Intraoperative hypothermia as a perioperative risk marker after percutaneous transhepatic cholangioscopic lithotripsy: a session-level cohort study

Background Inadvertent intraoperative hypothermia is a common and potentially modifiable perioperative exposure, but its association with postoperative complications after percutaneous transhepatic cholangioscopic lithotripsy (PTCSL) remains unclear. We examined this association within 30 days after PTCSL. Methods We conducted a single-center, session-level retrospective cohort study of PTCSL sessions performed between January 2016 and December 2025. Intraoperative hypothermia was defined as a lowest recorded temperature below 36.0 °C. The primary outcome was any postoperative complication occurring within 30 days after each session. Associations were estimated using modified Poisson generalized estimating equations with patient-clustered robust standard errors. Secondary analyses evaluated hypothermia severity, model-standardized risks, propensity-score overlap weighting, and prespecified sensitivity analyses. Results The study included 588 PTCSL sessions from 330 patients. Intraoperative hypothermia occurred in 195 sessions (33.2%), and postoperative complications occurred after 331 sessions (56.3%). The observed complication risk was 51.1% without hypothermia and 66.7% with hypothermia. In the primary preoperative-adjusted model, hypothermia was associated with a higher risk of postoperative complications (adjusted risk ratio [aRR], 1.335; 95% confidence interval [CI], 1.162–1.533; P < 0.001). The model-standardized risks were 50.3% (95% CI, 45.3%–55.5%) without hypothermia and 67.3% (95% CI, 60.5%–74.2%) with hypothermia, corresponding to a risk difference of 17.1 percentage points (95% CI, 8.9–25.2). After additional adjustment for procedure-related factors, the association remained statistically significant (aRR, 1.252; 95% CI, 1.089–1.440; P = 0.002). Grade 1 and Grade 2–3 hypothermia were both associated with higher postoperative risk. The overlap-weighted risk ratio was 1.313 (95% CI, 1.134–1.528), with a risk difference of 15.8 percentage points (95% CI, 7.2–24.7). Sensitivity analyses produced similar estimates. Conclusions Intraoperative hypothermia was associated with a higher risk of postoperative complications within 30 days after PTCSL. Recorded hypothermia may be a clinically informative and potentially modifiable perioperative marker. Prospective multicenter studies with standardized continuous temperature monitoring are needed to confirm these findings and evaluate whether targeted thermal management can improve postoperative outcomes.