To the Editor:--End-tidal carbon dioxide (ETCO2) monitoring can be used to detect air embolism, circuit disconnection, endotracheal tube kinking, and rebreathing. Recently, capnography assisted in the diagnosis of a tracheoesophageal fistula.
A 1-month-old, 3.2-kg infant was admitted with a diagnosis of gastroesophageal reflux. Increased temperature, respiratory rate, and leukocyte count and choking and brief cyanotic spells after feeding suggested aspiration pneumonia. Medical history included uncorrected cleft lip and palate and patent ductus arteriosus, patent foramen ovale, bilateral SVC, and right atrial and ventricular enlargement.
The infant was brought to the operating room for a Nissen fundoplication. Rapid sequence induction and intubation were done without positive-pressure ventilation by mask. Ventilation was begun via a 3.0 oral endotracheal tube, secured with the tip 9 cm at the alveolar ridge.
A central venous catheter was inserted. A chest x-ray showed the tip of the endotracheal tube to be midway between clavicles and carina. The incision was made, and after entering the abdomen, the surgeon noted air in the stomach. An oral-gastric suction catheter (position confirmed by surgeon) briefly emptied the stomach, but the stomach would refill. Ventilation peak pressure was reduced to 15 cmH2O out of concern that air leaked around the endotracheal tube, accumulated in the pharynx, and then moved down the esophagus into the stomach.
It was then noted that, when the oral-gastric tube was suctioned, the previously square ETCO2capnogram waveform changed to a reduced, rounded form (Figure 1). A presumption of tracheoesophageal fistula was made. This was confirmed by bronchoscopy and esophagoscopy, which showed an H-type connection about halfway between the anterior esophagus and posterior trachea.
We assumed that suctioning the oral-gastric tube drew air from the lungs, via the tracheoesophageal fistula, into the esophagus and stomach. The result was a reduction of expired carbon dioxide, shown by the changed ETCO2capnogram. Gastric dilitation may have been caused by inspiratory pressure and/or partial intubation of the fistula itself. 
Tom M. Fazlollah, C.R.N.A., Steven R. Tosone, M.D., Department of Anesthesiology, Egleston Children's Hospital, 1405 Clifton Road NE, Atlanta, Georgia 30322-1101.
(Accepted for publication July 7, 1995.)