Initial management and outcomes of nonmalignant esophageal perforations: A Nationwide Inpatient Sample analysis

Abstract

BackgroundManagement of esophageal perforation includes open surgery, minimally invasive surgery, and endoscopic stent placement. This study analyzed initial treatment and the associated short-term outcomes. MethodsA retrospective study using the National Inpatient Sample between October 2015 and December 2019 identified adults >18 years with esophageal perforation undergoing an initial nonelective esophageal procedure categorized into either open surgery, minimally invasive surgery, or endoscopic stent placement. Patients with esophageal cancer were excluded. Baseline characteristics and the van Walraven–weighted Elixhauser Comorbidity Index were identified. Outcomes included in-hospital mortality and postintervention complications. Univariable and multivariable Cox regression was used to compare in-hospital survival. ResultsIn total, 3,345 patients met inclusion criteria: the median age was 62 years (interquartile range 50–72 years), and 1,310 (39%) were female. Open procedure was pursued in 2,650 (79%), minimally invasive surgery in 310 (9%), and endoscopic stent placement in 385 (12%) with no differences in van Walraven–weighted Elixhauser Comorbidity Index or mortality. Patients who underwent minimally invasive surgery had a greater proportion of gastrointestinal complications (P = .006); otherwise, there were no differences in postintervention complications. In total, 380 (11%) patients died and were significantly older, with greater van Walraven–weighted Elixhauser Comorbidity Index, and had more postintervention complications. Univariable Cox regression identified age (hazard ratio 1.95, P < .001), van Walraven–weighted Elixhauser Comorbidity Index (hazard ratio 1.06, P < .001), stent placement (hazard ratio 1.93, P = .045), and transfer from a health facility (HR 2.40, P = .049) as associated with decreased in-hospital survival. Multivariable Cox regression revealed age (hazard ratio 1.041, P < .001) and van Walraven–weighted Elixhauser comorbidity index (hazard ratio 1.055, P < .001) were associated with decreased in-hospital survival. ConclusionPatients with esophageal perforation had an 11% in-hospital mortality rate and significant associated complications regardless of intervention. Increasing age and comorbidities are associated with poorer in-hospital survival.