Preoperative biliary drainage in patients performing pancreaticoduodenectomy : guidelines and real-life practice.

BACKGROUND AND AIM: Preoperative biliary drainage (PBD) in patients with pancreatic cancer remains debatable. The aim of this study was to analyse the indications for PBD in patients performing pancreaticoduodenectomy (PD) and to evaluate the impact of this procedure on postoperative outcome. METHODS: Observational retrospective cohort study of patients undergoing PD for pancreatic cancer. Clinical data and postoperative outcome, namely complications and 90-day mortality, were prospectively collected and compared between patients performing PBD or direct surgery (DS). RESULTS: Eighty-two patients were included: 40 underwent PBD and 42 performed DS. Major complications (27.5% vs 33.3%, P=0.156) and 90-day mortality (10% vs 16.7%, P=0.376) were similar between the two groups. There was a trend for higher mean total bilirubin in patients with PBD (P=0.073). The indication for PBD was suspicion of cholangitis/choledocholithiasis or need to perform neoadjuvant chemotherapy in 24 (60%) patients. In the remaining, elevated bilirubin was probably the only reason to perform PBD. Length of hospital stay was longer in PBD group (P=0.003). On multiple logistic regression, 90-day mortality was not related with preoperative bilirubin levels, biliary drainage or its indication, but solely with age (OR 1.15, 95%CI 1.05-1.31, P=0.008). CONCLUSIONS: PBD is often performed in patients undergoing PD without a formal indication, mainly due to high bilirubin levels. No increased morbidity/mortality was observed but length of hospital stay was prolonged in patients performing PBD.