Preoperative [18F]FDG-PET Metabolic Patterns and Surgical Outcome in Mesial Temporal Lobe Epilepsy: A Retrospective SPM-Assisted Atlas-Based ROI Analysis
TL;DR
Larger, adequately powered cohorts are needed before the potential clinical utility of quantitative FDG-PET analysis in presurgical evaluation can be determined, and the findings should be considered exploratory and hypothesis-generating rather than evidence that quantitative FDG-PET can discriminate surgical outcomes.
Abstract
Background: Interictal [18F]FDG-PET is commonly used for temporal lateralization in mesial temporal lobe epilepsy (mTLE). Whether atlas-based quantitative PET measures differ according to surgical outcome among otherwise concordant unilateral MRI-positive mTLE patients remains uncertain. This study compared preoperative FDG-PET ROI measures between 2-year Engel outcome groups using SPM-assisted spatial normalization followed by AAL atlas-based ROI analysis. Methods: Among 319 screened temporal lobe epilepsy surgery patients, 94 had evaluable preoperative FDG-PET images, and 62 patients with at least 2 years of postoperative follow-up were included in the total cohort. The primary quantitative analysis was restricted to 54 unilateral MRI-positive patients with hippocampal atrophy and mesial temporal sclerosis on MRI (Engel I, n = 45; Engel II–IV, n = 9). Cerebellum-normalized AAL ROI uptake ratios were compared using Mann–Whitney U tests in conventional left/right and ipsilateral/contralateral analyses. FDR-adjusted q values, effect sizes, and Hodges–Lehmann median differences with 95% confidence intervals were calculated. Reference-region and global mean normalization sensitivity analyses were also performed. Results: In the unilateral MRI-positive subgroup, 10 of 80 conventional left/right ROIs showed unadjusted differences between Engel I and Engel II–IV patients, predominantly involving left frontal-orbitofrontal, medial frontal, middle cingulate, and inferior parietal regions. After ipsilateral/contralateral transformation, eight ROIs showed unadjusted group differences, mainly in contralateral frontal-orbitofrontal regions, with additional cingulate and inferior parietal involvement. Uptake ratios were lower in Engel II–IV patients in all candidate ROIs. However, none of the ROI findings survived FDR correction. Raw cerebellar uptake did not differ between outcome groups, and global mean normalization did not yield FDR-corrected ROI findings. Conclusions: Quantitative ROI analysis identified several extratemporal metabolic differences between postoperative outcome groups; however, none remained significant after correction for multiple comparisons. These findings should therefore be considered exploratory and hypothesis-generating rather than evidence that quantitative FDG-PET can discriminate surgical outcomes. Larger, adequately powered cohorts are needed before the potential clinical utility of quantitative FDG-PET analysis in presurgical evaluation can be determined.