Reference Values of Arterial Stiffness−Related Indexes Derived from Pulse Wave Analysis

Abstract

The development of arterial applanation tonometry with modern pulse wave analysis (PWA), which includes estimation of central hemodynamic parameters based on the generalized pressure transfer function (GTF) included in the SphygmoCor ® system, has enabled the noninvasive evaluation of systemic arterial stiffness. Since the 1990s, this method had been applied to clinical studies and large-scale clinical trials, such as the Conduit Artery Function Evaluation (CAFE) substudy of the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT) (1). However, the absence of reference values for these parameters has prevented the use of PWA in routine clinical practice. The study of Wojciechowska et al. published in this journal (2) provided preliminary threshold values of arterial stiffness indexes, including peripheral as well as central augmentation indexes (AI’s), that were acquired using the SphygmoCor ® system, and determined by analyzing their distribution characteristics in a normal reference population of white Europeans. Although this was a pioneering study, care should be taken when interpreting the results due to the methodological issues described below. Chen et al. (3) extensively evaluated the use of GTF to estimate central aortic pressure waveform. Their study justified the application of the estimated central waveform only for the acquisition of systolic blood pressure and pulse pressure, and vascular compliance, if clinically acceptable accuracy were to be assured. The report also indicated the difficulty of estimating the central AI in the waveform reconstructed by the use of GTF. In contrast to systolic peak pressure and diastolic pressure decay, which are mostly determined by the lower frequency characteristics of GTF, the estimation of AI is related to higher frequency characteristics. As shown in this and the other reports (4), the variance of the pressure transfer function is larger for higher frequencies, which are essential to reproduce central AI precisely. Needless to say, as the GTF cannot be individualized, if the condition of the subject differs from the average condition in the reference population, there is no guarantee that the estimated aortic pressure waveform is identical with the actually measured one. Therefore, GTF-based estimation of the central AI is not accurate enough to replace the actual central AI. A similar criticism of GTF-based analysis also appeared in a more recent report by Millasseau et al. (5).