Journal Logo


Validation of noninvasive devices for central blood pressure assessment

a yet unresolved issue

Salvi, Paolo; Grillo, Andrea; Parati, Gianfranco

Author Information
doi: 10.1097/HJH.0000000000000970
  • Free

At the end of the twentieth century, the availability of transcutaneous tonometers, able to record pulse waveforms noninvasively, led to an increasing interest toward the possible importance of central SBP (cSBP) and central pulse pressure as cardiovascular prognostic indices. This increasing interest is based on the theoretical assumption that cSBP and central pulse pressure should be able to evaluate the actual hemodynamic load on the left ventricle in a given individual much better than peripheral blood pressure (BP) values measured at the brachial artery level by means of traditional sphygmomanometers [1–3].

In the first studies carried out on this issue, noninvasive central BP was detectable only by systems based on applanation tonometry. Actually, arterial pressure waveforms recorded noninvasively by arterial tonometry are largely superimposable to those recorded invasively, by means of an intra-arterial catheter. There are at present two well validated and reliable methods to record central pressure waves by arterial tonometry: a ‘direct’ method and an ‘indirect’ method [4]. The ‘direct’ method records pulse waves at common carotid artery; the ‘central’ BP values estimated with this approach are considered as a surrogate for aortic pressure because of the close proximity of these two arterial sites. This approach was shown to represent an easy and reproducible method to estimate central BP based on different types of tonometric devices. Initially, a pencil-type Millar tonometer incorporating a high-fidelity strain gauge transducer (Millar SPT-301 Pulse Wave Tonometer; Millar Inc., Houston, Texas, USA) connected to a chart recorder was used [1]. Subsequently, the PulsePen (DiaTecne srl, Milan, Italy) device was developed [5] followed, more recently, by the Complior Analyse (Alam Medical, Vincennes, France).

On the contrary, with the ‘indirect’ method, applanation tonometry is performed usually at the radial artery level. In this case, cSBP is calculated either via a regression equation employing the second systolic peak of a pulse waveform as the independent variable (method used by the Omron HEM-9000AI; Omron Healthcare Co. Ltd., Kyoto, Japan) or through a generalized transfer function (method used by the SphygmoCor device; AtCor Medical Pty Ltd, West Ryde, Australia).

The use of arterial tonometry in clinical research has led to important results. First, Safar et al.[1] showed, in patients with chronic kidney disease undergoing hemodialysis, that the disappearance of aortic to brachial pulse pressure amplification was a significant predictor of all-cause mortality, independent of age and other standard confounding factors, and that carotid pulse pressure was more powerful than brachial pulse pressure in the prediction of overall mortality. These conclusions were more recently confirmed by the results of the PARTAGE (Predictive values of blood pressure and arterial stiffness in institutionalized very aged population) study, which involved more than 1100 French nursing home residents older than 80 years, and showed that reduced peripheral pulse pressure amplification is related to an increased prevalence of cardiovascular disease [6] and is a real predictor of total mortality [7].

However, it was only after publication of the Conduit Artery Function Evaluation (CAFE) study [8], an ancillary study of the Anglo-Scandinavian Cardiac Outcomes Trial study, that cSBP started to be in the spotlight. The Anglo-Scandinavian Cardiac Outcomes Trial study showed a greater reduction in cardiovascular events in patients treated with a calcium channel blocker (amlodipine) compared with patients treated with a β-blocker (atenolol), without any difference being noted in the reduction of brachial SBP values between the two treatment groups, but with a significantly lower cSBP and central pulse pressure in patients randomized to receiving amlodipine compared with those receiving atenolol. This suggested the interpretation that the greater reduction in cardiovascular events in the group randomized to amlodipine was caused by a greater effect of this drug in lowering cSBP than in lowering brachial SBP. These results were also interpreted as a suggestion that cSBP is a more important cardiovascular prognostic factor than peripheral SBP values measured in the brachial artery by means of traditional sphygmomanometers, and raised the hypothesis that peripheral BP measurement might not always be the best method to assess the effects of drugs on the hemodynamic load exerted on the cardiovascular system of hypertensive patients. Following publication of the Conduit Artery Function Evaluation study, assessment of central BP has raised progressively increasing interest, accompanied by a parallel increase in the production and marketing of instruments for central BP assessment. This interest is testified by the increasing number of published studies comparing brachial and central BP, trying to define normal reference values for central BP, and to extend its assessment over 24 h. Other studies have also tried to redefine cardiovascular risk based on central BP values compared with peripheral BP, and based on their results, it has been proposed, in the management of hypertensive patients, to favor the choice of those antihypertensive drugs more active on central BP. On such a background, over the last decade, a continuous development in the technology of arterial stiffness and central BP assessment has occurred, resulting, among other outcomes, in the development and ‘validation’ of a new approach to investigate arterial wall properties, based on automated, cuff-based, oscillometric devices for the noninvasive assessment of central BP and of indices of arterial function derived from peripheral pulse wave analysis. Although most of these devices have been celebrated for their ability to provide simple and automated measurements of different hemodynamic parameters in a relatively fast and operator-independent manner, their ability to properly acquire aortic waveforms and to accurately calculate central aortic pressure, in particular in ambulatory conditions, is still a matter of debate. Important issues related to the algorithms used for transfer function that have been implemented in some of these devices are also still not completely resolved.

