Intake of coffee, tea, or soda is associated with an increased risk of gastroesophageal reflux (GER) symptoms, researchers report in the September issue of Clinical Gastroenterology and Hepatology. In contrast, consumption of water, juice, or milk is not associated with GER symptoms. Drinking water instead of coffee, tea, or soda reduces the risk of GER symptoms.
GER, characterized by symptoms of heartburn and acid regurgitation, affects more than 30% of the population in the United States each week and accounts for at least $18 billion in direct health care expenditures. GER reduces work time and health-related quality of life, and has been linked to complications such as peptic stricture and esophageal adenocarcinoma.
Beverage choice and frequency of intake have been associated with GER symptoms, but studies have found conflicting results and have methodological weaknesses. Raaj S. Mehta et al therefore investigated the association between coffee, tea, and soda intake and risk of GER symptoms among women enrolled in a large prospective cohort, the Nurses’ Health Study II. The cohort comprised 48,308 women, 42–62 years old, who were free of regular GER symptoms, without cancer, and not taking proton pump inhibitors or H2 receptor agonists. In 262,641 person-years of follow up, Mehta identified 7961 women who reported symptoms of GER once or more per week.
Overall, coffee, tea, and soda consumption were all associated with an increased risk of GER symptoms. After multivariable adjustment, hazard ratios (HRs) for women with the highest intake of each beverage (more than 6 servings/day) compared to women with the lowest intake (0 servings/day) were 1.34 for coffee (95% CI, 1.13–1.59; Ptrend < .0001), 1.26 for tea (95% CI, 1.03–1.55; Ptrend < .001), and 1.29 for soda (95% CI, 1.05–1.58; Ptrend < .0001).
In a substitution analysis, replacement of 2 servings/day of coffee, tea, or soda with 2 servings of water was associated with reduced risk of reflux symptoms: coffee HR, 0.96 (95% CI, 0.92–1.00); tea HR, 0.96 (95% CI, 0.92–1.00); and soda HR, 0.92 (95% CI, 0.89– 0.96). No meaningful interactions were identified between coffee, tea, or soda intake and GER risk factors such as including smoking, alcohol consumption, use of medications that may decrease lower esophageal sphincter tone, diabetes, or other dietary habits.
These findings indicate that risk of GER symptoms is beverage specific and independent of the number of servings of a beverage per day, and challenge the assumption that citrus juice is associated with risk of GER symptoms.
How could coffee, tea, and soda lead to GER symptoms? These beverages might cause hypotension of the lower esophageal sphincter. Caffeine has been reported to reduce lower esophageal sphincter tone. However, effects of coffee and soda did not differ significantly based on caffeine status, and there was a surprising stronger association between decaffeinated tea and risk for GER symptoms. So GER symptoms might be attributed to other unknown components in these beverages. Peppermint, a common ingredient in herbal decaffeinated teas, is thought to relax smooth muscle in the gastrointestinal tract, including the lower esophageal sphincter. Also, intake of carbonated beverages leads to the lowest intraesophageal pH, which might cause GER symptoms.
Strengths of this study include prospective collection of dietary data, minimizing recall bias, and that these data were well-validated and detailed enough to provide a range in servings per day of each beverage. The authors were able to assess incident GER symptoms by excluding those with GER symptoms before the start of the study and collect detailed data on potential confounders, including body mass index, use of postmenopausal hormones, and smoking. The authors were able to mutually adjust for other beverages to determine more clearly the specific effect of each beverage on the risk of GER symptoms. This study was the largest of its kind and is population based rather than hospital or clinic based.
A limitation of this study included use of a questionnaire to define GER symptoms rather than invasive tests, such as pH measurement or upper endoscopy. The authors admit they might have captured cases of functional heartburn or pathologic acid reflux disease. Also, dietary data were obtained from food frequency questionnaires, which do not capture all nutritional parameters related to acid reflux symptoms, including portion size or timing of meals in relation to bedtime. However, the study is unique in that it collected data on most potential dietary and lifestyle confounders, in a large population.
The authors suggest that clinicians consider advising patients with heartburn symptoms to try drinking more water in place of coffee, tea, and soda—especially those who consume 4 or more servings per day. Although these dietary modifications alone might not be sufficient to prevent GER symptoms, they could be effective when combined with other important lifestyle interventions, such as maintaining a healthy weight and not smoking.