Hyperplastic (inflammatory) polyps (HPs) of the gastric corpus and antrum typically develop in association with chronic gastritis. However, little is known regarding the etiology, pathologic features, and natural history of HPs of the gastroesophageal junction (GEJ). We have noted, anecdotally, that GEJ HPs often occur in patients without gastric pathology. The aim of this study was to evaluate the clinical, pathologic, and outcome features of patients with HPs of the GEJ, and to compare the data with a control group of individuals with HPs in the gastric corpus or antrum. One hundred thirty-four consecutive polyps of the GEJ were identified by a 5-year search through the pathology files of a major tertiary-care hospital. Of these, 46 (36%) polyps from 46 patients met the pathologic criteria for HPs and formed the basis of this study. The 46 study patients, and their polyps, were evaluated for a wide variety of clinical, endoscopic, and pathologic features including outcome on follow-up endoscopy. The findings were compared with 46 HPs from 46 patients of the distal stomach (antrum or corpus) that were obtained randomly from the same 5-year period. Compared with patients with gastric antral or corpus HPs, patients with HPs of the GEJ were significantly younger in age (mean age, 55.9 y vs. 63.0 y; P=0.04). Pathologically, GEJ HPs showed a significantly higher rate of multilayered epithelium (P=0.06) and association with Barrett esophagus (BE) (P=0.0001) compared with distal gastric HPs. All BE-associated GEJ HPs were associated with either ultrashort (<1 cm) or short segment (1 to 3 cm) BE. All other pathologic variables, including intestinal metaplasia, were similar to those of distal gastric HPs. In a subanalysis, BE-associated GEJ HPs (33% of all GEJ HPs) showed a higher male to female ratio and a higher rate of intestinal metaplasia compared with all other HPs. Furthermore, none of the BE-associated GEJ HPs were associated with chronic active gastritis versus the non-BE-associated GEJ HPs, although this was not statistically significant. Only 1 HP (from the GEJ) from both the study and control groups was associated with a neoplasm (signet-ring cell carcinoma). On follow-up, 1 patient with a GEJ HP and 4 with distal gastric HPs developed recurrent HPs and none of the patients from either patient group developed dysplasia or carcinoma. In conclusion, unlike HPs of the gastric corpus or antrum, a significant proportion of HPs of the GEJ arise in association with BE and without gastric pathology. In patients with BE, the columnar-lined segment is often ultrashort, and thus, an HP may be the first clinical/endoscopic manifestation of that disorder.