Evolution of outcomes of robotic bariatric surgery: first report based on MBSAQIP database.

BACKGROUND: Robotic surgery is increasingly being used in bariatric surgery; however, the benefits of robotic surgery in bariatrics remain controversial. OBJECTIVES: The objective of this study was to compare the outcomes of robotic bariatric surgery with laparoscopic surgery over a 3-year period between 2015 and 2017 using the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database. SETTING: University Hospital, United States. METHODS: Using the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database for the years 2015 to 2017, we included patients who underwent primary robotic or laparoscopic sleeve gastrectomy and Roux-en-Y gastric bypass procedures. Patients were divided into either robotic or laparoscopic groups. Primary outcomes included serious adverse events, organ space infection (OSI), readmissions, reoperations, and interventions at 30 days. Secondary outcomes included operation length and hospital stay. We performed propensity score matching based on clinically relevant preoperative variables to create balanced groups before analysis. We analyzed our data using separate Cochran-Mantel-Haenszel tests with year as the stratification variable and conducted subgroup analyses for robotic patients only using separate t tests for proportions, with P < .05 denoting statistical significance. RESULTS: Of the 315,647 patients available for comparison in the 2015 to 2017 Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program Participant User Files, there were 41,364 matched in the final data set. Using the Cochran-Mantel-Haenszel test, we found a significant association between year of performance and outcomes for OSI, 30 day-readmission, and intervention. The incidence of OSI after laparoscopic and robotic cases was .3% and .4%, respectively, in 2015 versus .2% and .3%, respectively, in 2017 (P = .04, odds ratio = 1.49). Thirty-day readmission for robotic cases was 5.2% in 2015 and 4.0% in 2017 (P < .05, odds ratio = 1.16). The incidence of 30-day intervention for robotic cases also dropped from 2.2% in 2015 to 1.3% in 2017 (P < .05, odds ratio = 1.37). Using a Student's t test, there was also a statistically significant decrease in serious adverse events in the robotic group between 2015 and 2017 (incidence of serious adverse events in 2015 was 5.2% versus 3.7% in 2017, P < .05). Rate of 30-day reoperation for the robotic group did change over time but was comparable to the laparoscopic group (1.4% versus 1.3%). CONCLUSIONS: Our study showed between 2015 and 2017 the outcomes of robotic bariatric surgery have improved as evidenced by the significant decrease in the rate of OSI, readmissions, and interventions at 30 days.