134 Clinical methods and pathophysiology

Aortic systolic pressure derived with different calibration

methods: associations to brachial systolic pressure in the
general population
Siegfried Wassertheurera, Bernhard Hametnera, Christopher C. Mayera,
Ahmed Hafezb, Kazuaki Negishic, Theodore G. Papaioannoud,
Athanase D. Protogeroue, James E. Sharmanc and Thomas Weberb

Background There is increasing evidence that the method (R2 = 0.937) (Williams’ test, P < 0.001 for comparison).
of calibration directly influences the association between Subgroup analysis showed the major influence of sex and
brachial systolic blood pressure (bSBP) and estimated aortic heart rate. The association between bSBP, aSBP1 (R2 = 0.83),
systolic blood pressure (aSBP) and subsequently affects and aSBP2 (R2 = 0.66), respectively, reduced significantly for
prognostic and diagnostic differentiation power of the latter. borderline hypertensives (P < 0.001 for comparison).
Objective The aim of this study was to investigate Conclusion In contrast to aSBP1, the association between
associations between different methods of systolic bSBP and aSBP2 is significantly less dominant and
pressure assessment in a large cohort and its comparison therefore aSBP2 may have potential prognostic superiority
with recently published evidence. over bSBP. Blood Press Monit 23:134–140 Copyright ©
2018 Wolters Kluwer Health, Inc. All rights reserved.
Participants and methods During a public health
campaign, cardiovascular hemodynamic data were assessed Blood Pressure Monitoring 2018, 23:134–140
using a validated oscillometric device in a pharmacy setting. Keywords: aortic blood pressure, arterial stiffness, brachial blood pressure,
The device measures bSBP, mean arterial pressure, and calibration
diastolic blood pressure (DBP), and records brachial a
Department of Biomedical Systems, Center for Health & Bioresources, Austrian
waveforms at the DBP level. aSBP1 was derived using bSBP Institute of Technology (AIT), Vienna, bDepartment of Cardiology, Klinikum Wels-
Grieskirchen, Wels, Austria, cMenzies Institute for Medical Research, University of
and DBP and aSBP2 using measured mean arterial pressure Tasmania, Hobart, Tasmania, Australia, dBiomedical Engineering Unit, First
and DBP for waveform calibration. In addition to pressures, Department of Cardiology, Medical School, Hippokration Hospital and
e
Cardiovascular Prevention and Research Unit, Department of Pathophysiology,
age, sex, and anthropometric data were recorded. Regression ‘Laiko’ Hospital, Medical School, National and Kapodistrian University of Athens,
analysis was carried out to investigate associations. Athens, Greece

Results A total of 7409 (5133/2276, female/male) individuals Correspondence to Siegfried Wassertheurer, MSc, PhD, Center for Health &
Bioresources, Austrian Institute of Technology (AIT), Donau-City Street 1,
with a median age of 54 years were sampled. aSBPs differed 1220 Vienna, Austria
significantly from bSBP (126.0 mmHg) for aSBP1 Tel: + 43 505 504 830; fax: + 43 505 504 840;
e-mail: [email protected]
(117.0 mmHg) and aSBP2 (127.5 mmHg, both P < 0.0001).
Regression analysis showed that aSBP2 (R2 = 0.853) is Received 9 October 2017 Revised 27 February 2018 Accepted 5 March 2018
significantly less associated with bSBP than aSBP1

