OBJECTIVES: The aim of this study was to determine the benefit of routine postoperative chest radiography after removal of esophageal foreign bodies in children. METHODS: Medical records were reviewed of all patients evaluated with an esophageal foreign body at a single children's hospital over 10 years. Operative records and imaging reports were reviewed for evidence of esophageal injury. RESULTS: Of 803 records identified, 690 were included. All underwent rigid esophagoscopy and foreign body removal. The most common items removed were coins (94%), food boluses (3%), and batteries (2%). The rate of esophageal injury was 1.3% (9 patients). No injuries were identified on chest radiographs done as routine or for concern of injury. Patients with operative findings suggestive of an esophageal injury (n = 105) were significantly more likely to have an injury (8.6% vs 0%, P = .0001). Of the 585 children who did not have physical evidence of injury, 40% (n = 235) received a routine chest radiograph. Regardless of the indication, no injuries were identified on chest films. CONCLUSIONS: We conclude that intraoperative findings during rigid esophagoscopy suggestive of an injury are predictive of esophageal perforation. Routine chest radiography is not warranted in children who do not meet this criterion. In patients with a concern for injury, we suggest that chest radiography should be deferred in favor of esophagram.
OBJECTIVES: The aim of this study was to determine the benefit of routine postoperative chest radiography after removal of esophageal foreign bodies in children. METHODS: Medical records were reviewed of all patients evaluated with an esophageal foreign body at a single children's hospital over 10 years. Operative records and imaging reports were reviewed for evidence of esophageal injury. RESULTS: Of 803 records identified, 690 were included. All underwent rigid esophagoscopy and foreign body removal. The most common items removed were coins (94%), food boluses (3%), and batteries (2%). The rate of esophageal injury was 1.3% (9 patients). No injuries were identified on chest radiographs done as routine or for concern of injury. Patients with operative findings suggestive of an esophageal injury (n = 105) were significantly more likely to have an injury (8.6% vs 0%, P = .0001). Of the 585 children who did not have physical evidence of injury, 40% (n = 235) received a routine chest radiograph. Regardless of the indication, no injuries were identified on chest films. CONCLUSIONS: We conclude that intraoperative findings during rigid esophagoscopy suggestive of an injury are predictive of esophageal perforation. Routine chest radiography is not warranted in children who do not meet this criterion. In patients with a concern for injury, we suggest that chest radiography should be deferred in favor of esophagram.