The modified Brain Injury Guidelines: safe, sensitive, but not yet specific.

Objective: The modified Brain Injury Guidelines (mBIG) have been determined safe, but statistical analyses are limited to lower-severity mBIG 1 and mBIG 2 populations. The aim of this study was to determine the sensitivity of mBIG criteria for the need for neurosurgical intervention and to evaluate associations between individual mBIG 3 radiographic criteria and need for intervention.

Methods: The records of all head trauma patients presenting to a level I trauma center from May 2020 to December 2023 were retrospectively reviewed. Patients without intracranial hemorrhage (ICH) on first CT imaging of the head and those who underwent intervention at an outside hospital were excluded. Radiographic stability versus progression on repeat CT imaging was collected. Clinical outcome measures were neurosurgical intervention, length of stay, and mortality. Patients were sorted based on published mBIG criteria and compared with prior studies reporting on intervention by the Brain Injury Guidelines (BIG) group.

Results: A total of 1128 patients with a mean age of 54.9 (SD 21.2) years, 67.7% male, were included. Most patients (69.7%) were in the mBIG 3 group. Ninety-seven patients (8.6%), all in the mBIG 3 cohort, received intervention after initial CT imaging. An additional 113 patients (10.0%) underwent intervention after some period of observation and at least one repeat CT study, of whom 112 were in the mBIG 3 group and 1 was in the mBIG 2 group (0.6% of mBIG 2 patients, 0.5% of those requiring intervention, and 0.09% of the entire study population). mBIG 3 criteria were 99.5% sensitive for the need for neurosurgical intervention. mBIG 2+3 criteria were 100.0% sensitive, consistent with the cumulative literature of all iterations of BIG. Their specificities were 37.2% and 18.1%, respectively. There was no correlation between intervention and anticoagulant/antiplatelet use (p = 0.069). Progression of ICH was associated with intervention within the mBIG 3 cohort (p < 0.001). In post hoc multivariate analysis of mBIG 3 patients with Glasgow Coma Scale scores of 13-15, significant associations were found between each specific mBIG 3 radiographic criterion and intervention, except intraparenchymal hemorrhage (IPH) (p = 0.205) and subarachnoid hemorrhage (SAH) (p = 0.274).

Conclusions: The mBIG 3 criteria are 99.5% sensitive for the need for neurosurgical intervention. Criteria requiring hospital admission (mBIG 2+3) are 100.0% sensitive. Radiographic mBIG 3 criteria for IPH and SAH alone are poor predictors for the need for neurosurgical intervention in patients with favorable neurological status. The recommendation against routine repeat CT head imaging in mBIG 1 and 2 patients is supported by the findings of this study.