Risk factors for percutaneous left ventricular assist device explant complications

INTRODUCTION: Percutaneous left ventricular assist device (pLVAD) explant remains nonstandardized with potential complications of bleeding and thrombosis. Explant settings include percutaneous techniques in the catheterization laboratory (CL), manually at bedside (MB), and surgically in the operating room (OR). OBJECTIVE: Identify high‐risk features for explant‐related complications, including indication for support, setting, and technique. METHODS: Postexplant bleeding and thrombosis/limb ischemia were identified following pLVAD removals over 2 years at a multicenter healthcare system. RESULTS: Of 156 patients, bleeding (n = 26 [17%]) and thrombosis (n = 9 [6%]) occurred more often in patients with the peripheral arterial disease (PAD), female gender, anemia, and cardiogenic shock. OR explants had a higher combined endpoint (4/8 [50%]) versus CL (23/133 [17%], p < 0.05) driven by transfusion. There was no difference between OR versus MB (5/15 [33%], p = 0.66) or CL versus MB (p = 0.62). In shock patients, there was no difference between CL (7/30 [23%]) versus MB (5/15 [33%], p = 0.5) and OR (4/7 [57%], p = 0.16); or MB versus OR (p = 0.38). Average length of stay was significantly lower in the CL group versus MB and OR (3.6 ± 33.2 vs. 18.4 ± 10.9 vs. 28.1 ± 15.8 days, p < 0.0001). Preclosure in shock patients (5/25 [20%] vs. 11/27 [41%], p = 0.1383) and crossover balloon occlusion technique (9/44 [16%] vs. 25/112 [22%]; p = 1) were not associated with higher combined endpoints versus control. CONCLUSION: Risk factors for pLVAD explant complications include PAD, female gender, and cardiogenic shock. There was no difference in complication rates between explant settings among cardiogenic shock patients, but shorter length of stay when performed in the CL. There was no difference in complication rates when using the crossover balloon occlusion technique.