An analysis of early postoperative complications following biliary reconstruction of major bile duct injuries using the Modified Accordion and Anatomic, Timing Of and Mechanism classifications

Abstract

Background: Few studies have reported patient outcome after surgical repair of bile duct injury using a standardized, validated classification system. This is the first analysis to investigate the correlation between the Anatomic, Timing Of and Mechanism classification of bile duct injury and severity of postoperative complications classified using the Modified Accordion Grading System. Methods: Patients undergoing index hepaticojejunostomy repair of bile duct injury in laparoscopic cholecystectomy at a tertiary referral center from 1993-2018 were included. Patient demographics, geographic distance from referral center, time to referral, Anatomic, Timing Of and Mechanism classification and highest Modified Accordion Grade complication were retrieved from a prospective database. The primary outcome was determined using correlation statistics to assess the relationship between level of injury and severity of postoperative complication. Results: One hundred and twenty-eight patients were included. There was no correlation between level of injury and severity of postoperative complication (rs(128) = –0.113, P = .203). Seventy (54.7%) patients had an injury less than 2 cm from the hepatic duct bifurcation and 52% of patients developed a postoperative complication, most mild to moderate in severity. Geographic distance resulted in substantial delays in referral (P < .001) but did not affect complication rate (P = .523). Conclusion: In this prospective analysis the short-term complication rate was higher than previous retrospective reports, but the distribution of the severity of complications and spectrum of injury type were similar. There was no correlation between severity of injury and postoperative complications. Geographic distance from referral center resulted in substantial differences in referral delay but had no statistically significant effect on outcome.