To determine if a discrepancy exists between subjective symptoms and the grade of endoscopic gastroesophageal reflux disease (GERD) in diabetes mellitus (DM) patients. All 2,884 patients who underwent esophagogastroduodenoscopy completed the modified Gastrointestinal Symptom Rating Scale (GSRS), an interview-based rating scale consisting of 16 items including a question on acid regurgitation. Patients were divided into DM and non-DM groups (1,135 and 1,749 patients, respectively). GERD was diagnosed endoscopically and graded according to the Los Angeles classification. Grade B or more severe GERD was defined as severe endoscopic GERD. The intergroup GSRS score was compared statistically. In severe endoscopic GERD patients, the prevalence of patients with a positive GSRS score in the acid regurgitation question was statistically lower in DM patients than non-DM patients. Of the 60 non-DM patients with severe endoscopic GERD, 40 patients (67%) had a positive GSRS score for acid regurgitation; however, of the 51 DM patients with severe endoscopic GERD, 23 patients (45%) had a positive GSRS score. Multivariate analysis showed that severe endoscopic GERD (OR: 2.01; 95% CI: 1.21-3.33; p = 0.0066), non-DM (OR: 0.74; 95% CI: 0.54-0.94; p = 0.0157), younger age (OR: 0.98; 95% CI: 0.97-0.99; p = 0.0125), and hiatal hernia (OR: 1.46; 95% CI: 1.12-1.90; p = 0.0042) were associated with acid regurgitation symptoms. There is a discrepancy between subjective symptoms and endoscopic GERD grade in DM patients. The ability of DM patients to feel acid regurgitation may be decreased.