Abstract
BACKGROUND: Despite considerable previous research, to what degree white matter lesions (WML) may be epileptogenic remains unclear. Therefore, the decision of initiating treatment with antiseizure medication (ASM) can be challenging in patients with only WML on neuroimaging. In this prospective study we assessed whether the prevalence, localization or severity of WML impact the risk of seizure recurrence in patients aged 60 years or older after first-time seizures. METHODS: Data was analyzed from 168 patients, aged ≥ 60 years-old who had experienced a previous unprovoked seizure and had either a potentially epileptogenic lesion or WML on neuroimaging. The frequency of seizure recurrence was documented after 6, 12, and 24 months. Pearson´s chi-square test of independence (categorical variables) and the independent Student´s t-test (continuous variables) were used to analyze intergroup differences. Binary logistic regressions were calculated to examine the influence of WML locations as a predictor of seizure recurrence. Kaplan-Meier survival analyses and log-rank statistics were performed to determine the cumulative recurrence rates between the groups. RESULTS: Fifteen patients had only potentially epileptogenic lesions on neuroimaging (EPI) and 93 showed WML only (OWML). Sixty patients showed both of them on neuroimaging (EWML). Frontal and parieto-occipital were the predominant WML locations. Neither severity nor location of WML had a significant impact on recurrence rates. The two-year cumulative probability of becoming seizure-free was significantly lower in the EPI group compared to the EWML (χ(2) [1] = 4.425, p = 0.035) and the OWML group (χ(2) [1] = 13.094, p < 0.001). A significant association between interictal epileptiform discharges in EEG and seizure recurrence was found in OWML patients (p = 0.004). CONCLUSION: We could not find any association between prevalence, severity or location of WML and seizure recurrence after first seizures in the elderly. Therefore, treatment with ASM should be started with caution in those patients. Our results show a trend of WML not having epileptogenic potential, but further studies are needed to get better evidence. TRIAL REGISTRATION: ClinicalTrials.gov Protocol Registration and Results, NCT06836687, AZ 199/17, release: 03/19/2024 retrospectively registered. https://register. CLINICALTRIALS: gov/prs/beta/studies/S000EBC700000025/recordSummary.