, amputation has to be undertaken as emergency, usually because of overwhelming infection.
However, for the majority, amputation is only performed following a full discussion of the risks and
benefits of all the treatment options with the person and their family. In some people, a decision
may be taken to offer end of life care rather than amputation.
Amputations for PAD are commonly undertaken at the following levels:

1. Toe: one or more toes is removed, often with the metatarsal heads (the knuckles of the digits)
2. Transmetatarsal: all of the toes removed together with the metatarsal heads
3. Trans-tibial: the leg is removed about a hands-breadth below the knee (known as below knee
amputation, BKA)
4. Trans-femoral: the leg is removed about a hands-breadth above the knee (known as above knee
amputation, AKA)
People with toe and transmetatarsal amputations often suffer little long-term disability and such
amputations are often carried out for diabetic foot problems or for tissue loss in limbs that have
undergone successful revascularisation. Amputations above the ankle level are considered “major”
and are the subject of this review. People with BKA will usually be fitted with a functioning prosthesis
in the hope that they will learn to walk, although in the long term the majority spent most of their
time in a wheelchair. People with AKA are usually wheelchair-bound in the long term.
It is not possible to develop guidelines for amputation that cover every eventuality. Furthermore, the
decision to proceed to, and the timing of, amputation is contingent upon the wishes of the person
and their family. As such, it is likely that there will be significant variations in practice between
individuals as to if and when amputation is performed.
There are some people in whom revascularisation by means of angioplasty or bypass surgery is
technically possible but in whom the risks and likely long-term outcomes of such intervention are so
high and poor respectively that the person may be best served by primary amputation. While it is
reasonable on clinical and economic grounds to try to avoid amputation in most people, it is
important to avoid the all too common situation where the person undergoes repeated unsuccessful
attempts at revascularisation only to end up losing their leg. Not only is such a situation devastating
for the person and their family, it also represents a potentially inappropriate use of resources.
Furthermore, there is a significant body of evidence to suggest that failed revascularisation adversely
affects amputation level.
The avoidance of amputation depends crucially upon prompt diagnosis of PAD (CLI) and referral to a
specialist vascular unit that is able to offer the full range of available treatments including
amputation where appropriate. Following amputation is important that people are offered
appropriate rehabilitation and limb fitting services so that the physical and psychological impact of
limb loss can be minimised.
Final version
267
PAD
12.2 Review question

What are the clinical indications for major amputation for the management of pain in people with
critical limb ischaemia and does major amputation improve the quality of life in people with critical
limb ischaemia?
12.2.1 Clinical evidence

A literature search was conducted for all study designs that considered major amputation in people
with PAD. No time limit was placed on the literature search and there were no limitations on sample
size. Indirect populations and emergency settings were excluded. No evidence was identified which
considered the clinical indications for major amputation for the management of pain in people with
critical limb ischaemia. One observational trial159 was identified which compared quality of life in
people before and after major amputation for PAD. This paper was included in the review. The trial
did not report outcome data for people with diabetes.
The quality and results of included studies are reported in Table 100. The mapped EQ-5D results are
reported in Table 101. The forest plots for each clinical outcome are reported in Appendix H.
Final version
268
PAD
Table 100: Clinical evidence profile: Quality of life after amputation critical limb ischaemia
No of
Quality assessment Effect
patients
Quality
Design Inconsistency Indirectness Imprecision Amputation Absolute
1159 observational serious(a) no serious no serious serious none 6 patients See Table 101 VERY
study inconsistency indirectness imprecision(b) LOW
1159 observational serious(a) no serious no serious serious none 6 patients See Table 101 VERY
study inconsistency indirectness imprecision(b) LOW
(a) Study only included 6 patients.
Table 101: SF 36 individual domain results and mapped EQ5D results

Physical General Social Mapped
Time point functioning Role physical Bodily pain health Vitality functioning Role emotion Mental health EQ5D
Admission(a) 29 9 35 38 38 66 24 67 0.484
(a)
6 months 31 25 23 37 37 62 30 52 0.412
(a)
12 months 24 20 21 35 23 37 11 41 0.329
(a) Paper did not report measure of uncertainty
Final version
269
PAD

The literature was reviewed for evidence related to the costs and consequences of amputation for
people with PAD. One cost-utility model was identified. Brothers et al 1999144 developed a decision
analytic model to compare three alternative treatment options in people with limb-threatening CLI:
primary bypass; primary amputation; and non-operative expectant management. Based on the
results of their study, amputation is excluded from the analysis by dominance (i.e. it is less effective
and more expensive than bypass). Bypass is cost effective at a cost of £4, 712 per QALY. This study is
summarised in the evidence profile below (Table 102).
In the absence of relevant UK data, the GDG considered both the procedural cost and the cost of care
associated with amputation in people with PAD. The GDG considered these costs relative to the cost
of bypass (Table 106) and non-operative pain management are presented in section 11.2.1.2. These
estimates were also used to assess the applicability of the costs used in the Brothers 1999144 study to
a UK setting.
Final version
270
PAD
Table 102: Economic study characteristics: Primary bypass versus primary amputation versus non-operative expectant management
Study Limitations Applicability Other comments costs QALYs effectiveness
Brothers Potentially Partially  Decision analytic model Primary bypass Primary bypass Primary bypass One- and two-way
1999144 serious applicable(b)  Population: people with first was £5, 466 resulted in a costs £4, 712 sensitivity analyses were
limitations(a) presentation of limb- more expensive gain of 1.16 per QALY gained performed to evaluate the
threatening ischaemia caused than non- QALYs compared to effect of varying expected
by tibial-peoneal artery operative compared to non-operative utility, incremental costs,
occlusive disease expectant non-operative expectant early patency, late patency
management(a) expectant management(c) and peri-operative
 Outcomes: QALYs
management(b) mortality rates. The
 Costs: Hospital, outpatient authors reported the
and physician charges results of these analyses in
obtained from patient graphical form only and
records did not excluded
 Perspective: USA hospital dominated options,
therefore, it is not possible
to analyse the results of
these analyses.
Based on threshold
analysis, the authors
concluded that primary
amputation becomes the
most cost-effective
strategy when primary
bypass patency is less than
11%. Expectant
management is the most
cost-effective treatment
when operative mortality
for revascularisation or
amputation exceeds 55%.
(a) Long-term patient survival, limb salvage rate, and primary and cumulative secondary patency rates were obtained from the results of retrospective analyses previously conducted by the
authors with no evidence of a systematic search; utility values were obtained from people with CLI rather than patients who had experienced each health state QALY gain was considered
only over a 5-year horizon, therefore, this study will underestimate the long-term effect of reduced operative mortality expected from both the expectant management and primary
amputation strategies; unclear method of QALY elicitation and valuation.
Final version
271
PAD
(b) USA hospital perspective.

(c) Primary amputation was £2, 186 more costly than non-operative expectant management.
(d) Primary amputation resulted in a gain of 0.06 QALYs compared to non-operative management.
(e) Primary amputation is excluded by extended dominance.
Final version
272
PAD
Table 103: Costs of amputation procedure

average unit quartile unit quartile unit
Currency code Currency description Activity cost cost cost
QZ11A Amputations with 559 £13, 943 £8, 656 £16, 844
major complications
QA11B Amputations without 2, 625 £9, 644 £7, 154 £10, 872
major complications
QZ11A Amputations with 1, 100 £199 £33 £256
major complications
QZ11B Amputations without 6, 770 £230 £161 £280
major complications
Total average cost(a)
Amputations with major complications £14, 044
Amputations without major complications £9, 733
(a) Assuming 55% of procedures performed for PAD are performed during non-elective admissions.
The GDG provided estimates of resource use based on their experience and the expertise of
physiotherapists, prosthetists and commissioners that they work with. For simplicity, these costs
were classified according to those that occur in the first year after amputation and those occurring in
subsequent years. The resource use and unit costs associated with each element of care in the year
following amputation are presented in Table 104. Costs associated with care in each subsequent year
are presented in Table 105.
Table 104: Cost of care in the first year following an amputation

Prosthetic limbs
55% of amputees are fitted with a prosthetic limb
£1, 850 per above the knee prosthetic limb (expert
opinion)
£2, 650 per below the knee prosthetic limb (expert
opinion)
3 prosthetist appointments per patient £343 per appointment (NHS Reference Costs)
Wheelchairs
45% of amputees use wheelchairs
50% of these are non-motorised (assumption) £58 per year per non-motorised wheelchair
50% of these are motorised (assumption) £287 per year per motorised wheelchair
Inpatient rehabilitation
1 assessment for rehabilitation per patient (expert £306 per assessment (NHS Reference costs)
opinion)
50 days of rehabilitation per patient (expert opinion) £290 per bed day for amputation rehabilitation (NHS
Reference costs)
Outpatient rehabilitation
1 assessment for rehabilitation per patient (expert £307 per assessment (NHS Reference costs)
opinion)
2 physiotherapists per class (expert opinion) £37 (x 2) per hour (PRSSU)
Final version
273
PAD
1 physiotherapy technician (expert opinion) £22 per hour (PRSSU)

