Epilepsy is the most common chronic canine neurological syndrome and has serious implication on the quality of life (QoL) of both the dogs and owners (1–4). Determining the etiology of canine epilepsy is important and has been the focus of the International Veterinary Epilepsy Taskforce (IVETF), leading to the creation of manuscripts addressing its classification, terminology, and diagnostics (5, 6).

The prevalence of canine epilepsy is estimated to range from 0.6 to 0.75 percent (7, 8), the majority of which is idiopathic, genetic or presumed genetic (9). There is a clear breed predisposition in hunting dogs, shepherds and molosser dogs (9, 10), and several recent studies focussed on the genetics of epilepsy (11–16). But there is also growing interest into the impact of epilepsy and the use of anti-seizure medication (ASM) on owners and their dogs. The goal of ASM is to reduce the number and severity of the seizures. However, the efficacy of various registered ASM's varies and ranges from 70 to 80% with phenobarbital (17, 18), 60% with bromide (17), and just 50% with imepitoin (19–21). Hence, up to 20–30% of all epileptic dogs do not respond favorably to registered ASM's and require multi-drug treatment with unlicensed ASM's such levetiracetam (22, 23), gabapentin (24) or zonisamide (25). These combined therapies introduce considerable side-effects (18, 26).

Hence, several studies have been published investigating the effect of both epilepsy and ASM treatment on the QoL of dogs and their owner (1, 4, 26, 27). Dogs may show behavioral and cognitive changes such as fear/anxiety, aggression, abnormal perception, inattention, and excitability (3, 28, 29). Not only will this have a negative effect on the QoL of the dog but also of the owner (3, 28).

But not all breeds respond equally to ASM. For instance, the outcome in BCs differs clearly from the general population (30). Up to 50% of all young male BCs are euthanized before the age of two as they do not respond to the ASM's (30, 31). Similar observations have been noted for Australian Shepherds (32), Italian Spinone (33), and Rottweiler (34). When there is a breed specific response to ASM's, there is usually also a breed specific QoL score for both dog and owner.

The aim of this pilot study was to investigate the QoL score of dogs suffering from idiopathic epilepsy and their owners and if possible, investigate whether a breed specific difference exists.

Owners, either from the Netherlands or Belgium, were asked to participate in a web based SurveyMonkey questionnaire. Owners were approached using social media, direct e-mailing, adverts on canine epilepsy fora and indirectly by spreading the questionnaire with the help of veterinarians in the Netherlands and of Belgium.

The questionnaire consisted of forty questions, based on earlier questionnaires used to evaluate owner perception of dogs with epilepsy in Ohio and Glasgow (27, 35) and the proposed Quality of Life assessment of Wessmann et al. (2). The questions focussed on general aspects of the dog—including breed, age and gender, the number and type of seizures (focal, generalized or other), effect of therapeutic modalities, side effects, the QoL of the dog, and the QoL of the owner. All questions were accompanied with an explanation of the used medical terms, a website page was available with an explanation in Dutch and examples of a tonic-clonic seizure and focal seizures were made available on YouTube. An example of the questionnaire (translated from Dutch into English) can be found in Appendix 1. The questionnaire contained both open and closed questions and some questions allowed multiple answers. The questions addressing the QoL were scored on a scale of 1 to 10. Two groups of questions were formulated. One group addressed the QoL of the dog, the second group the QoL of the owner.

Questions addressing the QoL of the dog:

Questions addressing the QoL of the owner:

All completed questionnaires were collected with the online survey development software SurveyMonkey within a timeframe of 2 months. Questionnaires were only analyzed if the dog had at least more than two, idiopathic epileptic seizures as described earlier (5) and the questionnaire had been completed. Owners were advised to contact us if there were questions and, on a voluntary base, could sent us the medical record of their dog. Results were analyzed with version 28 of the IBM SPSS Software and included descriptive statistics, chi-squared tests, t-tests and Pearson's correlation tests. Means ± standard deviations were calculated for all dogs and compared with the results of specific breeds. Each breed was compared with the entire group of dogs minus the breed to be tested. Breed scores were only calculated if at least six dogs were of the same breed. Results were considered statistically significant if the p-value was < 0.05.

A total of 622 owners (partly or completely) filled in the questionnaire. Based on the selection criteria, 402 questionnaires were suitable for further analysis.

The assessment of QoL in epileptic dogs and their owners has been previously studied in various countries (1, 3, 4, 26–28, 36) but limited information is available for breed specific QoL scores. To date, two breed specific studies addressing QoL score have been published. One was performed in 2016 in the Italian Spinone (33) and a second one recently in the Border Collie (30). Furthermore, breed comparisons for QoL of dogs and owners are, to the authors knowledge, limited. The aim of this pilot study was to investigate breed specific QoL scores for dogs suffering from idiopathic epilepsy and their owners. During a period of 2 months, 402 questionnaires were obtained from the owners of several dog breeds. In total, 14 breeds were overrepresented, which enabled a comparison of breed specific QoL scores of both owners and their epileptic dogs. Several interesting breed specific observations were noted. Compared with all breeds/dogs in this pilot study, the breed with the poorest statistically significant outcome appeared to be the Border Collie. This breed suffers not only from early onset seizuring (Figure 1), a low survival rate due to the severity of the seizures and the development of cluster seizures (Figures 2, 3), and up to 90% of all dogs required ASM and no < 47% was on a combination-therapy of ASMs. Sadly enough, this breed also had the lowest, statistically significant, QoL for both dog (Table 2) and owner (Table 3). Similar observations were noted in 2010 by Hulsmeyer et al. and just recently confirmed by our group (31). Wessmann et al. (1) reported that QoL was predominantly influenced by seizure frequency rather than the severity or clustering. Although not statistically significant, this was also the observation in this study. The Belgian Tervuren dog had the highest number of seizures (Table 1) compared to all other breeds/dogs, although this was not coupled with a high owner severity score (Table 3). In this study, the Border collie owners scored the severity of the seizures as high (Table 2).