Introduction variable for aSBP1 estimation, an inherent link is established

Noninvasive brachial systolic blood pressure (bSBP) proved systematically and a strong association between variables is
to be a powerful predictor in cardiovascular disease for over a predestined. Therefore, aSBP2 calibration waives the use of
century on the basis of the method introduced by Riva- bSBP and utilizes measured mean arterial pressure (MAP)
Rocci. The concept of aortic systolic blood pressure (aSBP) instead to attenuate the influence of bSBP. Among several
separate from bSBP evolved as a clinical concept within the studies, aSBP2 consistently showed less association with
last two decades [1]. Recently, it has been argued that the bSBP compared with traditional calibrated aSBP1, and fur-
strong association between both parameters a-priori limits thermore, aSBP2 estimates were closer to invasive measured
the incremental clinical value of aSBP [2]. In contrast, actual aortic pressures [3–6]. However, the investigated cohorts
clinical evidence suggests that this association is modifiable were limited in size and it is still unclear whether the
by the measurement procedure, in particular, by the cali- observed associations between pressures are preserved in
bration method [3–6]. Noninvasive systolic aortic pressure the general population.
estimates are commonly derived by either systolic and
diastolic (aSBP1) pressure or mean and diastolic pressure Therefore, the primary aim of this work is to investigate
(aSBP2) calibration [7]. By using bSBP as a direct input associations between different methods of aSBP assessment
1359-5237 Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/MBP.0000000000000319

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.

 Aortic systolic pressure Wassertheurer et al. 135

in a large cohort. We aimed to investigate the mechanisms of shape amplification into a physical unit such as ‘Torr’
systemic differences introduced by different calibrations and (mmHg) or ‘Pascal’ (Pa) requires an external (sphygmo)
to determine the inherent clinical potential and outcomes. manometer. As reported earlier [18], the Mobil-O-Graph
(IEM GmbH) device provides different methods of pulse
Participants and methods wave calibration. Method 1 (aSBP1) is based on brachial
Study population systolic and diastolic pressure, which is the most commonly
Participation in this study was voluntary and data used method in the literature independent of a specific
assessment was performed during a public health cam- device [13,14]. Method 2 (aSBP2) uses measured brachial
paign in Summer 2015 in the general population. Design MAP on the basis of the maximum magnitude of the
and procedures were adopted from dedicated preceding oscillogram and diastolic pressure. This feature of the
studies [8]. Participants had to be older than 18 years and Mobil-O-Graph devices has been validated successfully
were approached by healthcare personal in pharmacies against invasive solid-state catheters [9]. Furthermore, it is
and public places, and were asked whether they would important to consider that unlike peripheral SBP and DBP,
be interested in having their blood pressure measured. In various definitions for noninvasive MAP have been pro-
addition to blood pressure, heart rate, sex, height, weight, posed [19] that are not fully investigated here.
and age were recorded. Consent was provided for anon-
ymized processing of data and dissemination of results. Statistics
The study protocol adhered to the Declaration of All statistical analyses were carried out using MedCalc
Helsinki and was approved by the local ethics committee. Statistical Software, version 16.4.3 (MedCalc Software
Earlier published invasive validation data [9] were used bvba, Ostend, Belgium). For the continuous investiga-
for additional retrospective analyses to confirm observed tion of associations among variables, regression analysis
trends in noninvasive data. and Williams’ test were used. Furthermore, multiple
regression analysis was carried out to study determinants
Brachial and aortic blood pressure measurement of differences. For subgroup analysis, we categorized the
Blood pressure readings were performed in a dedicated, sample into age deciles (20–29, 30–39, 40–49, 50–59,
quiet area by healthcare professionals and trained para- 60–69, 70–79, 80 +) and brachial blood pressure cate-
medics after at least a minute of rest [10]. The Mobil-O- gories (optimal, <120/80; normal, > 120/80 to <130/85;
Graph PWA (IEM GmbH, Stolberg, Germany) device high normal, > 130/85 to <140/90; grade I hypertension,
used for the measurements is a validated oscillometric > 140/90 to <160/100; and grade II/III hypertension,
sphygmomanometer with a brachial cuff [11,12] and > 160/100 mmHg) [15]. Wilcoxon, Mann–Whitney, or
inbuilt ARCSolver (Austrian Institute of Technology, Kruskal–Wallis tests were used as appropriate. A
Vienna, Austria) pulse wave analysis to provide aSBP threshold value of 130 mmHg with respect to aSBP2 for
among other parameters [13]. For aSBP, this particular the presence of hypertension was defined [20].
configuration has been validated extensively against
invasive solid-state catheters and against tonometry with Results
respect to reproducibility, accuracy, and feasibility by Overall 7409 individuals (2276 men and 5133 women)
different groups [8,9,14]. Furthermore, it has been with a median age of 54 years (age range: 20–100 years)
cleared by the USA (Food and Drug Administration), were included in the analysis. Baseline results are shown
Europe (CE), and Japan (JPAL) regulatory authorities. in Table 1.

