Barrett's esophagus: diagnosis by double-contrast esophagography.

A blinded, retrospective study was performed to determine the role of double-contrast esophagography in diagnosing Barrett's esophagus. The study group consisted of 200 patients who had double-contrast esophagrams and endoscopy because of severe reflux symptoms. The radiographs were reviewed by two gastrointestinal radiologists who had no knowledge of the endoscopic findings. Patients were classified as being at high risk for Barrett's esophagus if the radiographs revealed a high stricture or ulcer or a reticular mucosal pattern; at moderate risk if the radiographs revealed a distal peptic stricture and/or reflux esophagitis; and at low risk if none of the aforementioned findings were present. When these radiologic criteria were used, 10 patients (5%) were thought to be at high risk, 73 (37%) at moderate risk, and 117 (58%) at low risk for Barrett's esophagus. Endoscopic correlation revealed biopsy-proved Barrett's mucosa in nine (90%) of 10 patients at high risk, in 12 (16%) of 73 at moderate risk, and in only one (1%) of 117 at low risk for Barrett's esophagus. Thus, endoscopy is clearly indicated for patients in the high-risk group. Because of the lower prevalence of Barrett's esophagus in the moderate-risk group, clinical judgment should be used in deciding when to perform endoscopy in these patients. However, most patients were in the low-risk group, and the prevalence of Barrett's esophagus was so low in this group that endoscopy does not appear to be warranted. Thus, the major value of double-contrast esophagography is its ability to separate patients into high-, moderate-, and low-risk groups for Barrett's esophagus to determine the relative need for endoscopy and biopsy.