Arterial Pulse Wave Velocity and Heart Rate

To the Editor: The concerns raised by Lantelme et al1 about the relationship between heart rate and pulse wave velocity (PWV) properly apply to the method they employed, not to the conventional method, nor to the long-established relationship between PWV and arterial stiffness. Lantelme et al used Complior, as did the 3 other studies (Lantelme references 8, 15, 17), which appeared to show increasing PWV with increasing heart rate. In Complior, the sensor used to detect the pulse produces a signal, which is related to the derivative of the pressure pulse. A proprietary algorithm is then used to identify the waveform in a proximal and in a peripheral artery, to measure the time difference between the 2 sites, and thereby to calculate pulse wave velocity from the distance between the sites. In the conventional method, PWV is measured from the time delay between the foot (sharp initial systolic upstroke) of the wave at the 2 sites. The theoretical and experimental basis for using PWV as a measure of arterial stiffness was established in the nineteenth century, and the earliest clinical studies were conducted in 1922.2 We are unaware of data similar to those presented by Lantelme, showing any significant relationship between heart rate and PWV using the conventional method, nor can we conceive any theoretic basis for such.3,4 Because increasing heart rate is associated with an in increase in the rate of systolic rise and abbreviated diastolic duration, the most likely explanation for Lantelme and coworkers’ finding is that the algorithm used to obtain transit time from the whole waveform is affected by these changes in waveform shape. The foot of the waveform is determined by the high frequency components in the pulse wave (greater than 10 Hz), and their contribution to the wave foot does not change with heart rate.3 The foot-to-foot velocity is related to the phase velocity of the high frequency components, and because the dynamic elastic properties of the arterial wall do not alter appreciably at high frequencies, it is inconceivable that there should be a passive effect on arterial stiffness simply by altering the input frequency. The alternate explanation the authors offer in terms of tachycardia resulting in vessel stiffening as a passive effect by shortening the time available for recoil then becomes difficult to understand. Complior is a convenient method for measuring PWV. Its initial evaluation5 did not include studies as comprehensive as those described by Lantelme et al. Unless these concerns can be addressed, the conventional method for measuring PWV from the wave foot is preferred when any change in heart rate is possible. The authors of this paper described carotid-femoral pulse wave velocity as a marker of atherosclerosis. This is simplistic and incorrect.3 Our earliest studies showed no difference in PWV between comparable groups of subjects in populations with high and low prevalence of atherosclerosis, but similar changes in each with aging.6 Megnien et al7 from Paris showed that aortic stiffening does not predict atherosclerotic disease in asymptomatic men at risk. Elevated arterial PWV is a measure of sclerosis, and this results most commonly from aging and hypertensive medial degeneration, not from intimal atherosclerosis.3 Christopher S. Hayward St Vincent’s Hospital Sydney, Australia E-mail [email protected]