Comparison of laparoscopic choledochotomy closure techniques

Background: Laparoscopic common bile duct exploration (CBDE) has traditionally been accompanied by T-tube drainage. However, other methods of choledochotomy closures have been reported. This study compared three laparoscopic methods of choledochotomy closure in a prospective, randomized fashion to determine which method should be the preferred technique. Methods: In this porcine model, 24 animals initially underwent laparoscopic common bile duct (CBD) clipping to simulate an obstruction. Two days later, the animals underwent laparoscopic clip removal and simulated CBDE through a 1.5-cm choledochotomy. The animals were then randomized to one of three groups: primary choledochotomy closure (group I), antegrade CBD stenting with primary closure (group II), or T-tube placement (group III). To assess for CBD stenoses and leaks, the animals were killed 2 months postoperatively, at which time a cholangiogram was performed and the bile duct harvested, The ratio of proximal CBD to choledochotomy site was assessed radiographically and histologically. Results: The operative time was significantly longer in group III (200 +/- 13 min, p < 0.05) than in group I (141 +/- 17 min) and group II (154 +/- 16 min). The ratio of the proximal CBD diameter to the choledochotomy site diameter by cholangiogram was 2.1:1.0 in group I, to 1.2:1.0 in group II, and 1.1:1.0 in group III (p < 0.01). The ratio of the proximal CBD intraluminal area to the choledochotomy site intraluminal area was 2.1:1.0 in group I compared to 1.1:1.0 in groups II and III (p < 0.01). None of the animals developed jaundice or sepsis. Conclusion: Significant stenoses were present at the choledochotomy site in the primary closure group, and T-tube placement resulted in prolonged operative times. We conclude that laparoscopic antegrade CBD stenting with primary closure of the choledochotomy site is the preferred technique after choledochotomy in an animal model. Further assessment in a clinical trial is warranted.