Re: An Antecolic Roux-En-Y Type Reconstruction Decreased Delayed Gastric Emptying After Pylorus-Preserving Pancreatoduodenectomy by Murakami et al.

Abstract

Dear Editor: We read with great interest the article by Murakami et al. from the Department of Surgery, Hiroshima University, Japan in the June issue of the Journal of Gastrointestinal Surgery. The authors presented a retrospective series of 132 consecutive pylorus-preserving pancreatoduodenectomies performed at their institution between 1994 and 2006. All patients received a pancreatogastrostomy, but two different reconstruction methods to obtain the digestive continuity: either a retrocolic Billroth I type reconstruction (1994–2000) or an antecolic Roux-en-Y reconstruction (2001–2006). In a multivariate analysis, the reconstruction method was the only factor influencing the occurrence of delayed gastric emptying with a significant benefit for the antecolic reconstruction (81% versus 10%; P<0.03). To our knowledge, this is the first comparative study to clarify the beneficiary effect of an antecolic reconstruction method in patients with pancreatogastrostomy. A recent meta-analysis of three randomized controlled trials comparing pancreatojejunostomy with pancreatogastrostomy showed an overall comparable delayed gastric emptying (DGE) rate for both reconstruction techniques [15.8% versus 13.9%; OR 0.85 (0.50; 1.44), P=0.54]. However, the authors should clearly state why they have used an end-to-end reconstruction compared to the most commonly used method of an end-to-side gastrojejunostomy as described by Delcore et al. and why they have changed their operative strategy in 2001. The given rate of 81% delayed gastric emptying in the retrocolic group with the used definition of the need of a nasogastric tube ≥10 days or an inability to tolerate ≥14 days seems to be extremely high. Even in the randomized controlled trial by Tani et al., which has been terminated due to the fact that an interim analysis revealed a clear benefit for the antecolic reconstruction method, the retrocolic reconstruction showed a DGE rate of only 50% using a bit stronger definition for DGE. This resembles again the necessity of clear definitions and grading of DGE, as it has recently been proposed by the International Study Group of Pancreatic Surgery (ISGPS). Furthermore, the authors seem to have overlooked the previously published article byHartel et al. which has already clearly outlined the superiority of an antecolic compared to a retrocolic reconstruction following pancreatoduodenectomy.