Room and equipment hire £15 per hour (expert opinion)
2 hours of class per week with 10 patients per class
8.5 weeks of rehabilitation for below the knee and
13 weeks for above the knee amputations
Wound care
2.5 nurse visits per week (expert opinion) £24 per home visit from a district nurse (PRSSU
2010) and £10 of wound care supplies used per
home visit (expert opinion)
90% have a non-complicated wound with an average
healing time of 12 weeks (expert opinion)
10% have a complicated wound with an average
healing time of 32 weeks (expert opinion)
Care home
36% of formerly independent patients require a care
home (assumption)
47 weeks per year (assumption) £986 per week (PRSSU 2010)
Community care & home modifications
64% of formerly independent patients remain in the
community (Taylor 2005, Larson 1998)
Half of patients remaining in the community will £296 per week (PRSSU 2010)
require care in the community (assumption)
All patients remaining in the community will have
some form of home modification (expert opinion)
1 concrete ramp £390 (PRSSU 2010)
3 grab rails £53 each (PRSSU 2010)
Relocation of toilet/other home renovation £1, 754 (PRSSU 2010)
Total average cost per patient in the first year following amputation = £28, 270
Table 105: Annual cost of care following the first year for patients with an amputation
Care home
36% of formerly independent patients require a care
home (assumption)
47 weeks per year (assumption) £986 per week (PRSSU 2010)
Community care
64% of formerly independent patients remain in the
community (Taylor 2005, Larson 1998)
Half of patients remaining in the community will £296 per week (PRSSU 2010)
require care in the community (assumption)
Wheelchair
45% of amputees use wheelchairs
50% of these are non-motorised (assumption) £58 per year per non-motorised wheelchair
50% of these are motorised (assumption) £287 per year per motorised wheelchair
Total average cost per patient = £23, 502
Final version
274
PAD
Table 106: Cost of bypass procedure

Currency average unit quartile unit quartile unit
code Currency description Activity cost cost cost
Elective inpatient (long stay) HRG data
QZ02A Lower limb arterial surgery with 3, 074 £6, 481 £4, 707 £7, 913
complications
QZ02B Lower limb arterial surgery without 1, 770 £4, 886 £3, 767 £5, 611
complications
Elective inpatient (long stay) excess bed day HRG data
QZ02A Lower limb arterial surgery with 1, 579 £302 £206 £327
complications
QZ02B Lower limb arterial surgery without 360 £217 £137 £276
complications
Total average cost – elective
Elective bypass with major complications £7, 009 (£5, 067 - £8, 485)
Elective bypass without complications £5, 954 (£4, 441 - £6, 969)
QZ02A Lower limb arterial surgery with 2, 768 £8, 229 £6, 187 £9, 948
complications
QZ02B Lower limb arterial surgery without 622 £6, 120 £4, 086 £7, 341
complications
QZ02A Lower limb arterial surgery with 8, 097 £232 £162 £298
complications
QZ02B Lower limb arterial surgery without 1, 014 £285 £189 £301
complications
Total average cost – Non elective
Elective bypass with major complications £8, 308 (£6, 241 - £10, 050)
Elective bypass without complications £6, 295 (£4, 202 - £7, 525)
Bypass (assuming 10% non elective)
Bypass with major complications £7, 139 (£5, 185 - £8, 641)
Bypass without complications £5, 988 (£4, 417 - £7, 025)
Source/Note: All costs obtained from 2009/10 NHS Reference Costs49
12.2.3.1 Clinical
Clinical indications for major amputation:
No clinical evidence was identified for the clinical indications for major amputation.
Quality of life in people before and after undergoing major amputation for PAD:
In patients with CLI who had had major amputation for PAD their quality of life as based on EQ5D
mapped68 from SF36 data reported in study159:
 Decreased from 0.484 to 0.412 between admission and 6 months [1 study, very low quality
evidence]159
Final version
275
PAD
 Decreased from 0.484 to 0.329 between admission and 12 months [1 study, very low quality
evidence]159
 Decreased from 0.412 to 0.329 between 6 months and 12 months [1 study, very low quality
evidence]159
12.2.3.2 Economic
No economic evidence was identified for the indications for amputation.
29.Do not offer major amputation to people with critical limb

ischaemia unless all options for revascularisation have been
Recommendation considered by a vascular multidisciplinary team.
Relative values of different The GDG considered quality of life as an important outcome. The small amount
outcomes of evidence available showed that quality of life falls after amputation. No
comparative data was available.
Trade off between clinical Major amputation is associated with high risk of mortality and morbidity and is
benefits and harms therefore considered as a last measure for the treatment of pain associated
with CLI. Specifically, the post-operative mortality rate for amputation is the
highest of all vascular procedures. People can further develop pressure sores,
phantom limb pain, and stump problems. In addition, further amputation is
common. There is also the loss of independence and emotional difficulties.
Economic considerations The GDG considered the cost of amputation compared to strategies such as
bypass surgery and non-operative pain management. They also considered the
results of the clinical review which found a decrease in quality of life following
amputation. However, there was no comparative clinical evidence of
alternative methods of management.
Based on the results of the clinical and economic review and clinical
experience of the GDG, the group thought that primary amputation is unlikely
to represent a cost-effective use of NHS resources, unless all other options
have been exhausted.
Quality of evidence One study reported on the change in quality of life following major
amputation.159 The study was graded as low quality by GRADE criteria. The
study included six patients undergoing major amputation. The study did not
define major amputation. The current review mapped the SF36 scores to
EQ5D. The results of this mapping show that there was a decrease for quality
of life at all study time-points.
It was noted that bodily pain was reported as worse after major amputation.
The study was not randomised and it was discussed that there may be a
patient selection bias i.e. patients reporting worse pain received amputation.
Other considerations The evidence available was extremely limited. It did not support the use of
amputation, but the GDG recognised that amputation may be necessary to
relieve severe symptoms that cannot be controlled in other ways and for
people with life-threatening disease in whom revascularisation is not an option
(patients with tissue loss, sepsis, infection and non-healing wounds). Bearing in
mind the cost and overall poor results of amputation, it was felt that patients
Final version
276
PAD
in whom this was being considered would usefully be discussed by a multi-

disciplinary team before proceeding to surgery.
It was agreed that the MDT membership was not part of the scope but the
GDG in discussion noted that most services within England and Wales already
have access to a multi-disciplinary team. This may vary by locality. There is a
quality indicator framework which describes the minimum requirements for
patients undergoing major amputation and the structure of multi-disciplinary
teams. The NICE Diabetic foot guideline (CG119) also recommends the key
professionals who should be involved in the multi-disciplinary team
management, some of whom may be useful in this setting e.g. tissue viability
and wound care specialist. The GDG discussed that some patients will still
undergo major amputation without being considered for revascularisation
(angioplasty and bypass) and wanted to make a recommendation which
discourages this practice.
Patient choice was also emphasised. Some people may wish to proceed
straight to amputation even where there are potential options for
revascularisation.