Notwithstanding these yet pending issues, on a theoretical basis, the potential usefulness of central BP assessment for an in-depth assessment of patients with conditions characterized by increased BP values cannot be denied. This seems to be the case also for those clinical conditions where an accurate central pulse wave analysis might be clinically informative providing an indirect estimate of myocardial function and cardiac work. Such theoretical assumptions have stimulated a large number of studies aimed at providing additional evidence in support of a future, a wider clinical use of central pulse wave analysis and of central BP assessment. In such a perspective, the demonstration that noninvasive assessment of central BP values is reliable, and that central arterial pressure waveform parameters derived noninvasively correspond to the ones invasively recorded in ascending aorta, is of utmost importance. For this reason, validation and reliability of devices for noninvasive assessment of central BP have become research priorities in recent years.

The systematic review and meta-analysis by Papaioannou et al.[9], published in this issue of the Journal of Hypertension, offers an important contribution in this field, by critically considering the available studies exploring the accuracy of commercial devices and methods for noninvasive estimation of aortic SBP. The study by Papaioannou et al.[9] has certainly the merit of having listed, grouped, and summarized validation studies performed so far to verify the reliability of noninvasive cSBP estimation by all commercially available, as compared to invasive intra-arterial measurements. The main conclusion of the authors is that automated recording of pulse waveforms, calibrated noninvasively by brachial mean and DBP values, seems a promising approach that can provide simpler and relatively more accurate noninvasive estimation of central aortic SBP.

In spite of its interest, however, the review study by Papaioannou et al.[9] is also affected by limitations, the most important one being the inclusion in the same meta-analysis of studies dealing with very different methods for assessing cSBP, with the risk of comparing pears and apples. Indeed, in the study by Papaioannou et al.[9], the results of both applanation tonometry and cuff-based oscillometric pulse wave detection and recording are compared in the same analysis and ranked at the same level. This is a questionable approach, given that there is a consistent difference between the solidity of the evidence currently available on validation of tonometric devices and the yet somehow preliminary evidence on the accuracy of systems based on oscillometry.

Indeed, when comparing cSBP values obtained from different devices with a variety of methodological approaches, both the scientific basis and the consistency of each individual method for cSBP assessment appear to be issues deserving greater attention. In this context, the reliability of tonometric methods is supported by a large series of studies. On one side, the methodology for cSBP estimated through the ‘direct’ tonometric method (PulsePen and Complior Analyse) is clear and well verifiable by any operator. On the other side, the ‘indirect’ tonometric method, which reconstructs the aortic pressure waveform from the radial artery through a transfer function, even if the algorithm used by devices such as the SphygmoCor is not disclosed by manufacturers, was shown to be reliable by a number of studies, as clearly highlighted in the meta-analysis by Papaioannou et al.[9]. On such a background, the discrepancy between cSBP measured by arterial tonometers and real cSBP values directly recorded in ascending aorta reported by different validation studies appears to be largely related to the calibration process of noninvasively assessed pulse pressure waveforms. Actually, the approach to arterial waveform calibration is probably the main limitation of currently available studies on arterial tonometry validation. This is related not only to the different choice, made in different studies, to perform such calibration by considering SBP and DBP rather than mean and DBP, but also to the difficulty in defining a real and accurate value of diastolic and mean arterial pressure to be used for calibration. The latter problem is likely to depend on the variable accuracy of oscillometric devices used for brachial BP measurement, even if such accuracy was considered acceptable according to international protocols. Since the first validation studies considered in this meta-analysis, in the attempt to increase the reliability of noninvasive central BP estimates many efforts have indeed been made by manufacturers to improve mean arterial pressure measurement, as an example by providing the possibility to define the real value of mean arterial pressure starting from the integral of the tonometric pressure wave recorded at the brachial artery.

At variance from what was available for ‘tonometric’ methods, there is still debate on the accuracy of methods estimating central BP through automated, cuff-based, oscillometric devices. An increasing number of these devices are available on the market, and are well received by clinicians because of their easy application, but the method they use to define central BP remains undisclosed, and thus difficult to be assessed for accuracy. Moreover, the pulse waves from which the estimation of cSBP is obtained are not always clearly shown, and the operator is not allowed to verify the reliability of signals recorded with some of these devices.

As a general remark, the almost totality of validation studies so far performed on devices for noninvasively assessing central BP have focused only on the comparison of crude cSBP values measured by the reference invasive versus the tested noninvasive method, with no simultaneous assessment of the reliability of the underlying methodology, an issue which would on the contrary need to be addressed [10]. It is only after performing a careful quality assessment of the recorded signal, and after verifying the reliability of the methodology employed, that noninvasively and indirectly obtained central pulse waves should be validated against reference aortic pulse waves obtained through an invasive approach. This should be done through the comparison not only between invasive and noninvasively estimated cSBP values but also between invasive and noninvasive waveforms, by considering the first 6 to 10 harmonics of the pulse waveforms recorded.

