Is there a Profile of Spontaneous Seizure-Alert Pet Dogs? A Survey of French People with Epilepsy

Abstract

Despite controversies and the lack of research, dogs are empirically selected and trained to perform as service dogs, in relation to the dogs’ and future owners’ characteristics. We assessed the characteristics of both humans and dogs in an unbiased population (not selected or trained) of spontaneous seizure-alert by pet dogs and investigated whether we could replicate previous findings. We addressed a self-reporting questionnaire to French people with epilepsy. We analyzed the general characteristics of the humans and pet dogs and their behaviors that could alert their owner before a seizure. In addition, we used the Monash Canine Personality Questionnaire refined to evaluate pet dogs’ personality through five different traits, and the Monash Dog-Owner Relationship scale to assess human–dog relationships. In line with previous reports, we found no particular factor, either pet-, people- or epilepsy-related that could be associated with the presence or absence of alert behaviors. Alert behaviors and circumstances were explored and three different alert patterns emerged. In terms of personality, seizure-alert pet dogs scored significantly higher than non-alerting dogs for the traits “Motivation” and “Training Focus” and lower for “Neuroticism”. The owner–dog bond score was significantly higher for seizure-alert dogs than for non-alerting dogs.

The Untrained Response of Pet Dogs to Human Epileptic Seizure

“…A further anomaly was the unexpected similarity of the dogs' reactions to the seizure and post-seizure odours, despite the post-seizure sample having been taken six hours after seizure. This outcome appears to contradict the findings of two recent studies (Maa et al 2021;Catala et al 2020). The former report 3-4 dogs distinguished between seizure and inter-seizure samples (3 hours post-seizure), with a probability of 93.7% (Maa et al 2021).…”

“…These findings have since been strengthened by a study which reports that seizures are associated with specific VOCs, which can be detected by trained seizure alert dogs more than an hour prior to a seizure's manifestation (Maa et al 2021). In a separate investigation, further support has emerged from reports that a distinct VOC seizure-related profile has been detected by ion flow tube spectrometry (SIFT-MS) up to four hours in advance of a seizure (Catala et al 2020). Taken together therefore, the results of this study and those of the published articles referred to, provide compelling support for the hypothesis that untrained dogs respond to epileptic seizures.…”

“…This inconsistency may be understood however, by considering the potential impact of environmental differences in sampling. For example, the volunteers in both the Maa et al (2021) study and that of Catala et al (2020) were all patients receiving treatment for epilepsy in medical centres with excellent ventilation and wellestablished clinical attention to cleanliness and hygiene. In the period following a seizure therefore, it is conceivable that seizure scent could have quickly dissipated or that the patients may have been given a change of clothes or been moved to a different ward.…”

“…In the period following a seizure therefore, it is conceivable that seizure scent could have quickly dissipated or that the patients may have been given a change of clothes or been moved to a different ward. In any of those circumstances, residual seizure odour would have been unlikely to linger, thus implying that the post-seizure samples used by those two studies, Maa et al (2021) study and Catala et al (2020), more accurately represented that phase of seizure. On the contrary, the sampling procedures used by the three epilepsy volunteers in this current study were not clinically derived, having been secured in the volunteers' own homes before, during and after a seizure event.…”

“…The emergence of three recently published studies provides compelling evidence to support this hypothesis; for example, report that medical detection dogs were successfully trained to alert on seizure-specific odour(s) and later Catala et al (2020) attest that data retrieved from SIFT-MS chemometric analysis illustrated the presence of pre-seizure-associated VOCs. Similarly, Maa et al (2021), delineate that seizure detection dogs demonstrated a predictive ability for seizures of up to 90 minutes prior to a seizure event.…”