Mortality and postoperative complications after different types of surgical reconstruction following pancreaticoduodenectomya systematic review with meta-analysis

摘要

BackgroundPancreaticoduodenectomy/PD is a technically demanding pancreatic resection. Options of surgical reconstruction include (1) the child reconstruction defined as pancreatojejunostomy/PJ followed by hepaticojejunostomy/HJ and the gastrojejunostomy/GJ the standard/s-Child, (2) the s-child reconstruction with an additional Braun enteroenterostomy BE-Child, or (3) Isolated-Roux-En-Y-pancreaticojejunostomy Iso-Roux-En-Y, in which the pancreas anastomosis is reconstructed in a separate loop after the GJ. Yet, the impact of these reconstruction methods on patients' outcome has not been sufficiently compared in a systematic manner.MethodsA systematic review and meta-analysis were conducted according to the Preferred-Reporting-Items-for-Systematic-review-and-Meta-Analysis/PRISMA-guidelines by screening Pubmed/Medline, Scopus, Cochrane Library and Web-of-Science. Articles meeting predefined criteria were extracted and meta-analysis was performed.ResultsNineteen studies were identified comparing BE-Child or Isolated-Roux-En-Y vs. s-Child. Compared to s-Child neither BE-Child (p=0.43) nor Iso-Roux-En-Y (p=0.94) displayed an impact on postoperative mortality, whereas BE-Child showed less postoperative complications (p=0.02). BE-Child (p=0.15) and Iso-Roux-En-Y (p=0.61) did not affect postoperative pancreatic fistula/POPF in general, but BE-Child was associated with a decrease of clinically relevant POPF (p=0.005), clinically relevant delayed gastric emptying/DGE B/C (p=0.004), bile leaks (p=0.01), and hospital stay (p=0.06). BE-Child entailed also an increased operation time (p=0.0002) with no impact on DGE A/B/C, hemorrhage, surgical site infections and pulmonary complications.ConclusionBE-Child is associated with a decreased risk for postoperative complications, particularly a decreased risk for clinically relevant DGE, POPF, and bile leaks, whereas Iso-Roux-En-Y does not seem to affect the clinical course after PD. Therefore, BE seems to be a valuable surgical method to improve patients' outcome after PD.