Goal-Directed Fluid Therapy Based on Stroke Volume Variations Improves Fluid Management and Gastrointestinal Perfusion in Patients Undergoing Major Orthopedic Surgery

BACKGROUND: The gastric intramucosal pH (pHi) and gastric mucosal-arterial CO2 gap (GAP) estimate visceral perfusion and predict outcome. Threshold values of these variables for use during resuscitation, however, remain poorly defined. The purpose of this study was to develop clinically derived cutoffs for both pHi and GAP for predicting death and multiple organ failure (MOF) in trauma patients. METHODS: This was a cohort study of 114 consecutive trauma patients who had pHi determined at 24 hours after intensive care unit admission. The corresponding GAP for each of these values of pHi was obtained through chart review. Receiver operating characteristic curves were constructed for both pHi and GAP with respect to death and MOF. These curves were used to determine the value of each variable that maximized the sum of sensitivity and specificity in predicting outcome. chi2 tests and odds ratios were used to determine if significant differences in outcome occurred above and below these cutoff values. RESULTS: Of 114 patients who had pHi determined at 24 hours after admission, 108 had corresponding GAP values available. The values of pHi and GAP that maximized sensitivity and specificity were 7.25 and 18 mm Hg, respectively. The odds ratio for pHi versus death was 4.6 and for pHi versus MOF was 4.3. The odds ratios for GAP versus death and MOF were 2.9 and 3.3, respectively. CONCLUSION: In trauma patients, the ability to predict death and MOF is maximized at values of pHi less than 7.25 and GAP greater than 18 mm Hg. These values represent clinically derived cutoffs that should be useful for evaluating the adequacy of intestinal perfusion during resuscitation.