To determine whether measurement of pepsin in throat sputum by immunoassay could be used as a sensitive and reliable method for detecting laryngopharyngeal reflux (LPR) compared with 24-hour double-probe (esophageal and pharyngeal) pH monitoring. Patients with clinical LPR undergoing pH monitoring provided throat sputum samples during the reflux-testing period for pepsin measurement using enzyme-linked immunoadsorbent assay. Pepsin assay results from 63 throat sputum samples obtained from 23 study subjects were compared with their pH monitoring data. Twenty-two percent (14/63) of the sputum samples correlated the presence of pepsin with LPR (pH < or = 4 at the pharyngeal probe), of which the median concentration of pepsin was 0.18 microg/mL (range 0.003-22 microg/mL). Seventy-eight percent (49/63) of the samples unassociated with (pharyngeal) reflux contained no detectible pepsin. Mean pH values for pepsin-positive samples were significantly lower than negative samples at both esophageal probe (pH 2.2 vs. pH 5.0) (P < .01) and the pharyngeal probe (pH 4.4 vs. pH 5.8) (P < .01). When the pepsin assay results were compared with the pharyngeal pH data for detecting reflux (events pH < or = 4), the pepsin immunoassay was 100% sensitive and 89% specific for LPR. Detection of pepsin in throat sputum by immunoassay appears to provide a sensitive, noninvasive method to detect LPR.