Outcomes of Abdominal Surgery in Patients With Mechanical Ventricular Assist Devices: A Multi-institutional Study.

Abstract

The aim of this study was to examine the outcomes of elective and emergent abdominal operations performed in end-stage heart failure patients supported with ventricular assist devices (VADs). With the growing volume of end-stage heart failure patients receiving VADs, an increasing number of these patients require surgery for noncardiac pathology. There is a paucity of studies on the safety of abdominal operations in this population. We performed a retrospective chart review across 3 hospitals of patients with VADs who underwent abdominal surgeries between 2003 and 2015. We used Chi-square, Fisher exact, and Mann-Whitney U tests for comparison of elective and emergent cases. Fifty-seven patients underwent 63 operations, of which 23 operations were elective, 24 were emergent, and 16 were emergently performed in the same admission as VAD placement and analyzed separately. Patients undergoing elective versus emergent procedures had similar comorbidities (Charlson score 2.9 vs 3.0). 43% versus 32% of patients had VADs as a destination therapy. Although perioperative anticoagulation approach was variable, holding warfarin and starting heparin/enoxaparin/bivalirudin bridge was most common (65% vs 54%). Although 2-fold higher in the emergent group (50 vs 100 mL, P = 0.06), median estimated blood loss was low. Postoperative bleeding requiring transfusion was not very common (13% vs 8%), whereas rate of ischemic cerebrovascular accident (4% each) and venous thromboembolism was low (0% vs 13%, P = 0.23). Thirty-day mortality rate was 4% versus 17%, P = 0.19. VAD patients have an acceptable risk profile for abdominal surgery.