OBJECTIVE
This research was aimed at studying the clinical and MRI predictors of coma duration, the intensity of critical care, and outcomes in patients with traumatic brain injury (TBI).
MATERIAL AND METHODS
309 patients with TBI of varying severity were included in the analysis, of whom 257 (86.7%) were treated in the intensive care unit (ICU), including 196 (63.4%) patients admitted in a comatose state lasting longer than 1 day. All patients underwent brain MRI within 21 days after the injury. MRI findings were classified according to MRI grading scale of brain damage level/localization proposed previously.
RESULTS
The proposed MRI grading significantly correlated with the Glasgow coma (GCS, r=–0.67; p<0.0001) and Glasgow outcome (0.69; p<0.001) scores in the entire group. In a subgroup of comatose patients (GCS<9) it correlated with coma duration (r=0.52; p<0.0001). Spearman correlation analysis showed a significant relationship between the MRI classification and a number of parameters: ICU length of stay (r=0.62; p<0.0001), the duration of mechanical ventilation (r=0.47; p<0.0001), the rate of mechanical ventilation, sedatives, analgesics, mannitol, a hypertonic solution of sodium chloride and vasopressors usage (p<0.01). A direct relationship was revealed between the grades of damage according to MRI classification and the need for the escalation of the main components of intensive care.
CONCLUSION
Our results support the hypothesis that the levels and localization of brain damage, estimated by the proposed MRI classification, might serve as predictors of coma duration, intensity and duration of intensive care and TBI outcomes. A prognosis based on clinical and neuroimaging data comparison can be valuable for planning and efficient use of the hospital beds and ICU resources, for optimizing the patient flow and timing for their further transfer to neurorehabilitation facilities.