The GDG highlighted this recommendation as a key priority for
implementation. The GDG were concerned that some patients maybe having
amputations for disease that could be treated if all possible options for
management were considered by a multi-disciplinary team. This
recommendation would have a high impact on reducing variation in care and
have a high impact on patient outcomes.
Final version
277
PAD
Glossary
13 Glossary
Abstract Summary of a study, which may be published alone or as an introduction to a
full scientific paper.
Algorithm (in guidelines) A flow chart of the clinical decision pathway described in the guideline, where
decision points are represented with boxes, linked with arrows.
Allocation concealment 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.
Applicability The degree to which the results of an observation, study or review are likely to
hold true in a particular clinical practice setting.
Arm (of a clinical study) Sub-section of individuals within a study who receive one particular
intervention, for example placebo arm
Association Statistical relationship between two or more events, characteristics or other
variables. The relationship may or may not be causal.
Baseline The initial set of measurements at the beginning of a study (after run-in period
where applicable), with which subsequent results are compared.
Bias 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.
Blinding Keeping the study participants, caregivers, researchers and outcome assessors
unaware about the interventions to which the participants have been
allocated in a study.
Carer (caregiver) Someone other than a health professional who is involved in caring for a
person with a medical condition.
Clinical efficacy The extent to which an intervention is active when studied under controlled
research conditions.
Clinical effectiveness The extent to which an intervention produces an overall health benefit in
routine clinical practice.
Clinician A healthcare professional providing direct patient care, for example doctor,
nurse or physiotherapist.
Cochrane Review The Cochrane Library consists of a regularly updated collection of evidence-
based medicine databases including the Cochrane Database of Systematic
Reviews (reviews of randomised controlled trials prepared by the Cochrane
Collaboration).
Cohort study 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.
Comorbidity Co-existence of more than one disease or an additional disease (other than
that being studied or treated) in an individual.
Comparability Similarity of the groups in characteristics likely to affect the study results (such
as health status or age).
Concordance 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.
Final version
278
PAD
Glossary
Confidence interval (CI) 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.
Confounding 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.
Consensus methods 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.
Control group 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.
Cost benefit analysis A type of economic evaluation where both costs and benefits of healthcare
treatment are measured in the same monetary units. If benefits exceed costs,
the evaluation would recommend providing the treatment.
Cost-consequences A type of economic evaluation where various health outcomes are reported in
analysis (CCA) addition to cost for each intervention, but there is no overall measure of
health gain.
Cost-effectiveness analysis An economic study design in which consequences of different interventions
(CEA) 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.
Cost-effectiveness model 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.
Cost-utility analysis (CUA) A form of cost-effectiveness analysis in which the units of effectiveness are
quality-adjusted life-years (QALYs).
Credible Interval The Bayesian equivalent of a confidence interval.
Decision analysis 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.
Discounting 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.
Dominance An intervention is said to be dominated if there is an alternative intervention
that is both less costly and more effective.
Drop-out A participant who withdraws from a trial before the end.
Economic evaluation Comparative analysis of alternative health strategies (interventions or
programmes) in terms of both their costs and consequences.
Effect (as in effect The observed association between interventions and outcomes or a statistic
measure, treatment effect, to summarise the strength of the observed association.
estimate of effect, effect
size)
Effectiveness See ‘Clinical effectiveness’.
Final version
279
PAD
Glossary
Efficacy See ‘Clinical efficacy’.

Epidemiological study 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.
EQ-5D (EuroQol-5D) A standardise instrument used to measure a health outcome. It provides a
single index value for health status.
Evidence Information on which a decision or guidance is based. Evidence is obtained
from a range of sources including randomised controlled trials, observational
studies, expert opinion (of clinical professionals and/or patients).
Exclusion criteria Explicit standards used to decide which studies should be excluded from
(literature review) consideration as potential sources of evidence.
Exclusion criteria (clinical Criteria that define who is not eligible to participate in a clinical study.
study)
Extended dominance 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.
Extrapolation In data analysis, predicting the value of a parameter outside the range of
observed values.
Follow-up 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.
Generalisability 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.
Harms Adverse effects of an intervention.
Health economics 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.
Health-related quality of A combination of an individual’s physical, mental and social well-being; not
life (HRQoL) merely the absence of disease.
Heterogeneity Or lack of The term is used in meta-analyses and systematic reviews when the results or
homogeneity. 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.
Imprecision Results are imprecise when studies include relatively few patients and few
events and thus have wide confidence intervals around the estimate of effect.
Inclusion criteria (literature Explicit criteria used to decide which studies should be considered as potential
review) sources of evidence.
Incremental analysis The analysis of additional costs and additional clinical outcomes with different
interventions.
Incremental cost The mean cost per patient associated with an intervention minus the mean
cost per patient associated with a comparator intervention.
Final version
280
PAD
Glossary
Incremental cost The difference in the mean costs in the population of interest divided by the
effectiveness ratio (ICER) differences in the mean outcomes in the population of interest for one
treatment compared with another.
Incremental net benefit The value (usually in monetary terms) of an intervention net of its cost
(INB) 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.
Indirectness The available evidence is different to the review question being addressed, in
terms of PICO (population, intervention, comparison and outcome).
Intention to treat analysis A strategy for analysing data from a randomised controlled trial. All
(ITT) 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.
Intervention Healthcare action intended to benefit the patient, for example, drug
treatment, surgical procedure, psychological therapy.
Intra-operative The period of time during a surgical procedure.
Kappa statistic A statistical measure of inter-rater agreement that takes into account the
agreement occurring by chance.
Length of stay The total number of days a participant stays in hospital.
Licence See ‘Product licence’.
Life-years gained Mean average years of life gained per person as a result of the intervention
compared with an alternative intervention.
Likelihood ratio 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.
Long-term care Residential care in a home that may include skilled nursing care and help with
everyday activities. This includes nursing homes and residential homes.
Markov model 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).
Meta-analysis 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.
Multivariate model A statistical model for analysis of the relationship between two or more
predictor (independent) variables and the outcome (dependent) variable.
Negative predictive value A measure of the usefulness of a screening/diagnostic test. It is the proportion
(NPV) [In of those with a negative test result who do not have the disease, and can be
screening/diagnostic tests:] interpreted as the probability that a negative test result is correct.
Number needed to treat The number of patients that who on average must be treated to prevent a
(NNT) single occurrence of the outcome of interest.
Observational study Retrospective or prospective study in which the investigator observes the
natural course of events with or without control groups; for example, cohort
studies and case–control studies.
Odds ratio 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.
Final version
281
PAD
Glossary
Opportunity cost 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.
Outcome 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’.
P-value 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’.
Perioperative The period from admission through surgery until discharge, encompassing the
pre-operative and post-operative periods.
Placebo An inactive and physically identical medication or procedure used as a
comparator in controlled clinical trials.
Positive predictive value In screening/diagnostic tests: A measure of the usefulness of a
(PPV) 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.
Postoperative Pertaining to the period after patients leave the operating theatre, following
surgery.
Post-test probability For diagnostic tests. The proportion of patients with that particular test result
who have the target disorder (post test odds/[1 + post-test odds]).
Power (statistical) 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.
Preoperative The period before surgery commences.
Pre-test probability 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.
Primary care Healthcare delivered to patients outside hospitals. Primary care covers a range
of services provided by general practitioners, nurses, dentists, pharmacists,
opticians and other healthcare professionals.
Primary outcome The outcome of greatest importance, usually the one in a study that the
power calculation is based on.
Product licence An authorisation from the MHRA to market a medicinal product.
Prognosis 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.
Prospective study 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.
Publication bias 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.
Final version
282
PAD
Glossary
Quality of life See ‘Health-related quality of life’.

