Abstracts

Rationale: A seizure dog is a dog that is trained to respond to a seizure in seizure disorder patients. However, until recently, there was no high-quality scientific evidence to support a seizure-sniffing capability in dogs. Using canines specially trained to detect 3 volatile organic compounds (VOCs) associated with electrographic seizures in a separate study, Maa et al. (Epilepsy & Behavior 2021) demonstrated that a significant number of seizures appear to be associated with the unique scent profile prior to clinical-electrical onset of the seizure itself in an EMU setting; The dogs had a 93.7% probability of correctly distinguishing between ictal and interictal sweat samples. The purpose of this abstract is to report on the safety and tolerability of having a scent-trained seizure alert service dog present for daily rounds an inpatient EMU.

Methods: A golden retriever, Reggie, underwent intensive training as a seizure alert dog by his owner and service dog handler (author MM), who is also an epilepsy fellowship-trained neurologist and attending in OHSU’s 5-bed EMU. The training was based on the methods employed by the techniques developed by Dr. Jennifer Cattet Ph.D., of Medical Assistance Canines, but adapted by use of seizure samples from EMU seizure patients. Patients were consented to donate ictal/interictal scent profile samples during EMU admissions for canine training purposes. Training consisted of approximately 8 months of daily 30 to 60 minute training sessions. The dog was trained to indicate the presence of seizure scent, after which then the dog will lead the handler to the location of the source of the ictal scent profile to confirm which patient in the EMU may be seizing.

Results: After the training period described above, Reggie joined MM on daily EMU rounds when on service in the EMU each month since August 2021. During that period, Reggie interacted with 4 to 6 EMU patients per week and there were no patient complaints or safety concerns related to the dogs’ presence. In fact, many patients reported positive interactions with the dog. If they owned a dog, they reported that being able to interact with Reggie during the hospital admission provided them with the canine attention they would get daily if at home allowing them to maintain positive morale during their EMU admission. Another consistent fact was that if Reggie was interacting a patient receiving a diagnosis of PNES, they would be accepting of their diagnosis in all cases. There were only two incidences of concern: (1) nursing was uncomfortable when a patient invited Reggie to come up on the hospital bed, and (2) a patient with psychogenic non-epileptic seizures (PNES) expressed frustration that the dog did not alert for a convulsive PNES and escalated the intensity and frequency of her events until she received her diagnosis of PNES with Reggie actively interacting with her and she was accepting of her diagnosis.

Conclusions: Presence of a seizure alert service dog on an inpatient EMU is safe and tolerable to patients. Future research will examine Reggie’s accuracy in indicating presence of seizure scent and a survey of patient perceptions of an EMU seizure dog and the dog’s responses to their seizures.

Funding: Not applicable