As a final consideration, above and beyond the problems related to the validation of noninvasive devices for central BP assessment available on the market, the main issue yet to be addressed from a clinical perspective in this fast developing field is the definition of the actual meaning of cSBP and the need to understand why and in which situation it should be measured. Indeed, on the background of the currently available evidence, international guidelines for hypertension management yet do not support central BP assessment to be used in routine clinical practice, and still consider such an assessment mostly for research purposes. In particular, the 2013 European Society of Hypertension/European Society of Cardiology guidelines for the management of arterial hypertension [11] recommend the measurement of central BP only in the evaluation of the isolated systolic hypertension in the young. Conversely, analysis of peripheral and central pulse waveforms and assessment of carotid–femoral pulse wave velocity, as allowed by arterial tonometers, may offer important information on hemodynamic mechanisms involved in arterial hypertension, including the study of timing and morphology of reflected waves [4,12], and may additionally provide an estimate of indices evaluating subendocardial flow supply : demand ratio [13,14]. This may allow a more tailored approach to the individual patient with high BP, with the possibility of a more precise and personalized management of the antihypertensive treatment. In such a broader approach to the evaluation of arterial wall properties, the estimate of cSBP should at present be considered only as one of multiple possibilities offered by central pulse wave analysis, in the frame of a more complete hemodynamic assessment of the hypertensive patient. In conclusion, although pulse wave velocity is already a solid index of hypertension-related organ damage [12,15] and has already been included in international guidelines for hypertension management [11], central BP estimation should still be considered mostly a research issue [16], including in such research also the identification of the best approach for a careful validation of devices and methodologies for its noninvasive assessment.


Conflicts of interest

P.S. is a consultant for DiaTecne s.r.l. The other authors have no conflicts.


1. Safar ME, Blacher J, Pannier B, Guerin AP, Marchais SJ, Guyonvarc’h PM, et al. Central pulse pressure and mortality in end-stage renal disease. Hypertension 2002; 39:735–738.
2. Pini R, Cavallini MC, Palmieri V, Marchionni N, Di Bari M, Devereux RB, et al. Central but not brachial blood pressure predicts cardiovascular events in an unselected geriatric population: the ICARe Dicomano Study. J Am Coll Cardiol 2008; 51:2432–2439.
3. Roman MJ, Devereux RB, Kizer JR, Lee ET, Galloway JM, Ali T, et al. Central pressure more strongly relates to vascular disease and outcome than does brachial pressure: the Strong Heart Study. Hypertension 2007; 50:197–203.
4. Salvi P. Pulse waves: how vascular hemodynamics affects blood pressure. Milan: Springer; 2012.
5. Salvi P, Lio G, Labat C, Ricci E, Pannier B, Benetos A. Validation of a new noninvasive portable tonometer for determining arterial pressure wave and pulse wave velocity: the PulsePen device. J Hypertens 2004; 22:2285–2293.
6. Salvi P, Safar ME, Labat C, Borghi C, Lacolley P, Benetos A. PARTAGE study investigators. Heart disease and changes in pulse wave velocity and pulse pressure amplification in the elderly over 80 years: the PARTAGE study. J Hypertens 2010; 28:2127–2133.
7. Benetos A, Gautier S, Labat C, Salvi P, Valbusa F, Marino F, et al. Mortality and cardiovascular events are best predicted by low central/peripheral pulse pressure amplification but not by high blood pressure levels in elderly nursing home subjects: the PARTAGE study. J Am Coll Cardiol 2012; 60:1503–1511.
8. Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D, et al. 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; 113:1213–1225.
9. Papaioannou TG, Karageorgopoulou TD, Sergentanis TN, Protogerou AD, Psaltopoulou T, Sharman JE, et al. Accuracy of commercial devices and methods for noninvasive estimation of aortic systolic blood pressure a systematic review and meta-analysis of invasive validation studies. J Hypertens 2016; 34:1237–1248.
10. Salvi P. Is validation of noninvasive hemodynamic measurement devices actually required? Hypertens Res 2014; 37:7–9.
11. Mancia G, Fagard R, Narkiewicz K, Redon J, Zanchetti A, Bohm M, et al. 2013 ESH/ESC guidelines for the management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). Eur Heart J 2013; 34:2159–2219.
12. Laurent S, Boutouyrie P, Asmar R, Gautier I, Laloux B, Guize L, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001; 37:1236–1241.
13. Salvi P, Parati G. Aortic stiffness and myocardial ischemia. J Hypertens 2015; 33:1767–1771.
14. Buckberg GD, Fixler DE, Archie JP, Hoffman JI. Experimental subendocardial ischemia in dogs with normal coronary arteries. Circ Res 1972; 30:67–81.
15. Laurent S, Boutouyrie P. Recent advances in arterial stiffness and wave reflection in human hypertension. Hypertension 2007; 49:1202–1206.
16. Agabiti-Rosei E, Mancia G, O’Rourke MF, Roman MJ, Safar ME, Smulyan H, et al. Central blood pressure measurements and antihypertensive therapy a consensus document. Hypertension 2007; 50:154–160.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.