Seizures and CNS hemorrhage: spontaneous intracerebral and aneurysmal subarachnoid hemorrhage.

BACKGROUND: Convulsive and nonconvulsive seizures frequently complicate acute brain injury particularly central nervous system hemorrhages and both have been associated with poor outcome. No randomized controlled trials have been conducted to guide decisions on seizure prophylaxis or treatment. The magnitude of additional injury from nonconvulsive seizures remains controversial and some argue that these epileptiform patterns primarily represent surrogate markers of severely injured brain. The deleterious effects of seizures on brain recovering from a recent injury have to be weighed against the deleterious effects of antiepileptic medications when making decisions on prophylaxis and treatment. REVIEW
SUMMARY: Currently seizure prophylaxis is not generally recommended for patients with spontaneous intracerebral hemorrhage (ICH) or aneurysmal subarachnoid hemorrhage (aSAH). However, short-term prophylaxis (during the acute critical illness) is commonly instituted for patients in whom seizures would likely lead to additional injury such as herniation or rebleeding. ICH or aSAH patients with seizures at the onset of their hemorrhage, patients with ICH in close proximity to the cortical surface, and aSAH patients with a poor clinical grade (poor neurologic examination and/or thick cisternal blood) are at high risk of seizures, especially nonconvulsive, and are frequently kept on short-term prophylaxis. Convulsive seizures occur in 7% to 17% of patients with spontaneous ICH and in between 6% and 26% of those with aneurysmal aSAH. These should be treated as soon as possible regardless of the underlying causative factors. Nonconvulsive seizures are seen in about 20% of patients with ICH and in 8% to 18% of those with aSAH. It is controversial how aggressively to treat nonconvulsive seizures.
CONCLUSION: Convulsive and nonconvulsive seizures are frequent after central nervous system hemorrhage and treatment is controversial, particularly for nonconvulsive seizures. Randomized controlled trials need to be conducted to better allow evidence-based guidelines for these common neurologic conditions.