Quality-adjusted life year An index of survival that is adjusted to account for the patient’s quality of life
(QALY) 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 cost-
utility analysis. The QALYs gained are the mean QALYs associated with one
treatment minus the mean QALYs associated with an alternative treatment.
Randomisation 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.
Randomised controlled A comparative study in which participants are randomly allocated to
trial (RCT) intervention and control groups and followed up to examine differences in
outcomes between the groups.
Receiver operated A graphical method of assessing the accuracy of a diagnostic test. Sensitivity Is
characteristic (ROC) curve 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.
Reference standard 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.
Relative risk (RR) 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).
Reporting bias See publication bias.
Resource implication The likely impact in terms of finance, workforce or other NHS resources.
Retrospective study A retrospective study deals with the present/ past and does not involve
studying future events. This contrasts with studies that are prospective.
Review question In guideline development, this term refers to the questions about treatment
and care that are formulated to guide the development of evidence-based
recommendations.
Secondary outcome An outcome used to evaluate additional effects of the intervention deemed a
priori as being less important than the primary outcomes.
Selection bias 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 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’
Sensitivity analysis 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
Final version
283
PAD
Glossary
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).
Significance (statistical) A result is deemed statistically significant if the probability of the result
occurring by chance is less than 1 in 20 (p <0.05).
Specificity 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.
Stakeholder Those with an interest in the use of the guideline. Stakeholders include
manufacturers, sponsors, healthcare professionals, and patient and carer
groups.
Systematic review 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.
Time horizon The time span over which costs and health outcomes are considered in a
decision analysis or economic evaluation.
Treatment allocation Assigning a participant to a particular arm of the trial.
Univariate Analysis which separately explores each variable in a data set.
Utility A measure of the strength of an individual’s 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.
Final version
284
PAD
Glossary
Final version
285
PAD
References
14 References
1 Fowkes FG, Housley E, Cawood EH, Macintyre CC, Ruckley CV, Prescott RJ. Edinburgh Artery
Study: prevalence of asymptomatic and symptomatic peripheral arterial disease in the general
population. International Journal of Epidemiology. 1991; 20(2):384-392
2 Kannel WB, Skinner JJ, Jr., Schwartz MJ, Shurtleff D. Intermittent claudication. Incidence in the
Framingham Study. Circulation. 1970; 41(5):875-883
3 American Diabetes Association. Peripheral arterial disease in people with diabetes. Diabetes
Care. 2003; 26(12):3333-3341
4 Leng GC, Lee A, Fowkes FG, Whiteman M, Dunbar J, Housley E et al. Incidence, natural history
and cardiovascular events in symptomatic and asymptomatic peripheral arterial disease in the
general population. International Journal of Epidemiology. 1996; 25(6):1172-1181
5 Hirsch AT, Haskal ZJ, Hertzer NR, Bakal CW, Creager MA, Halperin JL et al. ACC/AHA 2005
Practice Guidelines for the management of patients with peripheral arterial disease (lower
extremity, renal, mesenteric, and abdominal aortic): a collaborative report from the American
Association for Vascular Surgery/Society for Vascular Surgery, Society for Cardiovascular
Angiography and Interventions, Society for Vascular Medicine and Biology, Society of
Interventional Radiology, and the ACC/AHA Task Force on Practice Guidelines (Writing
Committee to Develop Guidelines for the Management of Patients With Peripheral Arterial
Disease): endorsed by the American Association of Cardiovascular and Pulmonary
Rehabilitation; National Heart, Lung, and Blood Institute; Society for Vascular Nursing;
TransAtlantic Inter-Society Consensus; and Vascular Disease Foundation. Circulation. 2006;
113(11):e463-e654
6 Park KB, Do YS, Kim DI, Kim DK, Kim YW, Shin SW et al. The TransAtlantic InterSociety
Consensus (TASC) classification system in iliac arterial stent placement: long-term patency and
clinical limitations.[Erratum appears in J Vasc Interv Radiol. 2007 May;18(5):695]. Journal of
Vascular & Interventional Radiology. 2007; 18(2):193-201
7 Muntner P, Wildman RP, Reynolds K, Desalvo KB, Chen J, Fonseca V. Relationship between
HbA1c level and peripheral arterial disease. Diabetes Care. 2005; 28(8):1981-1987
8 Norgren L, Hiatt W, Dormandy JA, Nehler MR, Harris KA, Fowkes F et al. Inter-Society
Consensus for the management of peripheral arterial disease. Journal of Vascular Surgery.
2007; 45(Suppl S):S5-S67
9 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/aboutnice/howwework/developingniceclinicalguidelines/clinicalguidel
inedevelopmentmethods/GuidelinesManual2009.jsp
10 National Institute for Health and Clinical Excellence. Guide to the methods of technology
appraisal. London: National Institute for Health and Clinical Excellence; 2008. Available from:
http://www.nice.org.uk/media/B52/A7/TAMethodsGuideUpdatedJune2008.pdf
11 Jaeschke R, Singer J, Guyatt GH. Measurement of health status: Ascertaining the minimal
clinically important difference. Controlled Clinical Trials. 1989; 10(4):407-415
Final version
286
PAD
References
12 Schunemann HJ, Puhan M, Goldstein R, Jaeschke R, Guyatt GH. Measurement properties and
interpretability of the Chronic respiratory disease questionnaire (CRQ). COPD: Journal of
Chronic Obstructive Pulmonary Disease. 2005; 2(1):81-89
13 Schunemann HJ, Guyatt GH. Commentary--goodbye M(C)ID! Hello MID, where do you come
from? Health Services Review. 2005; 40(2):593-597
14 GRADE Working Group. The Grading of Recommendations Assessment, Development and

Evaluation (GRADE) Working Group website. 2011. [Last accessed: 30 January 2012]
15 Organisation for Economic Co-operation and Development (OECD). OECD stat extracts:
purchasing power parities for GDP. 2010. Available from:
http://stats.oecd.org/Index.aspx?datasetcode=SNA_TABLE4 [Last accessed: 10 November
2010]
16 National Institute of Clinical Excellence. Social value judgements: principles for the
development of NICE guidance. 2008. Available from:
www.nice.org.uk/aboutnice/howwework/socialvaluejudgements.jsp
17 Gibson JM, Kenrick M. Pain and powerlessness: the experience of living with peripheral
vascular disease. Journal of Advanced Nursing. 1998; 27(4):737-745
18 Leech JE. Psychosocial and physiologic needs of patients with arterial occlusive disease during
the preoperative phase of hospitalization. Heart and Lung. 1982; 11(5):442-449
19 Treat-Jacobson D, Halverson SL, Ratchford A, Regensteiner JG, Lindquist R, Hirsch AT. A patient-
derived perspective of health-related quality of life with peripheral arterial disease. Journal of
Nursing Scholarship. 2002; 34(1):55-60
20 Wann-Hansson C, Hallberg IR, Klevsgard R, Andersson E. Patients' experiences of living with

peripheral arterial disease awaiting intervention: a qualitative study. International Journal of
Nursing Studies. 2005; 42(8):851-862
21 National Collaborating Centre for Mental Health. The treatment and management of
depression in adults with chronic physical health problems (partial update of CG23). Leicester
and London: The British Psychological Society and The Royal College of Psychiatrists, 2009
Available from: http://guidance.nice.org.uk/CG91
22 Hirsch AT, Criqui MH, Treat-Jacobson D, Regensteiner JG, Creager MA, Olin JW et al. Peripheral
arterial disease detection, awareness, and treatment in primary care. JAMA. 2001;
286(11):1317-1324
23 MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk
individuals: a randomised placebo controlled trial. Lancet. 2002; 360(9326):7-22
24 Lip GY FAU, Makin AJ. Treatment of hypertension in peripheral arterial disease. Cochrane
Database of Systematic Reviews. 2004; Issue 3:CD003075
25 Ostergren J, Sleight P, Dagenais G, Danisa K, Bosch J, Qilong Y et al. Impact of ramipril in

patients with evidence of clinical or subclinical peripheral arterial disease. European Heart
Journal. 2004; 25(1):17-24
26 Orchard TJ, Strandness DE, Jr. Assessment of peripheral vascular disease in diabetes. Report
and recommendations of an international workshop sponsored by the American Diabetes
Final version
287
PAD
References
Association and the American Heart Association September 18-20, 1992 New Orleans,
Louisiana. Circulation. 1993; 88(2):819-828
27 Antirhombotic Trialist' Collaboration. Collaborative meta-analysis of randomised trials of

antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk
patients. BMJ. 2002; 324(7330):141
28 Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA et al. Association of glycaemia
with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35):
prospective observational study. BMJ. 2000; 321(7258):405-412
29 Yusuf S, Sleight P, Pogue J, Davies R, Dagenais G. Effects of an angiotensin-converting-enzyme

inhibitor, ramipril, on cardiovascular events in high-risk patients. New England Journal of
Medicine. 2000; 342(3):145-153
30 National Institute for Health and Clinical Excellence. Clopidogrel and modified-release
dipyridamole in the prevention of occlusive vascular events. NICE technology appraisal
guidance 210. London: National Institute for Clinical Excellence (NICE), 2010 Available from:
http://guidance.nice.org.uk/TA210
31 Baxter GM, Polak JF. Lower limb colour flow imaging: A comparison with ankle: brachial
measurements and angiography. Clinical Radiology. 1993; 47(2):91-95
32 Janssen A. Pulsatility index is better than ankle-brachial doppler index for non-invasive
detection of critical limb ischaemia in diabetes. Vasa. 2005; 34(4):235-241
33 Premalatha G, Ravikumar R, Sanjay R, Deepa R, Mohan V. Comparison of colour duplex

ultrasound and ankle-brachial pressure index measurements in peripheral vascular disease in
type 2 diabetic patients with foot infections. Journal of the Association of Physicians of India.
2002; 50:1240-1244
34 Schroder F, Diehm N, Kareem S, Ames M, Pira A, Zwettler U et al. A modified calculation of

ankle-brachial pressure index is far more sensitive in the detection of peripheral arterial
disease. Journal of Vascular Surgery. 2006; 44(3):531-536
35 Guo X, Li J, Pang W, Zhao M, Luo Y, Sun Y et al. Sensitivity and specificity of ankle-brachial index
for detecting angiographic stenosis of peripheral arteries. Circulation Journal. 2008; 72(4):605-
610
36 Gornik HL, Garcia B, Wolski K, Jones DC, Macdonald KA, Fronek A. Validation of a method for
determination of the ankle-brachial index in the seated position. Journal of Vascular Surgery.
2008; 48(5):1204-1210
37 Collins R, Cranny G, Burch J, Guiar-Ibanez R, Craig D, Wright K et al. A systematic review of

duplex ultrasound, magnetic resonance angiography and computed tomography angiography
for the diagnosis and assessment of symptomatic, lower limb peripheral arterial disease.
Health Technology Assessment. 2007; 11(20):iii-iiv
38 Schernthaner R, Stadler A, Lomoschitz F, Weber M, Fleischmann D, Lammer J et al.

