The impact of bowel preparation on surgical site infection rates in gynecologic oncology surgery, post-hoc analysis from a randomized controlled trial (550)

Abstract

<b>Objectives:</b> To determine the relationship between oral antibiotic bowel preparation (OABP) and mechanical bowel preparation (MBP) with respect to surgical site infection (SSI) rates in patients with gynecologic cancer undergoing colorectal surgery. <b>Methods:</b> A post-hoc analysis was performed using data from a randomized controlled trial investigating the use of negative pressure wound therapy after laparotomy. Patients with presumed malignancy requiring laparotomy were enrolled from March 2016 to August 2019. For this analysis, we only included cases with confirmed gynecologic malignancy, undergoing gynecologic oncology surgeries that also required colorectal resections. The primary outcome was SSI within 30 (±5) days of surgery. Univariate and multivariable analyses with stepwise model selection for SSI were performed. <b>Results:</b> The cohort included 161 surgeries requiring one or more colorectal resections. Of these, 15 (9%) had no bowel preparation, 39 (24%) had OABP only, and 107 (67%) had OABP+MBP. There was no case in which MBP was used alone. The overall SSI rate was 19% (<i>n</i>=31) and varied significantly by bowel preparation type: 53% (<i>n</i>=8) in the no-preparation group, 21% (<i>n</i>=8) in the OABP alone group, and 14% (<i>n</i>=15) in the OABP+MBP group (p=0.003). The difference between OABP and OABP+MBP was non-significant (p=0.44). The median length of stay was nine days (range: 6-12) for patients with no preparation, six days (range: 5-8) in the OABP group, and seven days (range: 6-10) in the OABP+MBP group (p=0.045). The overall complication rate was 34% (<i>n</i>=54) and did not vary significantly by bowel preparation type: 53% (<i>n</i>=8) no preparation, 28% (<i>n</i>=11) OABP alone, 33% (<i>n</i>=35) OABP+MBP (p=0.23). The overall anastomotic leak rate was 1.2% (<i>n</i>=2), with one case occurring in the no bowel preparation cohort (6.7%), one in the OABP cohort (2.6%), and none in the OABP+MBP cohort. Univariate logistic regression found that OABP was associated with a decreased risk of SSI (odds ratio [OR]: 0.23, 95% CI: 0.06-0.80) compared with no bowel preparation; OABP+MBP was also associated with a decreased risk of SSI (OR: 0.14, 95% CI: 0.04-0.45) compared with no bowel preparation (p=0.004). On multivariate analysis, the use of a bowel preparation demonstrated a significant impact on SSI rates (p=0.004) when accounting for preoperative hemoglobin, partial thromboplastin time, and bowel resection type. <b>Conclusions:</b> The use of OABP is associated with a reduction in SSI rates. Adoption of bowel preparation prior to gynecologic oncology surgeries with anticipated colorectal resection reduces the risk of SSI. While the use of MBP alone has been abandoned, further investigation is needed to determine whether OABP is sufficient or whether OABP+MBP should be recommended.