Abstracts

Resective Epilepsy Surgery (RES) is a highly effective treatment for drug-resistant epilepsy that remains underutilized in the older adult population, in whom the incidence of epilepsy continues to rise. One factor driving this underutilization is concern for surgical complications such as subdural fluid collections (SFCs), which are more often encountered in older adults and may require intervention. This retrospective analysis examines the occurrence and potential risk factors of subdural fluid collections, including hygromas and hematomas, in a large single-center cohort of older adults undergoing RES.

We included patients aged > 60 years with drug-resistant epilepsy who underwent RES at a tertiary epilepsy center between 2000 and 2021. Patients who had previous neurosurgical resection or did not have MRI imaging within one year prior to RES were excluded. Automated SPM12 tissue segmentation was performed on preoperative T1-weighted MRIs to calculate brain parenchymal fraction (BPF), a normalized measure of whole brain volume. Forty two percent of BPF values were missing due to unavailable preoperative MRIs and were therefore estimated using multiple imputation methods. Clinically significant SFCs were identified on postoperative MRIs by a staff neurosurgeon. Age at surgery, sex, BPF, epilepsy duration, pre-operative intracranial EEG monitoring, and resected lobe were analyzed with regard to post-RES SFC development (with subgroups for hematoma and hygroma) and Engel 1 classification.

We identified 86 patients (41M, 45F) with a median age of 64 years (range:60-79). RES lobe distribution was as follows: 70 (81.4%) temporal, 8 (9.3%) frontal, 3 (3.5%) parietal, 4 (4.7%) multilobar, and 1 (1.2%) hemispherectomy. Of 24 patients who developed clinically significant SFC post-RES (28% of cohort), 13 (54%) developed hematoma, 7 (29%) developed hygroma, 4 (17%) developed both, and 7 (29%; 8% of study cohort) required surgical evacuation. 8 (73%) hygromas and 7 (41%) hematomas developed within 1 week following RES. 69 (80%) patients achieved Engel 1 outcome with an average follow-up time of 2.65 ± 0.23 years.

SFC was associated with older age at surgery (p=0.049). No association was found between SFC and Engel 1 outcome (p >0.05). Subdural hematoma was associated with older age at surgery (p=0.041) and lower BPF (p=0.042). Hygroma was not found to be associated with any of the tested factors. Pre-operative intracranial EEG monitoring was associated with lower rates of SFC (8.3% vs 33.9%, p=0.016) and hematoma (4.4% vs 25.4%, p=0.030). In multivariable analyses, no independent associations were found for SFC, hygroma, and hematoma occurrence.

In this cohort of older adults, RES demonstrated a high seizure freedom rate but approximately one in four older adults developed postoperative SFC. While SFC development does not impact seizure freedom, almost one in ten older adults required SFC drainage. Further investigation is required to identify potential characteristics that mitigate the risk of SFCs in older adults with epilepsy who require surgical treatment.