Aortic pulse wave velocity measured by pulse wave imaging (PWI): A comparison with applanation tonometry

Abstract

Background Arterial stiffness is a well-established indicator of cardiovascular disease outcome. Pulse Wave Velocity (PWV) is a surrogate for arterial stiffness that is measured either globally using carotid-to-femoral applanation tonometry or locally using biomedical imaging methods. Pulse Wave Imaging (PWI) is an ultrasound-based method for both qualitative visualization of the pulse wave propagation and quantitative estimation of arterial stiffness. The objective of this study is to assess the PWI performance in PWV estimation by comparing local abdominal aortic PWV values obtained by PWI to the carotid-femoral PWV measured by applanation tonometry. Methods A total of 18 subjects (age 18–66, 32.5 ± 14.5) with no history of cardiovascular disease were consecutively tested by both PWI and tonometry. Results The correlation coefficient r between values estimated with the two methods was found to be equal to 0.68. A linear regression yielded PWV PWI = 1.02∗ PWV tonometry +0.15. Tukey mean-difference plots indicated that PWV PWI was significantly lower than PWV tonometry (−0.3 m/s) at lower PWV values (PWV ≤ 7 m/s), whereas PWV PWI was significantly higher (+1.4 m/s) than PWV tonometry at higher PWV values (PWV > 7 m/s). Conclusions Despite the regional nature of the PWV PWI measurements, as opposed to the global PWV tonometry measurements, abdominal PWV PWI and carotid-femoral PWV tonometry values were found to be similar, with an average bias equal to 0.25 m/s. Such a bias and its variation with PWV may be partially explained by both physiological variations of PWV along the arterial tree and by the increasing uncertainty of the PWV estimate by PWI as PWV increases.