Abstracts

Rationale: 79-yo white male, a high-functioning mathematics professor emeritus and an accomplished flutist, underwent a left cochlear implant due to profound long-standing bilateral sensorineural hearing loss, L 0%, R 10% by audiology, with excellent outcome. However, five months after the surgery, he developed bouts of severe prolonged whole-body shaking immediately following certain auditory stimuli such as loud noises, voices, but not music. When the implant was off, almost any noise (except classical music) delivered to right ear caused these spells, and with implant set at 0.3 or above of maximum intensity, left ear was also susceptible. Previously active and athletic, he had to abandon most of his activities, since these frequent events were very disabling, terminating with a day-long holocephalic headache and prostration. He kept his implant mostly off, and his right ear firmly plugged. He was referred for evaluation by an epileptologist.Methods: Routine EEG was obtained. Long-term video EEG monitoring was performed, with multiple typical events captured, provoked incidentally, by environmental noises, or purposefully, by examiners. SPECT scans, ictal and interictal, were obtained, and an interictal FDG-PET study was done. SSEPs and BAERs were recorded with and without auditory provocation. MRI and MEG were not possible due to the implant.Results: Routine EEG revealed left temporal wicket waves and an asymmetry suggestive of left temporal breach rhythm. Video EEG monitoring captured episodes of mostly axial severe dysrythmic myoclonus with onset less than one second after auditory stimulus such as shutting the door, with his implant on or with his right ear unplugged. He was also reliably provoking spells by talking, and by other expected, non-startling noises. Spells came in clusters of 2-10 minutes of dysrythmic coarse whole-body jerking, accompanied later into the spell by lacrimation and facial flushing. He was able to talk between jerks. SPECT scan showed increased ictal uptake in right posterior insula and in the right mesial structures. PET revealed hypometabolism interictally in right mesiolateral temporal cortex. SSEPS were normal, BAERs were uninterpretable.Conclusions: The combination of SPECT and PET findings, as well as a definite clinical delay between the stimulus and the event, suggests that the spells are indeed due to cortically induced myoclonus with right temporal and insular foci. Areas 41 and 42, primary auditory cortex, probably 'ignite' the initial reflex axial myoclonus response; autonomic events such as flushing and lacrimation suggest secondary insular involvement. Given the temporal association with the cochlear implant, a causal relationship is not excluded, possibly with sensitization and epileptic recruitment of primary auditory cortex.