Pulse wave analysis detected by a computerized method using the cardio-ankle vascular index in aortic stenosis

Although pulsus parvus et tardus is a classical physical finding in aortic stenosis (AS), objective quantitative assessment of pulse waveform morphology has been limited. We developed a novel computerized pulse waveform parameter, initial upstroke time (iUT), derived from cardio-ankle vascular index (CAVI) pulse waveforms to quantify the delay in the initial systolic upstroke. We analyzed 158 patients with mild or greater AS who underwent CAVI measurement and echocardiography. Associations of conventional upstroke time (UT) and iUT with echocardiographic measures of AS severity, including aortic valve area (AVA), peak aortic jet velocity (Vmax), and mean pressure gradient (MPG), were evaluated using multivariable regression analyses. Diagnostic performance was assessed by receiver operating characteristic analysis. Both UT and iUT were significantly correlated with AVA and Vmax; however, iUT showed stronger correlations. In multivariable analyses including both parameters, iUT remained independently associated with AVA (standardized β = -0.488, P < 0.001) and Vmax (standardized β = 0.573, P < 0.001), whereas UT was not associated with Vmax and showed only a borderline association with AVA. Compared with UT, iUT also demonstrated superior diagnostic performance for identifying severe AS defined by Vmax ≥4.0 m/s (area under the curve [AUC], 0.843 vs. 0.678; P = 0.005) and MPG ≥ 40 mmHg (AUC, 0.795 vs. 0.637; P = 0.003), but not for AVA ≤ 1.0 cm². A novel computerized pulse waveform parameter reflecting the initial systolic upstroke was more strongly associated with echocardiographic measures of AS severity than conventional UT. Automated analysis of CAVI pulse waveforms may provide an objective adjunct for assessing AS without additional testing.