Multidetector CT angiography in the assessment of peripheral arterial occlusive disease:
accuracy in detecting the severity, number, and length of stenoses. European Radiology. 2008;
18(4):665-671
Final version
288
PAD
References
39 Kreitner KF, Kunz RP, Herber S, Martenstein S, Dorweiler B, Dueber C. MR angiography of the
pedal arteries with gadobenate dimeglumine, a contrast agent with increased relaxivity, and
comparison with selective intraarterial DSA. Journal of Magnetic Resonance Imaging. 2008;
27(1):78-85
40 Bueno A, Acin F, Canibano C, Fernandez-Casado JL, Castillo E. Diagnostic accuracy of contrast-

enhanced magnetic resonance angiography and duplex ultrasound in patients with peripheral
vascular disease. Vascular and Endovascular Surgery. 2010; 44(7):576-585
41 Eiberg JP, Gronvall Rasmussen JB, Hansen MA, Schroeder TV. Duplex ultrasound scanning of
peripheral arterial disease of the lower limb. European Journal of Vascular and Endovascular
Surgery. 2010; 40(4):507-512
42 Gjonnaess E, Morken B, Sandbaek G, Stranden E, Slagsvold CE, Jorgensen JJ et al. Gadolinium-

enhanced magnetic resonance angiography, colour duplex and digital subtraction angiography
of the lower limb arteries from the aorta to the tibio-peroneal trunk in patients with
intermittent claudication. European Journal of Vascular and Endovascular Surgery. 2006;
31(1):53-58
43 Kos S, Reisinger C, Aschwanden M, Bongartz GM, Jacob AL, Bilecen D. Pedal angiography in
peripheral arterial occlusive disease: first-pass i.v. contrast-enhanced MR angiography with
blood pool contrast medium versus intraarterial digital subtraction angiography. American
Journal of Roentgenology. 2009; 192(3):775-784
44 Napoli A, Anzidei M, Zaccagna F, Cavallo Marincola B, Zini C, Brachetti G et al. Peripheral

arterial occlusive disease: diagnostic performance and effect on therapeutic management of
64-Section CT angiography. Radiology. 2011; 261(3):976-986
45 Collins T, Lunos S. Home-based walking therapy improves walking ability and quality of life in
persons with diabetes mellitus and peripheral arterial disease. Vascular Medicine. 2010; 15
(2):155
46 de Vries M, Ouwendijk R, Flobbe K, Nelemans PJ, Kessels AG, Schurink GH et al. Peripheral
arterial disease: clinical and cost comparisons between duplex US and contrast-enhanced MR
angiography - a multicenter randomized trial. Radiology. 2006; 240:401-410
47 Kock MC, Adriaensen ME, Pattynama PM, Van Sambeek MR, van UH, Stijnen T et al. DSA versus
multi-detector row CT angiography in peripheral arterial disease: randomized controlled trial.
Radiology. 2005; 237(2):727-737
48 Ouwendijk R, de Vries M, Pattynama PM, Van Sambeek MRHM, De Haan MW, Stijnen T et al.
Imaging peripheral arterial disease: A randomized controlled trial comparing contrast-
enhanced MR angiography and multi-detector row CT angiography. Radiology. 2005;
236(3):1094-1103
49 Department of Health. NHS reference costs 2009-2010. 2011. Available from:

http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidan
ce/DH_123459 [Last accessed: 1 March 2011]
50 de Vries M, Ouwendijk R, Flobbe F, Nelemans PJ, Kessels AGH, Schurink GWH et al. Peripheral
artery disease: Clinical and cost comparison between duplex ultrasonography and contrast-
enhanced magnetic resonance angiography - A multicenter randomized trial. Nederlands
Tijdschrift Voor Geneeskunde. 2007; 151(32):1789-1794
Final version
289
PAD
References
51 Curtis, L. The costs of health & social care 2010. Canterbury: Personal Social Services Research
Unit, 2010
52 National Collaborating Centre for Primary Care. Clinical Guidelines and Evidence Review for
Lipid Modification: cardiovascular risk assessment and the primary and secondary prevention
of cardiovascular disease. London: National Collaborating Centre for Primary Care and Royal
College of General Practitioners, 2008 Available from: http://publications.nice.org.uk/lipid-
modification-cg67
53 Squires H, Simpson E, Meng Y, Harnan S, Stevens J, Wong R. Cilostazol, nadtidrofuryl oxalate,

pentoxifylline and inositol nicotinate for the treatment of intermittent claudication in people
with peripheral arterial disease. Health Technology Assessment. 2010;
54 Bendermacher BL, Willigendael EM, Teijink JA, Prins MH. Supervised exercise therapy versus
non-supervised exercise therapy for intermittent claudication. Cochrane Database of
Systematic Reviews. 2006; Issue 2:CD005263
55 Watson L, Ellis B, Leng GC. Exercise for intermittent claudication. Cochrane Database of
56 Cheetham DR, Burgess L, Ellis M, Williams A, Greenhalgh RM, Davies AH. Does supervised
exercise offer adjuvant benefit over exercise advice alone for the treatment of intermittent
claudication? A randomised trial. European Journal of Vascular and Endovascular Surgery.
2004; 27(1):17-23
57 Kakkos SK, Geroulakos G, Nicolaides AN. Improvement of the walking ability in intermittent
claudication due to superficial femoral artery occlusion with supervised exercise and
pneumatic foot and calf compression: a randomised controlled trial. European Journal of
Vascular and Endovascular Surgery. 2005; 30(2):164-175
58 Nicolai SP, Teijink JA, Prins MH. Multicenter randomized clinical trial of supervised exercise
therapy with or without feedback versus walking advice for intermittent claudication. Journal
of Vascular Surgery. 2010; 52(2):348-355
59 Pinto B, Marcus BH, Patterson RB, Roberts M, Colucci A, Braun C. On-site versus home exercise
programs: psychological benefits for individuals with arterial claudication. Journal of Aging and
Physical Activity. 1997; 5(4):311-328
60 Savage P, Ricci MA, Lynn M, Gardner A, Knight S, Brochu M et al. Effects of home versus
supervised exercise for patients with intermittent claudication. Journal of Cardiopulmonary
Rehabilitation. 2001; 21(3):152-157
61 Regensteiner JG, Meyer TJ, Krupski WC, Cranford LS, Hiatt WR. Hospital vs home-based
exercise rehabilitation for patients with peripheral arterial occlusive disease. Angiology. 1997;
48(4):291-300
62 Stewart AH, Smith FC, Baird RN, Lamont PM. Local versus systemic mechanisms underlying
supervised exercise training for intermittent claudication. Vascular and Endovascular Surgery.
2008; 42(4):314-320
63 Tew G, Nawaz S, Zwierska I, Saxton JM. Limb-specific and cross-transfer effects of arm-crank
exercise training in patients with symptomatic peripheral arterial disease. Clinical Science.
2009; 117(12):405-413
Final version
290
PAD
References
64 Tisi PV, Hulse M, Chulakadabba A, Gosling P, Shearman CP. Exercise training for intermittent
claudication: does it adversely affect biochemical markers of the exercise-induced
inflammatory response? European Journal of Vascular and Endovascular Surgery. 1997;
14(5):344-350
65 Treat-Jacobson D, Bronas UG, Leon AS. Efficacy of arm-ergometry versus treadmill exercise
training to improve walking distance in patients with claudication. Vascular Medicine. 2009;
14(3):203-213
66 Zwierska I, Walker RD, Choksy SA, Male JS, Pockley AG, Saxton JM. Upper- vs lower-limb
aerobic exercise rehabilitation in patients with symptomatic peripheral arterial disease: a
randomized controlled trial. Journal of Vascular Surgery. 2005; 42(6):1122-1130
67 Gardner AW, Parker DE, Montgomery PS, Scott KJ, Blevins SM. Efficacy of quantified home-
based exercise and supervised exercise in patients with intermittent claudication: a
randomized controlled trial. Circulation. 2011; 123(5):491-498
68 Ara R, Brazier J. Deriving an algorithm to convert the eight mean SF-36 dimension scores into a
mean EQ-5D preference-based score from published studies (where patient level data are not
available). Value in Health. 2008;(7):1131-1143
69 van Asselt AD, Nicolai SP, Joore MA, Prins MH, Teijink JA. Cost-effectiveness of exercise therapy
in patients with intermittent claudication: supervised exercise therapy versus a 'go home and
walk' advice. European Journal of Vascular and Endovascular Surgery. 2011; 41(1):97-103
70 Lee HL, Mehta T, Ray B, Heng MST, McCollum P, Chetter IC. A non-randomised controlled trial
of the clinical and cost effectiveness of a supervised exercise programme for claudication.
European Journal of Vascular and Endovascular Surgery. 2007; 33:202-207
71 Thompson PD. Exercise prescription and proscription for patients with coronary artery disease.
Circulation. 2005; 112(15):2354-2363
72 Menard JR, Smith HE, Riebe D, Braun CM, Blissmer B, Patterson RB. Long-term results of
peripheral arterial disease rehabilitation. Journal of Vascular Surgery. 2004; 39(6):1186-1192
73 Criqui MH, Langer RD, Fronek A, Feigelson HS, Klauber MR, McCann TJ et al. Mortality over a
period of 10 years in patients with peripheral arterial disease. New England Journal of
Medicine. 1992; 326(6):381-386
74 Heran BS, Chen JMH, Ebrahim S, Moxham T, Rees K, Thompson DR et al. Exercise-based cardiac
rehabilitation for coronary heart disease. Cochrane Database of Systematic Reviews. 2011;
Issue 7:CD001800
75 Anderson KM, Odell PM, Wilson PW, Kannel WB. Cardiovascular disease risk profiles. American
Heart Journal. 1991; 121(1 Pt 2):293-298
76 Payne R. Excel worksheet for calculating cardiovascular risk. 2010. Available from:
http://cvrisk.mvm.ed.ac.uk/calculator/excelcalc.htm [Last accessed: 25 January 2012]
77 Craig, R and Mindell, J. Health Survey for England 2006. Volume 1: Cardiovascular Disease and
Risk Factors in Adults. Leeds: The Information Centre, 2008
Final version
291
PAD
References
78 National Clinical Guideline Centre. Hypertension: the clinical managment of primary

