Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2022 Nov 11;64(2):59–68. doi: 10.1111/jsap.13568

Effect of an intervention of exercise on sleep and seizure frequency in idiopathic epileptic dogs

K Grady 1, S Cameron 1,✉, S P Kent 1, H Barnes Heller 2, M M Barry 1
PMCID: PMC10099787  PMID: 36368312

Abstract

Objective

The goal of this study was to compare sleep and seizure frequency between epileptic dogs prescribed a 20% activity increase and epileptic dogs not prescribed an activity increase.

Methods

Sixty‐nine dogs receiving anti‐epileptic drug therapy were enrolled in a 6‐month prospective, randomised, placebo‐controlled clinical trial with an intention‐to‐treat analysis. A canine activity monitoring device was used to measure activity levels and sleep scores.

Results

Using an intention‐to‐treat analysis, the treatment group had an average of 0.381 more seizures per month (95% CI: 0.09 to 0.68) compared with the control group, although the difference in seizure days per month was not statistically significant. In a subgroup analysis of dogs whose activity increased by at least 10%, partial compliers had 0.719 more seizures per month (95% CI: 0.22 to 1.22) and 0.581 seizure days per month (95% CI: 0.001 to 1.16) compared with the control group. Sleep scores increased by 1.2% in the treatment compared with the control group (95% CI: 0.2 to 2.3%).

Conclusions

Seizure frequency and sleep score increased slightly, but significantly, in dogs with idiopathic epilepsy prescribed an increase in activity, compared with a control group.

INTRODUCTION

Epilepsy is the most common chronic neurological disorder in humans and dogs, and the most common cause of seizures in dogs is idiopathic epilepsy (Podell et al. 1995, Berendt et al. 2004, Chandler 2006). In both dogs and people with epilepsy, anti‐epileptic drugs (AEDs) are the current mainstay of treatment; however, adverse effects are commonly associated with their use. In dogs, these adverse effects include polyuria, polydipsia, polyphagia, ataxia, restlessness and lethargy and are identified as a major contributor to decreased perception of quality of life by dog owners (Chang et al. 2006, Wessmann et al. 2016). There is also the continued cost of medications, drug monitoring and the commitment of giving medications on a regular schedule. Approximately 37%, 11% and 6% of dogs respond to first, second and third‐line AEDs, respectively, with a successful response to treatment defined as more than 50% reduction in seizure frequency (Packer et al. 2015).

Due to the limitations of AEDs, adjunctive non‐pharmacological treatments have been proposed to decrease seizure frequency and increase quality of life (Packer et al. 2019). In people with epilepsy, increased physical activity has been associated with higher quality of life and lower levels of anxiety, depression and medication side effects (Häfele et al. 2017). In addition, exercise has been associated with lower seizure frequency in some studies (Eriksen et al. 1994, Nakken 1999) and can reduce epileptiform electroencephalogram (EEG) discharges in people (Esquivel et al. 1991, Nakken et al. 1997, Arida et al. 2010). In rodent studies, exercise also delays and reduces the onset of pilocarpine‐induced seizures (Peixinho‐Pena et al. 2012, Campos et al. 2016, 2017). A variety of mechanisms, involving B‐endorphins, endogenous glucocorticoids, or melatonin, have been proposed to explain these findings, but they are currently poorly understood (Carrizosa‐moog et al. 2018). In addition, the positive effect of exercise is not consistent across studies. In particular, two interventional studies found that exercise treatments had no effect on seizure frequency in epileptic patients (Nakken et al. 1990, McAuley et al. 2001) and in two survey‐based studies, some participants reported seizures in association with exercise (Nakken 1999, Ablah et al. 2009).

In addition to seizures, people with epilepsy are more likely to have sleep disturbances (Kataria & Vaughn 2016). However, the effect of exercise on sleep quality is unclear in epileptic patients. One interventional study found that epileptic women reported fewer sleep disturbances following an exercise program (Eriksen et al. 1994), while another study on epileptic children found no effect of assigned exercise on objective sleep outcomes (Do et al. 2020).

In dogs with epilepsy, the effect of exercise on either seizure frequency or sleep quality is unknown. Consequently, the primary aim of this study was to investigate the effect of an increase in exercise on seizure frequency in dogs with idiopathic epilepsy. Our hypothesis was that dogs would have a significant reduction in seizure frequency during periods of prescribed physical activity compared with control periods of normal activity. Our secondary aim was to determine whether exercise improved sleep quality in dogs. We hypothesised that prescribed physical activity would lead to a significant reduction in seizure frequency, as well as improved sleep quality measured by decreased restlessness, in dogs with idiopathic epilepsy.

MATERIALS AND METHODS

The study design was a prospective, randomised, placebo‐controlled clinical trial with an intent‐to‐treat analysis over a 6‐month period. The protocol was approved by the Institutional Animal Care and Use Committee at the University of Wisconsin – Madison, School of Veterinary Medicine (#V006156). Owners gave informed consent for their dog to be enrolled as a study participant.

