Infective endocarditis (IE) is a serious and potentially life-threatening disease, and accurate diagnosis is essential. We performed a systematic review and meta-analysis to assess the diagnostic accuracy of transthoracic echocardiography (TTE), with transesophageal echocardiography (TEE) as the reference standard, in patients with suspected IE of the native valves. We performed a systematic search in MEDLINE, EMBASE and Cochrane Library searching for studies that enrolled adult patients with suspected native valves IE where data about both TTE and TEE could be extracted. We included 11 studies, for a total of 2209 patients. The overall sensitivity, specificity, negative and positive likelihood ratios (LR) of TTE are 0.71 (95% CI 0.56-0.82), 0.80 (95% CI 0.58-0.92), 0.37 (95% CI 0.20-0.68) and 3.56 (95% CI 1.3-9.72), respectively. The subgroup analyses of the studies considering different cut-off levels show that the strict negative criteria (i.e., managing indeterminate results as positive) have the highest sensitivity and the lowest LR-. On the contrary, when managing indeterminate results as negative (standard criteria), the specificity and LR+ are the highest. We observed no differences between the studies performed with older and more recent technologies. In conclusion, our study results support the use of a negative TTE as a single rule-out test in patients with a low pre-test probability. In selected cases, the use of strict negative criteria might exclude IE in intermediate-risk patients, and a positive TTE might be considered as a single rule-in test with no need for TEE if TEE results would not change the patient's management.
Keywords: Diagnostic accuracy; Endocarditis; Meta-analysis; Transesophageal echocardiography; Transthoracic echocardiography.