Predictive value of intramural pH and other risk factors for massive bleeding from stress ulceration.

We have devised and validated an indirect method for measuring the intramural pH in the stomachs of patients in the intensive care unit and have assessed its value in predicting the occurrence of massive hemorrhage from stress ulceration. The 103 patients, studied on an average of 3.8 +/- 0.6 consecutive days, were assigned to one of five groups: no bleeding, weakly positive guaiac, strongly positive guaiac, massive bleeding from an extragastric site, and massive bleeding from stress ulceration. The pH in gastric juice, arterial blood, and the wall of the stomach, PCO2 and PO2 in gastric juice and arterial blood, and [HCO3-] in arterial blood were recorded, together with history of treatment with antacids, cimetidine, and heparin and the number of risk factors known to be associated with bleeding from stress ulceration. Massive bleeding from stress ulceration occurred in 7 patients, all of whom were receiving antacids. Six of these patients died. A logistic analysis of the data obtained showed that the occurrence of bleeding from stress ulceration was best predicted by consideration of the combination of the number of risk factors and intramural pH (p less than 0.0001). Prediction by these variables was not improved by inclusion of any of the other variables recorded. Bleeding from stress ulceration was only seen in patients whose intramural pH had fallen below the lower limit of normality. The data indicate that intensive care unit patients would be better monitored if the intramural pH were measured, and that the occurrence of massive bleeding from stress ulceration would be better averted by maintaining the intramural pH at normal levels.