hypertension in adults. London: Royal College of Physicians, 2011 Available from:
http://guidance.nice.org.uk/CG127
79 Fowkes F, Murray GD, Butcher I, Heald CL, Lee RJ. Ankle brachial index combined with
Framingham risk score to predict cardiovascular events and mortality. JAMA. 2008; 300(2):197-
208
80 Lee CD, Folsom AR, Blair SN. Physical activity and stroke risk: a meta analysis. Stroke. 2003;
34(10):2475-2481
81 Briggs A, Claxton K, Sculpher M. Decision modelling for health economic evaluation. Oxford
University Press; 2006
82 Hiatt WR, Hirsch AT, Regensteiner JG, Brass EP, and the Vascular Clinial Trialists. Clinical trials
for claudication: assessment of exercise performance, functional status, and clinical end points.
Circulation Journal. 1995; 92:614-621
83 Guest JF, Davie AM, Clegg JP. Cost effectiveness of cilostazol compared with naftidrofuryl and
pentoxifylline in the treatment of intermittent claudication in the UK. Current Medical
Research and Opinion. 2005; 21(6):817-826
84 NHS Business Services Authority. NHS electronic drug tariff. 2011. Available from:
http://www.ppa.org.uk/edt/August_2011/mindex.htm [Last accessed: 1 August 2011]
85 Nylaende M, Abdelnoor M, Stranden E, Morken B, Sandbaek G, Risum O et al. The Oslo balloon
angioplasty versus conservative treatment study (OBACT)--the 2-years results of a single
centre, prospective, randomised study in patients with intermittent claudication. European
Journal of Vascular and Endovascular Surgery. 2007; 33(1):3-12
86 Nylaende M, Kroese AJ, Morken B, Stranden E, Sandbaek G, Lindahl AK et al. Beneficial effects
of 1-year optimal medical treatment with and without additional PTA on inflammatory markers
of atherosclerosis in patients with PAD. Results from the Oslo Balloon Angioplasty versus
Conservative Treatment (OBACT) study. Vascular Medicine. 2007; 12(4):275-283
87 Whyman MR, Fowkes FG, Kerracher EM, Gillespie IN, Lee A, Housley E et al. Is intermittent
claudication improved by percutaneous transluminal angioplasty? A randomized controlled
trial. Journal of Vascular Surgery. 1997; 26(4):551-557
88 Whyman MR, Fowkes FG, Kerracher EM, Gillespie IN, Lee A, Housley E et al. Randomised
controlled trial of percutaneous transluminal angioplasty for intermittent claudication.
European Journal of Vascular and Endovascular Surgery. 1996; 12(2):167-172
89 Greenhalgh RM, Belch JJ, Brown LC, Gaines PA, Gao L, Reise JA et al. The adjuvant benefit of
angioplasty in patients with mild to moderate intermittent claudication (MIMIC) managed by
supervised exercise, smoking cessation advice and best medical therapy: results from two
randomised trials for stenotic femoropopliteal and aortoiliac arterial disease. European Journal
of Vascular and Endovascular Surgery. 2008; 36(6):680-688
90 Mazari FA, Gulati S, Rahman MN, Lee HL, Mehta TA, McCollum P et al. Early outcomes from a
randomized, controlled trial of supervised exercise, angioplasty, and combined therapy in
intermittent claudication. Annals of Vascular Surgery. 2010; 24(1):69-79
Final version
292
PAD
References
91 Spronk S, Bosch JL, den Hoed PT, Veen HF, Pattynama PM, Hunink MG. Intermittent
claudication: clinical effectiveness of endovascular revascularization versus supervised
hospital-based exercise training--randomized controlled trial. Radiology. 2009; 250(2):586-595
92 Perkins JM, Collin J, Creasy TS, Fletcher EW, Morris PJ. Exercise training versus angioplasty for
stable claudication. Long and medium term results of a prospective, randomised trial.
European Journal of Vascular and Endovascular Surgery. 1996; 11(4):409-413
93 Creasy TS, McMillan PJ, Fletcher EW, Collin J, Morris PJ. Is percutaneous transluminal
angioplasty better than exercise for claudication? Preliminary results from a prospective
randomised trial. European Journal of Vascular Surgery. 1990; 4(2):135-140
94 Lundgren F, Dahllof AG, Lundholm K, Schersten T, Volkmann R. Intermittent claudication.

Surgical reconstruction or physical training? A prospective randomized trial of treatment
efficiency. Annals of Surgery. 1989; 209(3):346-355
95 Mazari FAK, Khan JA, Carradice D, Samuel N, Abdul Rahman MNA, Gulati S et al. Randomized
clinical trial of percutaneous transluminal angioplasty, supervised exercise and combined
treatment for intermittent claudication due to femoropopliteal arterial disease. British Journal
of Surgery. 2012; 99(1):39-48
96 Kruidenier LM, Nicola SP, Rouwet EV, Peters RJ, Prins MH, Teijink JAW. Additional supervised
exercise therapy after a percutaneous vascular intervention for peripheral arterial disease: A
randomized clinical trial. Journal of Vascular and Interventional Radiology. 2011; 22(7):961-968
97 Fowkes F, Leng GC. Bypass surgery for chronic lower limb ischaemia. Cochrane Database of
98 Greenhalgh RM. MIMIC Trials: Angioplasty effective in randomised controlled trials for
peripheral arterial disease. 2009. Available from: http://www.cxvascular.com/in-latest-
news?ccs=485&cs=4222 [Last accessed: 2 February 2009]
99 Mazari FA, Carradice D, Rahman MN, Khan JA, Mockford K, Mehta T et al. An analysis of
relationship between quality of life indices and clinical improvement following intervention in
patients with intermittent claudication due to femoropopliteal disease. Journal of Vascular
Surgery. 2010; 52(1):77-84
100 Spronk S, Bosch JL, den Hoed PT, Veen HF, Pattynama PM, Hunink MG. Cost-effectiveness of
endovascular revascularization compared to supervised hospital-based exercise training in
patients with intermittent claudication: a randomized controlled trial. Journal of Vascular
Surgery. 2008; 48(6):1472-1480
101 Kedora J, Hohmann S, Garrett W, Munschaur C, Theune B, Gable D. Randomized comparison of

percutaneous Viabahn stent grafts vs prosthetic femoral-popliteal bypass in the treatment of
superficial femoral arterial occlusive disease. Journal of Vascular Surgery. 2007; 45(1):10-16
102 McQuade K, Gable D, Pearl G, Theune B, Black S. Four-year randomized prospective

comparison of percutaneous ePTFE/nitinol self-expanding stent graft versus prosthetic
femoral-popliteal bypass in the treatment of superficial femoral artery occlusive disease.
Journal of Vascular Surgery. 2010; 52(3):584-590
103 McQuade K, Gable D, Hohman S, Pearl G, Theune B. Randomized comparison of ePTFE/nitinol

