Pulse wave analysis: what do the numbers mean?

Applanation tonometry, a simple noninvasive technique for recording a high-fidelity waveform from an accessible pulse, has facilitated pulse wave analysis in large population studies. Most interest has focused on estimating an aortic or “central” waveform by the application of a filter or generalized transfer function applied to a radial waveform.1 Systolic pressure and pulse pressure differ between central and peripheral arteries as a result of peripheral amplification. By contrast, there is little regional variation in mean and diastolic pressures within large arteries and conduit arteries. A synthesized aortic waveform can, thus, be calibrated on the assumption of equality of mean and diastolic pressures at central and peripheral sites and used to estimate central systolic and pulse pressures. The main limitation of this approach, at least for the estimation of central systolic blood pressure, is the accuracy of the peripheral blood pressure measurements used to calibrate the waveform.2 Most studies and a recent meta-analysis suggest that central pressures provide a better prediction of cardiovascular risk than peripheral pressures,3 a finding intuitively rationalized in terms of the load on the left ventricle and coronary and carotid arteries. In addition to central blood pressure, a variety of indices can be obtained relating to inflection points or shoulders in the arterial waveform. Central augmentation pressure (AP) is the height above the first systolic shoulder of the aortic waveform and central augmentation index (AI; cAI) the ratio of AP to central pulse pressure (Figure). AP is thought to relate to pressure wave reflection from the periphery of the circulation and cAI to be an index of pressure wave reflection. However, this interpretation has been challenged, and it has been suggested recently that cAI is determined mainly by the Windkessel or reservoir function of the aorta.4 Experimentally, AP and cAI can be changed independent of aortic stiffness by nitrovasodilation, which selectively dilates muscular arteries from which reflections may arise.5 These findings are more in line with reflection as the predominant determinant of AP and cAI. Irrespective of the determinants of cAI, it appears to be an independent predictor of cardiovascular events.3 Why this should be so is less obvious than for central systolic pressure. Suggestions (not mutually exclusive) are that it is a marker of dysfunction of the arterial wall and that its prognostic impact relates to loading of the left ventricle. Peripheral AI (pAI) is the ratio of the height of the late systolic shoulder in the peripheral (usually radial) pulse to the peripheral pulse pressure (Figure). Largely ignored in favor of cAI, there is now renewed interest in pAI. This is because it is recognized that cAI and pAI are closely related.6 It is possible that this is in part a consequence of the transfer function used to generate the aortic pulse from the radial pulse. In this regard it should be noted that, although the use of the transfer function to estimate central systolic blood pressure is well established, its use for estimation of central augmentation pressure is less well validated. Secondly, it is now appreciated that the late systolic shoulder of the radial waveform corresponds with central systolic blood pressure.7 It follows, therefore, that pAI is the ratio of central:peripheral pulse pressure, and, thus (even if its determinants are complex), it has a simple interpretation in terms of pulsatile load. Although it is clear that much remains to be elucidated with regard to the determinants of APs and indices, their interrelationship, relationship with central/peripheral blood pressure, and, above all, why they predict risk, can we use these measures for risk stratification and/or as targets for therapy? In the this issue of Hypertension, Chirinos et al8 have used individual data from a number of large population studies to provide normative population means for central (using a slightly different definition than the conventional one, see Figure) and radial AI adjusted for age, body height, and heart rate, which are important determinants of AI. This allows for a standardized comparison between values obtained in an individual and the population mean for subjects of the same age and characteristics as the individual. The deviation of AI from the mean can then be used as an index of risk for the individual. How much this will contribute to risk stratification for an individual and how AI compares with other biophysical measures of arterial wall structure/function remain to be clearly defined. It should also be noted that this approach effectively measures relative rather than absolute risk and, therefore, differs from the current approach to risk estimation. Perhaps more importantly than risk stratification, normative values potentially allow a targeted strategy for subjects confirmed to be at risk. Thus, for subjects in whom levels of peripheral blood pressure are at what are currently considered optimal levels but in whom pAI is elevated (indicative of raised central relative to peripheral pulse pressure), a therapy with a selective action to lower central pulse pressure might be beneficial. The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the King’s College London British Heart Foundation Centre, Department of Clinical Pharmacology, St Thomas’ Hospital, London, United Kingdom. Correspondence to Phil Chowienczyk, Department of Clinical Pharmacology, St Thomas’ Hospital, Lambeth Palace Road, London SE1 7EH, UK. E-mail [email protected] (Hypertension. 2011;57:1051-1052.) © 2011 American Heart Association, Inc.