Authors :
Presenting Author: David Vaughan, MBBS PhD FRACP – Florey Institute
Marty Bryant, BSc – Florey Institute; Victor Zhang, MBBS – Neurology – Austin Health; Yuliya Perchyonok, MBBS – Austin Health; Greg Fitt, MBBS – Austin Health; Chris Tailby, MPsych PhD – Florey Institute; Graeme Jackson, MD PhD – Florey Institute
Rationale:
Pathological subtypes of unilateral temporal lobe epilepsy (TLE) can be defined on MRI and show characteristic electrographic seizure dynamics (Zhang et al., 2022). We investigated whether there are also corresponding patterns of ictal perfusion in these groups and how cerebral perfusion evolves into the early post-ictal phase.
Methods:
Parametric mapping was applied to 105 paired ictal-interictal SPECT studies from 90 adults with unilateral TLE acquired at Austin Hospital in Melbourne between 2005 and 2022. All were temporal lobe seizures with no tonic-clonic phase, injected during the ictal (n=74) or early post-ictal phase (n=31). Injection time ratio was measured using video-EEG, as the time from earliest electroclinical seizure onset, divided by the total seizure duration until electrographic offset. MRI was used to determine TLE subtype (see Table 1). Statistical parametric mapping was performed separately for the ictal and post-ictal studies, with epileptic side, injection timing and SPECT camera included as regressors. To investigate, the ictal-to-postictal transition a region-of-interest (ROI) analysis was performed using an anatomical MRI-based parcellation.
Results:
Mean ictal injection time was 48 ± 20 s.d. seconds from seizure onset, and post-ictal injection time was 21 ± 20 seconds from seizure end, with no difference between groups. The hippocampal sclerosis group had longer seizures, mean duration 88 seconds vs 68 seconds, as previously reported (Zhang et al., 2022).
Similar patterns of ictal and post-ictal perfusion were seen across TLE subtypes. Overall, ictal hyper-perfusion (Fig. 1a) was lateralized toward the epileptic side at the anterior temporal lobe and adjacent brain regions, and involving the thalami and cerebellar dentate bilaterally. Prominent hypo-perfusion occurs during seizures throughout association cortex bilaterally, especially regions of the default mode network. In the early post-ictal phase (Fig. 1b), hyper-perfusion persists bilaterally at basal ganglia, cerebellum and ipsilaterally at piriform and brainstem. Post-ictal hypoperfusion extensively affects the ipsilateral cerebral cortex in the temporooccipital, frontal, parietal and cingulate regions.
ROI analysis (Fig. 1c) shows ictal hyper-perfusion peaking in the mid-ictal phase at the ipsilateral temporal lobe and basal ganglia, then post-ictally returning to baseline after approximately an additional +50% of the seizure duration.
Conclusions:
Perfusion during unilateral temporal lobe seizures reflects lateralised temporal cortico-subcortical network recruitment, but also more widespread deactivation of other cortical networks bilaterally. This pattern is similar for each of the MRI-defined TLE subtypes indicating a common progression of seizures. Understanding the group-level evolution of perfusion during temporal lobe seizures may assist interpretation of both individual SPECT studies and the relationship to ictal symptoms.
Funding:
The authors receive support from the Australian Government under the Medical Research Future Fund, Victorian Government Operational Infrastructure Support Grant, Victorian Biomedical Imaging Capability (VBIC), and the MASSIVE high-performance computing facility.