self-expanding stent graft vs prosthetic femoral-popliteal bypass in the treatment of superficial
femoral artery occlusive disease. Journal of Vascular Surgery. 2009; 49(1):109-115
Final version
293
PAD
References
104 Wilson SE, Wolf GL, Cross AP. Percutaneous transluminal angioplasty versus operation for
peripheral arteriosclerosis. Report of a prospective randomized trial in a selected group of
patients. Journal of Vascular Surgery. 1989; 9(1):1-9
105 Wolf GL, Wilson SE, Cross AP, Deupree RH, Stason WB. Surgery or balloon angioplasty for
peripheral vascular disease: a randomized clinical trial. Principal investigators and their
Associates of Veterans Administration Cooperative Study Number 199. Journal of Vascular and
Interventional Radiology. 1993; 4(5):639-648
106 Holm J, Arfvidsson B, Jivegard L, Lundgren F, Lundholm K, Schersten T et al. Chronic lower limb
ischaemia. A prospective randomised controlled study comparing the 1-year results of vascular
surgery and percutaneous transluminal angioplasty (PTA). European Journal of Vascular
Surgery. 1991; 5(5):517-522
107 van der Zaag ES, Legemate DA, Prins MH, Reekers JA, Jacobs MJ. Angioplasty or bypass for
superficial femoral artery disease? A randomised controlled trial. European Journal of Vascular
and Endovascular Surgery. 2004; 28(2):132-137
108 Bosch JL, Tetteroo E, Mali WP, Hunink MG. Iliac arterial occlusive disease: cost-effectiveness
analysis of stent placement versus percutaneous transluminal angioplasty. Dutch Iliac Stent
Trial Study Group. Radiology. 1998; 208(3):641-648
109 de Vries SO, Visser K, de Vries JA, Wong JB, Donaldson MC, Hunink MG. Intermittent
claudication: cost-effectiveness of revascularization versus exercise therapy. Radiology. 2002;
222(1):25-36
110 Hunink MG, Wong JB, Donaldson MC, Meyerovitz MF, de Vries J, Harrington DP.
Revascularization for femoropopliteal disease. A decision and cost-effectiveness analysis.
JAMA. 1995; 274(2):165-171
111 Visser K, de Vries SO, Kitslaar PJ, van Engelshoven JMA, Hunink MGM. Cost-effectiveness of
diagnostic imaging work-up and treatment for patients with intermittent claudication in the
Netherlands. European Journal of Vascular and Endovascular Surgery. 2003; 25:213-233
112 Bosch JL, Haaring C, Meyerovitz MF, Cullen KA, Hunink MG. Cost-effectiveness of percutaneous
treatment of iliac artery occlusive disease in the United States. American Journal of
Roentgenology. 2000; 175(2):517-521
113 Bosch JL, van der Graaf Y, Hunink MG. Health-related quality of life after angioplasty and stent
placement in patients with iliac artery occlusive disease: results of a randomized controlled
clinical trial. The Dutch Iliac Stent Trial Study Group. Circulation. 1999; 99(24):3155-3160
114 Dormandy J, Heeck L, Vig S. The natural history of claudication: risk to life and limb. Seminars in
Vascular Surgery. 1999; 12(2):123-137
115 Aulivola B, Craig RM. Decision making in the dysvascular lower extremity. Foot and Ankle
Clinics. 2010; 15(3):391-409
116 Axisa B, Fishwick G, Bolia A, Thompson MM, London NJM, Bell PRF et al. Complications
following peripheral angioplasty. Annals of the Royal College of Surgeons of England. 2002;
84:39-42
Final version
294
PAD
References
117 Cejna M, Thurnher S, Illiasch H, Horvath W, Waldenberger P, Hornik K et al. PTA versus Palmaz
stent placement in femoropopliteal artery obstructions: a multicenter prospective randomized
study. Journal of Vascular and Interventional Radiology. 2001; 12(1):23-31
118 Tetteroo E, van der Graaf Y, Bosch JL, van Engelen AD, Hunink MG, Eikelboom BC et al.
Randomised comparison of primary stent placement versus primary angioplasty followed by
selective stent placement in patients with iliac-artery occlusive disease. Dutch Iliac Stent Trial
Study Group. Lancet. 1998; 351(9110):1153-1159
119 Dick P, Wallner H, Sabeti S, Loewe C, Mlekusch W, Lammer J et al. Balloon angioplasty versus
stenting with nitinol stents in intermediate length superficial femoral artery lesions.
Catheterization and Cardiovascular Interventions. 2009; 74(7):1090-1095
120 Krankenberg H, Schluter M, Steinkamp HJ, Burgelin K, Scheinert D, Schulte KL et al. Nitinol
stent implantation versus percutaneous transluminal angioplasty in superficial femoral artery
lesions up to 10 cm in length: the femoral artery stenting trial (FAST). Circulation. 2007;
116(3):285-292
121 Schillinger M, Sabeti S, Loewe C, Dick P, Amighi J, Mlekusch W et al. Balloon angioplasty versus
implantation of nitinol stents in the superficial femoral artery. New England Journal of
Medicine. 2006; 354(18):1879-1888
122 Vroegindeweij D, Vos LD, Tielbeek A, Buth J, vd Bosch HC. Balloon angioplasty combined with
primary stenting versus balloon angioplasty alone in femoropopliteal obstructions: A
comparative randomized study. Cardiovascular and Interventional Radiology. 1997; 20(6):420-
425
123 Schillinger M, Minar E. Endovascular stent implantation for treatment of peripheral artery
disease. European Journal of Clinical Investigation. 2007; 37(3):165-170
124 Schillinger M, Sabeti S, Dick P, Amighi J, Mlekusch W, Schlager O et al. Sustained benefit at 2
years of primary femoropopliteal stenting compared with balloon angioplasty with optional
stenting. Circulation. 2007; 115(21):2745-2749
125 Currie IC, Wilson YG, Baird RN, Lamont PM. Treatment of intermittent claudication: the impact
on quality of life. European Journal of Vascular and Endovascular Surgery. 1995; 10(3):356-361
126 Goodacre S, Nicholl J, Dixon S, Cross E, Angelini K, Arnold J et al. Randomised controlled trial
and economic evaluation of a chest pain observation unit compared with routine care. BMJ.
2004; 328(7434):254
127 Tengs TO, Lin TH. A meta-analysis of quality-of-life estimates for stroke. PharmacoEconomics.
2003; 21(3):191-200
128 Sculpher M, Michaels J, McKenna M, Minor J. A cost-utility analysis of laser-assisted

angioplasty for peripheral arterial occlusions. International Journal of Technology Assessment
in Health Care. 1996; 12:104-125
129 Twine CP, Coulston J, Shandall A, McLain AD. Angioplasty versus stenting for superficial femoral
artery lesions. Cochrane Database of Systematic Reviews. 2009; Issue 2:CD006767
130 Bachoo P, Thorpe PA, Maxwell H, Welch K. Endovascular stents for intermittent claudication.
Cochrane Database of Systematic Reviews. 2010; Issue 1:CD003228
Final version
295
PAD
References
131 Sabeti S, Czerwenka-Wenkstetten A, Dick P, Schlager O, Amighi J, Mlekusch I et al. Quality of

life after balloon angioplasty versus stent implantation in the superficial femoral artery:
Findings from a randomized controlled trial. Journal of Endovascular Therapy. 2007; 14(4):431-
437
132 Grimm J, Muller-Hulsbeck S, Jahnke T, Hilbert C, Brossmann J, Heller M. Randomized study to

compare PTA alone versus PTA with Palmaz stent placement for femoropopliteal lesions.
Journal of Vascular and Interventional Radiology. 2001; 12(8):935-942
133 Klein WM, van der Graaf Y, Seegers J, Moll FL, Mali WP. Long-term cardiovascular morbidity,
mortality, and reintervention after endovascular treatment in patients with iliac artery disease:
The Dutch Iliac Stent Trial Study. Radiology. 2004; 232(2):491-498
134 Laird JR. Nitinol stent implantation versus balloon angioplasty for lesions in the superficial
femoral artery and proximal popliteal artery: Twelve-month results from the RESILIENT
randomized trial. Circulation: Cardiovascular Interventions. 2010; 3(3):267-276
135 Greenberg D, Rosenfield K, Garcia LA, Berezin RH, Lavelle T, Fogleman S et al. In-hospital costs
of self-expanding nitinol stent implantation versus balloon angioplasty in the femoropopliteal
artery (The VascuCoil Trial). Journal of Vascular and Interventional Radiology. 2004;
15(10):1065-1069
136 Dake MD, Ansel GM, Jaff MR, Ohki T, Saxon RR, Smouse HB et al. Paclitaxel-eluting stents show
superiority to balloon angioplasty and bare metal stents in femoropopliteal disease: twelve-
month Zilver PTX randomized study results. Circulation: Cardiovascular Interventions. 2011;
4(5):495-504
137 Klein WM, van der Graaf Y, Seegers J, Spithoven JH, Buskens E, Van Baal JG et al. Dutch Iliac
Stent Trial: Long-term results in patients randomized for primary or selective stent placement.
Radiology. 2006; 238(2):734-744
138 Rastan A, Tepe G, Krankenberg H, Zahorsky R, Beschorner U, Noory E et al. Sirolimus-eluting

