Validity and Reliability of Central Blood Pressure Estimated by Upper Arm Oscillometric Cuff Pressure

Transcription

1 nature publishing group Validity and Reliability of Central Blood Pressure Estimated by Upper Arm Oscillometric Cuff Pressure Rachel E.D. Climie 1,2, Martin G. Schultz 1, Sonja B. Nikolic 1, Kiran D.K. Ahuja 2, James W. Fell 2 and James E. Sharman 1 Background Noninvasive central blood pressure (BP) independently predicts mortality, but current methods are operator-dependent, requiring skill to obtain quality recordings. The aims of this study were first, to determine the validity of an automatic, upper arm oscillometric cuff method for estimating central BP ( ) by comparison with the noninvasive reference standard of radial tonometry (T CBP ). Second, we determined the intratest and intertest reliability of. Methods To assess validity, central BP was estimated by (Pulsecor R6.5B monitor) and compared with T CBP (SphygmoCor) in 47 participants free from cardiovascular disease (aged 57 ± 9 years) in supine, seated, and standing positions. Brachial mean arterial pressure (MAP) and diastolic BP (DBP) from the device were used to calibrate in both devices. Duplicate measures were recorded in each position on the same day to assess intratest reliability, and participants returned within 10 ± 7 days for repeat measurements to assess intertest reliability. Results There was a strong intraclass correlation (ICC = 0.987, P < 0.001) and small mean difference (1.2 ± 2.2 mm Hg) for central systolic BP (SBP) determined by compared with T CBP. Ninety-six percent of all comparisons (n = 495 acceptable recordings) were within 5 mm Hg. With respect to reliability, there were strong correlations but higher limits of agreement for the intratest (ICC = 0.975, P < 0.001, mean difference 0.6 ± 4.5 mm Hg) and intertest (ICC = 0.895, P < 0.001, mean difference 4.3 ± 8.0 mm Hg) comparisons. Conclusions Estimation of central SBP using cuff oscillometry is comparable to radial tonometry and has good reproducibility. As a noninvasive, relatively operator-independent method, may be as useful as T CBP for estimating central BP in clinical practice. Keywords: augmentation index; blood pressure; blood pressure monitor; hypertension; pressure waveform analysis; suprasystolic; tonometry American Journal of Hypertension, advance online publication 5 January 2012; doi: /ajh In recent years, several studies have reported that directly measured central blood pressure (BP) indices as well as noninvasively estimated central BP indices (e.g., central pulse pressure, augmentation index, and central systolic BP (SBP)) predict the onset of diabetes, cardiovascular events, and all-cause mortality independent of brachial BP measures. 1 3 The prognostic value of central BP has been recognized by expert consensus; 4,5 however, conventional brachial BP remains the standard method used to diagnose hypertension in clinical practice. One factor possibly contributing to the limited clinical adoption of central BP may be the relative complexity of the methods used to record central BP. Indeed, the most widely accepted central BP method is radial tonometry and synthesis of the aortic pressure waveform. While this technique has been well validated, 6 8 is highly reproducible, 9 and is generally regarded as the noninvasive reference standard, 10 the technique requires specialized 1 Menzies Research Institute Tasmania, University of Tasmania, Hobart, Australia; 2 School of Human Life Sciences, University of Tasmania, Launceston, Australia. Correspondence: James E. Sharman (James.Sharman@menzies.utas.edu.au) Received 19 July 2011; first decision 26 August 2011; accepted 7 November American Journal of Hypertension, Ltd. equipment, additional time, and a basic level of operator skill to obtain pressure waveforms of appropriate quality. Therefore, a relatively operator-independent method to acquire central BP would be a useful advancement. Recently a cuff method has been developed which enables an estimate of central BP from the upper arm oscillometric waveform acquired at suprasystolic pressures. 