Prevention of delayed gastric emptying after pylorus-preserving pancreatoduodenectomy with antecolic reconstruction, a long jejunal loop, and a jejuno-jejunostomy.

backgroundDelayed gastric emptying (DGE) is one of the major complications following pylorus-preserving pancreatoduodenectomy (PPPD). It leads to significant patient distress and prolonged hospitalization and therefore increased treatment costs. DGE etiology remains unclear but seems to be multifactorial. In order to decrease DGE rates, reconstruction methods have been modified. The presented retrospective study was to evaluate outcomes of different surgical techniques at our institution with special emphasis on retrocolic and antecolic reconstruction types. MATERIAL AND

methodsOne hundred thirteen consecutive patients underwent PPPD between September 2004 and December 2011 for periampullary and bile duct lesions of the pancreatic head and the papilla of Vater. These patients were reviewed for DGE occurrence and other factors. Four different types of reconstruction were applied: the classic retrocolic reconstruction using a short jejunal loop (short loop, n = 40) and three types of reconstructions using a long loop: one with a long loop and retrocolic duodenojejunostomy (n = 22), another with a long loop and an additional latero-lateral enterostomy (Braun's anastomosis, n = 23), and finally, an antecolic group with Braun's anastomosis (n = 28). Patients were reviewed for DGE incidence and severity following the International Study Group of Pancreatic Surgery definition of DGE.

resultsThe highest DGE occurrence was noted in the retrocolic group using a short jejunal loop (15 of 32 patients, 46.9%), whereas the reconstruction types using long loops showed a notable decrease: DGE occurred in 4 of 16 patients (25%) in the retrocolic group, in 6 of 21 patients (28.6%) in the retrocolic group with an additional latero-lateral enterostomy (Braun's anastomosis), and finally, only 1 of 22 patients (4.5%, p = 0.009) in the antecolic group with Braun's anastomosis presenting with DGE, grade A. However, neither hospitalization time nor days in the intensive care unit were significantly different. There was no difference in DGE rates between the retrocolic long-loop groups with and without Braun's anastomosis.

conclusionThe results of this retrospective study suggest that the antecolic route with a long jejunal loop and Braun's anastomosis minimizes DGE rates.