stents vs. bare-metal stents for treatment of focal lesions in infrapopliteal arteries: A double-
blind, multi-centre, randomized clinical trial. European Heart Journal. 2011; 32(18):2274-2281
139 Cox J, Koutroumanos N. Comparing coding between interventional radiologists and hospital
coding departments. Clinical Audit. 2010; 2:33-34
140 Twine CP, McLain A. Graft type for femoro-popliteal bypass surgery. Cochrane Database of
141 Klinkert P, Schepers A, Burger DH, Van Bockel JH, Breslau PJ. Vein versus
polytetrafluoroethylene in above-knee femoropopliteal bypass grafting: five-year results of a
randomized controlled trial. Journal of Vascular Surgery. 2003; 37(1):149-155
142 Burger DH, Kappetein AP, Van Bockel JH, Breslau PJ. A prospective randomized trial comparing
vein with polytetrafluoroethylene in above-knee femoropopliteal bypass grafting. Journal of
143 Bradbury AW, Adam DJ, Bell J, Forbes JF, Fowkes FGR, Gillespie I et al. Multicentre randomised
controlled trial of the clinical and cost-effectiveness of a bypass-surgery-first versus a balloon-
angioplasty-first revascularisation strategy for severe limb ischaemia due to infrainguinal
disease. The Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial. Health
Technology Assessment. 2010; 14(14):1-236
Final version
296
PAD
References
144 Brothers TE, Rios GA, Robison JG, Elliott BM. Justification of intervention for limb-threatening
ischemia: a surgical decision analysis. Cardiovascular Surgery. 1999; 7(1):62-69
145 Rand T, Basile A, Cejna M, Fleischmann D, Funovics M, Gschwendtner M et al. PTA versus
carbofilm-coated stents in infrapopliteal arteries: pilot study. Cardiovascular and Interventional
Radiology. 2006; 29(1):29-38
146 Rand T, Lammer J, Rabbia C, Maynar M, Zander T, Jahnke T et al. Percutaneous transluminal
angioplasty versus turbostatic carbon-coated stents in infrapopliteal arteries: InPeria II trial.
Radiology. 2011; 261(2):634-642
147 Randon C, Jacobs B, De Ryck F, Vermassen F. Angioplasty or primary stenting for infrapopliteal
lesions: results of a prospective randomized trial. Cardiovascular and Interventional Radiology.
2010; 33(2):260-269
148 Zdanowski Z, Albrechtsson U, Lundin A, Jonung T, Ribbe E, Thorne J et al. Percutaneous

transluminal angioplasty with or without stenting for femoropopliteal occlusions? A
randomized controlled study. International Angiology. 1999; 18(4):251-255
149 Brodmann M, Froehlich H, Dorr A, Gary T, Portugaller RH, Deutschmann H et al. Percutaneous
transluminal angioplasty versus primary stenting in infrapopliteal arteries in critical limb
ischemia. Vasa. 2011; 40(6):482-490
150 Duda SH, Pusich B, Richter G, Landwehr P, Oliva VL, Tielbeek A et al. Sirolimus-eluting stents for
the treatment of obstructive superficial femoral artery disease: six-month results. Circulation.
2002; 106(12):1505-1509
151 Duda SH, Bosiers M, Lammer J, Scheinert D, Zeller T, Oliva V et al. Drug-eluting and bare nitinol
stents for the treatment of atherosclerotic lesions in the superficial femoral artery: long-term
results from the SIROCCO trial. Journal of Endovascular Therapy. 2006; 13(6):701-710
152 Duda SH, Bosiers M, Lammer J, Scheinert D, Zeller T, Tielbeek A et al. Sirolimus-eluting versus
bare nitinol stent for obstructive superficial femoral artery disease: the SIROCCO II trial. Journal
of Vascular and Interventional Radiology. 2005; 16(3):331-338
153 Ballotta E. Prospective randomized study on bilateral above-knee femoropopliteal

revascularization: Polytetrafluoroethylene graft versus reversed saphenous vein. Journal of
154 Tilanus HW, Obertop H, van Urk H. Saphenous vein or PTFE for femoropopliteal bypass. A
prospective randomized trial. Annals of Surgery. 1985; 202(6):780-782
155 National Institute for Health and Clinical Excellence. Neuropathic pain: the pharmacological
management of neuropathic pain in adults in non-specialist settings. London: National
Institute for Clinical Excellence (NICE), 2010 Available from: http://guidance.nice.org.uk/CG96
156 National Institute for Health and Clinical Excellence. Spinal cord stimulation for chronic pain of
neuropathic or ischaemic origin. NICE technology appraisal guidance 159. London: National
Institute for Clinical Excellence (NICE), 2008 Available from: http://guidance.nice.org.uk/TA159
157 National Collaborating Centre for Primary Care. Early management of persistent non-specific
low back pain. London: Royal College of General Practitioners, 2009 Available from:
Final version
297
PAD
References
158 National Collaborating Centre for Chronic Conditions. The care and management of
osteoarthritis in adults. London: Royal College of Physicians, 2008 Available from:
159 Hernandez-Osma E, Cairols MA, Marti X, Barjau E, Riera S. Impact of treatment on the quality
of life in patients with critical limb ischaemia. European Journal of Vascular and Endovascular
Surgery. 2002; 23(6):491-494
Final version
298
PAD
Full appendices in separate document
Appendices
Appendix A: Full appendices in separate
document
End of document
Final version
299

PAD Guideline 2018

Uploaded by

Copyright:

Available Formats

PAD Guideline 2018

Uploaded by

Document Information

Original Title

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

PAD Guideline 2018

Uploaded by

Copyright:

Available Formats

Lower limb peripheral arterial

NICE Clinical Guideline 147

Published by the National Clinical Guideline Centre at

First published 2012

© National Clinical Guideline Centre - 2012

3.3.6 Grading the quality of clinical evidence............................................................... 29

6.2 Recommendation ................................................................................................................ 60

exercise, best medical treatment plus angioplasty ........................................... 132

10.4.2 Evidence statements ......................................................................................... 252

Appendices ............................................................................................................................... 299

Guideline development group members

Guideline development group – co-optee

National clinical guideline centre technical team

1.2 Specific risk factors

1.3 Definitions and classification systems

Table 1: Guideline definitions of peripheral arterial disease

Severe Limb Ischaemia

1.3.1 Classification of PAD based on symptom severity

1.3.2 Classification of PAD based on anatomical distribution of disease

Table 2: Broad anatomical definitions of peripheral arterial disease

1.3.3 Issues surrounding definitions of PAD

1.3.3.1 Ankle Brachial Pressure Index (APBI)

1.3.3.2 Use of classifications in clinical settings

1.4 Initial management

1.5 Secondary care

1.6 Importance to the NHS

2 Development of the guideline

NICE clinical guidelines can:

We produce our guidelines using the following steps:

The NCGC and NICE produce a number of versions of this guideline:

2.3 Who developed this guideline?

2.4 What this guideline covers

2.5 What this guideline does not cover

2.6 Relationships between the guideline and other NICE guidance

Related NICE Interventional Procedures:

Related NICE Clinical Guidelines:

 Obesity. NICE clinical guideline 43 (2006). Available from http://publications.nice.org.uk/obesity-

Related NICE Public Health Guidance:

3.1 Developing the review questions and outcomes

Table 3: List of guideline review questions

adults with critical limb ischaemia? survival (all)

3.1.1 Clinical outcomes not considered

3.1.2 Health related quality of life

3.2 Searching for evidence

3.2.1.1 Call for evidence

3.2.2 Health economic literature search

3.3 Evidence of effectiveness

o Observational studies: data presented as a range of values in GRADE profiles

3.3.2 Methods of combining clinical studies

3.3.2.1 Data synthesis for intervention reviews

3.3.2.2 Data synthesis for diagnostic test accuracy review

3.3.3 Type of studies

3.3.4 Types of analysis

3.3.5 Appraising the quality of evidence by outcomes

Table 4: Description of quality elements in GRADE for intervention studies

Table 5: Levels of quality elements in GRADE

Table 6: Overall quality of outcome evidence in GRADE

of effect and may change the estimate.

3.3.6 Grading the quality of clinical evidence