Epilepsy is one of the most common chronic neurological syndromes in dogs (1) and has serious implications on the quality of life (QoL) of both the dogs and owners (2–6). Regarding QoL, as assessed by owners of dogs with idiopathic epilepsy (IE), a recent study of our group demonstrated clear breed-related differences (6). QoL of both owner and dog was scored significantly lower by owners of Border Collies (BCs) compared with owners of other dog breeds, such as Golden Retriever or Labrador Retriever (6). The frequency of epileptic seizures (ES) and cluster seizures (CS) in Golden Retrievers and Labrador Retrievers was markedly lower than in BCs (6). The semiology in these three breeds is comparable (7), but the outcome is noticeably different. Half of all male BCs with IE that develop ES at 1 to 2 years of age die or are euthanized due to the lack of adequate seizure control (6, 8). As a genetic factor is suspected in numerous breeds with a high prevalence of IE (7), genetic studies have been performed in a number of breeds (9–12), including the BC (13). Several possible genetic associations have been found in this breed (13).

Anti-seizure medication (ASM) plays a central role in the treatment of canine IE. The efficacy of registered ASMs varies. Treatment success in dogs with IE is reported in 70 to 80% with phenobarbital (14, 15), 60% with bromide (14), and 50% with imepitoin (16–18). Like in humans with epilepsy, up to 20% to 30% of all epileptic dogs do not respond favorably to registered ASMs (19) and require additional or replacement therapy with unlicensed ASMs such as levetiracetam (20, 21), gabapentin (22), zonisamide (23), or felbamate (24).

Seizure-precipitating factors (SPFs) (also termed “triggers” or “provocative factors”) have been studied and reported in both humans (25–27) and dogs with IE (28). In humans, several SPFs have been identified, such as stress, age, gender, sleep deprivation, and weather (26, 29–33). In dogs stress, hormones, sleep deprivation, and the weather have been reported as SPFs (28, 34). The reduction of stress by itself does not prevent ES but can reduce the number of seizures.

Dogs are likely the most social companions of humans, although there are breed differences. The BC is described as a very sensitive dog¹ which has just recently been studied by comparing the Husky (an independent worker) and the BC (cooperative worker) (35). BCs have a strong focus on their owners, and this may have a clear influence on their and their owners' stress level. Importantly, veterinarians should be aware of their own impact on owners of dogs with IE (36).

In this article, we describe six unrelated BCs with IE, of which we believe that it is, in part, a genetic disorder, with several remarkable SPFs. By avoiding these SPFs, the outcome of these dogs improved dramatically. These cases illustrate that recognizing the SPF(s) and taking measures with regard to the management thereof can lead to a major reduction in ES frequency or even achieving seizure freedom.

The dogs described in this manuscript were all referred to one of the authors for consultation, management of IE, and/or further diagnostic testing. All six dogs are purebred BCs with a pedigree of the English Kennel Club,² Dutch Kennel Club,³ or the International Sheep Dog Society.⁴ All owners provided footage of their dog having an ES. The videos showed signs consistent with generalized tonic–clonic seizures (GTCS) (37). The possibility of focal ES evolving to become generalized could not be excluded based on these videos. All owners participated in two earlier studies published by our group (6, 34). The clinical workup was performed using the criteria published by the International Veterinary Epilepsy Task Force (IVETF) (38). All dogs underwent a physical neurological examination, CBC, serum chemistry, fasting and post-prandial bile acids, and urine examination to exclude metabolic causes of ES (reactive seizures) [tier I level of confidence that was idiopathic (presumably genetic) epilepsy] (38). In four dogs, a tier II level of confidence was reached [an unremarkable magnetic resonance imaging (MRI) study (Canon^® Vantage Elan, 1.5 Tesla) of the brain]. In two of these dogs a CSF examination was performed and found unremarkable (38). All blood examinations for the first three cases were performed at the University Laboratory of the Department of Clinical Sciences, Faculty of Veterinary Medicine, Utrecht University (Utrecht, The Netherlands). The last three cases were performed at the IDEXX laboratory Europe BV (Hoofddorp, The Netherlands). All owners agreed to participate in this study and the earlier studies by informed consent.

