Abstract / Summary
Cystobiliary fistula (CBF) is a clinically important complication of hepatic hydatid disease (HHD). It may be detected preoperatively or intraoperatively, whereas occult cystobiliary communication may become clinically apparent after surgery. Early identification of at-risk patients may facilitate perioperative assessment and management. We investigated risk factors associated with CBF in patients undergoing surgery for HHD. This multicenter retrospective study included patients who underwent surgical treatment for hepatic cystic echinococcosis (CE) at two hospitals from 1 December 2018 to 31 December 2024. All included patients had hepatic CE; we excluded patients with alveolar echinococcosis. We defined CBF as an abnormal communication between a hepatic hydatid cyst and the biliary tract based on intraoperative, pathological, and other perioperative evidence supporting cysto-biliary communication. We assessed postoperative bile leakage according to the International Study Group of Liver Surgery (ISGLS) criteria, but this alone was insufficient to establish the diagnosis of CBF. We used univariable logistic regression to screen candidate factors associated with CBF. We entered variables with clinical relevance and/or a univariable P value < 0.10 into a multivariable logistic regression model based on complete-case data. We further assessed model performance using receiver operating characteristic (ROC) curve analysis, calibration analysis, and decision curve analysis (DCA). We included 1,303 patients with complete data for the primary model, and 725 (55.6%) met the study definition of CBF. In multivariable logistic regression, preoperative cholangitis (OR, 2.88; 95% CI, 2.26–3.67; P < 0.001), intraoperative blood loss (OR, 1.09 per 100 mL increase; 95% CI, 1.06–1.12; P < 0.001), open surgery (OR, 2.31; 95% CI, 1.66–3.23; P < 0.001), and albumin (OR, 0.88 per 5 g/L increase; 95% CI, 0.79–0.96; P = 0.012) were independently associated with CBF, whereas age and alkaline phosphatase (ALP) were not. The multivariable model had an AUC of 0.720 and a Brier score of 0.212. DCA showed an overall positive net benefit across a threshold probability range of approximately 0.05–0.80. Preoperative cholangitis, intraoperative blood loss, open surgery, and lower albumin were independently associated with CBF in patients undergoing surgery for HHD, whereas age and ALP were not significantly associated after adjustment. These findings should be interpreted as associations rather than causal relationships, particularly because the temporal relationship between cholangitis and preceding biliary interventions could not be reliably established in all patients.