Does brain temperature correlate with intracranial pressure?

Objective: A positive correlation between brain temperature and intracranial pressure (ICP) has been proposed for patients under intensive care conditions. Design and Methods: Data were recorded at 5-minute intervals in patients under ICP monitoring conditions. Brain temperature: combined ICP/temperature probe (Raumedic), core temperature: indwelling urinary catheter with temperature probe (Rusch). The correlation between brain temperature and ICP was assessed by computing an estimated mean correlation coefficient (r(e)) and by a time series analysis. Patients: Forty consecutive neurosurgical patients receiving intensive care therapy for trauma, cerebrovascular malformation, and spontaneous hemorrhage were studied. A total of 48,892 measurements (9778 h) were analyzed. No additional interventions were performed. Results: The median ICP was 14 mm Hg (range: - 13 to 167). The brain temperature (median 38 degrees C; range 23.2 to 42.1) was 0.3 degrees C (range: -3.6 to 2.6) higher than the core temperature (median 37.7 degrees C; range 16.6 to 42.0), P < 0.001. The mean Pearson correlation between ICP and brain temperature in all patients was r(e) = 0.13 (P < 0.05); the time series analysis (assuming a possible lagged correlation between ICP and brain temperature) revealed a mean correlation of 0.05 +/- 0.25 (P < 0.05). Both correlation coefficients indicate that any relationship between brain temperature and ICP accounts for less than 2% of the variability [coefficient of determination (r(2)) < 0.02]. Conclusions: These data do not support the notion of a clinically useful correlation between brain temperature and ICP.