Purpose: Few studies have addressed the clinicopathologic aspects of heterotopic gastric mucosa in the proximal esophagus (“inlet patch”). Reliable prevalence data are not available due to the small size of published series. This study was designed to determine the prevalence of the inlet patch and its clinical and histopathologic associations in a large nationwide series of patients.
Methods: Using the database of a gastrointestinal pathology reference laboratory, we analyzed the histopathologic reports of esophageal biopsies from patients who had an EGD in private US endoscopy centers from 1/2008 through 12/2009. We then extracted demographic, clinical, and endoscopic data from all patients who had a histopathologic diagnosis of inlet patch.
Results: There were 173, 809 unique patients with esophageal biopsies (median age 57 years, range 1-104; 47.1% male). Inlet patch was diagnosed in 569 patients (median age 56 years, range 8-87; 53.4% male). Thus, the prevalence was 0.33%. The most common indications for EGD in patients with an inlet patch were GERD (53.2%, vs. 60.8% in patients without inlet patch, OR 0.73, 95%CI 0.63-0.87) and dysphagia/odynophagia (19.6%, vs. 21.9% in patients without inlet patch, not significant). Endoscopically, inlet patch was suspected in 266 (48%) of the 569 patients who had one confirmed histologically, versus 232 (0.2%) in those who did not. Both hiatal hernia (19.0% vs. 27.1%; OR 0.63, 95% CI 0.51-0.78) and Barrett's mucosa (14.4% vs. 20.1%, OR 0.67,95%CI 0.53-0.84) were less common in patients with an inlet patch. No significant associations were detected with other esophageal, gastric, or duodenal pathologies, including gastric heterotopia in the duodenum. H. Pylori gastritis was present in 22 patients with an inlet patch and 5 of them had organisms in the heterotopic tissue. Intestinal-type epithelium within the inlet patch was found in 8 patients (1.4%), none of whom showed dysplastic changes.
Conclusion: The prevalence of inlet patch in this series (0.3%) was at the lower end of previously reported values (0.1%-10%). There was a strong association between the endoscopic suspicion of an inlet patch and its histopathologic confirmation (positive predictive value 0.53). No clinically relevant associations were detected. Since there was no relationship with Barrett's esophagus and no dysplasia was detected, surveillance for these patients cannot be recommended. However, the rarity of intestinal metaplasia in our series (˜1%) may have limited our ability to detect dysplasia. Since intestinal-type epithelium in the lower esophagus (Barrett's mucosa) is a precursor of dysplasia and carcinoma, it may be prudent to follow up patients in whom intestinal-type epithelium is diagnosed in an inlet patch.