11 This approach involves a simple to operate, press button device similar to conventional brachial cuff BP oscillometry, and may have greater potential for integration into clinical practice compared with radial tonometry. Whether the oscillometric cuff estimates central BP with similar precision to radial tonometry has never been assessed and was the first aim of our study. Additionally, we sought to determine the intratest and intertest reliability of the method. To address the first aim, central BPs derived from the new oscillometric method ( ) in three different postural positions (supine, seated, and standing) were compared with those derived by radial tonometry (T CBP ). 10 Intratest reliability was assessed by comparison of consecutively obtained measures within one clinic visit, and the intertest reliability was assessed by comparison of measures obtained on separate days. AMERICAN JOURNAL OF HYPERTENSION 1

2 Central Blood Pressure Validation Methods Study participants. A total of 47 consecutive participants (n = 27 men) were recruited for the study via local advertisements. Participants were not included if presenting with a condition known or suspected to affect the quality of brachial or central BP measures (e.g., arrhythmia, valvular disease, or coronary artery disease), or if they were identified as requiring medical treatment. As such, exclusion criteria included pregnancy, cardiac arrhythmia or a clinical history of cardiovascular disease. Hypertension was defined as clinic brachial BP 140/90 mm Hg, self-reported hypertension as diagnosed by a physician, or use of antihypertensive medications. Diabetes mellitus was determined by self-report or previous diagnosis by a physician. Study protocol. Participants attended the Menzies Research Institute Tasmania for BP assessment on two occasions separated by 10 ± 7 days. Before each visit, participants were asked to refrain from alcohol consumption and exercise on the day of testing, to avoid consuming heavy meals (i.e., were in a postabsorptive state), smoking, and caffeine-containing products in the 3 hours prior to testing. As recommended by guidelines, 12,13 a familiarization session was provided at the beginning of the first visit which included an induction to the study protocol and testing equipment as well as recording of standard anthropometric variables. All BP measurements were recorded from the left arm using an appropriately sized cuff in accordance with guidelines. 14 A third visit to the clinic was undertaken after an overnight fast. During this visit, a blood sample was taken from the antecubital fossa (for standard blood biochemistry), and participants were fitted with a 24-hour ambulatory BP device to assess BP control. The study was approved by the University of Tasmania Human Research Ethics Committee, and all participants signed informed consent. Validation study (comparison of T CBP ). Central BP was estimated in all study participants at each visit, by two operators and by each method in a systematic format as recommended. 13 Prior to undertaking measurements for the study, operators were trained in measuring central BP using both techniques and were required to perform 20 recordings deemed to be of acceptable quality according to specification of each device (as described below). For the study, measurements were taken after a minimum of 5-min rest in three postural positions: first while supine (on a hospital bed with no pillow), second while seated (with back supported, feet flat on the floor and arm supported with a pillow so that the BP cuff and forearm was at approximately the same level as the heart), and finally while standing (with the arm supported on an adjustable bench so that the BP cuff and forearm was at approximately the same level as the heart). The order of T CBP measurements was randomly determined at each visit (i.e., either T CBP or was taken first, followed by or T CBP, respectively). Both methods were repeated in the same order as recommended. 