Because of the potential for complications from a coin lodged in the esophagus, prompt removal is usually recommended. Soprano and Mandl developed a decision analysis model to compare the clinical outcomes and costs of different approaches to removal of a coin in the esophagus.

The authors' decision tree had four limbs that represented four management strategies: (1) endoscopic removal under general anesthesia, (2) esophageal bougienage, (3) outpatient observation for 12 to 24 hours to determine if the coin passes into the stomach spontaneously and (4) inpatient observation for 12 to 24 hours to determine if the coin passes into the stomach spontaneously. The hypothetic patient was a child younger than 18 years who had swallowed a single coin within the past 24 hours, had no history of esophageal disease and no respiratory distress at the time of presentation. Bougienage was defined as advancement of a bougie dilator from the mouth to the stomach in an upright, nonsedated patient.

Data on the success and complication rates of the different strategies were obtained from studies reported in the medical literature. The Medline search uncovered six articles about endoscopic removal of esophageal coins; these articles included 362 patients. The procedure resulted in advancement of the coin to the stomach in all of the patients. Two studies of esophageal bougienage were found. The procedure advanced the coin into the stomach in all 77 patients.

Data from the authors' previous study of spontaneous passage of esophageal coins were used to determine the success rate with this approach. This study revealed a 28 percent rate of spontaneous passage of the coin from the esophagus into the stomach within 12 hours.

Analysis of the data revealed overall complication rates of 5.8 percent for endoscopic removal, 4.2 percent for each of the observation strategies and zero percent for esophageal bougienage. On sensitivity analysis, bougienage had a complication rate of 4.4 percent when the procedure was 94 percent successful.

Fees at the authors' hospital were used to calculate the costs of the four different strategies. Endoscopic removal was found to cost $3,297 and esophageal bougienage $382, for a difference of $2,915. The cost of outpatient observation was $2,439, and the cost of inpatient observation was $3,141.

The authors conclude that esophageal bougienage to advance an esophageal coin into the stomach is associated with the lowest complication rate and the lowest cost. Compared with the other strategies evaluated, use of bougienage would result in significant reductions in complications and costs. Compared with the complication rate and cost of endoscopic removal, outpatient observation to allow the coin to pass spontaneously would result in a reduction in complications and cost. The authors point out that patients who do not meet the criteria of their hypothetic patient (presentation within 24 hours of coin ingestion, no history of esophageal problems and no respiratory distress) are unlikely to pass the coin from the esophagus to the stomach. Endoscopic removal is the preferred approach in patients who do not meet the relatively strict criteria of their hypothetic case.