RESUMO
OBJECTIVE: To perform a systematic review and meta-analysis of studies evaluating anticoagulation during the early postoperative period following mechanical heart valve implantation. METHODS: Five literature databases were searched to assess the rates of bleeding and thromboembolic events among patients receiving oral anticoagulation (OAC), both with and without bridging anticoagulation therapy with unfractionated heparin (UFH) or subcutaneous low molecular weight heparin (LMWH). The studies' results were pooled via a mixed effects meta-analysis. Heterogeneity (I(2) ) and publication bias were both evaluated. RESULTS: Twenty-three studies including 9534 patients were included. The bleeding rates were 1.8% (95% confidence interval CI 1.0-3.3) in the group receiving OAC, 2.2% (95% CI 0.9-5.3) in the OAC + UFH group, and 5.5% (95% CI 2.9-10.4) in the OAC + LMWH group (P = 0.042). The thromboembolic event rate was 2.1% (95% CI 1.5-2.9) in the group receiving OAC, as compared with 1.1% (95% CI 0.7-1.8) when the bridging therapy groups were combined as follows: OAC + UFH and OAC + LMWH (P = 0.035). Most of the analyses showed moderate heterogeneity and negative test results for publication bias. CONCLUSIONS: Bridging therapy following cardiac valve surgery was associated with a lower thromboembolic event rate, although the difference was small, with considerable overlap of the CIs. Direct comparisons are missing. Bridging therapy with UFH appears to be safe; however, this observation has a risk of bias. Early bridging therapy with LMWH appears to be associated with consistently high bleeding rates across multiple analyses. On the basis of the quality of the included studies, more trials are necessary to establish the clinical relevance of bridging therapy and the safety of LMWH.