An analogous poor outcome, like that of the Border Collie, was expected for the Australian Shepherd. An earlier study describing the characteristics of idiopathic epilepsy in this breed (32) was comparable with a similar study in the Border Collie (31). Although the owners of the Australian Shepherds score their dogs' and their own QoL low, it was not statistically significant compared with the other breeds. Surprisingly, the survival outcome in the Australian Shepherd was clearly different to that of the Border Collie (Figure 2). The only clear difference between these two breeds was, in this study, the age of onset which was in the Border collie much lower. But this was in two other breeds also statistically significant: the French Bulldog and the Boxer. The survival time in both breeds was also low, although this difference was not statistically significant. As far as the Border Collie is concerned this observation has just recently been confirmed in another study (30).

The only breed, in this study, with a similar, but not statistically significant, low survival rate was the French Bulldog (Figures 1, 2). Surprisingly, the French Bulldog, Boxer and Border Collie all had a statistically significant early age of seizure onset (Table 1). The number of seizures was low in the French Bulldog and Boxer, but the severity of seizures was scored as high, although this difference was not statistically significant for the French Bulldog. In contrast to the earlier study of Wessmann et al. (1), we found that there is most likely a combination of factors that influence the QoL in these breeds rather than just the number of seizures. Next to the seizure numbers, early age of seizure onset and the severity of the seizures were also predictors of outcome in some breeds. The number of clusters was, except for the Drentsche Patrijshond, apparently not important. The Drentsche Patrijshond, a Dutch hunting dog, had the highest age of onset and the highest number of cluster seizures. However, the owners scored their dog's and their own QoL comparable to that of other dog owners/dogs. Hence, clustering appears to not be a predominant factor for QoL.

The QoL was scored statistically significantly higher for the Chihuahua, Dachshund and Golden retriever than the other breeds. The QoL score of the owners of Golden retrievers and Labrador retrievers was least affected by the epilepsy (Table 3). Owners of the Chihuahua, Dachshund and the Drentsche Patrijshond also reported an overall low impact of their dog's seizures on their QoL. Surprisingly, the number of seizures was high in the Chihuahua and Dachshund compared to the Labrador retriever, Golden retriever and Drentsche Patrijshond (Table 2). However, the number of seizures was not the predominant factor influencing the QoL score.

The effects of ASM side effects on QoL were also assessed. The acceptability of ASM side effects was statistically significantly higher in the Chihuahua and Golden retriever (Table 2). In other breeds, the ASM side effects appear to have no major influence on the QoL score of the dog. The same could be concluded for the number of visits to the consulting veterinarian/specialist (Table 1). There were statistically significantly fewer evening or night emergency visits in the Golden retriever and Labrador retriever than in other breeds, while the most emergency visits were reported in the Border Collie and French Bulldog (Table 2).

In the Border Collie, French Bulldog, and Boxer, costs of care were scored as least acceptable, although this was not statistically significant (Table 3). These costs were not due to the number of needed (evening/nightly) veterinary visits, as these were comparable to all other breeds/dogs (Table 1) but the owners scored them as higher compared to the other breeds/dogs (Table 3).

An important factor that influences the QoL scores of an owner is the ability to lead a normal life. The owners of Border Collies scored being able to leave the dog alone statistically significantly lower compared to other breeds, and this question also scored low in breeds such as the Boxer and French Bulldog. Owners of these breeds scored their own QoL low. This question also scored low for the Chihuahua owners, although this breed had the highest QoL score (Table 2).

This seems to suggest that there is not only a breed specific, but also an owner-specific impact on QoL. Owners choose their dogs for various reasons. A Border Collie owner may plan to use their dog for other tasks than a Chihuahua owner. Expectations, or disappointments, will have a clear influence on the outcome. Owners usually choose their dogs based on breed-specific traits, which may be strongly impacted by epilepsy-related factors. It is interesting to note that owners of some breeds accept disease-related changes more easily than others. This warrants further research.

This study has limitations. A significant limitation of any on-line questionnaire is that the population studied is self-selected. There is a possibility that owners with good QoL and good seizure control are underrepresented in the study. As there is a clear breed difference we expect that this bias is minimal. And although we were able to collect 402 questionnaires, the number of questionnaires obtained were low for five of the 14 breeds. Obtaining more filled-in questionnaires may have allowed a broader breed-specific analysis. However, seven breeds were clearly overrepresented to provide first insights into breed specific QoL scores for both dogs and owners.

This is the first study reporting breed-specific differences in the perceived QoL of dogs with idiopathic epilepsy and their owners. In this study, Border Collies and their owners had statistically significantly lower QoL scores than other dog breeds. What influences the QoL score differs for each breed, although the age of seizure onset, the number of seizures and their severity have a negative influence on the perceived QoL. Owners of Labrador retrievers and Golden retrievers reported a minimal impact of the seizures on their pet's QoL, with Chihuahua and Dachshund owners also reporting an overall good QoL of their dogs. Costs of medical care, veterinary visits, and the inability to leave the dog alone may be of negative influence in breeds that already have a negative outcome.

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

Ethical review and approval was not required for the animal study because it is a questionnaire web-based study. Written informed consent was obtained from the owners for the participation of their animals in this study.

PM was responsible for the study conception and data collection in co-operation with co-authors. Statistical analysis, data analysis, and manuscript writing were performed by MH and PM. PM supervised data analysis and manuscript editing, in co-operation with MP, MD, SB, and KS. All authors contributed to the article and approved the submitted version.