King's College London

Research portal

Echocardiography versus computed tomography and cardiac magnetic resonance for the detection of left heart thrombosis: a systematic review and meta-analysis

Research output: Contribution to journalReview article

Alberto Aimo, Eleni Kollia, Georgios Ntritsos, Andrea Barison, Pier Giorgio Masci, Stefano Figliozzi, Dimitrios Klettas, Kimon Stamatelopoulos, Dimitrios Delialis, Michele Emdin, Georgios Georgiopoulos

Original languageEnglish
JournalClinical Research In Cardiology
Accepted/In press1 Jan 2020

King's Authors


Background: Accurate and reproducible diagnostic techniques are essential to detect left-sided cardiac thrombi [either in the left ventricle (LV) or in the left atrial appendage (LAA)] and to guide the onset and duration of antithrombotic treatment while minimizing the risk for thromboembolic and hemorrhagic events. Methods: We conducted a systematic review and meta-analysis aiming to compare the diagnostic performance of transthoracic echocardiography (TTE) vs. cardiac magnetic resonance (CMR) for the detection of LV thrombi, and transesophageal echocardiography (TEE) vs. computed tomography (CT) for the detection of LAA thrombi. Results: Six studies were included in the first meta-analysis (TTE vs. CMR for LV thrombosis). Pooled sensitivity and specificity values were 62% [95% confidence interval (CI), 37–81%] and 97% (95% CI, 94–99%). The shape of the hierarchical summary receiver operating characteristic (HSROC) curve and the area under the curve (AUC) of 0.96 suggested a high accuracy. Ten studies were included in the second meta-analysis (CT versus TEE for LAA thrombosis). The pooled values of sensitivity and specificity were 97% (95% CI, 77–100%) and 94% (95% CI, 87–98%). The pooled diagnostic odds ratio (DOR) was 500 (95% CI, 52–4810), and the pooled likelihood ratios (LR + and LR−) were 17% (95% CI, 7–40%) and 3% (95% CI, 0–28%). The shape of the HSROC curve and 0.99 AUC suggested a high accuracy of CT vs. TEE. Conclusions: TTE is a fair alternative to DE-CMR for the identification of LV thrombi, while CT has a good accuracy compared to TEE for the detection of LAA thrombosis. PROSPERO registration: CRD42020185842.

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454