Postoperative monitoring of body temperature in patients with brain trauma: a prospective multicenter study

Abstract

Objective To investigate the relationship of brain temperature (intra-ventricle, brain parenchyma, or subdural space) to axillary temperature, rectal temperature and bladder temperature in postoperative patients with brain trauma, and to explore the prognosis value of the brain temperature through a multicenter prospective observational study. Methods A total of 83 patients with brain trauma at 11 hospitals were enrolled to this study from March 2016 to December 2016. The data including patients' brain temperatures, axillary temperatures, rectal temperatures and bladder temperatures were hourly documented postoperatively until withdrawal of monitoring device of brain temperatures. The changes of the values were then further analyzed. Results Compared with the axillary temperature, the subdural temperature difference [0.10(-0.40, 0.40)℃] was less than intraventricular temperature difference [0.30(-0.20, 0.90)℃], which was less than parenchymal temperature difference [0.60(0.20, 1.00)℃](both P<0.05). Compared with rectal temperature and bladder temperature, the subdural temperature difference [-0.20(-0.40, 0.00)℃, -0.10(-0.30, 0.00)℃, respectively] was less than parenchymal temperature difference [0.10(-0.10, 0.20)℃, 0.10(0.00, 0.20)℃, respectively], which was less than intraventricular temperature difference [0.20(0.10, 0.30)℃, 0.20(0.10, 0.30)℃, respectively] (all P<0.05). No matter whether hypothermia treatment was administered, the difference between brain temperature and axillary temperature (ΔTBr-Ax) was larger than that between brain temperature and rectal temperature (ΔTBr-Re) and that between brain temperature and bladder temperature (ΔTBr-Bl) (both P<0.05). When brain temperature was higher than 38 ℃, ΔTBr-Ax was larger than ΔTBr-Re , which was larger than ΔTBr-Bl (both P<0.05). When brain temperature was less than or equal to 38 ℃, ΔTBr-Ax was larger than ΔTBr-Bl, which was larger than ΔTBr-Re (both P<0.05). When the fluctuation of brain temperature within 12 hours post operation was more than 1 ℃, the preoperative Glasgow coma scale (GCS) and the extended Glasgow outcome scale(GOS-E) at discharge were significantly lower compared with those of cases whose brain temperature fluctuation was less than or equal to 1 ℃ (both P<0.05). Conclusions For patients with brain trauma, brain temperature monitoring seems to have a very significant clinical value. When monitoring of brain temperature is not available, the bladder temperature might serve as the best reflection of brain temperature. Key words: Craniocerebral trauma; Body temperature; Multicenter study; Prospective studies