13 Intratest and intertest reliability study (variance in between measurement 1 and measurement 2 and between visit 1 and visit 2). Intratest reliability was determined by comparing the two recordings taken in each position within ~5 min of each other by a single operator. The oscillometric cuff was removed and then repositioned at the upper arm before the next measurement was taken. Intertest reliability was determined by comparing data taken under the same conditions in each posture, but on separate days. T CBP estimation. Arterial pressure waveforms were acquired by radial applanation tonometry using a hand-held tonometer (SPC-301; Millar Instruments, Houston, TX). The aortic pressure waveform was synthesized and central BP estimated using a validated 6 8 and reproducible 9 generalized transfer function incorporated into customized software (SphygmoCor 8.1; AtCor Medical, Sydney, Australia). The quality of T CBP recordings was determined by visual inspection and also by an inbuilt operator index. We decided a priori that an operator index of <75% was poor quality and would be excluded from the analysis. Where possible, a repeat measure was acquired when a reading was deemed to be poor quality. Augmentation pressure was calculated from the aortic pressure waveform as P1 (early systolic wave) P2 (late systolic shoulder) and indexed to pulse pressure (SBP diastolic BP (DBP)) for calculation of augmentation index. estimation. The was determined using a R6.5B vascular monitor (Pulsecor, Auckland, New Zealand) and algorithm version MR1517 with an appropriate sized cuff positioned over the left upper arm. This device has an inbuilt validated oscillometric brachial BP monitor (Welch Allyn, Skaneateles Falls, NY). The process of estimating central BP starts with a conventional oscillometric measurement of brachial BP. Within 3 s of cuff deflation, the cuff reinflates and holds for 10 s at a pressure ~30 mm Hg above the measured brachial SBP (suprasystolic pressure). During this held-inflation period, suprasystolic pressure signals are recorded by a high fidelity pressure transducer. From the suprasystolic waveform, the central pressure waveform was derived in the time-domain from the relationship between the total oscillatory pressure in the aorta and the total oscillatory pressure under the occlusion cuff (as previously described). 11 This estimation was scaling-independent (i.e., the estimated aortic wave shape did not depend on brachial BP). Aortic pulse pressure (AoPP), aortic mean arterial pressure (MAP) (AoM), and DBP (D) were calculated from the estimated central pressure waveform. Central SBP was then calculated using the following equation: where brachial MAP and DBP are oscillometric BPs derived from the Welch Allyn device. AoPP brachial MAP brachial DBP + brachial DBP AoM D ( ) We decided a priori to exclude recordings of poor quality, which was defined as having signal-to-noise ratio of 2 AMERICAN JOURNAL OF HYPERTENSION

3 Central Blood Pressure Validation <10 db or inconsistent or poor quality waveforms determined by visual inspection. As with T CBP, a repeat measure was acquired when a reading was deemed to be poor quality. Augmentation index was calculated from the suprasystolic P2 P0 signal on the basis of the formula: P1 P0, where P0 is end-diastolic pressure, P1 is early systolic pressure, and P2 is late systolic pressure. Calibration of devices. Waveforms from the T CBP devices were calibrated with the same MAP and DBP acquired by the device for each individual measurement. This method ensured a direct comparison of transfer functions to derive central BP between devices and was not affected by potential inaccuracies of noninvasive upper arm BP. MAP, brachial SBP, and DBP were all derived by oscillometry using the validated algorithm of the Welch Allyn BP monitor. Statistical analysis. All data were analyzed using SPSS for Windows software version 17.0 (SPSS, Chicago, IL). Data are presented as mean ± s.d. unless otherwise stated, and P < 0.05 was considered statistically significant. To assess the validity and reliability of, intraclass correlations (ICCs) and Bland Altman analyses were performed. Relationships between central BP recordings by the different methods (T CBP ) were assessed by ICC two-way random model of absolute agreement, and two-way mixed model of absolute agreement were performed to assess relationships between measurements (measurement 1 and measurement 2) and visits (visit 1 