Accuracy in the diagnosis of short-segment Barrett's esophagus: the role of endoscopic experience.

BACKGROUND: The diagnosis of Barrett's esophagus is based on histologic demonstration of specialized intestinal metaplasia. Experience may be important in the endoscopic recognition of Barrett's esophagus, including in regard to appropriate procurement of biopsy specimens. The aim of this study was to assess factors that may influence accuracy in the diagnosis of short-segment Barrett's esophagus (SSB). METHODS: Endoscopy reports pertaining to procedures performed over a 1-year period that included esophageal biopsies because of suspected intestinal metaplasia were reviewed. Barrett's epithelium involving less than 2 cm of the distal esophagus was considered SSB; greater than 2 cm was considered long-segment (LSB). Endoscopists were regarded as "more experienced" if they had completed training more than 5 years earlier and "less experienced" if the time elapsed since the completion of training was less than 5 years. RESULTS: More and less experienced endoscopists both obtained esophageal biopsy specimens because of suspected Barrett's esophagus at the same rate (14%). Length of suspected Barrett's epithelium was not predicted by symptoms or demographic data. Endoscopically, patients with SSB had significantly fewer (64.2% vs. 90.8%) and smaller (2.9 +/- 0.1 vs. 3.5 +/- 0.2 cm) hiatal hernias compared with those with LSB (p < 0.05). Suspected SSB was histologically confirmed in 38.4% (True SSB), whereas LSB was confirmed in 75% (True SSB) (p < 0.05). More experienced endoscopists were significantly more likely to obtain histologic confirmation of SSB than less experienced endoscopists (48.6% vs. 29.5%; p = 0.02, nominal significance from univariate hypothesis testing; correction for multiple testing of data removed significance at the p < 0.05 level; OR = 2.26). CONCLUSION: With greater experience, an endoscopist is more likely to diagnose SSB. This may be due to more accurate procurement of adequate tissue samples, which thereby results in a greater yield of histopathologic diagnoses.

Prevalence of Barrett's esophagus in Hispanics is similar to Caucasians.

Barrett's esophagus (BE) is considered to be a disease of white males with a prevalence ranging from 0.5 to 4.0% in patients undergoing upper endoscopy (EGD) for any indication, and from 12 to 15% in patients with gastroesophageal reflux disease (GERD). The prevalence of BE in Hispanics is not known, but it is assumed to be lower. The aims of this study were to determine the prevalence of BE in Hispanic patients and to compare demographic and endoscopic characteristics with Caucasian patients with BE. Records of patients undergoing an EGD between October 1993 and October 1996 were retrospectively reviewed. Patients were included in the study only if they had columnar-appearing esophageal mucosa at endoscopy and intestinal metaplasia with Alcian blue-staining goblet cells on biopsy. An extensive chart review was performed in patients with BE. There were 75 new cases of BE discovered: 60 (80%) were Caucasians, 6 (8%) Hispanics, 1 (1.4%) Native American, and 8 (10.6%) patients with either unknown or unconfirmed ethnicity. Of the 75 patients, 74 male, and the mean age was 65 +/- 11.4 years (range 36-92 years). The prevalence of BE in Caucasians and Hispanics undergoing EGD for any reason was similar (5.3% and 3.8%, respectively, P = 0.563). The prevalence of BE in patients presenting with GERD symptoms was also similar between Caucasians and Hispanics (25% and 16%, respectively, P = 0.304). The two groups did not differ significantly with respect to age, symptoms, habits, or endoscopic findings. In conclusion, the prevalence of BE among Hispanic patients is similar to Caucasian patients, an unexpected finding.