Influence of experience on intra- and interindividual variability in assessing peripheral endothelial dysfunction with high resolution ultrasound

The non-invasive evaluation of endothelial dysfunction with high-resolution ultrasound has become a widely accepted tool in determination of high-risk subjects for early atherosclerosis. Furthermore it is often used as intermediate outcome in intervention studies. Aim: We examined the influence of examiner experience on intra-and inter-individual variability in the measurement of flow-associated vasodilation (FAD) independent of automated analysis systems. Method: FAD was measured on two occasions in 7 and 8 subjects respectively (mean age 32 +/-3 years) by two investigators after different prior training procedures with a 13 MHz linear transducer (LA14A, ESAOTE Biomedica). Results: The intra-individual variability expressed as median absolute difference in the measurements of FAD was 1.1 % (range from 0.03 % to 3.2 %) for examiner one with an experience of more than 50 FAD measurements through former studies and 2.9 % (range from 1.6 % to 9.2 %) for examiner two with only 10 training examinations under supervision. By a further training period of two months,, with an increase of examinations of additional 20 measurements by both examiners, the intra-observer variability could be dropped to 0.9 % (range from 0.03 % to 1.3 %) for examiner two (p = 0.0025) with no significant change for examiner one (median 0.6 % with a range from 0.14 % to 3.7 %). As expected, the inter-individual variability was not influenced by this further training (median 1.0 % with a range of 0.5 % to 3.6 % versus a median of 1.6 % with a range from 0.15 % to 7.5 %). Conclusion: 30 training measurements of FAD under supervision should be regarded as minimum requirement for valid determination of endothelial function. The reachable result for the intraobserver variability is thereby within the range of computed analysis systems.