OBJECTIVE: The objective of the study was to determine if hypothermia in pediatric trauma patients is associated with increased mortality. METHODS: We reviewed the charts of level 1 trauma patients aged 3 months to 17 years who presented between September 2006 and March 2008. We analyzed data for patients with temperatures recorded within 30 minutes of arrival to the pediatric emergency department. Logistic regression models were used to test for associations of hypothermia with death while adjusting for mode of transport, season of year, and presence of intracranial pathology as documented by an abnormal head computed tomographic scan. RESULTS: Of the 226 level 1 trauma patients presenting during the study period, 190 met inclusion criteria. Twenty-one patients (11%) died. The odds ratio (OR) of a hypothermic patient dying was 9.2 times that of a normothermic patient when adjusting for seasonal variation (95% confidence interval [CI], 3.2-26.2; P < 0.0001). The OR of a hypothermic patient dying was 8.7 times that of a normothermic patient when adjusting for mode of transport (ground vs air) (95% CI, 3.1-24.6; P < 0.0001). Although it did not reach statistical significance, there was a trend toward an association between hypothermia and the presence of traumatic brain injury as evidenced by an abnormal head computed tomographic scan (OR = 2.4; 95% CI, 0.9-6.0; P = .07). CONCLUSIONS: Hypothermia is a risk factor for increased mortality in pediatric trauma patients. This pilot study warrants a more detailed, multicenter analysis to assess the impact of hypothermia in the pediatric trauma patient.
OBJECTIVE: The objective of the study was to determine if hypothermia in pediatric traumapatients is associated with increased mortality. METHODS: We reviewed the charts of level 1 traumapatients aged 3 months to 17 years who presented between September 2006 and March 2008. We analyzed data for patients with temperatures recorded within 30 minutes of arrival to the pediatric emergency department. Logistic regression models were used to test for associations of hypothermia with death while adjusting for mode of transport, season of year, and presence of intracranial pathology as documented by an abnormal head computed tomographic scan. RESULTS: Of the 226 level 1 traumapatients presenting during the study period, 190 met inclusion criteria. Twenty-one patients (11%) died. The odds ratio (OR) of a hypothermicpatient dying was 9.2 times that of a normothermic patient when adjusting for seasonal variation (95% confidence interval [CI], 3.2-26.2; P < 0.0001). The OR of a hypothermicpatient dying was 8.7 times that of a normothermic patient when adjusting for mode of transport (ground vs air) (95% CI, 3.1-24.6; P < 0.0001). Although it did not reach statistical significance, there was a trend toward an association between hypothermia and the presence of traumatic brain injury as evidenced by an abnormal head computed tomographic scan (OR = 2.4; 95% CI, 0.9-6.0; P = .07). CONCLUSIONS:Hypothermia is a risk factor for increased mortality in pediatric traumapatients. This pilot study warrants a more detailed, multicenter analysis to assess the impact of hypothermia in the pediatric traumapatient.