Abstracts

Rationale: Qualitative Positron emission tomography (PET) has been an integral component of the pre-surgical evaluation of patients with drug-resistant temporal lobe epilepsy (TLE), with regional temporal hypometabolism consistently predicting favorable outcomes. The association of Quantitative PET analysis remains to be validated. Our aim is to assess quantitative PET in the evaluation of patients with TLE and evaluate its association with epilepsy surgery outcomes.Methods: 218 patients who underwent temporal lobectomy for intractable TLE between 2003 and 2008 at the Cleveland Clinic were reviewed. PET findings were assessed qualitatively via visual analysis and quantitatively using SPM8 software and PET Scenium for the Standardized Uptake Value (SUV) of sub-regions of interest. Images were co-registered onto 3D-MPRAGE and PET images were spatially normalized to the Montreal Neurological Institute space and segmented. Finally, an asymmetry index (AI) quantifying brain hypometabolism was calculated. A favorable seizure outcome was defined as Engel 1 after surgery.Results: 49% were female and 78% were adults. The mean age at onset was 15.7 years (SD=14.9), mean monthly seizure frequency was 21.7 (SD= 38.5), 68% had a history of GTC seizures. MRI was abnormal in 84%, with MTS representing the most common finding (47% of all patients). Prior history of generalized tonic-clonic seizure (OR: 2.8, p=0.003) and having more than one seizure semiology (OR: 2.9, p=0.003) were associated with seizure recurrence. On qualitative PET analysis, neither bilateral/non-localizable metabolism, nor unilateral temporal hypometabolism correlated with seizure outcome in the overall cohort. A subanalysis stratifying for the side of surgery, extension of temporal hypometabolism to the ipsilateral extra-temporal regions was associated with worse seizure outcome when compared with focal temporal hypometabolism in patients who had a LEFT temporal resection (OR: 3.49; p=0.03). With RIGHT-sided surgery, no such correlation was noted. Quantitative PET analysis was evaluated by means of t-test to correlate the AI with Engel outcome. Statistically significant difference was found in LEFT-sided surgeries with a higher mean AI of the hippocampus and amygdala and a lower mean AI of the precuneus in the favorable outcome (Engel 1) group. In RIGHT-sided surgeries, a significant difference was noted with a higher mean AI of the anterior cingulate/paracingulate and and a lower mean AI of the temporal pole in the favorable outcome group (Engel 1).(table 1).Conclusions: Our results confirm the favorable prognostic value of restricted temporal hypometabolism noted on qualitative visual PET analysis in patients undergoing epilepsy surgery. Quantitative PET may offer complimentary information. The hemispheric variation of relevant quantitative regions of hypometabolism may be due to differences in epileptic networks or differences in surgical strategies by side of surgery. Our data provide promising results supporting further exploration of any additional prognostic value provided by quantitative PET in the context of TLE surgery.