Abstracts

Rationale: Rasmussen Encephalitis (RE) is a rare, chronic, progressive disorder characterized by drug-resistant focal seizures, sometimes evolving into epilepsia partialis continua (EPC). RE leads to progressive motor and cognitive decline, often necessitating hemispheric disconnection (HD) with a high risk of significant disabilities, especially in late-onset RE due to reduced neuroplasticity. Responsive neurostimulation (RNS) offers an alternative approach. However, data on the use of RNS for RE or focal motor status epilepticus are limited, and no reports were found on RNS specific for RE-related EPC.


Methods: A 17-year-old right-handed male presented with sporadic left frontal focal seizures, which evolved into EPC affecting the right side of his face within a year. MRI revealed FLAIR hyperintensities with corresponding volume loss, PET hypometabolism, and SPECT hypoperfusion in the left frontal lobe, insula, mesial temporal lobe, caudate, and anterior lentiform nuclei. fMRI confirmed left hemispheric dominance. Intracranial EEG (icEEG) captured multiple seizures originating from the left perisylvian cortical region with frontal operculum involvement. Cortical mapping was unsuccessful due to EPC causing dramatic slurring and pausing of speech. A brain biopsy showed chronic inflammation confirming RE diagnosis. Despite best medical treatment, his EPC progressed to constant 24-hour right lower face and tongue twitching, severe dysarthria, progressive cognitive decline, and frequent head deviation with right upper extremity involvement. HD and resection were rejected due to the risk of debilitating deficits. Instead, RNS was chosen. During surgery, severe subdural adhesions from previous grid placement were encountered. Using previous icEEG data supplemented by intraoperative ECoG guidance, two strip electrodes were positioned over the left face motor cortex.

Results: The RNS registered near-continuous epileptiform discharges and frequent long episodes (LE). A lead-to-lead stimulation pathway was initiated to counteract regional cortical seizures. Adjustments in stimulation parameters resulted in >50% reduction in the spread of EPC to the right shoulder and neck during the 30-month follow-up period, though facial EPC persisted despite LE count decreasing from 4000-6000/day to 600/day (although the detection threshold has changed over time). Currently, the stimulation is at 6 mA (3 µC/cm², 200 Hz, 160 μs, 100 ms bursts) with battery voltage at 3.03 V. Low-frequency stimulation (7 Hz with 5000 ms) trials were also tolerated. Cognitive function, speech, and motor abilities remained stable. No device- or stimulation-related adverse effects or complications occurred despite ongoing rituximab treatment. No variability in LE was noted with periodic rituximab infusion, but initiation of cenobamate caused an additional decrease in LE.

Conclusions: RNS may help maintain cognitive and motor stability in RE-related EPC cases where HD is not feasible, particularly in dominant hemispheric RE. Further research is needed to determine if a corticothalamic RNS approach could provide additional efficacy in treating EPC.

Funding: None.