Antibiotic Choice Is Independently Associated With Risk of Surgical Site Infection After Colectomy A Population-Based Cohort Study

Objective: To determine which perioperative care practices are associated with decreased risk of surgical site infection (SSI) after colectomy surgery. Background: Optimization of perioperative care has been a common strategy for improving surgical safety, but the relationship between process measure compliance and surgical complication rates is controversial. Methods: This is a retrospective cohort study performed within the Michigan Surgical Quality Collaborative (MSQC), an organization of hospitals that prospectively collects patient data, processes of care, and 30-day outcomes. Patients undergoing colectomy surgery (n = 4331) were studied. Factors potentially associated with SSI were tested using univariate statistical tests, and a hierarchical generalized linear model was created to test for independent associations between processes of care and SSI, while adjusting for patient risk factors and clustering of patients within hospitals. Results: Several perioperative care practices were independently associated with lower risk of SSI after adjustment for patient risk, procedure type/duration, and clustering of patients by hospital site. Best practices include selection of a Surgical Care Improvement Project (SCIP-2)-compliant prophylactic intravenous antibiotic, postoperative normothermia, postoperative day 1 glucose control, and oral antibiotics given when bowel prep used (SCIP-1 was not significant). Further, several specific prophylactic antibiotic choices were independently associated with lower SSI rates, including cefazolin/metronidazole, ciprofloxacin/metronidazole, and ertapenem. Conclusions: In Michigan, several perioperative care practices are independently associated with decreased risk of SSI after colectomy, including SCIP-2-compliant prophylactic antibiotics, postoperative normothermia, glucose control, and oral antibiotics. Furthermore, specific prophylactic antibiotic choices are associated with lower risk of SSI. These results account for patient factors and unmeasured hospital effects, suggesting that dissemination of these perioperative care practices may decrease SSI rates.