Systematic review and meta-analysis of technique for closure of the pancreatic remnant after distal pancreatectomy

Background: Appropriate closure of the pancreatic remnant after distal pancreatectomy is still debated. A variety of procedures have been recommended to reduce the frequency of pancreatic fistula. This review quantitatively compares the available techniques. Methods: Original articles and abstracts published up to the end of June 2004 were searched without language restriction in the Cochrane Controlled Trials Register, Medline and Embase. Three reviewers independently assessed each study's eligibility and quality, and extracted the data. A random effects model was performed using weighted odds ratios. Results: Only ten of 262 articles could be included, two randomized clinical trials and eight observational studies. Reported postoperative morbidity varied from 13.3 to 64 per cent. The primary outcome measure, pancreatic fistula rate, occurred within the range 0-60.9 per cent. Meta-analysis of the six studies comparing stapler versus hand-sutured closure showed a non-significant combined odds ratio for occurrence of a pancreatic fistula of 0.66 (95 per cent confidence interval 0.35 to 1.26, P = 0.21) in favour of stapler closure. Conclusions: The quality and quantity of information extracted from the available trials are insufficient to enable any firm conclusion to be drawn on the optimal surgical technique of pancreatic stump closure; there is a trend in favour of the stapling technique.