Preoperative Body Mass Index < 45 Kg/m2 Predicts Clinical Success after Bariatric Surgery in Patients with Ventricular Assist Devices

Abstract

Purpose Although the outcomes of bariatric surgery (BS) in patients with ventricular assist devices (VADs) are promising, the evidence is limited to case reports or very small retrospective cohorts. Hence, no study has tried or been able to identify potential predictors of successful outcomes in this high-risk population, which could aid patient selection. Methods This study combined the individual participant data (IPD) of patients with VADs who underwent BS at Ochsner Medical Center (OMC) and patients of references identified through a systematic search in ClinicalTrials.gov, Cochrane, Embase, and PubMed. Patients who underwent BS during VAD support, as staged interventions, were included if their IPD for post-BS BMI were available. Datasets from the literature search and OMC were merged in an IPD meta-analysis to identify predictors of the composite outcome: body mass index (BMI) Results Among the 38 patients, 18 (58%) were male, the mean age was 42 (±12.5) years, and 33 (86.9%) underwent sleeve gastrectomy, while 5 (13.2%) had Roux-en-Y gastric bypass. During a follow-up of 20 (12-30) months, 30 patients achieved the composite outcome, among whom 22 were listed for HT, 17 underwent HT, and 3 experienced recovery and VAD explantation. Time to composite outcome was 10 (3-17) months. Among all baseline variables, preoperative BMI was the only statistically significant predictor for achieving the composite outcome (p Conclusion Preoperative BMI was the only statistically significant predictor for the composite clinical outcome due to its impact in postoperative BMI during follow-up. Most patients received HT. This reflects the important role of a BMI ≥35 kg/m2 as a contraindication for HT. It is likely that more patients, especially those with shorter follow-up, would have reached the composite outcome after primary references were published or conclusion of this study. Future larger studies may be able to identify additional predictors of clinical outcomes.