Discordance between Carotid-Femoral and Brachial-Ankle Pulse Wave Velocity in Identifying Elevated Central Arterial Stiffness

Introduction: Carotid-femoral pulse wave velocity (cfPWV) and brachial-ankle pulse wave velocity (baPWV) are both used to assess arterial stiffness, but their agreement in identifying elevated central arterial stiffness remains uncertain. Methods: This prospective observational study included 222 adults who underwent standardized baseline vascular function testing. The primary phenotype analysis included 205 participants with available cfPWV, maximum bilateral baPWV (baPWVmax), and minimum ankle-brachial index data. Elevated central arterial stiffness was defined as cfPWV ≥10 m/s and elevated baPWV as baPWVmax ≥14 m/s. Associations between cfPWV and baPWVmax were assessed using Spearman correlation and residual correlation after adjustment for age, sex, and systolic blood pressure. Threshold agreement, receiver operating characteristic analysis, and alternative baPWVmax thresholds were evaluated. Results: cfPWV and baPWVmax showed a moderate positive correlation (Spearman ρ = 0.36, p <0.001), but their residual correlation was no longer significant after adjustment for age, sex, and systolic blood pressure (ρ = −0.05, p = 0.456). baPWVmax showed limited discrimination for elevated cfPWV, with an area under the curve of 0.669 (95% confidence interval, 0.592–0.745). At the commonly used baPWV threshold of 14 m/s, sensitivity was 40.3%, specificity was 75.9%, and Cohen’s κ was 0.17. Overall, 75 of 205 participants (36.6%) had discordant cfPWV-baPWV classification. The most frequent discordant phenotype was cfPWV-only elevation, which accounted for 43 participants (21.0%) and 59.7% of those with elevated cfPWV. The Youden-derived baPWV threshold was 11.75 m/s, which improved sensitivity to 70.8% but reduced specificity to 60.2%. Discordance persisted using alternative baPWV thresholds and after excluding participants with an ankle-brachial index <0.90. Conclusion: cfPWV and baPWV are correlated at the group level but show limited interchangeability for individual threshold-based classification. A baPWV value below 14 m/s should not be interpreted as excluding elevated central arterial stiffness. baPWV remains a practical vascular risk assessment tool, but cfPWV is still needed when the clinical or research focus is central arterial stiffness.