Improved survival in children with esophageal perforation

Objective: To analyze the cause, location, signs and symptoms, presence of underlying disease, time inter val to diagnosis, treatment, and morbidity and mortality in 24 children (19 boys and 5 girls) with esophageal perforation who were treated from 1975 to 1995. Design: Data were collected retrospectively from hospital and office records. Setting: A tertiary care children's hospital. Results: The average age at diagnosis was 58 months (range, 1 day to 19 years). Fourteen children had underlying esophageal disease (atresia, n=7 and gastroesophageal reflux, n=7). Iatrogenic perforations occurred in 17 children: 8 during dilatation, 5 during an antireflux procedure, 2 during endoscopy, and 2 after passage of a feeding tube. Trauma was the cause of perforation in 6 children. In 2 cases the cause was unknown. Perforation occurred in the thoracic esophagus in 12 cases, abdominal esophagus in 7, cervical esophagus in 5, and involved both the thoracic and abdominal esophagus in 1. Signs and symptoms included dysphagia (15 patients), dyspnea (14), fever (12), cyanosis (8), abdominal pain (6), chest pain (5), and subcutaneous emphysema (3). Management of esophageal perforation included nonoperative management (7 patients), drainage alone (1), primary closure (16), and resection and diversion (1). Two perforations occurred in 1 child. Complications occurred in 11 (44%) of the 25 cases and were more common after delayed diagnosis (73%). The average hospital stay was 20 days. There was I death (4%) attributed to esophageal perforation. Conclusions: Morbidity and mortality are directly related to delays in diagnosis and therapy. Most cases of esophageal perforation in children can be closed primarily and the esophagus salvaged despite delayed presentation. The mortality rate in children with esophageal perforation (4%) is significantly less than that for adults (25%-50%).