and visit 2). The relationship between - and T CBP -determined augmentation index was assessed by Pearson s correlations because different equations are used to derive augmentation index between methods, and this results in substantially different values. Therefore, the ICC test for agreement is not appropriate. RESULTS Clinical characteristics Participant characteristics are summarized in Table 1. None of the participants were current smokers. Antihypertensive medication included angiotensin-converting enzyme inhibitors, angiotensin receptor blocker, and β blockers, and insulin-sensitive agents included metformin and sulfonylurea. In one participant, measurements of central BP (by both methods) were only recorded during visit 1. In 20 participants, at least one recording was excluded because of poor quality. However, in all 46 participants who attended both visits, it was possible to obtain at least three (of a possible six) measures at each visit. Validation (comparison of T CBP ) There were highly significant correlations between T CBP and estimation of central SBP in supine, seated, and standing positions. Furthermore, there were small mean differences in central SBP estimation between methods (Figure 1 and Table 2). Bland Altman analysis of the two methods showed that overall, estimation of central SBP was slightly, but Oscillometric central SBP ( ) (mm Hg) Table 1 Characteristics of the study participants (n = 47) Variables Mean ± s.d. or n (%) Age (years) 57 ± 9 Body mass index (kg/m 2 ) 27.2 ± 5.1 Waist:hip (ratio) 1.0 ± 0.2 Upper arm circumference (cm) 31.5 ± 8.2 Brachial systolic blood pressure (mm Hg) 123 ± 14.6 Brachial diastolic blood pressure (mm Hg) 68 ± Hour ambulatory systolic blood pressure (mm Hg) 131 ± Hour ambulatory diastolic blood pressure (mm Hg) 77 ± 6.7 Diabetes mellitus, n (%) 18 (38) Hypertension, n (%) 14 (30) Hypercholesterolemia (%) 13 (28) Antihypertensive medications, n (%) 15 (32) Insulin or insulin-sensitive agents, n (%) 22 (47) Statin, n (%) 10 (21) Glucose (mmol/l) 6.2 ± 1.8 Insulin (µiu/ml) 8.5 ± 7.4 Total cholesterol (mmol/l) 5.5 ± 1.1 High-density lipoprotein cholesterol (mmol/l) 1.7 ± 0.4 Triglycerides (mmol/l) 1.1 ± R 2 = Radial tonometry central SBP (T CBP ) (mm Hg) Supine Seated Standing y = x Figure 1 Comparison of central systolic blood pressure (SBP) estimated by radial tonometry (T CBP ) and oscillometry ( ) in supine, seated, and standing positions (n = 495). The solid line is the trend line and the dashed line is the line of identity. not significantly, higher than the corresponding T CBP estimation; however, there was no systematic bias (Figure 2). The limits of agreement were 4.4 mm Hg, and 96% of comparisons with T CBP were within 5 mm Hg (Figure 3). There was a significant correlation between the - and T CBP -determined augmentation index measured in all three positions (r = 0.684, P < 0.001). AMERICAN JOURNAL OF HYPERTENSION 3

4 Central Blood Pressure Validation Table 2 Comparison of central SBP estimated by radial applanation tonometry (T CBP ) and upper arm oscillometric cuff pressure ( ) in 47 participants Participants and position Central SBP by T CBP (mm Hg) Central SBP by (mm Hg) Mean difference ICC P value Regression equation (y) All (n = 495) 99.4 ± ± ± < x Supine (n = 178) 98.1 ± ± ± < x 0.1 Seated (n = 175) ± ± ± < x Standing (n = 143) 99.8 ± ± ± < x 0.9 Data expressed as mean ± s.d. P value refers to ICC between devices. BP, blood pressure; ICC, intraclass correlations; n, total number of measures recorded;, oscillometric cuff method for estimating central BP; SBP, systolic blood pressure; T CBP, radial tonometry method for estimating central BP. Oscillometric central SBP ( ) -radial tonometry central SBP (T CBP ) (mm Hg) Supine Seated Standing Average central SBP estimated by and T CBP (mm Hg) Mean difference Figure 2 Bland Altman plot of the mean values and differences between central systolic blood pressure (SBP) estimated by radial tonometry (T CBP ) and oscillometry ( ) in supine, seated, and standing positions (n = 495). Cumulative percentage (%) Average difference in central SBP estimated by and T CBP (mm Hg) Figure 3 Cumulative percentage of absolute difference in