# Dynamics of early electroencephalographic patterns and epileptic seizures in acute intracerebral hemorrhage: A prospective controlled study

## TL;DR

This study identifies early EEG patterns that predict seizures in patients with brain hemorrhages, suggesting early EEG monitoring could help guide personalized treatment.

## Contribution

The study introduces rhythmic and periodic EEG patterns as early biomarkers for seizures in acute intracerebral hemorrhage, offering a potential window for intervention.

## Key findings

- Rhythmic and periodic EEG patterns (RPPs) were associated with increased seizure risk and preceded seizures by over 12 hours in most cases.
- Patients with RPPs who received levetiracetam had a lower seizure rate compared to those untreated.
- Incorporating RPPs into a predictive model improved seizure risk prediction, though the improvement was not statistically significant.

## Abstract

Acute symptomatic seizures (ASyS) occur in up to 30% of patients with intracerebral hemorrhage (ICH) when continuous electroencephalography (cEEG) is used, potentially worsening outcomes. Identification of early EEG biomarkers of ASyS may help guide personalized antiseizure medication (ASM) prophylaxis. Here, we aimed to describe early interictal EEG patterns, their dynamics, and their association with seizure risk, considering the effect of prophylactic levetiracetam.

This prospective analysis used data from the PEACH phase 3 trial (2017–2020), which enrolled adults with acute spontaneous supratentorial ICH, randomized to receive levetiracetam or placebo. Patients underwent systematic 48‐h cEEG within 48 h of symptom onset. Electrographic seizures and interictal EEG patterns were analyzed using standardized terminology of the American Clinical Neurophysiology Society. Associations between rhythmic and periodic patterns (RPPs) and seizures with clinical and radiological variables were assessed using univariate analyses. We also conducted exploratory testing of the CAV (cortical involvement, age < 65 years, volume > 10 mL) score for predicting ASyS, incorporating RPPs and ASM exposure.

Forty‐two patients were included (median [Q1–Q3] age = 72 [60–79] years, 29% women), 19 in the levetiracetam group. Interictal EEG abnormalities were common and not influenced by ASM, including background asymmetry (73%), sporadic epileptiform discharges (62%), and RPPs (52%). RPPs were associated with ICH volume (p = .039) and cortical involvement (p = .003). Among patients with RPPs, 50% developed ASyS (20% in those treated with ASM vs. 75% in untreated patients, p = .030). Most patients (91.7%) with seizures had RPPs that preceded seizures, in >90% cases by 12 (Q1–Q3 = 4–25) h. Integrating RPPs into the CAV model led to an improvement of ASyS prediction (area under the curve = .949 vs. .918, p = .53) that was statistically nonsignificant.

RPPs are strong markers of ictogenesis in acute ICH and precede ASyS, thus offering a potential therapeutic window. These findings support the use of early cEEG for risk stratification and personalized ASM prophylaxis.