Anastomotic Leak after Esophagectomy - Analysis of the STS General Thoracic Surgery Database.

Background: Anastomotic leak after esophagectomy is a major cause of morbidity and mortality. We sought to identify the prevalence of anastomotic leak, stratified by operative approach and disease etiology, as well as risk factors for leak.

Methods: A retrospective cohort analysis using the STS General Thoracic Surgery Database was conducted on patients who underwent esophagectomy with gastric reconstruction between 2009-2021. Baseline characteristics and postoperative outcomes were compared between patients who did and did not develop an anastomotic leak. Multivariable mixed effects logistic regression models identified risk factors for leak.

Results: Of 18,419 patients, 3,416 (19%) developed an anastomotic leak. Patients who leaked had more comorbidities including obesity and diabetes. There was no difference in leak based on disease etiology (p=0.435.) Patients with anastomotic leak had increased 30-day mortality (7% vs 4%, p<0.001), reoperation (58% vs 10%, p<0.001), and longer lengths of stay (18 vs 10 days, p<0.001). On multivariable analysis, obesity (OR 1.27[1.16-1.38], p<0.001), diabetes (OR 1.14[1.04-1.25], p=0.006), and smoking (OR 1.26[1.15-1.37], p<0.001) were independently predictive of anastomotic leak. Compared to an open 2-field, a transhiatal (OR 1.35[1.17-1.55], p<0.001) or 3-field esophagectomy (OR 1.46[1.25-1.70], p<0.001) was more likely to leak. A robotic approach was associated with an increased risk of leak (OR 1.28[1.03-1.08], p<0.001), however lost significance in a modern subgroup from 2018-2021.

Conclusions: Obesity, diabetes, smoking, pulmonary hypertension, and a cervical anastomosis are risk factors for anastomotic leak regardless of disease etiology. These important clinical risk factors identify an opportunity for modifiable risk reduction with aggressive medical optimization perioperatively.