To determine the relationship of tonometrically measured gastric intramucosal pH to clinically accepted indices of systemic oxygenation. Prospective, nonintervention study. Medical and surgical ICUs of a university hospital. Critically ill patients (n = 22) with pulmonary artery catheters in place who also required nasogastric suctioning. Tonometric measurements of gastric intramucosal pH were compared with concurrently obtained systemic indices of tissue oxygenation. These variables included oxygen delivery (DO2), oxygen consumption (VO2), oxygen extraction ratio, arterial lactate concentrations, mixed venous PO2, and mixed venous pH. The study period ranged from the time of insertion of the pulmonary artery catheter to the time of its removal. We classified patients who were alive by the end of the study as survivors and those patients who died with the pulmonary artery catheter in place as nonsurvivors. Both groups had similar levels of DO2 during the study period, but nonsurvivors had greater levels of VO2, oxygen extraction ratio, and serum lactate concentrations than survivors. Gastric intramucosal pH, mixed venous pH, and mixed venous PO2 values were lower in nonsurvivors. Both groups demonstrated the phenomenon of oxygen supply dependency. When the final measurements taken before the removal of the pulmonary artery catheter or death were compared, only gastric intramucosal pH and mixed venous pH showed differences between the groups, being lower in nonsurvivors. All the patients who died, except for one, had final gastric intramucosal pH values of less than 7.32. In this group of patients, death was associated with increased tissue needs for oxygen that were not adequately satisfied by the available levels of oxygen supply. We also conclude that tonometrically measured gastric intramucosal pH is a useful noninvasive adjunct to current methods of monitoring systemic oxygenation.