Different effects of dietary factors on reflux esophagitis and non-erosive reflux disease in 11,690 Korean subjects.

ABSTRACT

BACKGROUND: Although dietary factors seem to be associated with gastroesophageal reflux symptoms, their effects on reflux esophagitis and non-erosive reflux disease (NERD) are unclear. We evaluate dietary effects on NERD and reflux esophagitis. METHODS: A total of 11,690 health check-up persons completed questionnaires for reflux symptoms and 3-day recordings for dietary intake and underwent esophagogastroduodenoscopy from 2004 to 2008. Multiple logistic regression with odds ratio (OR) and 95% confidence interval (CI) was used to evaluate the relationship of dietary components with NERD or reflux esophagitis. RESULTS: Prevalence of NERD and reflux esophagitis was 7.7 and 7.2%, respectively. In adjusted analysis, highest quartile of beans (OR 0.78, 95% CI 0.64-0.95), 3rd quartile of vegetables (OR 0.74, 95% CI 0.60-0.91), 4th quartile of fruit (OR 0.78, 95% CI 0.64-0.95), 4th quartile of egg (OR 0.78, 95% CI 0.64-0.96), and 3rd quartile of fish (OR 0.80, 95% CI 0.66-0.98), and 4th quartile of milk (OR 0.78, 95% CI 0.65-0.94) reduced NERD. Reflux esophagitis had no association with food groups, whereas it was related with men, absence of H. pylori, hiatal hernia, BMI, and total energy intake. Furthermore, dietary effect on NERD was similar in men and women, whereas highest tertile of potato (OR 1.91) and milk (OR 1.87) increased reflux esophagitis only in women. CONCLUSIONS: While many food groups affected NERD, reflux esophagitis was associated with BMI and total energy intake rather than dietary component. These results may suggest different approaches toward dietary management of NERD and reflux esophagitis.