central systolic blood pressure (SBP) between the methods of radial tonometry (T CBP ) and oscillometry ( ). The dotted line indicates that 96% of central SBP values measured by were within 5 mm Hg of T CBP. Intratest reliability (variance in between measurement 1 and measurement 2) A total of 476 consecutive central BP recordings for measurement 1 and 2 were available for the analysis. There were highly significant correlations for data between measurement 1 and measurement 2 in supine, seated, and standing positions, and there were small mean differences in central SBP estimation between measurements (Table 3). There was a significant correlation between the first and the second measurement of augmentation index derived from in all positions (ICC = 0.951, P < 0.001). The overall mean difference between augmentation index measurement 1 and measurement 2 was 1.8 ± 18.4%. Intertest reliability (variance in between visit 1 and visit 2) Intertest comparisons are presented in Table 4. There were a total of 440 measures available from visit 1 and visit 2, and there were significant correlations between visits in supine, seated, and standing positions. On the basis of these findings, we estimate that 32 participants would be required in each of two independent samples to detect a 5 mm Hg difference with α = 0.05 and β = There was a significant correlation in augmentation index estimated by between visit 1 and visit 2 in all positions (ICC = 0.841, P < 0.001), and the overall mean difference was 12.8 ± 27.0%. Discussion Many studies have shown the independent prognostic importance of central BP indices acquired by radial (T CBP ) and carotid applanation tonometry. Despite this, the technique is not widely accepted and has limitations for integration into clinical practice because (among other things) there is the requirement of specialized equipment, technical precision, and dedicated time to obtain waveforms of appropriate quality. The main aim of this study was to determine the validity of a noninvasive upper arm oscillometric BP cuff method ( ) to estimate central BP by comparison with T CBP. There was strong agreement between methods suggesting that the algorithm is at least as accurate as T CBP for central BP estimation. Second, we sought to assess the intratest and intertest reliability of. There was excellent intratest reliability of, and while there was higher variation with respect to intertest reliability, measures were within acceptable limits of agreement. 4 AMERICAN JOURNAL OF HYPERTENSION

5 Central Blood Pressure Validation Table 3 Intratest reliability data for comparison between two measures of central SBP estimated by upper arm oscillometric cuff pressure recorded 5 min apart Participants and position Central SBP measurement 1 (mm Hg) Central SBP measurement 2 (mm Hg) Mean difference ICC P value Regression equation (y) All (n = 238) ± ± ± < x Supine (n = 87) 99.8 ± ± ± < x Seated (n = 86) ± ± ± < x Standing (n = 65) ± ± ± < x Data expressed as mean ± s.d. P value refers to ICC between measurements. BP, blood pressure; ICC, intraclass correlations; n, total number of measures recorded; SBP, systolic BP. Table 4 Intertest reliability data for comparison of central SBP estimated by upper arm oscillometric cuff pressure at visit 1 and visit 2 Participants and position Central SBP at visit 1 (mm Hg) Central SBP at visit 2 (mm Hg) Mean difference ICC P value Regression equation (y) All (n = 220) ± ± ± < x Supine (n = 84) ± ± ± < x Seated (n = 81) ± ± ± < x Standing (n = 55) ± ± ± < x Data expressed as mean ± s.d. P value refers to ICC between visits. BP, blood pressure; ICC, intraclass correlations; n, total number of measures recorded; SBP, systolic BP. The method estimates central BP from a brachial suprasystolic waveform using an algorithm validated against intra-aortic pressures. This invasive work was carried out in a relatively small population (n = 22) under stable conditions with no inducement of hemodynamic perturbations. 