Evaluation of unexplained chest pain by the gastroenterologist: a continuing dilemma.

Unexplained chest pain continues to be a common problem in clinical practice. When cardiac disease has been ruled out, the patient is often referred to the gastroenterologist for further testing, principally to rule in or rule out an esophageal etiology. The majority of investigators in this field believe that gastroesophageal reflux disease is the most common etiology that can be identified by esophageal testing in these patients. Motility abnormalities and provokable chest pain by agents such as edrophonium are seen in lower frequency. A response to empiric antisecretory therapy (proton pump inhibitor trial) has been the standard of practice in assessing and managing these patients. As esophagogastroduodenoscopy has in general been of low yield, 24-hour ambulatory esophageal pH monitoring has been used as the “test of choice” when diagnostic evaluation is needed. Unfortunately, despite multiple attempts to do so, agreement as to the value of ambulatory pH monitoring and, in fact, the best way to use it has been lacking. Because of some inconsistencies in results, many have recommended the addition of a calculated symptom index, in which the number of chest pain episodes that occur in proximity to an esophageal pH drop to below 4, are divided by the total number of chest pain episodes. In perhaps the largest and most carefully performed study to date, Hewson et al, evaluated 100 consecutive patients referred by cardiologists to the esophageal laboratory for undiagnosed causes of chest pain. Patients received esophageal manometry, Bernstein testing, edrophonium provocation, and 24-hour esophageal pH monitoring for the presence of abnormal acid exposure and calculation of symptom index. In this study, ambulatory pH monitoring was abnormal in 48% with a positive symptom index in 60%. It must be noted, however, that the authors accepted any correlation of reflux with chest pain as a positive symptom index and, as such, probably inflated its value. When the original criteria of Weiner et al were used (positive symptom index defined as greater than 50%) 24 out of 100 (25%) had a positive symptom index. The pH study, with or without calculation of the symptom index, identified more patients with probable or so-called definite esophageal etiologies for chest pain than manometry and provocative tests. Therefore, the authors suggested that the pH study with symptom index was the most valuable. However, because response to any therapeutic intervention was not evaluated, the retrospective ultimate value of the pH study and symptom index cannot be ascertained. Several similar retrospective and prospective studies have evaluated pH monitoring in unexplained chest pain patients and have found similar results for the percentage of patients with a positive 24-hour study, but varying numbers of patients with a positive symptom index. The present study by Dekel, et al has carefully evaluated 94 consecutive patients referred to their laboratory after a cardiac evaluation determined no etiology for the chest pain. All of these patients had chest pain greater than or equal to 3 times per week. Upper endoscopy and 24-hour esophageal pH monitoring with symptom index were performed in