BACKGROUND: Esophageal cancer is currently one of the high-risk malignant tumors worldwide, posing a serious threat to human health. This study aimed to analyse the causes of postoperative mortality and intrathoracic anastomotic leakage(IAL) after esophagectomy. METHODS: A retrospective analysis was conducted on 172 patients with esophageal cancer resection and focused on the preoperative and postoperative indicators. Cox regression analysis was performed to identify factors affected IAL and evaluated the potential factors on postoperative mortality. The Kaplan-Meier curve was applied to evaluate the effect of leakage on postoperative mortality after propensity score matching. RESULTS: Univariable and multivariable Cox regression analysis showed that infection and high BMI were significant risk factors for IAL, patients with BMI over 24 kg/m in IAL group was two times higher than that of the group without IAL (95% CI = 1.01-6.38; P = 0.048). When patients were infected, the hazard ratios(HRs) of anastomotic leakage was twice that of patients without infection (95% CI = 1.22-4.70; P = 0.011). On the other hand, IAL was a significant cause of postoperative mortality, the 40-day postoperative mortality rate in the leakage group was significantly higher than the non leakage group (28.95% in leakage group vs. 7.46% in non leakage group, P<0.01). After propensity score matching, IAL still significantly affected postoperative mortality. The total length of hospital stay of the leakage group was inevitably longer than that of the non leakage group (22.19 ± 10.79 vs. 15.27 ± 8.59). CONCLUSION: IAL was a significant cause of death in patients underwent esophageal cancer resection. Patients with high BMI over 24 kg/m and infection may be more prone to developing IAL after esophagectomy. IAL inevitably prolonged the length of hospital stay and increased postoperative mortality.