A prospective observational study collected temperature data from 51 patients in 11 neurosurgical centers and follow-up outcome information at 6 months in 49 patients. Brain temperature (Tbr) was measured directly by an intraventricular temperature sensor. Axillary temperature (Tax) and rectal temperature (Tre) were measured by electric thermometers. Tbr was 0.4 to 1.5°C higher than body temperature. Tre correlated well with the Tbr (coefficient: 0.7378; p < 0.05). Among all patients, Glasgow Coma Scale (GCS) scores on admission were significantly lower in the patients with post-operatively extreme peak temperature (Tpeak, < 37°C or >39°C in first 24 h) and major temperature variation (Tvari > 1°C in first 12 h; p < 0.05, p < 0.01, respectively). Among the patients with no temperature intervention, the extreme Tpeak group showed a lower Glasgow Outcome Scale-Extended (GOS-E) score at 6 months (p < 0.05) with lower GCS scores on admission (p < 0.01), compared with the moderate Tpeak group. Remarkably, the major Tvari group showed significantly lower GOS-E scores (p < 0.05) with the same GCS scores as the minor Tvari group. Thus, Tre is the better candidate to estimate Tbr. Spontaneously extreme Tpeak in TBI represents both more serious injury on admission and worse prognosis, and Tvari might be used as a novel prognostic parameter in TBI. Brain temperature is therefore one of the critical indicators evaluating injury severity, prognostication, and monitoring in the management of TBI. This prospective observational study has been registered in ClinicalTrials.gov ( https://clinicaltrials.gov ), and the registration number is NCT03068143.
Keywords: body temperature; brain temperature; multi-center study; prospective studies; traumatic brain injury; treatment outcome.