Variability and reproducibility of flow-mediated dilatation in a multicentre clinical trial.

Abstract

The aim of this study was to assess the reproducibility of flow-mediated dilatation (FMD) in a multicentre setting. This study was performed as part of the dal-VESSEL trial in which FMD was measured in 19 vascular imaging centres in six European countries. A subgroup of patients who were allocated in the placebo group and scanned twice at each trial time point (substudy) was analysed. Intra-sonographer variability was calculated from FMD measurements 48 h apart. Centre variability and short-, medium-, and long-term reproducibility of FMD were calculated at 48 h and at 3 and 9 months intervals, respectively. Intra- and inter-reader variability was assessed by re-analysing the FMD images by three certified readers at two time intervals, 7 days apart. Sixty-seven patients were included. Variability between centres was comparable at 48 h and 3 months interval but almost doubled at 9 months. The mean absolute difference in %FMD was 1.04, 0.99, and 1.45% at the three time intervals, respectively. Curves were generated to indicate the number of patients required for adequate power in crossover and parallel study designs. This study demonstrates for the first time that in a multicentre setting reproducible FMD measurements can be achieved for short- and medium-term evaluation, which are comparable with those reported from specialized laboratories. These findings justify the use of FMD as an outcome measure for short- and medium-term assessment of pharmacological interventions.