2-[18F]FDG PET patterns in people with drug resistant mesial temporal lobe epilepsy with hippocampal sclerosis

Abstract-Text (inklusive Referenzen und Bildunterschriften)

Objectives: To identify the value of presurgical 2-[18F]FDG patterns in patients with drug resistant mesial temporal lobe epilepsy (MTLE) due to hippocampal sclerosis (HS) who underwent a selective amygdalohippocampectomy (SAHE) based on post-surgical long-term seizure outcome.

Methods: In this retrospective study we included the data of 172 patients with drug resistant MTLE due to HS who underwent presurgical evaluation and subsequent SAHE between July 1st, 1994 and December 31st, 2019 at the Epilepsy Surgery Service of the Medical University of Vienna and were followed at least for 24 months. The surgical outcome was classified according to the ILAE Outcome classification (class 1 and 1a, meaning seizure freedom, SZF, and class 2-6 meaning seizure recurrence, SZR). 2‑[18F]FDG PET studies were analyzed visually and semiquantitatively using the Brain Registration and AnalySiS (BRASSTM) tool. We conducted a t-test based analysis of the glucose metabolism of the 47 ROIs.

Results: 97 patients (47 female; 57 with left MTLE) with a mean age of 39 years (range 16 ± 62 years) were included in the final analyses. The mean age of onset of seizures was 13.7 ± 11 years (min 0.3, max 47); the mean disease duration constitutes 25.28 ± 14.16 years (min 3.46, max 60.98). At the latest follow-up visit 11.0 ± 5.5 years (min 2, max 20) after surgery a total of 69.1% of patients were seizure-free (ILAE class 1 and 1a).

Visual 2‑[18F]FDG PET imaging analysis: Patients who had no hypoperfusion in the mesial temporal area or who showed an involvement of extratemporal regions revealed a trend towards lower successful surgical outcome rates (only 58.3% achieved SZF compared to patients with concordant temporal lobe tracer uptake abnormalities (72.6%), although results were not significant (p = .208).

Quantitative 2‑[18F]FDG PET imaging analysis: 73 patients were included in the quantitative analysis. Patients with left and right MTLE were analyzed separately. In left MTLE patients the ipsilateral extratemporal hypometabolism frontal medial (p =.023), and left orbitofrontal cortex (p =.035) and in right MTLE patients the right inferior parietal cortex (although not significant p =.07) were associated with a worse surgical outcome.

We could not find predictive or prognostic values of any clinical parameters at any time of postoperative assessment on the seizure outcome.

Conclusion: The extent of temporal hypometabolism does not appear to have an influence on the surgical outcome. Extratemporal involvement had the biggest influence on a negative surgical outcome in both left and right MTLSE patients" groups. The fact that patterns of abnormal extratemporal 2-[18F]FDG metabolism tend to correlate with worse surgical outcomes should prompt clinicians into viewing focal epilepsy (especially MTLE due to HS) as a brain network disease, in which connectivity to extratemporal brain regions seems to play a big role.

Acknowledgements:

We would like to thank Doris Daxberger, Satu Messeritsch and Katja Sebesta for their support in data collection.