11 To our knowledge, the current study is the first to determine the validation of in response to differing hemodynamic conditions that are relevant to clinical practice (supine, seated, and standing postures), and the device performed well in each position. Indeed, the mean difference and standard deviation for comparison of the two methods was well within the standard required by the Association for the Advancement of Medical Instrumentation (5 ± 8 mm Hg) 22 and the device would obtain an A grade rating according to the British Hypertension Society classification for BP devices. 23 Having said this, many participants required a repeat measure because of an initial poor quality recording. Furthermore, while there was strong intratest agreement for augmentation index, there was more scatter and wider limits of agreement for the intertest assessment of reliability. These data suggest that further refinement on the stability of the suprasystolic pressure signal may be required. Several other methods to estimate central BP from an upper arm cuff BP signal have recently been reported The Arteriograph device employs a similar suprasystolic oscillometric cuff method as in the current study, but there are differences as to how central SBP is derived. Both (Pulsecor) and Arteriograph work on the principle of analyzing the pressure waves within an over pressurized cuff that is placed over the brachial artery and occluded at 30 mm Hg 11 to 35 mm Hg 25 above SBP. In this stop-flow condition, the energy of the pressure pulse wave is propagated through the skin into the cuff, and this suprasystolic signal is recorded over 10 s by a piezoelectric sensor pneumatically connected to the cuff. After consideration of brachial SBP level, the Arteriograph method calculates central SBP from the amplitude of the late systolic wave. 25 Derivation of aortic pulse wave velocity is also claimed to be achievable from the time interval between the first and second systolic peaks on the suprasystolic waveform. Indeed, excellent correlation between invasive and estimated aortic pulse wave velocity were reported with this method. 25 It has been suggested that more work is required to determine the accuracy and understand the working principles underlying the Arteriograph technique. 27,28 These comments could also apply to the Pulsecor technique that estimates central SBP by analysis of the cuff suprasystolic signal using an entirely different approach (see methods) to Arteriograph. However, the Pulsecor device does not calculate an estimate of aortic pulse wave velocity. It would certainly be of interest to compare these devices in future studies. Another device (ARCSolver) has been used to derive central BP from the brachialis diastolic pressure signal. This method correlated closely to central SBP derived by T CBP with similar limits of agreement to our study (±2 s.d. 6 mm Hg) and had good reproducibility. 26 Certainly, these are encouraging studies with respect to future refinement of risk stratification related to BP and simplification of methods to estimate central BP. However, the studies are limited by the lack of assessment under changed hemodynamic conditions (e.g., posture, drugs, or exercise), and these types of examinations are required to determine the robustness and generalizability of the methods. Indeed, Cheng and colleagues 24 recently validated a technique for estimating central SBP from the brachial oscillometric signal, and this performed well in comparison to invasive BP measurement at baseline and in response to sublingual nitroglycerine. AMERICAN JOURNAL OF HYPERTENSION 5

6 Central Blood Pressure Validation Strengths and limitations A strength of our study was that participants were assessed under different, and clinically relevant, postures. We also studied both men and women, with different morbidities (i.e., hypertension, diabetes, hypercholesterolemia) across a good range of ages and body sizes. Thus, the findings would appear to have reasonable external validity. On the other hand, the study would have been strengthened by inclusion of more participants. Furthermore, we have also excluded higher-risk participants with existing arrhythmias and cardiovascular disease, so the results will not necessarily be applicable to these patients. Another limitation is that we have used a noninvasive method as the reference for comparison with. This approach only enables comparison of transfer functions between devices. While the T CBP algorithm to derive central BP appears to be well validated 6 8 and has good reproducibility, 9 some have criticized the method, particularly with respect to accurate determination of central augmentation index. 