Empiric therapy for gastroesophageal reflux disease

Background: In the absence of highly specific symptoms and without esophageal erosions, long-term pH monitoring is necessary for diagnosing gastroesophageal reflux disease. This method, however, is not generally available. Objective: To determine whether gastroesophageal reflux disease can be diagnosed empirically by acid suppression in patients with normal results of endoscopy. Methods: We studied 33 consecutive outpatients with pathologic findings on pH monitoring who had symptoms compatible with gastroesophageal reflux disease and normal results of esophagogastroduodenoscopy, particularly a normal appearance of the esophageal mucosa. The severity of symptoms was graded on a visual analog scale from 1 to 10 by the patient. The patients were treated for at least 7 days with either ranitidine, 150 mg twice daily (patients 1 through 10), omeprazole, 40 mg/d (patients 11 through 21), or omeprazole, 40 mg twice daily (patients 22 through 33). A reassessment of symptoms and second pH monitoring were performed during the last day of treatment. Results: Omeprazole, 40 mg/d, significantly reduced the severity of symptoms from 7.1 (range, 4 to 9) to 3.7 (0 to 8) and the reflux measure mean acidity from 0.98 mmol/L (0.21 to 76 mmol/L) to 0.02 mmol/L (0 to 0.47 mmol/L). Omeprazole, 40 mg twice daily, significantly reduced the severity of symptoms from 6.8 (3 to 10) to 0.6 (0 to 2) and the mean acidity from 0.38 mmol/L (0.13 to 8.5 mmol/L) to 0.01 mmol/L (0 to 0.14 mmol/L). Both doses of omeprazole were superior to ranitidine, 150 mg twice daily. When a 75% reduction of symptoms was defined as positive, the "omeprazole test" with 40 mg twice daily had a sensitivity of 83.3%, whereas the sensitivity with 40 mg/d was only 27.2%. Conclusion: In practice, the diagnosis of gastroesophageal reflux disease can be ruled out if symptoms do not improve with a limited course of high-dose proton pump inhibitors. (Arch Intern Med. 1995;155:1808-1812)