Calibration Associations between systolic pressures

In the absence of an invasive pressure, there is no alter- Regression analysis showed that aSBP2 (R2 = 0.85) is
native to the use of a noninvasive calibration mechanism, significantly (Williams’ test, all P < 0.001 for comparison)
although the method of calibration strongly influences the less associated with bSBP than aSBP1 (R2 = 0.94) as
desired magnitude of absolute values provided by a method shown in Fig. 1. In the selective subgroup analysis with
[3,7]. Actual methods for the noninvasive quantification of respect to bSBP, aSBP2 always showed a significantly
aSBP often use a generalized transfer function [13,15,16] to lower R2 than aSBP1 for each pressure and age category.
convert the shape of recorded peripheral pulse waves into Stratification for sex showed significant effects between
estimates of aortic pulse wave shapes. Various transfer calibrations and groups as shown in Fig. 2. Furthermore,
function such as methods from different research groups it is notable that the categorization of blood pressure
have shown their ability to predict aortic pressure on the categories ‘normal’, ‘high normal’, and ‘grade 1 hyper-
basis of invasive pressure reading and the concept of per- tension’, representing 4536 patients or about 61% of our
ipheral waveform transformation [17,18]. By definition, this cohort, leads to associations between bSBP, aSBP1, and
waveform transformation is based on relative numbers and aSBP2 of R2 = 0.83 and 0.66, respectively. Differences in
is invariant to absolute pressure levels. Typically, positive sex (male/female) are consistent and again more pro-
aortic to peripheral pressure wave amplification is observed. nounced for aSBP2 (R2 = 0.59 vs. 0.70) than for aSBP1
The noninvasive quantification of this dimensionless wave (R2 = 0.81 vs. 0.83).

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.

136 Blood Pressure Monitoring 2018, Vol 23 No 3

Magnitudes of systolic pressure earlier catheter data [6]. As shown in Fig. 4a and b, the same
Table 1 shows that the mode of calibration obviously trends could be observed for invasive aortic data compared
influences the magnitude of aSBP estimation and even with traditional (brachial oscillometric) bSBP.
shows apparently higher aSBP2 compared with brachial
This difference in absolute pressure values may lead to
readings (Wilcoxon test, P < 0.001). misleading classifications of hypertension between methods
In contrast, aSBP1 typically shows a positive aortic to per- as exemplarily illustrated in the upper left rectangle of
ipheral systolic amplification. Figure 3a shows that the Fig. 1 showing 1721 individuals (∼23% of the total cohort)
apparent negative aortic to peripheral systolic amplification being normotensive according to bSBP (≤ 140 mmHg), but
for aSBP2 is not constant over the lifespan. The influence of being hypertensive as stratified by aSBP2 (≥130 mmHg).
pressure levels is presented by the Bland–Altman plot in
Fig. 3b, showing a slight inverse trend and a mean difference Analysis of differences
of about 1 mmHg with 95% limits of agreement between As reported in Table 2, the continuous analysis of the
bSBP and aSBP2 of about ± 13 mmHg. To invasively con- determinants of differences between aSBP1 and the
firm the trends depicted in Fig. 3a and b, we reanalyzed mean pressure-derived aSBP2 showed the differential
influence of tested confounders. Heart rate and sex
turned out to be clinically relevant determinants in this
Table 1 Baseline data particular relation, whereas the influence of age and
Male (N = 2276) Female (N = 5133) actual blood pressure levels did not differ significantly
between calibrations. Heart rate and age showed a weak,
2.5–97.5 2.5–97.5
Median percentiles Median percentiles but significant inverse relation in our data.
Age (years) 54 24–83 54 20–81
HR (1/min) 76 54–104 76 58–103 Discussion
MAP (oscillations) 106 85–133 99 79–129 The presented study evaluated the relation between bSBP
(mmHg)
DBP (mmHg) 84 63–109 77 58–102 and aSBP, with a focus on the influence of different calibration
bSBP (mmHg) 131 107–168 124 100–165 methods for aSBP estimation in a cohort of 7409 participants.
aSBP1 (mmHg) 121 98–153 115 92–153
aSBP2 (oscillations) 133 109–173 125 101–167
(mmHg) Associations between systolic pressures
aSBP, aortic systolic blood pressure; bSBP, brachial systolic blood pressure; The first finding was that the associations between bSBP
DBP, diastolic blood pressure; HR, heart rate; MAP, mean arterial pressure. and aSBP2 on a large cohort level did not change markedly

