Gastroesophageal reflux disease (GERD), with its cardinal symptom, heartburn, is the most common disorder of the esophagus in the West. Comparatively, GERD is less common in Singapore but its frequency in the population is increasing. Although the vast majority of patients with GERD have heartburn and acid regurgitation, GERD can present in atypical ways, including as a non-cardiac chest pain (NCCP). We have previously shown that GERD is a common cause of NCCP in Singapore. Up to 40% of our patients with NCCP had endoscopic esophagitis, abnormal 24-hour pH monitoring results, and/or a positive acid perfusion test. These tests, although diagnostic, are costly, labour intensive, and not always readily available in the primary care setting. A trial of high-dose proton pump inhibitor (e.g. omeprazole 60 mg daily) has been proposed as a simple, safe, non-invasive and reliable means to diagnose GERD in Western patients with NCCP. We have not used the test routinely in our practice. This study will evaluate the use of a short course of esomeprazole, the S-isomer of omeprazole, as a diagnostic test for detecting GERD in patients with NCCP. The hypothesis is that in NCCP patients with GERD, esomeprazole will resolve their symptoms. Consecutive patients diagnosed with NCCP at the National University Hospital, Singapore, will be invited to participate in the study. Eligible patients will be randomly assigned to receive either esomeprazole (40 mg o d) for 14 days, or comparable dose of placebo at a similar schedule for 14 days, in a double-blinded fashion. At the start of the study, all subjects will complete a baseline symptom assessment. Symptoms will be scored on a graded scale based on severity. During the study weeks, each patient will record his/her own daily symptoms. The patient will be assessed again after the 14-day treatment. The primary outcome measure will be the change in symptom score after initiation of treatment.

Inclusion Criteria:

• patients with recurrent chest pain of more than three months duration with either (1) normal or non-obstructive coronary arteries (<50% luminal diameter narrowing), (2) normal dobutamine stress echocardiography or (3) a negative exercise electrocardiogram and a cardiologist's evaluation that symptoms are not cardiac in origin.

Exclusion Criteria:

• Patients will be excluded if they are < 18 or > 70 years old, are pregnant, have a medical contraindication for esomeprazole therapy, have already been empirically treated with an antireflux regimen, report a history of peptic ulcer disease or gastrointestinal surgery, or are unwilling or unable to provide informed consent. In addition, patients who are unable to fully complete all stages of the study will be excluded.