Background: Surgical or percutaneous occlusion of the left atrial appendage (LAA) is increasingly used for thromboembolic protection in atrial fibrillation. Incomplete LAA closure may increase risk of thrombosis and thromboembolism, and therefore approaches to address residual communications are needed. Objective: To analyze the technique of closing an incompletely occluded LAA and subsequent patient outcomes. Methods: We performed a retrospective analysis of 5 consecutive patients who presented for completion of LAA closure. Results: Four patients were male, mean age 75, average CHA2DS2-VASc score 5.4, and four had prior surgical LAA ligation. One patient had previously had a WATCHMAN device placed for whom a 3D printed model was created from preprocedural imaging data to guide Amplatzer occluder device selection for closure. The residual LAA communication maximal diameter averaged 6.2 mm (range 5-8mm). In 4 of 5 cases, an ablation catheter was used to enter the LAA. The residual LAA communication was closed with either an Amplatzer occluder (n=3) or a WATCHMAN device (n=2). No procedural complications occurred, and no residual leak remained afterwards. No neurologic events occurred during follow up (average 603 days, range 155-1177 days). Anticoagulation or dual antiplatelet therapy was stopped following a transesophageal echo (TEE) ³ 6 weeks after the procedure demonstrated no residual communication in 4 of 5 patients, and after 20 weeks in the fifth patient without a follow up TEE. Conclusion: Large residual LAA communications after LAA occlusion attempts can be successfully and safely closed percutaneously using either Amplatzer occluder devices or WATCHMAN devices.