We encountered six BCs suffering from IE in which remarkable observations were made regarding SPFs. The first dog became seizure-free once the owner started with music therapy and a learning game. The dog had been unresponsive to the chosen treatment and apparently only became seizure-free after the start of music therapy and gaming. When the owner was forced to stop the learning game, the dog had a seizure again and replaced the bromide with phenobarbital (14, 41). It is possible that the phenobarbital is, in this dog, less effective than the potassium bromide, but this is in contrast to what has been published earlier (14, 41). It appears that this dog only became seizure-free after she started with the game of object retrieval. It can be coincidental, but only when the owner stopped the game of object retrieval, the dog started, after 8 years of being seizure-free, to seizure again.

In case number 2, the surviving dog stopped seizuring once the other dog had died. It is unlikely that the surviving dog was suffering from psychogenic non-epileptic seizures (PNES) (42), as this has not, to the best of authors' knowledge, been described in dogs. PNES are considered to be personality pathology, psychiatric illness, and experiential and behavioral manifestations of depression (42). Moreover, we hypothesize that the relation between these two dogs, or the possible stress this situation has caused for the owners, caused such a level of stress that both dogs became refractory. Chronic stress has been described as a possible SPF in humans (43, 44).

The third case illustrates most likely the sensitivity of the BC breed. We concluded after our work-up that this dog had an idiopathic (presumably genetic) epilepsy. Hence, the dog was genetically prone to develop epilepsy, but the environment of the dog was the SPF to show the seizures. Again, this is unlikely to be a dog with PNES. After relocation, this dog became seizure-free and behaved normally. It is possible that the new owner missed a single seizure, but as the dog was suffering from CS, it is not very likely that they missed complete clusters.

In case 4, it is possible that the stress the dog encounters in the busy city is an SPF which is absent on the island. It is also possible that the owner plays a role. The owner may be more relaxed on the island. However, whatever the SPF is, the result was remarkable. Both cases 4 and 5 kept on having seizures, but in case 4, it appeared to have a relation with the domestic environment, and in case 5 with the work of the owner. In the latter case, there seems to be a direct relation between the level of owner interaction and seizures in his dog. In case 6, it was not just the act of throwing and playing with a ball but rather where the owners would throw the ball. In cases 4, 5, and 6, the owners tested these relationships repeatedly and found consistent results, which seem to contradict that these observations are mere chance.

Stress as an SPF has been described in several studies addressing epilepsy in humans (25–27, 30, 32). The number of studies on dogs is limited. Stress was identified as one of the SPFs in a small number of dogs, according to Forsgård et al. (28), and our group identified stress as a factor in a study with 116 BCs (34). Moreover, most likely, there are breed differences as well. In a recent study, we observed a clear breed (6). We hypothesize that although epilepsy in many breeds is most likely of genetic origin (7), there are striking differences in possible SPFs for the different breeds. Dogs, that are judged by the breeders and breed clubs as more sensitive in nature, like the BC, Australian Shepherd and Belgian Shepherd, may be more susceptible to SPFs as described than less sensitive breeds such as the Dutch Partridge dog, Labrador retriever and Golden retriever. This might even have implications for genetic studies as phenotyping may be more difficult.

Just recently, it was observed that the relationship between the owner and veterinary surgeon plays a role in the management of dogs with IE (36). These six cases illustrate that we also need to take into account the domestic situation of the dog. There is an influence of both the owner and the environment. The owner or environment does not cause epilepsy but maybe (part of) an SPF.

The limitation of this study is the mere fact that it is a very small group of dogs. However, they are all purebred BCs, and in all six dogs, we identified a possible SPF.

Veterinarians treating IE in dogs need to consider breed, their own relation with the owner, and the domestic situation or environment in which the dog is housed. Preferably larger prospective studies are needed to further investigate the nature and role of SPFs in dogs with epilepsy.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical approval was not required for the studies involving animals in accordance with the local legislation and institutional requirements because the study describes six referred patients. Written informed consent was obtained from the owners for the participation of their animals in this study.

PM: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Writing–original draft, Writing–review and editing. KS: Data curation, Investigation, Validation, Writing–review and editing.