To determine the relation between the sensation of pain in gastroesophageal reflux and the pH of the refluxate, we studied 25 individuals with symptomatic gastroesophageal reflux and positive Bernstein tests. We quantitatively assessed the sensitivity of the esophageal mucosa to pain associated with the intraesophageal infusion of eight different HCl solutions (pH 1, 1.5, 2, 2.5, 3, 4, 5, and 6). Test solutions were infused at 8 ml/min through an eight-lumen catheter with the orifices placed 5 cm above the lower esophageal sphincter. Each subject received all eight solutions in a double-blind randomized fashion. The time-to-pain onset increased with increasing pH; i.e., there was a highly significant difference between the time-to-pain and pH (p less than 0.001), with the time-to-pain significantly longer with increasing pH (r = 0.77). In addition to more rapid onset of pain, all subjects experienced pain with the pH 1 and 1.5 solutions, 80% had pain with the pH 2.0 solution, and half had pain with solutions of pH 2.5-6. Fifteen of these subjects underwent 24-h pH monitoring and these tests were examined for factors associated with pain. Only 64% of all pain episodes were associated with a pH drop of less than 4; the lowest pH obtained was not different between episodes with and without pain. Reflux episodes resulting in pain were significantly longer than those without pain and were more often associated with a recently preceding painful episode. Overall, none of the data from the 24-h pH monitoring was useful for predicting pain. The acid infusion studies and the 24-h pH data, taken together, suggest episodes of pain sensitize the patient for subsequent pain.