Contusion Volume is a Cross-cohort Predictor of Delayed Seizures after Traumatic Brain Injury

Area of Science:

• Neuroscience
• Radiology
• Clinical Neurology

Background:

• Predicting long-term outcomes after traumatic brain injury (TBI) is clinically challenging.
• Acute computed tomography (CT) scans reveal hemorrhagic contusions, but their predictive value for sequelae is uncertain.

Purpose of the Study:

• To determine if acute CT-measured hemorrhagic contusion volume predicts long-term sequelae after TBI.
• To investigate the association between contusion volume and various patient outcomes, including delayed seizures and global functioning.

Main Methods:

• Utilized a TBI Model Systems (TBIMS) screening cohort (n=345) with automated lesion segmentation (BLAST-CT) on acute brain CT scans.
• Screened 198 outcome variables at 1-year post-injury for association with contusion volume, controlling for multiple comparisons.
• Validated significant findings in external cohorts: Transforming Research and Clinical Knowledge in TBI (TRACK-TBI) and Vietnam Head Injury Study (VHIS).

Main Results:

• Acute hemorrhagic contusion volume significantly predicted delayed seizures (C-index=0.81, P_FWE=0.007), but not global functioning measures (C-index=0.55, P_FWE=1).
• A 30 cc contusion volume threshold effectively discriminated patients with vs. without delayed seizures (OR 12.6).
• This association was replicated in TRACK-TBI and VHIS cohorts, with contusions >30 cc linked to increased seizure risk.

Conclusions:

• CT-derived contusion volume is a robust predictor of delayed seizures following TBI, unlike global functioning measures.
• A 30 cc volume threshold can enhance epilepsy prediction models and aid in selecting patient populations for clinical trials.