Pulse wave velocity (PWV) has been shown to be a reliable marker for arterial stiffness. Its accuracy depends heavily on the estimation of the transit time (TT) between proximally and distally measured pulse waveforms. Several methods of determining the TT exist, but no consensus for the standardization of one of them has been reached yet. In this work, various state-of-the-art TT estimation methods and the resulting PWV values are examined.

In total, 118 pair-wise measurements using applanation tonometry on the carotid and femoral arteries from 59 patients over a wide age range (21–88 years), recorded by the SphygmoCor system, were examined. For the TT estimation, two traditional methods based on intersecting tangents (IT) and the maximum systolic upstroke (MSU) were used. Furthermore, the two recently proposed methods “diastole-patching” (DP), which compares the region in the foot of the proximal waveform, and the cross correlation of the complete waveform (CC) were compared.

The resulting PWVs differed significantly (p<0.05, Bonferroni corrected paired T-test) between the various methods, with exception of MSU vs. DP. The means ± standard deviations were 8.2±1.8m/s using IT, 7.6±1.9m/s for the MSU, 7.8±2.2m/s using DP, and 9.9±6.7m/s by calculation of the CC. Comparisons of single measurements lead to differences of even more than 6m/s.

Although the 2013 ESH/ESC Guidelines for the management of arterial hypertension suggest a threshold of 10m/s as an estimate of alterations of aortic function, no evaluation method is mentioned. Our results suggest that an agreement on the optimal TT estimation may be useful.