Fig. 1

Regression on the cohort level. Associations between bSBP, aSBP1, and aSBP2, respectively, show different offsets and slopes (right). The upper
left quarter represents those readings that are normotensive according to bSBP (dotted vertical line, ≤ 140 mmHg, x-axis), but hypertensive according
to aSBP1 (left) and aSBP2 (center) (dotted horizontal line, ≥ 30 mmHg, y-axis). aSBP, aortic systolic blood pressure; bSBP, brachial systolic blood
pressure; HTN, hypertension.

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.

 Aortic systolic pressure Wassertheurer et al. 137

Fig. 2

Associations between bSBP and aSBP1 (lighter gray bars) and bSBP and aSBP2 (darker gray bars) by sex and age groups (a, b) as well as
hypertension classes (c, d). The numbers on top of the bars represent the according regression coefficients (R2). aSBP, aortic systolic blood
pressure; bSBP, brachial systolic blood pressure.

(Fig. 1) compared with earlier reported data in small the oscillometric method, but not the Korotkoff method,
populations and therefore is highly relevant as the con- can directly measure MAP noninvasively. This leads to
firmation of this disparity in large cohorts provides a formal inherent methodological offsets. Given a pressure-
rationale for earlier published results with respect to invariant transfer function, different pressure measure-
hypertensive end organ damage and all-cause mortality ment paradigms will lead to consistent results within each
[4–6,18,21]. According to these data, the most promising particular paradigm (aSBP1 and bSBP Korotkoff-based),
application of aSBP2 is among people with borderline but will be misleading if methods are mixed (aSBP2
hypertension (e.g. upper normal to grade 1 hypertension) MAP-based and bSBP Korotkoff-based). In contrast,
because of the observation that only 66% of aSBP2 are the results of the determinants of differences analysis
explained by bSBP in this subgroup. presented in Table 2 underpin that differences between
aSBP1 and aSBP2 in relation to bSBP are also because
Magnitudes of systolic pressures of a different sensitivity to heart rate and sex, and
Mediated by heart rate and sex, the second key finding that aSBP2 obviously depends more on the actual wave
was that different calibrations finally led to different shape than aSBP1. Similar effects have been reported and
classifications in hypertension. The observed influence explained earlier for different calibrations with respect to
of the calibration method on the magnitude of aSBPs pulse pressure amplification [25]. The latter seems to be a
reported in Table 1 as well as its causes, including the more direct successor of bSBP from a mathematical view-
nonphysiological effects of apparent negative central on point. This novel numerical exploration is therefore in line
peripheral pressure amplification, have already been with physiological evidence showing that female waveforms
observed and particularly explained in the recent litera- differ significantly in shape compared with those from
ture [18,22–24]. In contrast, brachial systolic and diastolic males [1,26]. Furthermore, systolic pressure amplification
pressures actually used in clinical routine relate to a dif- among bSBP and aSBP2 (as illustrated in Fig. 3a) shows a
ferent reference standard (Korotkoff) than measured nonlinear pattern over the investigated age decades. In the
MAP (invasive pressure by catheter). Furthermore, only younger individuals, the brachial cuff overestimates aSBP2

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.

