Risk Factors and Consequences of Anastomotic Leakage After Esophagectomy for Cancer

Research output: Contribution to journalArticleAcademicpeer-review

31 Citations (Scopus)

Abstract

Background: Identifying predictors of anastomotic leakage can contribute to prevention of this common complication after esophagectomy. This study identified predictors for anastomotic leakage and assessed the influence of anastomotic leakage on short-term outcomes and long-term survival. Methods: A retrospective cohort study was conducted of consecutive patients who underwent esophagectomy in the Amsterdam University Medical Centers, location Amsterdam Medical Center, between 1993 and 2019. Multilevel logistic and Cox regression models were used to assess predictors for anastomotic leakage and survival, and an operation year-level random effects was considered for the unmeasured characteristics at year of operation. Results: Included were 1539 patients, and anastomotic leakage developed in 288 (19%). Predictors for developing anastomotic leakage after a transthoracic esophagectomy were a higher body mass index and a cervical anastomosis. Diabetes mellitus type 2 and chronic obstructive pulmonary disease were predictors for anastomotic leakage after a transhiatal esophagectomy. Median intensive care unit and hospital stay was longer for patients with anastomotic leakage than for patients without anastomotic leakage (both P < .001 for transthoracic esophagectomy, P = .010 and P < .001, respectively, for transhiatal esophagectomy). A higher percentage of patients with anastomotic leakage died within 30 days (3.8% vs 1.9%, P = .050). However, anastomotic leakage did not significantly influence long-term survival (hazard ratio, 0.994; 95% CI, 0.849-1.176; P = .994). Conclusions: Higher body mass index, cervical anastomosis, diabetes mellitus, and chronic obstructive pulmonary disease are predictors for anastomotic leakage after esophagectomy. Anastomotic leakage is associated with worse short-term outcomes, but long-term survival was not influenced. Future studies should focus on patient optimization, accurate patient selection, and development of tools in risk assessment.

Original languageEnglish
Pages (from-to)255-263
Number of pages9
JournalAnnals of Thoracic Surgery
Volume112
Issue number1
Early online date2021
DOIs
Publication statusPublished - Jul 2021

Cite this