Pancreaticojejunostomy versus pancreaticogastrostomy after pancreaticoduodenectomy: a systematic review and meta-analysis.

BACKGROUND AND OBJECTIVES The method of pancreatic reconstruction after pancreaticoduodenectomy (PD) is closely associated with postoperative morbidity, mortality, and patient's quality of life. The objective of this study is to evaluate which anastomosis approach - pancreaticogastrostomy (PG) or pancreaticojejunostomy (PJ) is a better option of choice in terms of postoperative complications. METHODS Articles comparing PG and PJ that were published by July 2011 were retrieved and subjected to a systematic review and meta-analysis. RESULTS Four randomized controlled trials (RCTs) and 22 observational clinical studies (OCSs) were included. RCTs showed that the PG group had significantly lower incidence rates of postoperative intra-abdominal fluid collection (p = 0.003, relative risk (RR) 0.50, 95% CI 0.31-0.79) and multiple intra-abdominal complications (p = 0.0007, RR 0.26, 95% CI 0.12-0.56) than the PJ group. OCSs demonstrated significant differences between PG and PJ in terms of frequencies of postoperative biliary fistula, intra-abdominal fluid collection, pancreatic fistula, morbidity, and mortality. The overall analysis revealed significant differences in frequencies of intra-luminal hemorrhage (p = 0.03, OR 2.82, 95% CI 1.08-7.33) and grade B/C pancreatic fistula (p = 0.002, OR 0.42, 95% CI 0.24-0.73) between the two groups. CONCLUSIONS Current literature has no adequate evidence to prove that PG is superior to PJ for patients undergoing PD in terms of postoperative complications. A standardized classification of pancreatic fistula and other intra-abdominal complications may enable an objective, valid comparison between PG and PJ.