The effect of an empirical trial of high-dose lansoprazole on symptom response of patients with non-cardiac chest pain--a randomized, double-blind, placebo-controlled, crossover trial

Abstract

Background: Empirical trial with high-dose omeprazole has been shown to be a sensitive tool for diagnosing patients with gastro-oesophageal reflux disease-related non-cardiac chest pain.

Aim: To determine the clinical value of an empirical trial of high-dose lansoprazole in detecting patients with gastro-oesophageal reflux disease-related non-cardiac chest pain.

Methods: Patients who were referred by a cardiologist after a comprehensive evaluation, with at least three episodes per week of unexplained chest pain as the predominant symptom, were enrolled into the study. Oesophageal mucosal disease was determined by upper endoscopy followed by 24-h oesophageal pH monitoring to assess acid exposure. Patients were then randomized to either placebo or lansoprazole 60 mg am and 30 mg pm for 7 days. After a washout period of 1 week, patients crossed over to the other arm of the study for an additional 7 days. Patients completed a daily diary assessing severity and frequency of chest pain as the predominant symptom throughout the baseline treatment and washout periods. The lansoprazole empirical trial was considered diagnostic if chest pain score improved > or =50% than baseline.

Results: Of the 40 patients with non-cardiac chest pain that were enrolled, 18 (45%) had erosive oesophagitis and/or abnormal pH test (gastro-oesophageal reflux disease-positive) and 22 (55%) had both tests negative (gastro-oesophageal reflux disease-negative). Of the gastro-oesophageal reflux disease-positive patients, 14 (78%) had significantly higher symptom improvement on lansoprazole than on placebo (22%) (P = 0.0143). Of the gastro-oesophageal reflux disease-negative group, two (9.1%) markedly improved on the medication and eight (36.3%) on placebo (P = 0.75). The sensitivity and specificity of the lansoprazole empirical trial was 78 and 80%, respectively. By day 2, 12 (85.7%) of the gastro-oesophageal reflux disease-related non-cardiac chest pain responders had either complete or almost complete symptom resolution.

Conclusions: The lansoprazole empirical trial is highly sensitive and specific for diagnosing gastro-oesophageal reflux disease-related non-cardiac chest pain patients. The trial enables diagnosing most of the responders within the first 2 days and thus a shorter duration of therapy may be considered in a subset of non-cardiac chest pain patients.