End-to-end cervical esophagogastric anastomoses are associated with a higher number of strictures compared with end-to-side anastomoses

J Gastrointest Surg. 2013 May;17(5):872-6. doi: 10.1007/s11605-013-2159-8. Epub 2013 Feb 12.

Abstract

Background: Leakage and benign strictures occur frequently after esophagectomy. The objective of this study was to analyze the outcome of hand-sewn end-to-end versus end-to-side cervical esophagogastric anastomoses.

Methods: A series of 390 consecutive patients who underwent esophagectomy with gastric conduit reconstruction was analyzed.

Results: The end-to-end technique was performed in 112 (29 %) patients and the end-to-side in 278 (71 %) patients. Anastomotic leakage occurred in 20 (18 %) patients with an end-to-end anastomosis versus 58 (21 %) patients with an end-to-side anastomosis (p = 0.50). A higher incidence in anastomotic strictures was seen in end-to-end anastomoses (48 (43 %)) compared with end-to-side anastomoses (89 (32 %); p = 0.04). Moreover, a median of 11 (7-17) dilations was necessary in patients with a benign anastomotic stricture in the end-to-end group compared with four (2-8) dilations in patients with a benign anastomotic stricture in the end-to-end group (p < 0.036). After multivariate analysis, the difference in anastomotic leakage rates remained nonsignificant (p = 0.74), whereas anastomotic stricture rate and number of dilations were higher in the end-to-end group (p = 0.03 and p = 0.01, respectively).

Conclusion: The technique of anastomosis is not significantly related to anastomotic leakage rate. However, patients with end-to-end anastomoses develop postoperative strictures more frequently, requiring a higher number of dilations compared to end-to-side anastomoses.

MeSH terms

  • Aged
  • Anastomosis, Surgical / methods*
  • Anastomotic Leak / epidemiology*
  • Chi-Square Distribution
  • Esophageal Neoplasms / surgery*
  • Esophageal Stenosis / epidemiology*
  • Esophagectomy / methods*
  • Female
  • Humans
  • Incidence
  • Linear Models
  • Male
  • Middle Aged
  • Netherlands / epidemiology
  • Postoperative Complications / epidemiology*
  • Statistics, Nonparametric
  • Treatment Outcome