The Incidence of Adenocarcinoma in Barrett’s Esophagus


Synopsis: This study is consistent with other recent reports and confirms a greatly increased risk for adenocarcinoma among patients with Barrett’s esophagus.

Source: Drewitz DJ, et al. Am J Gastroenterol 1997;92: 212-221.

Between 1982 and 1995, drewitz et al followed 170 patients with Barrett’s esophagus. None of the patients had adenocarcinoma at initial endoscopy or within six months of entry. The average follow-up period was 57 months. Adenocarcinoma developed in four patients, a frequency of one per 200 patient years of follow-up. All four cases of carcinoma were in males who had long-standing heartburn, and the average age at the time of cancer diagnosis was 60 years.


Barrett’s esophagus is being increasingly recognized among patients with long-standing gastroesophageal reflux disease. It appears to be particularly common among middle-aged white males and is clearly associated with an increased risk for adenocarcinoma of the esophagus. The risk for carcinoma appears to be confined to patients who have histological evidence of intestinal metaplasia—some, indeed, now limit the definition of Barrett’s to those with this pathological type. What is disputed, and of considerable clinical importance, is the relative risk for carcinoma among patients with Barrett’s. Is it sufficiently high to merit frequent surveillance with endoscopy and biopsy in these patients? Are there particular patients with Barrett’s at increased risk for carcinoma? In previous studies, the risk for Barrett’s has varied from as high as one in 55 to as low as one in 441 patient years on follow-up. These highly variable figures have led to considerable discrepancies in recommendation for follow-up interval.

The importance of this particular study lies in the fact that it was prospective, the patient population was clearly defined, and the protocol for definition of Barrett’s and follow-up of the patients was rigidly standardized. While the incidence of new adenocarcinoma detected in this study was low—and lower than previously reported—it is consistent with some other recent reports and does confirm a greatly increased risk for adenocarcinoma among patients with Barrett’s. Of interest, all of the patients who developed cancer had been diagnosed with dysplasia before they were diagnosed with cancer. The average duration of time from the recognition of any grade of dysplasia until cancer diagnosis was 19 months. Esophagectomy for a cure was performed in two of the four patients. The other two were not considered surgical candidates and eventually died of esophageal cancer. These results would appear to support the concept of surveillance but leave the time interval unclear. At present, some advocate yearly endoscopy, while others recommended it every 2-4 years. Of concern, carcinoma was diagnosed in one patient with what is described as short segment Barrett’s (i.e., a Barrett’s segment of < 3 cm in length). We clearly need more of these prospective, carefully performed studies to clarify our approach to Barrett’s.


1. Para M, et al. Increasing incidence of adenocarcinoma of the esophagus and esophagogastric junction. Gastroenterol 1993;104:510-513.

2. Spechler SJ, et al. Prevalence of metaplasia at the gastroesophageal junction. Lancet 1994;34:1553-1536.