Epilepsy is rarely thought of as a downstream consequence of survivable brain injuries, yet for millions of stroke, trauma, and hemorrhage survivors, it represents one of the most disabling long-term complications. Understanding precisely how much each injury type elevates risk — and whether those elevations are comparable — has clinical consequences for monitoring, prophylaxis discussions, and patient counseling across neurology and rehabilitation medicine.
This systematic review and meta-analysis pooled data from 120 observational studies published through March 2026, quantifying epilepsy risk across four distinct acquired brain injury (ABI) categories. Ischemic brain injury (IBI) showed the strongest association with subsequent epilepsy, with a pooled odds ratio of 2.40 (95% CI: 2.17–2.65), and location-based subclassifications within this group drove particularly elevated odds. Traumatic brain injury (TBI) followed closely, yielding an OR of 2.28 (95% CI: 2.14–2.43) with consistent elevations across all TBI subtypes assessed. Hemorrhagic brain injury (HBI) conferred a meaningful but comparatively lower OR of 1.53 (95% CI: 1.43–1.63), with intracerebral and unspecified hemorrhagic events contributing most prominently. Immune-related brain injury (IrBI) also reached statistical significance, rounding out a finding that no major ABI category is epilepsy-neutral.
The magnitude of ischemic injury's association is particularly notable given that stroke is already among the leading causes of adult-onset epilepsy — this meta-analytic confirmation consolidates what had been a fragmented and sometimes contradictory literature. The 120-study pool is a genuine methodological strength, lending statistical power that individual cohort studies cannot achieve. That said, the reliance on observational designs means residual confounding — particularly from pre-injury neurological vulnerability, medication use, and injury severity gradations — cannot be excluded. The heterogeneity across study populations, diagnostic criteria for epilepsy, and follow-up durations likely inflates confidence intervals and may partially explain the variation across subtypes. Whether these associations reflect shared epileptogenic mechanisms, such as cortical scarring, neuroinflammation, or glutamatergic excitotoxicity, or simply injury severity differences, remains an open question. For rehabilitation and neurology teams, these pooled estimates reinforce the case for sustained seizure surveillance in ABI survivors well beyond the acute phase.