29 As such, future studies should compare with invasive central BP, rather than T CBP. A comparison of this nature would also lessen the problem in the current study of calibrating both T CBP with potentially inaccurate noninvasive MAP and DBP. 30 In summary, this study found substantial equivalence between central SBP estimated by cuff oscillometry ( ) and T CBP. Furthermore, had good intratest and intertest reliability. Together, this data provide preliminary support for the notion that may be as useful a technique as T CBP for the estimation of central BP. The merit of compared with T CBP is that it is relatively operator independent and does not require specialized tonometry equipment or careful precision to record arterial waveforms. Moreover, uses a conventional upper arm cuff method that is well accepted and possibly more likely to be integrated into clinical practice into the future. Acknowledgment: J.E.S. was supported by a National Health and Medical Research Council Career Development Award (reference ). Disclosure: J.E.S. has research collaborations with AtCor Medical and some equipment for this study was loaned by Pulsecor Limited. The other authors declared no conflict of interest. 1. Chen JY, Chou CH, Lee YL, Tsai WC, Lin CC, Huang YY, Chen JH. Association of central aortic pressures indexes with development of diabetes mellitus in essential hypertension. Am J Hypertens 2010; 23: Vlachopoulos C, Aznaouridis K, O Rourke MF, Safar ME, Baou K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with central haemodynamics: a systematic review and meta-analysis. Eur Heart J 2010; 31: Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D, Hughes AD, Thurston H, O Rourke M; CAFE Investigators; Anglo-Scandinavian Cardiac Outcomes Trial Investigators; CAFE Steering Committee and Writing Committee. Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study. Circulation 2006; 113: Agabiti-Rosei E, Mancia G, O Rourke MF, Roman MJ, Safar ME, Smulyan H, Wang JG, Wilkinson IB, Williams B, Vlachopoulos C. Central blood pressure measurements and antihypertensive therapy: a consensus document. Hypertension 2007; 50: Avolio AP, Van Bortel LM, Boutouyrie P, Cockcroft JR, McEniery CM, Protogerou AD, Roman MJ, Safar ME, Segers P, Smulyan H. Role of pulse pressure amplification in arterial hypertension: experts opinion and review of the data. Hypertension 2009; 54: Chen CH, Nevo E, Fetics B, Pak PH, Yin FC, Maughan WL, Kass DA. Estimation of central aortic pressure waveform by mathematical transformation of radial tonometry pressure. Validation of generalized transfer function. Circulation 1997; 95: Pauca AL, O Rourke MF, Kon ND. Prospective evaluation of a method for estimating ascending aortic pressure from the radial artery pressure waveform. Hypertension 2001; 38: Sharman JE, Lim R, Qasem AM, Coombes JS, Burgess MI, Franco J, Garrahy P, Wilkinson IB, Marwick TH. Validation of a generalized transfer function to noninvasively derive central blood pressure during exercise. Hypertension 2006; 47: Holland DJ, Sacre JW, McFarlane SJ, Coombes JS, Sharman JE. Pulse wave analysis is a reproducible technique for measuring central blood pressure during hemodynamic perturbations induced by exercise. Am J Hypertens 2008; 21: Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, Pannier B, Vlachopoulos C, Wilkinson I, Struijker-Boudier H; European Network for Noninvasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27: Lowe A, Harrison W, El-Aklouk E, Ruygrok P, Al-Jumaily AM. Non-invasive modelbased estimation of aortic pulse pressure using suprasystolic brachial pressure waveforms. J Biomech 2009; 42: O Brien E, Atkins N, Stergiou G, Karpettas N, Parati G, Asmar R, Imai Y, Wang J, Mengden T, Shennan A; Working Group on Blood Pressure Monitoring of the European Society of Hypertension. European Society of Hypertension International Protocol revision 2010 for the