138 Blood Pressure Monitoring 2018, Vol 23 No 3

Fig. 3 Fig. 4

Differences between bSBP and aSBP2 over age groups (a) and
pressure range (b). Younger and individuals with lower pressures are Scatterplot (a) of differences between noninvasive oscillometric brachial
overestimated by bSBP compared with aSBP2 and vice versa. aSBP, systolic pressure and invasive aortic catheter pressure over age shows
aortic systolic blood pressure; bSBP, brachial systolic blood pressure. that younger individuals’ blood pressure is overestimated by brachial
cuff measurement compared with aortic catheter measurement and
vice versa. (b) Bland–Altman plot shows that brachial cuff measurement
overestimates lower systolic pressure compared with aortic catheter
readings. Trends in both plots are significant. aSBP, aortic systolic blood
significantly, but around the age of 40, this association is pressure; bSBP, brachial systolic blood pressure.
inverted. A similar effect can be observed in Fig. 4a for the
invasive data. Furthermore, Figs 3b and 4b show that tradi-
tional brachial cuff measurements seem to overestimate at
lower systolic pressures, but underestimate at higher systolic Table 2 Determinants of differences between aSBP1 and aSBP2
pressures compared with aSBP2. Similar observations with
aSBP1 − aSBP2 Coefficient SE Rpartial P
respect to brachial cuff and invasive pressures have been
reported by several invasive studies [27–29]. These results Constant − 36.0170
Age − 0.03394 0.004223 − 0.093 < 0.001
are remarkable with respect to pressure classification and the Sex 5.7493 0.9101 0.073 < 0.001
discussion on spurious hypertension in young individuals. Heart rate 0.4435 0.01997 0.249 < 0.001
Sex × heart rate − 0.05635 0.01168 −0.055 < 0.001

Analysis of differences aSBP, aortic systolic blood pressure.

The third key finding was that calibration influences the
differences between aSBP1 and aSBP2 for a given aortic relation between peripheral pressure, calibration mode, and
waveform (Table 2) and that heart rate and sex play a transfer function. In other words, not only the magnitude
predominant role in this association. With respect to the and amplitude of the peripheral pressure but also the cho-
contribution of heart rate, a change of 10 beats/min for a sen transfer function affect the resulting aortic pressure.
given patient will affect the absolute difference between The calibration method itself contributes independently
aSBP1 and aSBP2 by 4.4 mmHg. This reflects a remarkable toward the result.

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.

