The diagnosis of reflux oesophagitis: an evaluation of five investigative procedures.

Five methods of diagnosis have been compared prospectively in 43 patients referred for suspected reflux oesophagitis. A final diagnosis of reflux oesophagitis, as defined by the presence of at least two of the three features of typical symptoms, abnormal endoscopic findings, and abnormal findings on oesophageal biopsy, was made in 27 patients. Observer error in the interpretation of endoscopic and histological appearances was small. Measurement of resting pressure of the lower oesophageal sphincter (LES) failed to identify individual patients with reflux oesophagitis, although the mean pressure in 26 patients with oesophagitis (10.1 +/- 5.2 mm Hg) was significantly lower that in 13 patients without oesophagitis (16.8 +/- 10.2 mm Hg, P less than 0.005). Barium studies were unhelpful, as a hiatus hernia was present in only 14 and barium reflux in only 11 of the 27 patients with oesophagitis. Acid perfusion (Bernstein test) was positive at 15 minutes in 23 of the 27 oesophagitis patients, but was falsely positive in seven of the 14 patients without oesophagitis. By accepting only those responses to acid perfusion which were positive at or before seven minutes, the false positive responses were reduced to one out of the 14 patients. Typical symptoms and/or an early positive Bernstein response will correctly identify most patients with reflux oesophagitis, but the diagnosis should be confirmed by endoscopy and biopsy when important therapeutic decisions are pending.