Body temperature has been identified as an essential element in the assessment of the patient requiring critical care. In the care of both civilian and military patients with traumatic injuries, focus has been placed on hypothermia upon patient admission to the hospital [1–13]. In these patients, hypothermia is associated with increases in coagulopathy, hemorrhage, multiple organ failure, length of hospital stay, and mortality [4, 6–11, 13]. Correction of hypothermia in the period of initial treatment is vigorously advocated and is associated with improved outcomes [14].

The presence of hyperthermia in the injured patient has received less attention than that of hypothermia. Malone et al. noted an increase in mortality in trauma patients if they had either hypothermia (< 36°C) or hyperthermia (> 38°C) [15]. Martin et al., in an analysis of the National Trauma Data Bank (NTDB), focused on hypothermia, and demonstrated a threefold increase in mortality in temperatures above 38°C [9]. At present, there are no guidelines on the initial treatment of the trauma patient admitted with hyperthermia after injury.

Martin et al. focused on trauma patients with hypothermia [9]. However, review of their NTDB data suggested that hyperthermia could also be associated with increased mortality. Hyperthermia may occur as a result of infection, high ambient temperatures, or increased physical effort. In elevated ambient temperature environments, the risk of primary hyperthermia can be exacerbated by heavy work or exercise. Present US military conflicts in Iraq and Afghanistan are occurring in countries with persistently high environmental temperatures. We hypothesized that in the present military conflicts, the incidence of hyperthermia in trauma patients upon admission would be greater than in a civilian setting. Further, the increase in the frequency of hyperthermia was expected to be associated with an increase in mortality and, therefore, correction of body temperature should be considered.

NTDB data on 621,081 patients were available for analysis. Significant increases in mortality were noted in this patient set when body temperatures were < 36°C or > 38°C (Figure 1). Thus, deviations beyond these temperatures were used to define hypothermia and hyperthermia, respectively.

In the military population, the Barell index for traumatic brain injury (TBI) was recorded and the anatomical injury score (AIS) for brain trauma was coded. The incidence of moderate or severe brain injury with the Barell index in patients was 55% in the hypothermia group, 36% in the normal group, and 49% in the hyperthermia group. With an AIS > 2, there was a 53% rate of brain injury for the hypothermia group, 35% for the normal group, and 56% for the hyperthermia group. For both measures of brain injury, the rates were increased in patients with hypothermia and hyperthermia compared to normal. There were no differences between hypothermia and hyperthermia in the incidence of TBI.

Hypothermia is a well-defined contributor to poor outcomes in patients with traumatic injuries [1–13]. The presence of hypothermia is associated with coagulopathy and increased blood transfusion requirements as well as mortality. The influence of hyperthermia on trauma mortality has not often been addressed. In an evaluation of using the systemic inflammatory response syndrome (SIRS) score to assess outcomes in patients with traumatic injuries, others noted that altered temperature (either hypothermia or hyperthermia) was a significant predictor of outcome [15–18]. In the present study, we referenced the NTDB data to determine the range of normal temperature, based on when the rate of mortality increased. Although the NTDB and other registries have numerous limitations, these databases allow access to information of large numbers of patients to describe the population of interest. A range for normal values was found to be 36° to 38°C using data from over 600,000 patients. The hypothermic threshold was identical to that in earlier studies of this data set [9, 13]. These studies of hypothermia suggested increases in mortality below body temperatures of 36°C; however, the hyperthermic threshold was not previously defined. In the present study of military and civilian populations, the impact of hypothermia or hyperthermia was associated with similar increases in mortality, compared to patients with temperatures in the range of 36° to 38°C.

There were differences in the incidence rates of hypothermia and hyperthermia between the military and the civilian populations. The greater incidence of hyperthermia in the military population and the reduction in hypothermia may be the result of the ambient environment, differences in clothing worn, and the patients' activities at the time of injury. The average ambient temperature over the year in San Antonio where the civilian population was evaluated was 20°C, with an average high of 27°C and a low of 14°C. In Iraq, where 95% of the military population was studied, the average annual environmental temperature was 22°C, with the average high and low temperature of 30° and 15°C, respectively (UN World Meteorological Organization). Average temperatures did not appear significantly different between the two study sites. In theater, soldiers wear personnel protective equipment and heavy clothing, which limit heat loss. Furthermore, there may be significant differences in other factors; for example, at the time of injury, soldiers are often on foot and engaged in physical activity in contrast to civilians who may be riding in air-conditioned automobiles. The greater occurrence of hyperthermia may also be the result of increased incidence of brain injuries [19]. However, within the military population, the incidence of severe traumatic injury, while greater than in patients with normal temperatures, was not different between patients with hypothermia or hyperthermia. The increase in ambient temperature, protective clothing worn, types of injuries, and the work activities experienced by the military population may have contributed to the increased incidence of hyperthermia and the reduced incidence of hypothermia in contrast to the incidence observed in the civilian population.

A greater probability of mortality in the civilian population was associated with hyperthermia and may be the result of a number of factors. There are distinct differences in age between civilian and military casualties [20]. Civilian casualties are generally older and, on average, are affected by comorbid disease factors. The cause of injury in civilian patients is often blunt in contrast to penetrating, with or without explosion, in military casualties. In addition, the extent of injury may have been different between the populations. Though a group difference in outcome between the military and civilian subjects was observed, the increase in mortality with alterations in body temperature was noted in both populations.

On admission, neither population showed a difference between temperature groups in physiological indices (such as systolic blood pressure and RTS), which are normally used acutely in evaluating the status of a patient. Thus, the differences in body temperature may be a delineating physiological factor in assessing patient status and subsequent outcome, as previously suggested by others [9, 13, 16, 17].

This is a retrospective study with the inherent limitations associated with this methodology. Registry data were used that are not the primary source, which may include transcription errors. That said, the size of the patient population studies was an attempt to minimize these errors. Finally, we only used patients in which complete data were available and outcomes were known. This may have led to some selection bias.

Deviations in body temperature above or below the range of 36° to 38°C in the patients admitted to the hospital with traumatic injuries are indicative of a poor outcome and, therefore, should be addressed as soon as possible. This indication is similar to the findings observed in patients with an increased SIRS due to increased body temperature upon admission [15–18]. In the care of the patient with traumatic injuries, emphasis has been placed on avoiding hypothermia; however, hyperthermia is similarly detrimental. Hyperthermia may be the result of ambient temperature, clothing worn, or work environment as in the present military conflicts. Although a greater incidence rate of hyperthermia was observed in the military group compared to the civilian group, a greater mortality rate was not apparent. The presence of hypothermia or hyperthermia should be considered in the initial treatment of the patient with traumatic injuries and corrected to the normal range of 36° to 38°C.