validation of blood pressure measuring devices in adults. Blood Press Monit 2010; 15: Wilkinson IB, McEniery CM, Schillaci G, Boutouyrie P, Segers P, Donald A, Chowienczyk P. ARTERY Society guidelines for validation of non-invasive haemodynamic measurement devices: Part 1, arterial pulse wave velocity. Artery Research 2010; 4(2): Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ; National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National High Blood Pressure Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003; 289: Huang CM, Wang KL, Cheng HM, Chuang SY, Sung SH, Yu WC, Ting CT, Lakatta EG, Yin FC, Chou P, Chen CH. Central versus ambulatory blood pressure in the prediction of all-cause and cardiovascular mortalities. J Hypertens 2011; 29: Roman MJ, Devereux RB, Kizer JR, Lee ET, Galloway JM, Ali T, Umans JG, Howard BV. Central pressure more strongly relates to vascular disease and outcome than does brachial pressure: the Strong Heart Study. Hypertension 2007; 50: Roman MJ, Devereux RB, Kizer JR, Okin PM, Lee ET, Wang W, Umans JG, Calhoun D, Howard BV. High central pulse pressure is independently associated with adverse cardiovascular outcome the strong heart study. J Am Coll Cardiol 2009; 54: Weber T, Auer J, O rourke MF, Kvas E, Lassnig E, Lamm G, Stark N, Rammer M, Eber B. Increased arterial wave reflections predict severe cardiovascular events in patients undergoing percutaneous coronary interventions. Eur Heart J 2005; 26: London GM, Blacher J, Pannier B, Guérin AP, Marchais SJ, Safar ME. Arterial wave reflections and survival in end-stage renal failure. Hypertension 2001; 38: Pini R, Cavallini MC, Palmieri V, Marchionni N, Di Bari M, Devereux RB, Masotti G, Roman MJ. Central but not brachial blood pressure predicts cardiovascular events in an unselected geriatric population: the ICARe Dicomano Study. J Am Coll Cardiol 2008; 51: Safar ME, Blacher J, Pannier B, Guerin AP, Marchais SJ, Guyonvarc h PM, London GM. Central pulse pressure and mortality in end-stage renal disease. Hypertension 2002; 39: White WB, Berson AS, Robbins C, Jamieson MJ, Prisant LM, Roccella E, Sheps SG. National standard for measurement of resting and ambulatory blood pressures with automated sphygmomanometers. Hypertension 1993; 21: O Brien E, Petrie J, Littler W, de Swiet M, Padfield PL, Altman DG, Bland M, Coats A, Atkins N. An outline of the revised British Hypertension Society protocol for the evaluation of blood pressure measuring devices. J Hypertens 1993; 11: Cheng HM, Wang KL, Chen YH, Lin SJ, Chen LC, Sung SH, Ding PY, Yu WC, Chen JW, Chen CH. Estimation of central systolic blood pressure using an oscillometric blood pressure monitor. Hypertens Res 2010; 33: AMERICAN JOURNAL OF HYPERTENSION

7 Central Blood Pressure Validation 25. Horváth IG, Németh A, Lenkey Z, Alessandri N, Tufano F, Kis P, Gaszner B, Cziráki A. Invasive validation of a new oscillometric device (Arteriograph) for measuring augmentation index, central blood pressure and aortic pulse wave velocity. J Hypertens 2010; 28: Wassertheurer S, Kropf J, Weber T, van der Giet M, Baulmann J, Ammer M, Hametner B, Mayer CC, Eber B, Magometschnigg D. A new oscillometric method for pulse wave analysis: comparison with a common tonometric method. J Hum Hypertens 2010; 24: Parati G, De Buyzere M. Evaluating aortic stiffness through an arm cuff oscillometric device: is validation against invasive measurements enough? J Hypertens 2010; 28: Segers P, Kips J, Trachet B, Swillens A, Vermeersch SJ, Mahieu D, Rietzschel E, De Buyzere M, Van Bortel L. Limitations and pitfalls of non-invasive measurement of arterial pressure wave reflections and pulse wave velocity. Artery Research 2009; 3(2): Hope SA, Meredith IT, Tay D, Cameron JD. Generalizability of a radial-aortic transfer function for the derivation of central aortic waveform parameters. J Hypertens 2007; 25: Mahieu D, Kips J, Rietzschel ER, De Buyzere ML, Verbeke F, Gillebert TC, De Backer GG, De Bacquer D, Verdonck P, Van Bortel LM, Segers P; Asklepios investigators. Noninvasive assessment of central and peripheral arterial pressure (waveforms): implications of calibration methods. J Hypertens 2010; 28: AMERICAN JOURNAL OF HYPERTENSION 7