Effects and Clinical Characteristics of Intracranial Pressure Monitoring-Targeted Management for Subsets of Traumatic Brain Injury: An Observational Multicenter Study

Objectives: To evaluate the efficacy of traumatic brain injury management guided by intracranial pressure monitoring and to explore the specific subgroups for which intracranial pressure monitoring might be significantly associated with improved outcomes based on a classification of the various traumatic brain injury pathophysiologies using the clinical features and CT scans. Design: Retrospective observational multicenter study. Setting: Twenty-two hospitals (16 level I trauma centers and six level II trauma centers) in nine provinces in China. Patients: Moderate or severe traumatic brain injury patients who were more than 14 years old. Interventions: Intracranial pressure monitoring. Measurements and Main Results: All data were collected by physicians from medical records. The 6-month mortality and favorable outcome were assessed with the Glasgow Outcome Scale Extended score. An intracranial pressure monitor was inserted into 838 patients (58.1%). The mean duration of intracranial pressure monitoring was 4.443.65 days. The significant predictors of intracranial pressure monitoring included the mechanism of injury, a Glasgow Coma Scale score of 9-12 at admission that dropped to a score of 3-8 within 24 hours after injury, a Marshall CT classification of III-IV, the presence of a major extracranial injury, subdural hematoma, intraparenchymal lesions, trauma center level, and intracranial pressure monitoring utilization of hospital. The intracranial pressure monitoring and no intracranial pressure monitoring groups did not significantly differ in terms of complications. For the total sample, the placement of intracranial pressure monitoring was not associated with either 6-month mortality (16.9% vs 20.5%; p = 0.086) or 6-month unfavorable outcome (49.4% vs 45.8%; p = 0.175). For patients with a Glasgow Coma Scale score of 3-8 at admission, intracranial pressure monitoring was also not significantly associated with 6-month mortality (20.9% vs 26.0%; p = 0.053) or an unfavorable outcome (56.9% vs 55.5%; p = 0.646). Multivariate logistic regression analyses showed that intracranial pressure monitoring resulted in a significantly lower 6-month mortality for patients who had a Glasgow Coma Scale score of 3-5 at admission (adjusted odds ratio, 0.57; 95% CI, 0.36-0.90; adjusted p = 0.016), those who had a Glasgow Coma Scale score of 9-12 at admission that dropped to 3-8 within 24 hours after injury (adjusted odds ratio, 0.28; 95% CI, 0.08-0.96; adjusted p = 0.043), and those who had a probability of death at 6 months greater than 0.6 (adjusted odds ratio, 0.55; 95% CI, 0.32-0.94; adjusted p = 0.029). Conclusions: There were multiple differences between the intracranial pressure monitoring and no intracranial pressure monitoring groups regarding patient characteristics, injury severity, characteristics of CT scan, and hospital type. Intracranial pressure monitoring in conjunction with intracranial pressure-targeted therapies is significantly associated with lower mortality in some special traumatic brain injury subgroups. The prospective randomized controlled trials specifically investigating these subgroups will be required to further characterize the effects of intracranial pressure monitoring on behavioral outcomes in patients with traumatic brain injury.