Introduction: The purpose of this study was to evaluate intraobserver and interobserver reproducibility of carotid-femoral pulse wave velocity (PWV) with the Complior method, by comparing two generations of the device and the manual method.
Methods: We studied 31 participants (9 women) with a mean age of 47.67±12.55 years, who were submitted to sequential measures of carotid-femoral PWV by two experienced operators. The evaluations were made by randomly switching between the manual method, the Complior first generation, and the Complior SP (third generation).
Results: The analysis of concordance of measurements of PWV showed strong intraobserver and interobserver correlations. The intraobserver correlation coefficients were 0.984, 0.977, and 0.966, respectively, for manual evaluation, the Complior first generation, and the Complior SP (P<0.0001). In the case of interobserver reproducibility, we found correlations of 0.940, 0.979, and 0.974, respectively, for manual evaluation, the Complior first generation, and Complior SP (P<0.0001). Good indicators of reproducibility were also a withdrawal of the Bland-Altman analysis, with most values within 2 standard deviations. The mean difference±SD for intraobserver and interobserver agreement was -0.006±0.05 and -0.06±0.09 for manual determination; -0.02±0.06 and -0.04±0.06 for Complior first generation; and 0.16±0.09 and 0.15±0.07 for the Complior SP.
Conclusion: The data showed a high reproducibility of the Complior method, documented by excellent correlations and low mean differences in measurements within and between observers. These data showed unequivocally that this method met the quality requirements for its inclusion in integrated clinical follow-up programs.