Reliability of measurement of endothelial function across multiple institutions and establishment of reference values in Japanese.

AIMS For the standardization of flow-mediated vasodilatation (FMD) assessment as a clinical tool, validation of its reliability across multiple institutions and the establishment of normal/reference values based on reliable data from multiple institutions are needed. METHODS AND RESULTS In Study 1, assessment of FMD (scan recording and analysis) using an ultrasonographic semi-automatic measuring system (sFMD) was conducted at 18 participating institutions (sFMD-INST) (n = 981). All of the brachial arterial scans were also analyzed at a core laboratory (sFMD-COLB). After 111 subjects with inadequate sFMD recordings were excluded (n = 880), the correlation between the sFMD-INST and sFMD-COLB improved from R = 0.725 to R = 0.838 (p < 0.001). In Study 2, based on good-quality sFMD data obtained from 6660 subjects without cardiovascular disease (CVD) and 729 subjects with CVD from 27 institutions, reference values of sFMD are proposed by the Framingham risk score (FRS)-based risk categories and according to gender and age. The receiver-operating characteristic curve analysis revealed a significant power of sFMD values in reference ranges to discriminate between subjects with and without CVD (e.g., area under curve = 0.64 in the FRS-low risk group). CONCLUSIONS When the analysis was limited to cases with clear sFMD recordings, the reliability of the sFMD assessment (scan and its analysis) conducted in individual institutions appeared to be acceptable. Reference sFMD values (lower cuff occlusion) for the Japanese population are proposed based on reliable data derived from multiple institutions, and the reference values may identify patients without advanced vascular damage.