Coronary Flow Velocity Reserve Assessed by Transthoracic Doppler: The iPOWER Study: Factors Influencing Feasibility and Quality

Background Coronary flow velocity reserve (CFVR) measured by transthoracic Doppler echocardiography is a noninvasive measure of microvascular function, but it has not achieved widespread use, mainly because of concerns of validity and feasibility. The aim of this study was to describe the feasibility and factors associated with the quality of CFVR obtained in a large prospective study of women suspected of having microvascular disease. Methods Women with angina-like chest pain and no obstructive coronary artery disease on coronary angiography (<50% stenosis) were consecutively examined by transthoracic Doppler echocardiography of the left anterior descending coronary artery to measure CFVR (n = 947). Quality was evaluated on the basis of (1) identification of the left anterior descending coronary artery, (2) maintained probe position throughout the examination, (3) visibility and configuration of the left anterior descending coronary artery in two-dimensional color Doppler mode, and (4) gradual, consistent increases of characteristic, well-defined flow velocity curves in pulsed-wave mode. Results The mean age (SD) was 62.1 ± 9.7 years. On the basis of the evaluations, patients were divided into four groups according to quality score: nonfeasible (n = 28 [3%]), low quality (n = 80 [8%]), medium quality (n = 451 [48%]), and high quality (n = 388 [41%]). Quality score was associated with diabetes (P < .01), body mass index (P = .02), waist circumference (P = .05), nonsignificant atherosclerosis on coronary angiography (P = .03), and operator experience (P < .01). Low examination quality was associated with lower CFVR (P = .03), also after multivariate adjustment. Conclusions Transthoracic Doppler echocardiographic measurement of CFVR is highly feasible and of good quality in experienced hands. However, CFVR is possibly underestimated when examination quality is low. Awareness of pitfalls and potential bias may improve the validity and interpretation of the measures obtained.