Background
Postoperative morbidity after laparoscopic bariatric surgery is considered higher for patients undergoing revisional versus primary procedures. The objective of this retrospective cohort study was to compare outcomes between patients undergoing primary versus revisional robotically assisted laparoscopic (RAL) Roux-en-Y gastric bypass (RYGB).
Methods
Data of all patients who underwent RAL primary and revisional RYGB between 2009 and 2019 at two accredited, high-volume bariatric surgery centers—the Memorial Hermann – Texas Medical Center, Houston, TX, and the Tower Health, Reading Hospital, Reading, PA, were analyzed. Primary outcomes were early (< 30 days) and overall postoperative complications. Secondary outcomes included intraoperative complications, operative times, conversions to laparotomy, length of hospital stay, early (< 30 days) postoperative readmissions and deaths.
Results
Data of 1072 patients were analyzed, including 806 primary and 266 revisional RAL RYGB procedures. Longer operative times (203 versus 154 min, P < 0.001), increased number of readmissions for oral intolerance (10.5% versus 6.7%, P = 0.046) and higher rate of gastrojejunal stricture (6.4% versus 2.7%, P = 0.013) were found in the revisional group. Gastrointestinal leak rates were 0.2% for the primary versus 1.1% for the revisional group (P = 0.101). Early (< 30 days) reoperations rates were 2.2% for the primary versus 1.1% for the revisional group (P = 0.318). There were no statistically significant differences between groups in overall and severe complication rates.
Conclusion
Patients undergoing RAL primary and revisional RYGB had comparable overall outcomes, with a non-significant higher early complication rate in the revisional group. Despite the study being underpowered to detect differences in specific complication rates, the morbidity seen in the revisional RYGB group remains markedly below literature reports of revisional laparoscopic RYGB and might suggest a benefit of robotic assistance. Further prospective studies are needed to confirm these results.