 Aortic systolic pressure Wassertheurer et al. 139

Potential clinical consequences C.M. are inventors (not holder) of a patent that is used in the
The clinical outcome on the basis of the analyses is ARCSolver method.
therefore the observed apparent inconclusive classi-
fication of 1721 patients with respect to hypertension. Conflicts of interest
This subsequently means that classification of risk rela- There are no conflicts of interest.
ted to BP may be significantly different because these
individuals are labeled as normotensive on the basis of
References
cuff bSBP, but in fact may have hypertension accord- 1 Nichols WW, O’Rourke MF. McDonald’s blood flow in arteries, 4th ed.
ing to aSBP2 thresholds. This has potential clinical London, UK: Arnold; 1998.
impact, especially in the light of a recently published 2 Mitchell GF. Central pressure should not be used in clinical practice. Artery
Res 2015; 9:8–13.
meta-analysis on the accuracy of brachial blood pressure 3 Papaioannou TG, Karageorgopoulou TD, Sergentanis TN, Protogerou AD,
measurement in comparison with invasive catheter Psaltopoulou T, Sharman JE, et al. Accuracy of commercial devices and
measurements [30]. The mentioned work reports that methods for noninvasive estimation of aortic systolic blood pressure a
systematic review and meta-analysis of invasive validation studies.
the accuracy of traditionally brachial cuff techniques is J Hypertens 2016; 34:1237–1248.
significantly lower in the range between 130 and 4 Negishi K, Yang H, Wang Y, Nolan MT, Negishi T, Pathan F, et al. Importance
160 mmHg. Misleading classification of hypertension is of calibration method in central blood pressure for cardiac structural
abnormalities. Am J Hypertens 2016; 29:1070–1076.
likely to be a clinical consequence also with respect 5 Protogerou AD, Argyris AA, Papaioannou TG, Kollias GE, Konstantonis GD,
to proper therapy. Another important point is the con- Nasothimiou E, et al. Left-ventricular hypertrophy is associated better with
sideration of different classifications whenever patient 24-h aortic pressure than 24-h brachial pressure in hypertensive patients: the
SAFAR study. J Hypertens 2014; 32:1805–1814.
matching or aggregation to comparable groups before 6 Nakagomi A, Okada S, Shoji T, Kobayashi Y. Crucial effect of calibration
prospective tests or studies is needed. Therefore, further methods on the association between central pulsatile indices and coronary
investigations and research addressing this issue seem atherosclerosis. Am J Hypertens 2017; 30:24–27.
7 Mahieu D, Kips J, Rietzschel ER, de Buyzere ML, Verbeke F, Gillebert TC,
worthwhile and promising. As discussed, several recent et al. Asklepios Investigators. Noninvasive assessment of central and
studies showed the incremental prognostic value of peripheral arterial pressure (waveforms): implications of calibration methods.
aSBP2 over bSBP. The data of this study suggest that J Hypertens 2010; 28:300–305.
8 Nunan D, Wassertheurer S, Lasserson D, Hametner B, Fleming S, Ward A,
this physiologically relevant superiority can be shown Heneghan C. Assessment of central haemomodynamics from a brachial cuff
because of the reduced correlation between bSBP and in a community setting. BMC Cardiovasc Disord 2012; 12:48.
aSBP2 compared with aSBP1. This highlights that aSBP2 9 Weber T, Wassertheurer S, Rammer M, Maurer E, Hametner B, Mayer CC,
et al. Validation of a brachial cuff-based method for estimating central systolic
has a higher degree of freedom (i.e. variability) from the blood pressure. Hypertension 2011; 58:825–832.
cuff-based bSBP than suggested initially by aSBP1 [31]. 10 Mancia G, Fagard R, Narkiewicz K, Redón J, Zanchetti A, Böhm M, et al. Task
For these reasons, we strongly recommend that in future Force Members. 2013 ESH/ESC guidelines for the management of arterial
hypertension: the Task Force for the management of arterial hypertension of
research studies, aSBP should not be calculated mathe- the European Society of Hypertension (ESH) and of the European Society of
matically (by adding a systematic error), but measured Cardiology (ESC). J Hypertens 2013; 31:1281–1357.
by the available methodology of DBP and mean blood 11 Franssen PM, Imholz BP. Evaluation of the Mobil-O-Graph new generation
ABPM device using the ESH criteria. Blood Press Monit 2010;
pressure (aSBP2) calibration. 15:229–231.
12 Weiss W, Tölle M, Zidek W, van der Giet M. Validation of the mobil-O-Graph:
24 h-blood pressure measurement device. Blood Press Monit 2010;
Limitations 15:225–228.
It needs to be considered that the study is cross sectional 13 Wassertheurer S, Kropf J, Weber T, van der Giet M, Baulmann J, Ammer M,
in nature and patient selection was uncontrolled; thus, no et al. A new oscillometric method for pulse wave analysis: comparison with a
common tonometric method. J Hum Hypertens 2010; 24:498–504.
information for smoking, dietary habits, or activities 14 Weiss W, Gohlisch C, Harsch-Gladisch C, Tölle M, Zidek W, van der Giet M.
before measurement is known. Furthermore, the results Oscillometric estimation of central blood pressure: validation of the Mobil-O-
may be device dependent. Graph in comparison with the SphygmoCor device. Blood Press Monit
2012; 17:128–131.
15 Karamanoglu M, O’Rourke MF, Avolio AP, Kelly RP. An analysis of the
relationship between central aortic and peripheral upper limp pressure
Conclusion
waves in man. Eur Heart J 1993; 14:160–167.
The presented findings potentially influence future 16 Chen C, Nevo E, Fetics B, Pak P, Yin F, Maughan L, et al. Estimation of
considerations on diagnostic thresholds, patient matching central aortic pressure waveform by mathematical transformation of radial
tonometry pressure: validation of generalised transfer function. Circulation
in case–control settings, reference values as well as efforts
1997; 95:1827–1836.
toward invasive validation procedures for MAP. To our 17 Pauca AL, O’Rourke MF, Kon ND. Prospective evaluation of a method for
knowledge, this is the first study in which this behavior estimating ascending aortic pressure from the radial artery pressure
waveform. Hypertension 2001; 38:932–937.
has been analyzed and reported in this context, and the 18 Wassertheurer S, Baumann M. Assessment of systolic aortic pressure and
findings warrant further investigations. its association to all cause mortality critically depends on waveform
calibration. J Hypertens 2015; 33:1884–1888.
19 Papaioannou TG, Protogerou AD, Vrachatis D, Konstantonis G, Aissopou E,
Acknowledgements Argyris A, et al. Mean arterial pressure values calculated using seven different
This work was supported by the program ‘Talente’ by the methods and their associations with target organ deterioration in a single-
center study of 1878 individuals. Hypertens Res 2016; 3:640–647.
Austrian Ministry for Transport, Innovation and Technology 20 Cheng HM, Chuang SY, Sung SH, Yu WC, Pearson A, Lakatta EG, et al.
BMVIT (FEMCOR Project, FFG no. 849808). S.W. and Derivation and validation of diagnostic thresholds for central blood pressure

