Duration of surgical antibiotic prophylaxis and surgical site infection in orthopaedic surgery: a prospective cohort study

Background:The optimal duration for surgical antibiotic prophylaxis (SAP) for preventing surgical site infection (SSI) in orthopaedic surgeries remains poorly supported by high-level evidence. This study aimed to assess the association between SAP duration and the occurrence of SSI within one year postoperatively. Methods:This prospective cohort study was based on the database from Surgical Site Infection Surveillance and Improvement Project (SISIP) of a tertiary orthopaedic university hospital from October 2014 to December 2020. The main outcome was SSI, defined according to the CDC/NHSN criteria, determined by a review of index hospitalization medical records, microbiology laboratory reports, and readmission records for SSI treatment within one year after discharge. Adjusted generalized additive models (GAMs) were performed to assess the relationships between SAP duration and SSI, determine the cut-off point of SAP duration, and estimate the relative contribution of each included variable across the overall cohort and the three subgroups (open fracture, closed fracture, and non-traumatic group). Multivariable logistic regression models were used to estimate the association between prolonging SAP duration based on the cut-off point and SSI. Results:There were 37 046 patients (55.1% male) included, with an overall SSI incidence of 2.35% (871/37 046). In adjusted GAMs, no statistically significant relationships were observed in the overall cohort and open or closed group (P>0.05), but a non-linear relationship was exhibited in the non-traumatic group (P=0.03); the cut-off point was 2.4 days for the overall cohort and 3.6 days (open), 2.6 days (closed), 1.1 days (non-trauma) for three subgroups. In adjusted logistic regression, prolonging SAP duration did not demonstrate a statistically significant protective effect in overall cohort (aOR=0.868; 95% CI, 0.746-1.011) and three groups (open: aOR=0.867; 95% CI, 0.668-1.124; closed: aOR=0.925; 95% CI, 0.754-1.135; non-trauma: aOR=1.184; 95% CI, 0.832-1.683). The relative contribution ranks of SAP duration were 21st overall among 34 factors, 14th for open fractures, 28th for closed fractures, and 3rd for the non-traumatic group among 33 factors. Conclusion:Prolonged postoperative SAP duration has no protective effect against SSI in orthopaedic surgery. Our findings support current guidelines against the practice of continuing SAP postoperatively.