Files
Abstract
Importance: Civilian penetrating brain injury (PBI) is associated with high mortality. However, scant literature is available to guide neurocritical care monitoring and management of PBI.
Objective: To examine the association of intracranial pressure (ICP) monitoring with mortality, intensive care unit (ICU) length of stay (LOS), and dispositional outcomes in patients with severe PBI.
Design, setting, and participants: This comparative effectiveness research study analyzed data from the Trauma Quality Improvement Program of the National Trauma Data Bank in the US from January 1, 2017, to December 31, 2019. Patients with PBI were identified, and those aged 16 and 60 years who met these inclusion criteria were included: ICU LOS of more than 2 days, Glasgow Coma Scale (GCS) score lower than 9 on arrival and at 24 hours, and Abbreviated Injury Scale score of 3 to 5 for the head region and lower than 3 for other body regions. Patients with bilaterally fixed pupils or incomplete data were excluded. A 1:1 propensity score (PS) matching was used to create a subgroup of patients. Patients were divided into 2 groups: with vs without ICP monitoring. Data analysis was conducted between September and December 2022.
Exposures: Intracranial pressure monitoring vs no monitoring.
Main outcomes and measures: Outcomes were mortality, rate of withdrawal, ICU LOS, and dispositional outcome. Measures were age, initial systolic blood pressure, initial oxygen saturation level on a pulse oximeter, first-recorded GCS score, GCS score at 24 hours, Abbreviated Injury Scale score, midline shift, and pupillary reactivity.
Results: A total of 596 patients (505 males [84.7%]; mean [SD] age, 32.2 [12.3] years) were included, among whom 220 (36.9%) died and 288 (48.3%) had ICP monitoring. The PS matching yielded 466 patients (233 in each group with vs without ICP monitoring). Overall mortality was 35.8%; 72 patients with ICP monitoring (30.9%) died compared with 95 patients (40.8%) without ICP monitoring . Patients with ICP monitoring were more likely to survive (odds ratio [OR], 1.54; 95% CI, 1.05-2.25; P = .03; number needed to treat, 10). No difference in favorable discharge disposition was observed. The PS-weighted analysis included all 596 patients and found that patients with ICP monitoring were more likely to survive than those without (OR, 1.40; 95% CI, 1.10-1.78; P = .005). The E-value for the OR calculated from the PS-matched data set was 1.79. In addition, ICP monitoring vs no monitoring was associated with an increase in median (IQR) ICU LOS (15.0 [8.0-21.0] days vs 7.0 [4.0-12.0] days; P < .001).
Conclusions and relevance: In this comparative effectiveness research study, PBI management guided by ICP monitoring was associated with decreased mortality and increased ICU LOS, challenging the notion of universally poor outcomes after civilian PBI. Randomized clinical trials that evaluate the efficacy of ICP monitoring in PBI are warranted.