We investigated the inter-day repeatability of newly proposed indexes of cardiovascular health based on the absolute (PWVABS) or relative (PWVRATIO) difference in pulse wave velocity (PWV) between the carotid–femoral (PWVCF) and carotid–radial (PWVCR) segments compared with PWVCF alone and with systolic blood pressure (BPSYST) over multiple days.
All variables were measured in triplicates in a group of 16 young, healthy individuals (8 men/8 women, body mass 66 ± 14 kg, height 1.7 ± 0.1 m, age 27 ± 3 years old, all mean ± SD) on 6 different occasions, always at the same time of the day (± 1 hour) and with at least 48 hours between assessments.
Values did not differ significantly over the 6 different visits for any of the parameters (all Ps > 0.08). Intraclass coefficients (ICCs) ranged from 0.52 (PWVRATIO) to 0.8 (PWVCF). The ICC 95% confidence interval of PWVRATIO (0.30–0.77) and PWVABS (0.33–0.78) did not overlap with that of PWVCF (0.79–0.95), indicating lower repeatability. The coefficient of variation (CV) of PWVABS was over 5-fold larger than the next largest one (46.7 vs. 8.1% for PWVABS and PWVRATIO, respectively), whereas PWVCF and BPSYST showed the lowest CVs (4.6 and 4.0%, respectively).
Between PWVABS and PWVRATIO, the relative method showed better repeatability and might be more sensitive to longitudinal changes in patients. Nonetheless, indexes of arterial stiffness-based PWV gradients between different segments show lower inter-day repeatability compared with PWVCF or BPSYST, suggesting that the latter are still preferable in clinical practice.