Battle of the buttress: 5-year propensity-matched analysis of staple-line reinforcement techniques from the MBSAQIP database

Background Vertical sleeve gastrectomy (VSG) has demonstrated to be safe; however, controversy remains on how to decrease major complications, particularly bleeding and leaks. There are variations in staple-line reinforcement techniques, including no reinforcement, oversewing, and buttressing. We sought to evaluate the effect of those methods on post-operative complications using the Metabolic and Bariatric Surgery Accreditation Quality Initiative Program (MBSAQIP) database. Methods The MBSAQIP was queried for patients who underwent VSG during 2015–2019. A propensity-matched analysis was performed between different staple-line reinforcement (SLR) methods, specifically No reinforcement (NR), Oversewing (OS), and Buttressing (BR). The primary outcome of interest was complications within 30 days. Results A total of 513,354 VSG cases were analyzed. The cohort was majority female (79.0%), with mean (SD) age of 44.2 ± 11.9 years and mean BMI of 45 ± 7.8 kg/m 2 . Frequency of SLR methods used was 54%BR, 25.6%NR, 10.8% BR + OS, and 9.8%OS. There were no differences in rate of leaks among SLR methods. Compared to NR, BR was associated with decreased rate of reoperations, overall bleeding, and major bleeding ( p < 0.05) but prolonged operative time and length of stay (LOS) ( p < 0.05). OS was associated with decreased overall bleeding ( p < 0.05) but prolonged operative times and length of stay ( p < 0.05) compared to NR. Compared to BR, OS was associated with increased operative times, LOS, and rates of post-operative ventilator use, pneumonia, and venous thrombosis ( p < 0.05). Patients with bleeding were associated with lower rate of BR (56% vs 61%) and higher rate of NR (34% vs 28%) compared to patients with no bleeding. Bleeding was associated with a greater frequency of leaks (4.4% vs 0.3%), along with higher morbidity and mortality ( p < 0.05). Conclusions Of the reinforcement methods evaluated, BR and OS were both associated with decreased bleeding despite longer operative times. No method was found to significantly reduce incidence of leaks; however, bleeding was associated with increased incidence of leaks, morbidity, and mortality. The liberal use of SLR techniques is recommended for further optimization of patient outcomes after VSG.