Study population and enrollment

Dogs of any breed, sex or weight were included in the study if they were between the ages of 1 and 9 years old, had been diagnosed with presumptive or confirmed idiopathic epilepsy (using Tier I or Tier II criteria as described by the International Veterinary Epilepsy Task Force Consensus Proposal, including having their first seizure documented between ages 6 months and 6 years of age), had a seizure history more than 1 year and had a history of more than two generalised seizures. In addition, the dog's seizure frequency had to be more than 1 seizure every 3 months, on average, while on AED therapy. Any drug or drug combination of AEDs was acceptable. Dogs with other co‐morbidities, such as hypothyroidism or diabetes mellitus, not related to epilepsy or AED effects were excluded.

At the time of enrolment, intake forms were completed by the owners with detailed information regarding the dog's seizure history, seizure frequency, AED medications (including drug name, dose and frequency), rescue protocols, other medications or supplements regularly administered, regular routine (including sleep schedule) and food (brand and amount per day). Dogs were randomly assigned to an exercise treatment group or control group using a computer‐based random number generator.

A canine activity monitoring device (CAMD) was attached to each dog's collar (FitBark™, Kansas City, MO). The CAMD uses a 3D accelerometer to track activity levels and takes multiple readings per second, which is then stored in 1‐minute data segments – called Bark Points and referred to here as activity points (WWW Document 2018b). The device transmits data to a paired smartphone via Bluetooth® (Bluetooth SIG, Kirkland, WA) technology. Each owner downloaded the CAMD application (app) to their smartphone, enrolled their dog via the app and registered our hospital as their veterinarian. This step allowed our study team to have continuous access to the dog's activity data through the manufacturer's website. Owners were instructed to leave the CAMD on the collar and to leave the collar on their dog except for baths or swimming. They were asked to sync the CAMD to their smartphone or tablet by opening the app at least one time per week. In addition to activity, sleep scores were calculated automatically by the CAMD as the percentage of time spent restless over a 4‐hour window of expected sleep, which could be tailored to the owner's usual sleep schedule. For example, if a dog was moving or restless for 1 of the 4 hours during the sleep window, then a sleep score of 75% would be automatically calculated. The same study team members recorded all data entries, including activity monitoring and sleep scores, for the duration of the study.

Study design

Once enrolled, owners were instructed to continue their normal routine with their dogs for the first 3 months and were asked to keep a detailed seizure log during the entire study period. The length of seizure, time of seizure and type of seizure (either partial or generalised) were recorded by the owner.

First study phase (0 to 3 months): for all dogs, baseline (normal) activity and sleep scores were monitored by the CAMD. At the end of the 3 months, weekly activity averages were calculated for each individual dog.

Second study phase (4 to 6 months): owners of dogs in the activity treatment group were asked to increase their average weekly activity by 20%. A specific target of activity points was given to the owner for a weekly goal – based on the baseline activity data from their dog. An increase of 20% was chosen based on averaging prescribed activity and exercise programs used in human epilepsy and lab animal models (Ablah et al. 2009; Campos et al. 2017; Carrizosa‐moog et al. 2018; Esquivel et al. 1991; McAuley et al. 2001; Nakken et al. 1997; Peixinho‐Pena et al. 2012). Owners of dogs in the control group were asked to continue their normal routine. All owners were asked to give their dog a placebo (lactose powder) pill PO q24h during the 4‐ to 6‐month study period and were informed that they were receiving either a nutraceutical or placebo pill. This was done to mask the study hypothesis and prevent the control group owners from increasing their dog's activity as well. Periodically, activity levels for the treatment group were checked via the CAMD website, and owners not being compliant with the prescribed activity increase were contacted via a one‐time reminder e‐mail.

Outcome measures

Activity (recorded as activity points) and sleep (recorded as a sleep score) were obtained weekly for each dog from the CAMD manufacturer's website for the 6‐month duration of the study and were recorded by a study team member. Sleep scores were a continuous range from 0% to 100%, calculated as the percentage of time spent moving during a predetermined 4‐hour window. Seizure logs were collected from owners throughout the study period. Number of seizures per month and number of seizure days per month were used as outcome measures.

Data analyses

First, we determined the association between the assigned treatment group and measured activities. A mixed‐effect, linear regression model was fit to the monthly activity points within a difference‐in‐difference framework to estimate the effect of treatment on activity points. This model controlled for age, sex and spay/neuter status and contains random‐effect intercepts for each dog to account for other individual factors. The difference‐in‐difference framework provided a way to compare the differences in treated and control dogs between the before‐treatment and after‐treatment periods (Abadie 2005). Reported activity point averages were estimated by predicting activity points based on the model and averaging over other covariates. Hypothesis testing used the Satterthwaite estimation of degrees of freedom.

Next, we determined the relationship between activity level and seizure frequency. Total number of seizures per month and seizure days per month over the 6‐month study period were obtained for all dogs. A Poisson regression mixed‐effect model controlling for age, sex and spay/neuter status was performed to compare number of seizures per month and seizure days per month incidence between treatment and control dogs. Poisson regression was chosen over Negative‐binomial regression because over‐dispersion was not present in the data.

