121 Analysis of Post-traumatic Epilepsy After Severe Traumatic Brain Injury in a Modern Series Identifies an Increasing Incidence and Novel Risk Factors. (1st April 2022)

• Record Type: Journal Article
• Authors: Pease, Matthew; Elmer, Jonathan; Nwachuku, Enyinna L.; González-Martínez, Jorge A.; Puccio, Ava; Castellano, James; Okonkwo, David O.
• Is Part Of: Neurosurgery. Volume 68(2022)Supplement 1
• Page Start: 34
• Page End: 35
• Publication Date: 2022-04-01
• Subjects: Nervous system -- Surgery -- Periodicals; 617.48005
• DOI: 10.1227/NEU.0000000000001880_121
• Languages: English
• ISSNs: 0148-396X

Abstract

Abstract : INTRODUCTION: Post-traumatic epilepsy (PTE) accounts for 20% of symptomatic epilepsy with rates highest after severe Traumatic Brain Injury (TBI). PTE has not been well studied in modern cohorts and its risk factors are poorly understood. METHODS: We retrospectively reviewed severe TBI patients from a prospective database treated from 2002 through 2018 at a single level 1 trauma center, excluding those with alcohol withdrawal seizures, prior epilepsy, or death at discharge. Using definitions from the International League Against Epilepsy, a seizure was defined as a witnessed clinical event confirmed to be a seizure by the treating healthcare team or electroencephalography, and PTE as two or more seizures. We collected timing of seizures, follow up, demographic, clinical information, and Glasgow Outcomes Scale (GOS). Because more severe injury predisposes to both death and PTE, we developed a competing hazards model to identify independent predictors of PTE. RESULTS: We identified 393 patients after exclusions. The rate of early seizures within 7 days was 2.8% (11). The cumulative incidence of PTE was 11% by one year and 25% overall, with 29% of PTE patients requiring multiple anti-seizure medications.Patients with PTE had more severe injuries with higher rates of unreactive pupils, subdural hematomas, obliteration of the basal cisterns, skull fractures, Marshall CT classification, decompressive hemicraniectomies (DHC), lobectomies, and shunting, as well as increased length of stay (LOS), younger age, and intracranial infections (all p < 0.01). Our adjusted model identified age (subdistribution hazard 0.75; 95% confidence interval [95% CI]: 0.62 – 0.93]), DHC (3.3; 95% CI: 2.0-5.5), lobectomy (2.2; 95% CI: 2.6 – 4.2), and infection (1.9; 95% CI: 1.1 – 3.3) as predictors of PTE. Patients with PTE had worse functional outcomes (GOS 1-3 versus 4-5) at 6, 12, and 24 months (p < 0.03). CONCLUSION: The incidence of PTE after severe TBI is higher than historical cohorts suggest and continues to increase though fifteen years post-trauma. Patients with PTE have more severe injuries with worse functional outcomes.