ABSTRACT
BACKGROUND AND AIMS: Malignant esophagorespiratory fistulas (MERFs) usually are managed by the placement of self-expandable metal stents (SEMSs) but with conflicting results. This study aimed to identify risk factors associated with clinical failure after SEMS placement for the treatment of MERFs. METHODS: This was a retrospective analysis of a prospectively maintained database used at a tertiary-care cancer hospital, with patients treated with SEMS placement for MERFs between January 2009 and February 2016. Logistic regression was used to identify predictive factors for clinical outcomes and to estimate the odds ratio (OR) and the 95% confidence interval (CI). The Kaplan-Meier method was used for survival analysis, and comparisons were made by using the log-rank test. RESULTS: A total of 71 patients (55 male, mean age 59 years) were included in the study, and 70 were considered for the final analysis (1 failed stent insertion). Clinical failure occurred in 44% of patients. An Eastern Cooperative Oncology Group (ECOG) performance status of 3 or 4 and fistula development during esophageal cancer treatment were associated with an increased risk of clinical failure. ECOG status of 3 or 4, pulmonary infection at the time of SEMS placement, and prior radiation therapy were predictive factors associated with lower overall survival. Dysphagia scores improved significantly 15 days after stent insertion. The overall stent-related adverse event rate was 30%. Stent migration and occlusion caused by tumor overgrowth were the most common adverse events. CONCLUSION: SEMS placement is a reasonable treatment option for MERFs; however, ECOG status of 3 or 4 and fistula development during esophageal cancer treatment may be independent predictors of clinical failure after stent placement.