Reproducibility of brachial ultrasonography and flow-mediated dilatation (FMD) for assessing endothelial function

High-resolution brachial artery ultrasonography is used to study vasodilator response induced by physiologic reactive hyperaemia. We examined the reproducibility of measuring flow-mediated dilatation (FMD) on two occasions.

To determine the degree of variability of this technique in our vascular laboratory for the design of clinical research studies.

Nineteen subjects were studied on two separate occasions using an Acuson 128 ultrasound device and a 7.0 MHz linear array transducer. Reactive hyperaemia was induced in the brachial artery by inflation and release of a blood pressure cuff. Nitrate-induced dilatation was assessed in 11 of the 19 subjects. Measurements were made by two observers blinded to subject details.

The 11 subjects given sublingual GTN during the first ultrasound study had a mean nitrate-induced dilatation of 20.7% (sd 9.6). The mean vessel diameter of 3.78 mm (sd 0.7) at rest and 3.89 mm (sd 0.7) during reactive hyperaemia yielded a mean FMD of only 3.0% (sd 2.7). The mean difference in FMD within-observers was 0.13% (sd 2.07), between-observers 0.06% (sd 2.17) and between-studies was 0.57% (sd 6.83).

The reproducibility of FMD measured by brachial artery ultrasound was poor and likely to render the measurements inaccurate for clinical research in our hands. Between-study variation contributed the largest proportion of total study variability. We suggest that investigators using this technique conduct their own careful reproducibility studies in order to avoid the misinterpretation of 'negative' studies.