Reproducibility of the Cardio-Ankle Vascular Index (CAVI) in Caucasians (SAPALDIA 3)

Abstract

Purpose: This nested study of the second follow-up of the Swiss Cohort Study on Air Pollution and Lung and Heart Diseases in Adults (SAPALDIA 3) aimed at analysing the reproducibility of the arterial stiffness marker cardio-ankle vascular index (CAVI) in Caucasians. Methods: In 105 persons aged 50-80 yrs (50 males, 55 females, 63.0±6.9 yrs) two measurements of CAVI were taken on each of two days within 3 months using a VaseraVS-1500 vascular screening system (Fukuda Denshi) in supine position after 15 min of rest. The analyses involved mean coefficients of variation (CV), (figur presented) Bland-Altman Plots (BAP) and intraclass correlation coefficients (ICC). In order to allow for a fieldworker effect on the measurement of CAVI a mixed linear model with a fixed effect of time (i.e., first and second measurement day) and random effects for subject and time within subject (measurement date) and fieldworker was applied. Results: The CV was 4.4% and the ICC reflecting the ratio of between and within subject variability was 0.803 and when excluding subjects with differing fieldworker 4.3% and 0.805, respectively (N=82). The limits of agreement of the BAP were -1.155 to 1.445 with a mean difference of 0.145 (CI 0.019-0.271). The BAP displayed no particular dispersion pattern. The mixed linear model revealed that the random subject effect accounted for 75.7% of the variability of CAVI measurements, the subject-specific time effect for 14.1%, the fieldworker effect for 5.7% and short-term variability for 4.5%. Conclusions: These results are in line with previous findings in Asian populations and show that the measurement of CAVI is highly reproducible also in Caucasians. Since the fieldworker effect is negligible CAVI can be regarded as an easy-to-apply tool for population wide application.