Abstracts

Rationale: Tumor related epilepsy (TRE) is relatively common as 25-50 % of cerebral tumors manifest with seizures. Basic characteristics of TRE are poorly understood, and the degree of seizure control with antiepileptic drugs (AEDs) has not been well described. We reviewed our series of patients with TRE to better characterize the seizure types and degree to which they are medically refractory. Epilepsy related to brain tumors is compared to a larger series of symptomatic localization-related epilepsy (SLE). Methods: We queried the UVA epilepsy database to identify the study cohort with TRE. Basic demographics, seizure type, and EEG characteristics were recorded. To estimate the severity of epilepsy, we noted the number of AEDS currently and ever used, the occurrence of status epilepticus, and the duration or presence of seizure freedom. Tumor pathology and the type of surgical intervention is noted for each patient. Findings were statistically compared to SLE patients without TRE.Results: The TRE cohort includes 76 patients with a mean age of 44.7 years and 37 of the patients are males. Median seizure frequency was 0.16 seizures/mo while using a median of 2 antiepileptic medications. 34 patients were seizure free at the time of last followup. The common seizure classifications for the group include: complex partial with secondary generalization (20), complex partial seizures only (10), and simple partial seizures only (13). Of the 36 patients with preoperative EEG testing, 22 had epileptiform discharges. 47 patients had postoperative EEG performed and 27 of these demonstrated epileptiform discharges. Tumor location was: frontal lobe (25), temporal lobe (21), frontotemporal (4), mesial temporal (3), parietal (10), and 2 parietal-occipital. Tumor pathologies are: 14 meningiomas, 35 gliomas, 4 DNETS, 1 ganglioglioma, and a variety of other uncommon pathologies. 18 of the gliomas were low-grade and 11 were high-grade. Preoperative EDs did not correlate with seizure freedom but a trend was noted toward lack of postoperative EDs and seizure freedom (P=0.09). Rate of seizure freedom did not correlate with high grade gliomas, low-grade gliomas, or history of status epilepticus. There was no difference in seizure freedom between gliomas and meningiomas. The DNET and oligoastrocytoma tumor types were more likely to come under seizure control with surgery and antiepileptic medications. There was no relationship between tumor type and the development of status epilepticus. Most of the patients (52) were using 1 or 2 antiepileptic medications. In comparing SLE to TRE, a difference in the type of seizure represented. Conclusions: We did not find a statistical correlation between EDs found on pre or postop EEG and the development of seizure freedom. While most cases of TRE are treated with one or two AEDS, only about half of the cases result in seizure freedom. The predominant types of seizures noted with cerebral tumors are simple and complex partial seizures. Secondary generalizations are common. Mixed seizure types are also not uncommonly seen. TRE is different from other types of lesional epilepsy.