Observed versus Expected Ischemic and Bleeding Events Following Left Atrial Appendage Occlusion

Data on the efficacy of left atrial appendage occlusion (LAAO) in clinical practice are limited. We performed a systematic review and meta-analysis of observational studies that reported observed versus expected rates of ischemic strokes and/or major bleeding following LAAO. Our primary end points were the pooled relative risk reduction (RRR) in ischemic stroke and major bleeding with corresponding 95% confidence intervals compared with what was expected by the CHA2DS2-VASc and HASBLED scores, respectively. Twenty-nine studies including 11,071 patients (age 74.0 +/- 8.7 years, 60% males) met the inclusion criteria. The mean CHA2DS2-VASc score was 4.22 +/- 1.48, and the mean HASBLED score was 3.04 +/- 1.16. During 19,567 patient-year follow-up, 290 of 11,071 patients (2.62%) suffered an acute ischemic stroke. This represented a 73.6% (95% confidence interval 68.9-78.2%) RRR in ischemic strokes compared with what was expected based on the CHA2DS2-VASc score. A total of 26 studies reported observed versus expected major bleeding (10,056 patients; age 74.0 +/- 8.7, 60% males). During 16,967 patient-year follow-up, 404 of 10,056 patients (4.0%) suffered a major bleeding event. This represented a 55% (95% confidence interval 44.2% to -65.9%) RRR in major bleeding compared with what was expected based on the HASBLED score. These estimates were consistent across subgroups stratified according to age, CHADS2VASc, HASBLED scores and type of LAAO device used. In conclusion, LAAO is associated with a favorable observed/expected ratio with regards to ischemic stroke and major bleeding in clinical practice. Future clinical trials remain essential to further assess the efficacy of LAAO via a direct comparison with oral anticoagulation. (C) 2020 Elsevier Inc. All rights reserved.