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.

140 Blood Pressure Monitoring 2018, Vol 23 No 3

measurements based on long-term cardiovascular risks. J Am Coll Cardiol 27 Finnegan TP, Spence JD, Wong DG, Wells GA. Blood pressure
2013; 62:1780–1787. measurement in the elderly: correlation of arterial stiffness with
21 Weber T, Wassertheurer S, Schmidt-Trucksäss A, Rodilla E, Ablasser C, differences between intra-arterial and cuff pressures. J Hypertens
Jankowski P, et al. Relationship between 24-hour ambulatory central systolic 1985; 3:231–235.
blood pressure and left ventricular mass: a prospective multicenter study. 28 Fagher B, Magnusson J, Thulin T. Direct and indirect blood pressure in
Hypertension 2017; 70:1157–1164. normotensive and hypertensive subjects. Intern Med 1994; 236:85–90.
22 Papaioannou TG, Vavuranakis M, Tousoulis D. Calibration of noninvasive 29 Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves J, Hill MN, et al.
central blood pressure devices and negative aortic-to-brachial systolic Subcommittee of Professional and Public Education of the American Heart
pressure amplification. Kidney Int 2017; 91:253–254.
Association Council on High Blood Pressure Research. Recommendations
23 Wassertheurer S, Hametner B, Sharman J, Weber T. Systolic blood pressure
for blood pressure measurement in humans and experimental animals: Part
amplification and waveform calibration. Hypertens Res 2017; 40:518.
1: blood pressure measurement in humans: a statement for professionals
24 Boutouyrie P, London GM, Sharman JE. Estimating central blood pressure in
the extreme vascular phenotype of advanced kidney disease. Kidney Int from the Subcommittee of Professional and Public Education of the
2016; 90:736–739. American Heart Association Council on High Blood Pressure Research.
25 Agnoletti D, Zhang Y, Salvi P, Borghi C, Topouchian J, Safar ME. Pulse Hypertension 2005; 45:142–161.
pressure amplification, pressure waveform calibration and clinical 30 Picone D, Schultz MG, Otahal P, Aakhus S, Al-Jumaily AM, Black JA, et al.
applications. Atherosclerosis 2012; 224:108–112. Accuracy of cuff-measured blood pressure: systematic reviews and meta-
26 Segers P, Mahieu D, Kips J, Rietzschel E, de Buyzere M, de Bacquer D, analyses. J Am Coll Cardiol 2017; 70:572–586.
et al. Asklepios Investigators. Amplification of the pressure pulse in the 31 Chemla D, Hébert JL, Aptecar E, Mazoit JX, Zamani K, Frank R, et al.
upper limb in healthy, middle-aged men and women. Hypertension 2009; Empirical estimates of mean aortic pressure: advantages, drawbacks and
54:414–420. implications for pressure redundancy. Clin Sci (Lond) 2002; 103:7–13.

Copyright r 2018 Wolters Kluwer Health, Inc. All rights reserved.