BACKGROUND AND OBJECTIVES: To examine the effectiveness of intracranial pressure (ICP)–guided management on mortality and dispositional outcomes of patients with severe traumatic brain injury (sTBI). METHODS: Comparative effectiveness study with propensity score matching and mediation analysis, using the Trauma Quality Improvement Program of the National Trauma Data Bank from 2017 to 2022. Included sTBI patients with abbreviated injury severity 3 to 5 in the head region and <3 in other regions, Glasgow Coma Scale <9, age 16 to 60 years, and at least one reactive pupil. Matching was based on demographics, presenting clinical characteristics, and preexisting conditions. Exposures: ICP monitoring vs no monitoring. Outcomes: Mortality, hospital discharge disposition, and withdrawal of life sustaining treatments (WLST). Measures: Age, initial blood pressure and oxygen saturation, 24 hours highest Glasgow Coma Scale score, computed tomography midline shift, and pupillary reactivity. Mediators: neurosurgical intervention after the first 24 hours, hospital complications, and WLST. RESULTS: From 10 851 patients, 4769 received ICP monitoring matched with 6082 who did not. Mortality 23.0% ICP monitoring group vs 17.3% for the no-ICP group. ICP monitored patients were more likely to die during their hospital stay (odds ratio OR 1.278, CI 1.139-1.434, P < .001, E 1.874). ICP monitored patients were less likely to have a favorable discharge outcome (OR 0.705, CI 0.642-0.775, P < .001, E 2.189). With WLST as the mediator, the direct effect of ICP monitoring on mortality was no longer significant (OR 1.011, CI 0.995-1.030, P = .16); the indirect effect of WLST accounted for most of the total effect of ICP on mortality (proportion mediated 0.714, P -value <.001). CONCLUSION: ICP monitoring was associated with increased mortality and worse discharge outcomes in patients with sTBI. Increased mortality was largely mediated by WLST. These results question the effectiveness of ICP-guided management and highlight the major impact that decisions to WLST have on treatment effects and patient outcomes.
Lazaridis et al. (Fri,) studied this question.