Purpose <p>For unclear reasons, up to 50% of subjects with gastroesophageal reflux disease have an inadequate symptomatic response to treatment with a proton pump inhibitor. Our primary objective was to examine esophageal acid in subjects who experienced heartburn following a standard meal that failed to respond to a proton pump inhibitor.</p> Methods <p>We analyzed data from a previous randomized crossover trial of proton pump inhibitors in 27 subjects with gastroesophageal reflux disease and baseline esophageal pH &lt; 4 for more than 10% of a 24-h recording. Esophageal pH and heartburn severity were measured over 3&#xa0;h following a standard meal at baseline and after omeprazole.</p> Results <p>Relationships between heartburn severity and esophageal acid were identified using the quadrant method. Twelve subjects had high values for heartburn severity with omeprazole, indicating a failed response; 75% of these subjects had low esophageal acid reflecting esophageal hyperalgesia or a cause other than esophageal acid. In subjects with high esophageal acid that failed to improve with omeprazole, heartburn severity improved with omeprazole in 57%, reflecting esophageal hypoalgesia or silent esophageal reflux.</p> Conclusions <p>Failure of heartburn to respond to omeprazole was common in subjects with gastroesophageal reflux disease who exhibited high baseline esophageal acid exposure and most of these failures were associated with low esophageal acid.</p>