Abstract Background In pancreatic surgery, morbidity and mortality remain high. Major complications include postoperative pancreatic fistula (POPF) and postpancreatectomy haemorrhage (PPH), among others. One of the few modifiable factors is the reconstruction technique used to anastomose the pancreatic remnant to the gastrointestinal tract. Aims To compare the outcomes of pancreaticogastrostomy (PG) and pancreaticojejunostomy (PJ) after pancreatoduodenectomy. Methods A systematic literature search was performed in PubMed, Web of Science, and the Cochrane Central Register of Controlled Trials to identify randomised controlled trials (RCTs) comparing PG with PJ. Outcomes included mortality, overall complications, POPF, PPH, operation time, and length of hospital stay. Pooled estimates were calculated using a random-effects model. Risk of bias was assessed using the Cochrane RoB 2.0 tool. Certainty of evidence for each outcome was subsequently rated according to the GRADE approach. Results Thirteen RCTs including 2030 patients were included. There was no difference in mortality (OR 0.98, 95%-CI: 0.61 to 1.59, p=0.94, moderate certainty of evidence) or overall complications (OR 1.13, 95%-CI: 0.70 to 1.80, p=0.62, low certainty of evidence). POPF was less frequent after PG (OR 0.69, 95%-CI: 0.50 to 0.95, p=0.02, very low certainty of evidence). In contrast, PPH occurred less frequently after PJ (OR 1.52, 95%-CI: 1.13 to 2.05, p0.01, low certainty of evidence). There was no difference in operation time (MD 4.2 min, 95%-CI: −4.8 to 13.1, p=0.36, low certainty of evidence) or length of hospital stay (MD 4.0 days, 95%-CI: −2.1 to 10.1, p=0.20, very low certainty of evidence). Conclusion Pancreaticogastrostomy and pancreaticojejunostomy yield overall comparable outcomes after pancreatoduodenectomy. Given the similar mortality and overall complication rates, both techniques can be considered equivalent options and should be performed according to surgeon expertise and institutional preference.
Gllareva et al. (Mon,) studied this question.