Cumulative seizure occurrence and the predictors of seizure in low- and high-grade gliomas
DOI:
https://doi.org/10.54029/2023swzKeywords:
brain tumor, glioma, epilepsy, cumulative incidenceAbstract
Objective: We aimed to determine the cumulative seizure occurrence, 1-year, and 5-years of preoperative seizures in gliomas, and an update on the predictors.
Methods: This was a retrospective analysis of 239 patients with histopathologically confirmed gliomas in University Malaya Medical Centre, Malaysia, between 2008-2020. Kaplan-Meier curves were used to determine the cumulative incidence of seizures. Logistic regression was performed to determine the predictors of preoperative seizures.
Results: A total of 80/239 patients (33.5%) had preoperative seizures. They were more commonly seen in those with younger age of presentation (40.0% in those <40 years old vs. 26.9% in ≥40 years old, p<0.05) and low-grade tumors (42.2% vs. 28.8% in high-grade tumors). Those with cortical involvement, especially the frontal lobe, or without focal deficit, headache, nausea, or vomiting were more likely to have seizures preoperatively. Logistic regression identified three significant predictors for preoperative seizure: absence of focal deficits at presentation (OR 6.090, 95% CI 3.110-11.925, p<0.001), cortical location (OR 3.834, 95% CI 1.363-10.786, p<0.05) and absence of headache at presentation (OR 2.487, 95% CI 1.139-5.431, p<0.05). The cumulative incidence of seizure was 29% at one year and 32% at 5-year for gliomas. Specifically, the seizure incidence was higher in low-grade gliomas (39% at 1-year) and certain tumor types such as ganglioglioma (50%), oligodendroglioma (48%), and astrocytoma (45%).
Conclusion: The cumulative incidence of preoperative seizures in low-grade gliomas and certain tumor types is high. The predictors included cortical involvement and absence of focal neurological deficit or headache at presentation.