LAA Occlusion Devices vs. Medical Therapy in AF

Study Questions:

What is the relative safety and efficacy of left atrial appendage (LAA) occlusion devices as compared to medical therapy in patients with nonvalvular atrial fibrillation (AF)?


A network meta-analysis was performed using 19 randomized controlled trials. These trials either compared warfarin to placebo, antiplatelet therapy, or one of the novel oral anticoagulants (NOACs). Two randomized trials examining the efficacy and safety of LAA occlusion using the Watchman device were also included.


Among 87,831 patients in the 19 studies, 36,645 were assigned to warfarin, 43,314 to NOACs, 6,215 to antiplatelet therapy, 925 to placebo, and 732 were included in the LAA occlusion trials. On direct comparison, there was a mortality benefit with NOACs as compared to warfarin; there was no difference in this outcome between NOACs and LAA occlusion. Warfarin was superior to both antiplatelet therapy and placebo. With respect to thromboembolism, NOAC treatment was superior to warfarin, whereas LAA occlusion was equivalent. On indirect comparison, LAA occlusion was associated with a lower risk of mortality as compared to placebo or antiplatelet therapy, but was not different than that associated with NOACs. In terms of thromboembolism, LAA occlusion was superior to placebo and antiplatelet therapy, but again was not different than NOACs. There was less bleeding in the NOAC group as compared to warfarin; there was no significant difference between warfarin and LAA occlusion. The risk of gastrointestinal (GI) bleeding was higher in patients taking NOAC as compared to warfarin. There was no difference in major bleeding between LAA occlusion and placebo, antiplatelet therapy, or NOAC therapy. There was less GI bleeding associated with LAA occlusion versus NOACs.


The authors concluded that LAA occlusion is associated with a lower risk of mortality and thromboembolism as compared to placebo or antiplatelet therapy, and is equivalent to NOAC therapy.


In the absence of direct comparisons between LAA occlusion and NOACs, the authors performed a network meta-analysis in an attempt to make indirect comparisons. The findings support the efficacy and safety profile of LAA devices. Obviously, these findings should be interpreted with caution since they were not drawn from randomized controlled trials.

< Back to Listings