Dogs, in addition to the loyal companionship they offer to their human owners, were trained over the last century to perform a multitude of difficult tasks based on their out of norm olfactory and visual capabilities.
On a large scale, dogs have been trained by police and rescuers to detect illicit substances and retrieve victims during extreme rescue operations. On a much smaller scale, special training programs were developed to enable dogs to be of great assistance to disabled individuals offering them a sense of independence and liberation from human assistance.
In this context, service dogs were trained to assist individuals with visual or hearing impairment as well as individuals with a number of mental disorders. Lately, various schools successfully trained dogs to detect hypoglycemic episodes in patients suffering from diabetes; thus, enabling them to take the necessary safety measures before the occurrence of a syncope.
In addition to this, various cases were reported where domestic dogs showed alerting behaviours before the occurrence of a seizure in epileptic patients. This solicited the interest of multiple researchers fuelling a multitude of research with an aim to verify these claims and uncover the scientific reason behind this aptitude. So, are dogs gifted with the ability to sense epileptic seizures or are these occurrences just mere coincidences?
A number of research articles published in the American Academy of Neurology Journal supported the aforementioned claim by conducting exploratory studies on diversified samples. The chosen samples were divided into two major categories: dogs that underwent special training and domestic dogs claimed to be showing attention-seeking behaviours before the occurrence of a seizure.
A study done by Strong et al. (2002)  reported that dogs that underwent a special training program showed specific and reliable alerting behaviour such as excessive barking and jumping for a duration of 30 minutes on average before the occurrence of seizures. On the other hand, pet dogs were able to alert the owner at times ranging between 30 seconds and 45 minutes before a seizure with an average of 3 minutes. A second study performed by Kirton et al. (2004)  on a different sample further asserted the findings showing that the dogs reacted on average 31 minutes before an onset for tonic-clonic seizures for trained dogs and 2.5 minutes for untrained house pet dogs before the ictal event both with an accuracy of more than 80% and a specificity of 100% as no false alerts were reported during the study.
Even though the claim that dogs can predict upcoming seizures was verified by the aforementioned studies, the single or multiple cues triggering this response were still unknown due to the high variability in the timing of the alerting behaviours in relation to the time of the episodes. Hence, a number of plausible theories arose to explain the phenomenon without excluding an equally plausible multimodality hypothesis.
The 300 million olfactory receptors of dogs compared to 6 million in a human nose, are very sensitive to even the slightest variations in the smell of the atmosphere surrounding them. Some scientists claim that dogs may base their alerting behaviour before the occurrence of a seizure solely on that olfactory power. During a seizure, the human body releases neurotransmitters and hormones that alter the body smell impairing the normal activity of the nervous system and offers an olfactory cue to the dog. This is a similar process to the smell modifications that occur with hypoglycaemia in diabetes for which dogs have been proven to respond consistently . Other scientists believe in the fact that dogs respond to a visual cue from the patient where an impending seizure could be linked to a behavioural or postural change as well as a change in motor activities.
This can enable dogs who have mastered the art of reading human body language to notice this change and predict the episode. In addition to this, studies conducted by Haut et al. (2007)  and Litt et al. (2007)  have shown that some epileptic patients have the ability to recognise an upcoming seizure by experiencing prodromes. A prodrome is a pre-ictal phenomenon defined as alterations preceding a seizure. Hence, this can point to the eventuality that dogs rely on prodromes that occur between 5 minutes and 24 hours before the seizure to predict the occurrence of an episode.
A number of surveys showed that the presence of seizure alerting dogs in households with epileptic children significantly increased the patients overall quality of life (QOL). According to Kirton et al. (2008) , 82% of families with trained SRD reported a major improvement to their QOL while the remaining 18% experienced a moderate improvement. Additionally, a study conducted by Strong et al. (2002)  monitoring the patients’ tonic-clonic seizure frequency over a 48-week period (12-week baseline, 12-week training period, and 24-week follow up) showed that by comparing the baseline frequency and the last 12-week follow-up, an average reduction of 43% in seizures was noted. Nonetheless, some studies suggest that this decrease is strongly associated with a mood change and can be correlated to the effects that dogs have on the mental health of their owners . Hence, the specificity of the effect of alert dogs as well as the pet effect in the context of epilepsy is still to be confirmed.
 A Seizure Response Dog (SRD) is a dog that displays attention-seeking and alerting behavior during a seizure episode of a human epileptic patient
 The Quality of Life (QOL) in the medical domain is the patient’s ability to enjoy normal life activities
Youssef, Consultant, Leyton UK
- Strong V, Brown S, Huyton M, Coyle H. Effect of trained Seizure Alert Dogs on frequency of tonic-clonic seizures. https://doi.org/10.1053/seiz.2001.0656
- Kirton A, Wirrell E, Zhang J, Hamiwka L. Seizure-alerting and -response behaviours in dogs living with epileptic children. https://doi.org/10.1212/WNL.62.12.2303
- Hardin DS, Anderson W, Cattet J. Dogs Can Be Successfully Trained to Alert to Hypoglycaemia Samples from Patients with Type 1 Diabetes. https://doi.org/10.1007/s13300-015-0135-x
- Haut SR, Hall CB, LeValley AJ, Lipton RB. Can patients with epilepsy predict their seizures? https://doi.org/10.121 2/01.wnl.0000252352.26421.13
- Litt B, Krieger A. Of seizure prediction, statistics, and dogs: a cautionary tale. https://doi.org/10.121 2/01.wnl.0000255912.43452.12
- Kirton A, Winter A, Wirrell E, Snead OC. Seizure response dogs: Evaluation of a formal training program. https://doi.org/10.1016/j.yebeh.2008.05.011
- Johnson EK, Jones JE, Seidenberg M, Hermann BP. The relative impact of anxiety, depression, and clinical seizure features on health-related quality of life in epilepsy. https://doi.org/10.1111/j.0013-9580.2004.47003.x
- Catala A, Cousillas H, Hausberger M, Grandgeorge M (2018) Dog alerting and/or responding to epileptic seizures: A scoping review. https://doi.org/10.1371/journal.pone.0208280
Read more on the same subject: