Clinical utility of pH paper versus pH meter in the measurement of critical gastric pH in stress ulcer prophylaxis.

Abstract

To evaluate the clinical utility of measuring gastric pH with a pH meter vs. pH paper in critical care patients. Prospective comparison of gastric pH measurements, using both pH meter and pH paper. Surgical intensive care unit (ICU) at a rural Midwestern university medical center. Fifty-one patients who received therapy for prophylaxis of stress ulcers in the surgical ICU. Therapy for stress ulcer prophylaxis was monitored. The pH of 985 gastric samples, taken from 51 patients, was measured with both pH meter and pH paper. The pH meter and pH paper measures demonstrated a concordance correlation coefficient of .896. The mean difference between the two measures (pH paper - pH meter) was estimated to be between -0.4 and 1.4, suggesting a positive bias for the paper. The prevalence of events representing clinically relevant differences between the pH meter and pH paper in the measurement of the same gastric sample was calculated. The frequency with which each of the events occurred consecutively (or, in one case, two nearly consecutive events on the same day) was also calculated. Bias in a clinically relevant range was estimated. A set of "probability profiles" was constructed. A hand-held pH meter and pH paper are not interchangeable measures of gastric pH. The pH paper exhibits an appreciable positive bias compared with a hand-held pH meter in the clinically relevant range of 2 to 6. More research is needed to determine if that bias affects treatment outcomes. We recommend the use of a pH meter for patients who demonstrate pH readings of < or = 4, consecutive with readings of < or = 5.