To evaluate the relationship of early hypothermia to multiple organ failure and mortality in a prospectively-collected database of severely injured trauma patients.
This prospective observational study was performed at 7 level I trauma centers over a 16-month period. Severely injured trauma patients with signs of hypoperfusion (eg, base deficit, hypotension) and need for blood transfusion during their early hospital course were followed for 24 hours with near infrared spectroscopy-derived tissue oxygen saturation (StO2) and other variables for 28 days to evaluate outcomes including multiple organ dysfunction syndrome (MODS) and death. Early hypothermia was defined as the presence of a temperature ≥35°C anytime within the first 6 hours of hospitalization. Comparisons between groups were made using the Wilcoxon Two-Sample test for continuous variables and either the Fisher exact or χ2 test for categorical variables. Multivariate logistic regression was utilized to understand the effect of hypothermia on outcome (MODS and mortality).
Hypothermia was very common in this cohort of patients, present in 43% of patients enrolled (155/359). Hypothermic patients were 3 times more likely than normothermic patients to develop MODS (21% vs. 9%, P = 0.003). Hypothermic patients did not have an increased incidence of mortality (16% vs. 12%, P= 0.2826). Base deficit in hypothermic patients did not discriminate between patients who did or did not develop MODS (9.8 + 4.6 mEq/L vs. 9.4 + 4.4 mEq/L), but had good discrimination for mortality in both hypothermic and normothermic patients. Significant predictors of MODS using multivariate analysis included minimum StO2 (P= 0.0014) and hypothermia (P = 0.0371). Predictors for mortality using multivariate analysis included minimum StO2 (P= 0.0021) and base deficit (P= 0.0454), but not hypothermia (P= 0.5289). Hypothermia remained a significant risk factor for MODS when systolic blood pressure, volume of fluid, and volume of blood infused were included in the multivariate model.
Hypothermia is common in severely injured trauma patients (nearly half of patients in this series) and is a significant risk factor for MODS but not mortality. The predictive value of base deficit for development of MODS is blunted in the presence of hypothermia. A low StO2 value predicts MODS and mortality in trauma patients and is a durable measure in both normothermic and hypothermic patient groups.
Severely injured trauma patients admitted to 7 level I trauma centers were prospectively observed to evaluate outcomes. Hypothermia was common, occurring in 43% of patients (155/359). Hypothermic patients were 3 times more likely to develop multiple organ dysfunction syndrome, but did not demonstrate increased incidence in mortality. With multivariate analysis, multiple organ dysfunction syndrome was predicted by minimum StO2 and hypothermia, and mortality was predicted by minimum StO2 and base deficit but not hypothermia.
From the *North Memorial Medical Center and Department of Surgery, University of Minnesota, Minneapolis, Minnesota; †Technomics Research, LLC, Long Lake, Minnesota; ‡Division of General Surgery, St. Michael's Hospital, Toronto, Canada; §Department of Surgery, University of Texas Health Science Center, Houston, Texas; ¶Department of Surgery, University of Arizona, Tucson, Arizona; ∥Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania; **Department of Surgery, University of Colorado, Denver, Colorado; and ††Department of Surgery, University of Texas Health Science Center, San Antonio, Texas.
This study was previously presented in poster form at the 27th annual ISICEM meeting in Brussels, Belgium on March 27, 2007.
Reprints: Greg J. Beilman, MD, Division of Surgical Critical Care/Trauma, Department of Surgery, University of Minnesota, MMC 11, 420 Delaware Street SE, Minneapolis, MN, 55455. E-mail: firstname.lastname@example.org.
In the article by Beilman, et al. Ann Surg. 2009; 249(5):845-850, line 7 in the abstract methods section should read as, “Early hypothermia was defined as the presence of a temperature <35°C anytime within the first 6 hours of hospitalization”. Also, line 5 of the results section should read as, “Base deficit in hypothermic patients did not discriminate between patients who did or did not develop MODS (9.8 ± 4.6 mEq/L vs. 9.4 ± 4.4 mEq/L). In contrast, base deficit in hypothermic patients discriminated with respect to mortality (14.6 ± 7.2 mEq/L versus 9.5 ± 4.5 mEq/L; P = 0.0021), but this effect was not observed in normothermic patients.” Finally, line 6 of the conclusion on p. 850 should read as, ”Base deficit has good discrimination for mortality in hypothermic patients, but not in normothermic patients. In contrast, the discriminatory effect of base deficit for development of MODS is blunted in the presence of hypothermia.” The Authors and the publisher apologize for the errors.