PREOPERATIVE MECHANICAL AND ORAL ANTIBIOTIC PREP REDUCES ANASTOMOTIC LEAK INDEPENDENT OF RESECTION TYPE
Hiromichi Miyagaki, Dimitre Stefanov, Xiaohong Yan, Gregory A. Wasserman*, Neil Mitra, Yanni Hedjar, Hmc Shantha Kumara, Richard L. Whelan
Surgery, Lenox Hill Hospital, New York, NY
Background:
Anastomotic leak after colectomy is a feared complication associated with major morbidity and mortality. Preoperative mechanical bowel preparation (MBP) and per oral antibiotics (POA) have been shown in some studies to lower the rate of leaks, skin and soft tissue infections and deep organ space infection after colectomy, however, other studies support no prep or a POA prep alone. This study's purpose was to revisit this controversial issue by assessing the impact of different bowel preparation approaches on anastomotic leak rates in a large population of colorectal resection patients (right and partial colectomy, LAR) as well as to identify independent risk factors for anastomotic leak.
Methods
A retrospective analysis of the National Surgical Quality Improvement (NSQIP) database from 2016-2020 was done. Patients were included if the surgical indication was "Colon cancer" or "Colon cancer with obstruction" and said patients underwent a colectomy (CPT codes: 44145, 44146, 44207, and 44208 (LAR); 44140, 44141, 44144 and 44204 (PC); 44160 and 44205 (RHC)). Moribund pts were excluded as were pts in whom the case was listed as emergent or if sepsis was present. The primary outcome tracked was anastomotic leak. Regression analysis was performed to identify independent risk factors
Results
46,071 cases met inclusion criteria; no bowel prep, 16.72%; MBP alone, 16.16%; POA alone, 5.56%; POA and MBP, 61.57%. The overall leak rate was 2.5% (1,157 patients): R colon, 1.88%, Partial, 2.37, and LAR 3.14. Overall, the combination of MBP & POA was associated with a 40% reduction in leak rate (OR=0.60, 95% CI 0.52) vs. no prep; MBP & POA was also associated with lower leak rates vs. MBP alone (OR=0.72, 95% CI 0.63-0.83, p<.0001), or POA alone (OR=0.83, 95% CI 0.72-0.97, p=0.02). The same trend was observed for each resection type; by investigating 1st order interactions we concluded the effect of POA and MBP were likely similar in the 3 resection categories as regards leaks. Multivariable logistic regression analysis identified several factors including male sex, current smoker, case length >167 minutes, >10% weight loss in 6 months prior to surgery, and preoperative chemotherapy that were associated with a higher chance of anastomotic leak. Of note, 78% of obstructed pts received a bowel prep.
Conclusion
This NSQIP database study showed that the combination of preop mechanical and antibiotic bowel preparation together significantly reduced the incidence of anastomotic leak following colorectal resection when compared to no prep or to either MBP or POA alone. Although not conclusive, analysis suggests this is true for right colectomy as well as partial and LAR. Prospective studies are warranted.
Back to 2023 Posters