EComment. Oesophagectomy: Could the anastomotic location be an independent prognostic factor?

Abstract

Despite vast advancements in surgical technique, oesophagectomy maintains a relatively high mortality and peri-operative morbidity rates. The systematic review by Kayani et al. attempted to investigate the effect of anastomosis location on post-operative outcomes and found only nine related papers from 1950 to 2011 [1]. About 40% of post-operative deaths are related to anastomotic leaks [2]. Studies comparing the outcome of cervical and intra-thoracic anastomoses following ooesophageal resection are small in size and poorly standardized with respect to surgical approach and anastomotic technique. The purpose of this letter is to put our experience into this clinical context. In our recent series of unpublished 63 oesophagectomies for cancer, 46 were performed with a cervical semi-mechanical anastomosis (group A) and 17 with intra-thoracic anastomosis by circular stapler (group B). In all cases, the alimentary tract was reconstructed through the transposition of a gastric tubule with the exception of five cases, in which the left colon was used in an anti-peristaltic way. The dehiscence rate was 11% in group A and 8% in group B (p Log Rank 0.11). Overall 30-day mortality was 14.5%. In group A, 24.5% of patients with anastomotic leak died before the 30th post operative day (POD), whilst this percentage topped at 60% in group B. Our findings are not in complete agreement with the conclusions of Kayani et al. [1] and suggest that a transhiatal oesophagectomy with a cervical anastomosis should be performed, since cervical fistulas are more easily treated conservatively, with good results of early endoscopic approaches. Furthermore, it is subject to a lower mortality rate within the 30th POD. We think that our data could improve the debate on better surgical approaches for oesophageal cancer treatment.