Stroke after cardiac surgery may be caused by emboli emerging from an atherosclerotic ascending aorta (AA). Epiaortic ultrasound scanning (EUS), the current ‘gold’ standard for detecting AA atherosclerosis, has not gained widespread use because there is a lack of optimized ultrasound devices, it lengthens the procedure, it endangers sterility, and there is a false belief by many surgeons that palpation is as sensitive as EUS. Furthermore there is no clear evidence proving that the use of epiaortic scanning changes outcome in cardiac surgery. Various researchers investigated the ability of transesophageal echocardiography (TEE) to discriminate between the presence and absence of AA atherosclerosis. It is acknowledged that TEE has limited value in this, but it has never been supported by a meta-analysis estimating the true diagnostic accuracy of TEE based on all quantitative evidence. We aimed to do this using state-of-the-art methodology of diagnostic meta-analyses.
We searched multiple databases for studies comparing TEE vs. EUS for detection of atherosclerosis. A random-effects bivariate meta-regression model was used to obtain summary estimates of sensitivity and specificity, incorporating the correlation between sensitivity and specificity as well as covariates to explore heterogeneity across studies.
We extracted six studies with a total of 346 patients, of whom 419 aortic segments were analyzed, including 100 segments with atherosclerosis +AFs-median prevalence 25+ACU- (range 17-62+ACU-)+AF0-. Summary estimates of sensitivity and specificity were 21+ACU- (95+ACU- CI 13-32+ACU-) and 99+ACU- (96-99+ACU-), respectively.
Because of the low sensitivity of TEE for the detection of AA atherosclerosis, a negative test result requires verification by additional testing using epiaortic scanning. In case of a positive test result, AA atherosclerosis can be considered as present, and less manipulative strategies might be indicated.