Comparison of Gastric Intramucosal pH and Standard Perfusional Measurements in Pediatric Septic Shock

Objective: To examine the utility of gastric tonometry as an early indicator of morbidity and mortality in pediatric patients with septic shock. Design: Prospective clinical study. Setting: Multidisciplinary pediatric ICU. Patients: Eight critically ill pediatric patients with septic shock and with pulmonary artery and gastric tonometry catheters in place. Interventions: Standard perfusional data including cardiac index, oxygen content (arterial, mixed venous), oxygen delivery, oxygen consumption, urine output, capillary refill, and lactate value were measured every 4 h. Intramucosal pH (pHi) was calculated from simultaneous tonometric measurements of gastric intramucosal Pco 2 (Pico 2 ). Mean pHi values were compared between survivors and nonsurvivors. Tonometric data were compared with standard perfusional data by regression analysis. Low pHi values ( Measurements and main results: Sets (n=108) of paired data were collected from § patients (age 7 to 168 months) representing 4 deaths. The mean pHi in nonsurvivors (7.32 ± 0.18) was significantly lower than that in survivors (7.48 ± 0.07). Both pHi and Pico 2 were correlated with urine output by regression analysis. The pHi, but not Pico 2 correlated with extraction ratio and lactate level. There were no other significant correlations noted. Analysis of the ability of pHi values to predict adverse clinical events revealed that pHi less than 7.35 accurately predicted 26 of 39 concurrent or subsequent adverse clinical events. However, 24 of 50 pHi values less than 7.35 showed no association with similar adverse events, either concurrent or subsequent. The pHi had a sensitivity of 67%, a specificity of 74%, a positive predictive value of 52%, and an efficiency of 72% in predicting these adverse events. Conclusion: Use of gastric tonometry in pediatric septic shock patients appears to distinguish survivors from nonsurvivors. However, in general, tonometric assessment of splanchnic perfusion correlates poorly with standard clinical, hemodynamic, oxygen transport, or metabolic measurements of perfusion. Low pHi measurements are not predictive of concurrent or subsequent adverse clinical events.