Bariatric surgery in patients with ventricular assist devices: a matched analysis of MBSAQIP.

Abstract

Background: Patients with obesity and heart failure managed with ventricular assist devices (VADs) are a medically complex population who could benefit from metabolic and bariatric surgery (MBS), but are often considered prohibitively high risk.

Objectives: We wished to report the outcomes following MBS in patients with VADs in a contemporary North American cohort.

Setting: Accredited MBSAQIP centers in United States and Canada in 2022.

Methods: We used the 2022 MBSQIP participant user file (PUF) to identify all patients who had a preoperative ventricular assist device. We reported the baseline characteristics, hospitalization course, 30-day complications, and weight loss outcomes. We performed 1:1 nearest neighbor Mahalanobis distance matching without replacement to identify patients with similar comorbidities, and used g-computation to estimate the independent effect of VAD therapy on key outcomes.

Results: Of the 120 patients undergoing MBS with a history of VAD, 78 underwent sleeve gastrectomy and 17, Roux-en-Y gastric bypass. The median postoperative length of stay was 2 days, and with 119 patients surviving to hospital discharge, the overall 30-day risk of death was .8%. There were no recorded pulmonary embolism, stroke, cardiac arrest, or myocardial infarction events. Compared to their matched counterparts, VAD patients had an increased risk of length of stay >7 days (risk ratio 1.40, 95% CI: .91-2.14) and required more units of blood transfusion (mean difference .19, 95% CI: .06-.32). VAD therapy was not statistically significantly associated with an increased risk of death (risk ratio 1.04, 95% CI .76-1.42).

Conclusions: Early outcomes after bariatric surgery in patients with VADs suggest feasibility and safety. Larger studies with well-defined selection criteria and longer term follow up are needed to confirm these results.