Withholding Pre-emptive Antibiotics After Pancreaticoduodenectomy in Patients with Biliary Colonization: A...

Pre-emptive antibiotic therapy is frequently prescribed after pancreaticoduodenectomy when intraoperative bile cultures are positive, despite the absence of clinical infection. Evidence supporting this practice remains limited. We assessed the association between withholding pre-emptive antibiotics and postoperative outcomes in patients with biliary colonization.

This multicenter retrospective cohort study included 301 consecutive adults undergoing pancreaticoduodenectomy with positive intraoperative bile cultures in four French tertiary centers between January 2014 and December 2024. Overall, 89 patients received PE-ATB, and 212 did not. Propensity score matching at a 1:2 ratio yielded 138 patients, including 46 treated patients and 92 controls. The primary outcome was 5-year overall survival. Secondary postoperative outcomes were examined using prespecified exploratory non-inferiority margins.

Preoperative biliary stenting independently predicted PE-ATB use (odds ratio, 4.11; 95% confidence interval, 1.26-13.40;P =  0.019). In the matched cohort, withholding PE-ATB was compatible with the non-inferiority margin for intra-abdominal abscess, pancreatic fistula, transfusion requirements, and antibiotic duration. Overall postoperative infection did not differ significantly between groups (odds ratio, 1.42; 95% confidence interval, 0.66-3.07; P = 0.376). Five-year mortality was higher in the group without PE-ATB in the matched cohort (hazard ratio, 2.02; 95% confidence interval, 1.01-4.04; P = 0.048), whereas no survival difference was observed in the overall unmatched cohort.

Withholding PE-ATB was not associated with worse early postoperative outcomes and was associated with reduced antibiotic exposure. The survival difference observed only after matching should be considered cautiously and may reflect residual confounding. Prospective randomized trials are needed.