Objective: Guidelines recommend a >10m/s threshold of carotid-femoral pulse wave velocity (cf-PWV) to detect hypertension mediated organ damage. However, the >10m/s threshold was derived from the cf-PWV distribution across cross-sectional studies with limited contribution of prospective cohort studies. Using data from the International Database of Central Arterial Properties for Risk Stratification (IDCARS), we aimed to determine an outcome-based threshold for cf-PWV.

Design and method: Adults (≥18 years) recruited by eight IDCARS centres and followed up for 6 months or longer qualified for inclusion in the current subject-level meta-analysis. Cf-PWV was measured using the Sphygmocor device (Itasca, IL). Pulse transit distance was determined by subtracting the distance between the carotid sampling site and the suprasternal notch from the distance between the carotid and femoral sample sites. The primary endpoint consisted of fatal and nonfatal cardiovascular events. Total and cardiovascular mortality and fatal and nonfatal coronary events were secondary endpoints. The cf-PWV threshold was determined by multivariable Cox regression with average risk based on established risk factors in the whole cohort as reference.

Results: After 5.0 years of follow-up (median), of 3494 participants (mean age: 52.2 years; women: 55.3%; mean cf-PWV: 7.8 m/s), 156 (4.5%) experienced the primary endpoint, and 104 (3.0%) died, 25 (0.7%) of cardiovascular disease and 77 (2.2%) had a coronary event. After whole adjusting for sex, age, systolic blood pressure, heart rate and other cardiovascular risk factors, the outcome-driven cf-PWV thresholds were 8.6, 9.1, 8.7 and 8.6 m/s for a composite cardiovascular endpoint, all-cause and cardiovascular mortality and a coronary event, respectively. A cf-PWV ≥9 m/s (n=752, 21.5%), compared with slower velocities, had an increased risk of a composite cardiovascular endpoint (HR: 1.75; 95% CI: 1.20-2.54, P=0.004).

Conclusions: The outcome-based threshold of ≥9 m/s for cf-PWV, which was present in 20% of the IDCARS participants, is close to the cut-off limit proposed by guidelines based on other approaches.