Background: The method of estimating distance traveled by the pulse wave, used in the calculation of pulse wave velocity (PWV), is not standardized. Our objective was to assess whether different methods of distance measurement influenced the association of PWV to cardiovascular mortality in hemodialysis patients.

Methods 98 chronic hemodialysis patients had their PWV measured using three methods for distance estimation; PWV1: suprasternal notch-to-femoral site minus suprasternal notch-to-carotid site, PWV2: carotid-to-femoral site, PWV3: carotid-to-femoral site minus suprasternal notch-to-carotid site. Carotid-to-femoral distance was used to approximate torso length. Patients were followed for a median of 30 months and the association of PWV and cardiovascular mortality was assessed using survival analysis before and after stratification for torso length.

Results: The three methods resulted in significantly different PWV values. During follow up 50 patients died, 32 of cardiovascular causes. In log-rank tests only tertiles of PWV1 was significantly related to outcome (p-values 0.017, 0.257, 0.137, for PWV1, PWV2 and PWV3, respectively). In adjusted Cox proportional hazards regression only PWV1 was related to cardiovascular mortality. In stratified analysis, however, among patients with below median torso length all PWV values were related to outcome, while in patients with above median torso length none of the PWV methods resulted in significant relationship to outcome.

Conclusions PWV calculated using suprasternal notch-to-femoral distance minus suprasternal notch-to-carotid distance provides the strongest relationship to cardiovascular mortality. Longer torso weakens the predictive value of PWV, possibly due to more tortuosity of the aorta hence more error introduced when using surface tape measurements.