The purpose of the study was to compare the intraobserver and interobserver reliability and agreement for measurement of flow-mediated dilation (FMD) between B-mode method and echo-tracking (ET) method. Twenty healthy volunteers (mean age, 31.6 +/- 9.2 years) underwent ultrasound examination by both B-mode and ET methods. Baseline brachial artery diameter, post-cuff release diameter, and FMD percent were assessed by each sonologist on 2 consecutive days. Reliability was assessed by intraclass correlation coefficients, and Bland-Altman plots were used to visually compare measurement bias and agreement by the 2 ultrasound methods. A total of 40 pairs of data were available for analysis. Excellent intraobserver and interobserver reliability values were found for all variables assessed by the 2 methods. The intraclass correlation coefficient values were higher for ET in both intraobserver and interobserver evaluations, but only for interobserver evaluations for post-cuff release diameter and FMD was there no overlap in the 95% confidence interval. The Bland-Altman plots showed that in 95% of the measurements, the percentage difference between ET and B-mode ultrasound techniques was within 18.1%, 19.4%, and 17.3% for baseline brachial artery diameter, post-cuff release diameter, and FMD percent, respectively. The results suggest that ET and B-mode methods are reproducible in assessing the FMD. The ET method improves the reliability of FMD assessment, but we cannot determine which measurement is better for FMD.