Effect of Hypothermia in the Emergency Department on the Outcome of Trauma Patients: A Cross-Sectional Analysis

Abstract

This study aimed to assess whether hypothermia is an independent predictor of mortality in trauma patients in the condition of defining hypothermia as body temperatures of <36 °C. Data of all hospitalized adult trauma patients recorded in the Trauma Registry System at a level I trauma center between 1 January 2009 and 12 December 2015 were retrospectively reviewed. A multivariate logistic regression analysis was performed in order to identify factors related to mortality. In addition, hypothermia and normothermia were defined as temperatures <36 °C and from 36 °C to 38 °C, respectively. Propensity score-matched study groups of hypothermia and normothermia patients in a 1:1 ratio were grouped for mortality assessment after adjusting for potential confounders such as age, sex, preexisting comorbidities, and injury severity score (ISS). Of 23,705 enrolled patients, a total of 401 hypothermic patients and 13,368 normothermic patients were included in this study. Only 3.0% of patients had hypothermia upon arrival at the emergency department (ED). Compared to normothermic patients, hypothermic patients had a significantly higher rate of abbreviated injury scale (AIS) scores of ≥3 in the head/neck, thorax, and abdomen and higher ISS. The mortality rate in hypothermic patients was significantly higher than that in normothermic patients (13.5% vs. 2.3%, odds ratio (OR): 6.6, 95% confidence interval (CI): 4.86–9.01, p < 0.001). Of the 399 well-balanced propensity score-matched pairs, there was no significant difference in mortality (13.0% vs. 9.3%, OR: 1.5, 95% CI: 0.94–2.29, p = 0.115). However, multivariate logistic regression analysis revealed that patients with low body temperature were significantly associated with the mortality outcome. This study revealed that low body temperature is associated with the mortality outcome in the multivariate logistic regression analysis but not in the propensity score matching (PSM) model that compared patients with hypothermia defined as body temperatures of <36 °C to those who had normothermia. These contradicting observations indicated the limitation of the traditional definition of body temperature for the diagnosis of hypothermia. Prospective randomized control trials are needed to determine the relationship between hypothermia following trauma and the clinical outcome.

The Effect of Thermal Care Workshop on EMS Staff Readiness in managing Accidental Hypothermia in Trauma Patients

“…Hypothermia refers to a central temperature below 35 degrees Centigrade [9]. In trauma patients, accidental hypothermia refers to a drop in central body temperature below 36 degrees Centigrade and is categorized as: mild (34-36°C), moderate (32-34°C), and severe (below 32°C) [7,10,11]. Hypothermia can be caused by trauma, infection or other diseases.…”

The Impact of Accidental Hypothermia on Mortality in Trauma Patients Overall and Patients with Traumatic Brain Injury Specifically: A Systematic Review and Meta‐Analysis

“…Temperature measurement methods were specified in six studies and comprised nasopharyngeal [ 39 ], esophageal [ 1 , 7 , 31 , 32 , 41 ], bladder [ 7 , 31 , 32 ], rectal [ 32 , 39 ], tympanic [ 32 , 41 ], axillary [ 32 ], and oral [ 32 ] temperature probes. Eight studies did not describe the exact method of temperature measurement [ 6 , 8 , 35 – 38 , 40 , 42 ].…”

“…Included patients were predominantly male (range 71–79%) [ 6 , 31 , 35 , 36 , 39 , 41 , 42 ]. Eight studies compared the age of the normothermic and hypothermic patient groups [ 7 , 8 , 31 , 35 – 37 , 39 , 40 ]. In the studies by Bukur et al and Martin et al, hypothermic patients were significantly older than normothermic patients [ 8 , 35 ].…”

“…In sensitivity analysis including studies with the same definition of hypothermia (temperature < 35 °C, four studies, 703,372 patients [ 8 , 31 , 37 , 41 ]), studies with similar patient inclusion criteria regarding the ISS (ISS ≥ 16, three studies, 3,068 patients [ 31 , 37 , 41 ]), studies published during the last 10 years (four studies, 35,834 patients [ 31 , 35 , 40 , 41 ]), and studies with more than 10,000 patients (four studies, 773,616 patients [ 8 , 32 , 35 , 40 ]), hypothermia on admission was associated with significantly higher mortality, too (OR 6.15 [95%CI 3.43–11.05], I 2 = 90%; OR 5.02 [95%CI 3.35–7.52], I 2 = 39%; OR 3.76 [95%CI 1.48–9.52], I 2 = 96%; and OR 5.55 [95%CI 2.23–3.84], I 2 = 100%).…”

“…Eleven studies adjusted for the ISS [7,8,31,[35][36][37][38][39][40][41][42] and two studies for the NISS [6,32] in multivariable analysis for mortality. In eleven studies, hypothermia was reported as an independent predictor for mortality, even after adjustment for the injury severity and other covariates [6,8,32,[35][36][37][38][39][40][41][42]. In two studies, the effect of hypothermia on mortality did not remain significant after adjustment in multivariable analysis [7,31] (Table 3).…”