A sensitivity analysis has been suggested by an anonymous peer reviewer. Within the 3‐month study period, a multi‐variable Poisson regression was built for total number of seizures with covariates for treatment indicator, 3‐month seizure total within the baseline, age, sex and spay/neuter status. The same analysis was then performed for seizure days per month.

Finally, we evaluated the relationship between daytime activity levels and sleep scores. Average sleep scores per month were determined over the 6‐month study period for all dogs. A mixed‐effect, linear regression model was fit to the monthly sleep scores with a difference‐in‐difference framework. This model controlled for age, sex, spay/neuter status and dosages (in mg/kg/day) of eight AEDs (phenobarbital, potassium bromide, zonisamide, levetiracetam – intermediate release, levetiracetam – extended release, gabapentin, topiramate and imepitoin). It also contained a random intercept term for each dog to account for other undetermined factors. Hypothesis testing used the Satterthwaite estimation of degrees of freedom. Additionally, a sensitivity analysis has been suggested by an anonymous peer reviewer. A multi‐variable linear regression was built on 3‐month average sleep score in the study period with covariates for treatment indicator, 3‐month seizure total from baseline, age, sex, spay/neuter status and dosages (in mg/kg/day) of six AEDs.

Sample size analysis was calculated using the standard definition of a successful seizure treatment, which is considered a 50% (or more) reduction in seizure frequency. Previously published studies have shown refractory epileptic dogs to have an average of 2.5 seizures per month, with a standard deviation of approximately 1.8. Therefore, a reduction of 1.25 seizures per month was used as the delta in the power calculation. Using a significance level of 0.05 and a power of 0.8, sample size was calculated to be 34 dogs per group. However, a high drop‐out rate was expected due to the longevity of the study and the high rate of death and possible euthanasia in this subpopulation of epileptic dogs. Therefore, a goal of 40 dogs per group (80 dogs total) was used. All statistical analysis was done in R version 4.0.0 (R Core Team 2020).

RESULTS

Study population

A total of 82 epileptic dogs were enrolled in the study. Three owners did not turn in seizure logs by the time of data analysis and those cases were excluded. Three dogs discontinued AED therapy during the study and were excluded. During the 24‐week study period, six dogs were euthanased – four due to refractory seizures, one due to liver failure, one due to an unknown cause and one dog was removed from the study because the owner was concerned that the CAMD may be causing an increased seizure frequency. These seven dogs were euthanased or discontinued less than 20 weeks into the study and were excluded from the analysis.

Demographic information is summarised in Table 1, including number of dogs per group, breeds, age, sex and number of AEDs received. All dogs received the lactose placebo capsule during months 4 to 6 of the study. Three owners elected not to give the placebo due to perceived allergy, gastrointestinal upset or delivery concerns.

Table 1.

Demographic information

Group Number of dogs Sex Mean age at time of enrollment, years Mean age at time of first seizure, years Mean weight, kg Number of AEDs received Breeds
Control 32 3 M, 1F, 23 NM, 5 SF 5.36 (range: 1.5 to 8.67) 2.85 (range: 0.5 to 7) 27.4 (range: 5.45 to 84) 1 (n=10) 2 (n=10) 3 (n=9) 4 (n=3) Cardigan Welsh corgi (n=1), Border collie (n=2), Boston terrier (n=1), Bulldog (n=1) Bullmastiff (n=1), cocker spaniel (n=2), English springer spaniel (n=3), French bulldog (n=1), German shorthair pointer (n=1), golden retriever (n=2), Jack Russell terrier (n=1), Labrador retriever (n=2), mixed breed (n=10), Pointer (n=1), Rhodesian Ridgeback (n=1), Saint Bernard (n=1), Shiba Inu (n=1)

Treatment

 37 2 M, 1F, 21NM, 13 SF 4.53 (range: 1.33 to 8.16) 2.76 (range: 0.33 to 7) 24.5 (range 5 to 79.2) 1 (n=14) 2 (n=8) 3 (n=12 4 (n=3) Australian shepherd (n=6), beagle (n=1), Border collie (n=3), Cavalier King Charles spaniel (n=2), French bulldog (n=3), German shepherd dog (n=2), German shorthair pointer (n=1), golden retriever (n=1), great Pyrenees (n=1), Keeshond (n=1), Labrador retriever (n=2), mixed breed (n=13), Vizsla (n=1)
Open in a new tab

M Intact male, F Intact female, NM Neutered male, SF Spayed female

Prescribed activity

Compared with control dogs, treatment dogs had an estimated mean increase in activity of 219 activity points [95% confidence interval (CI)=−103 to 541] during the prescribed activity increase, although this did not reach statistical significance (P=0.18, Fig 1). This effect compares the difference (between treatment and control groups) in activity level changes across study periods. Dogs in the assigned activity treatment group in months 4 to 6 had an estimated increase in activity during the exercise prescription period (mean, 7319 activity points; 95% CI=6268 to 8370), relative to their baseline activity level (mean 7264 activity points; 95% CI=6213 to 8314), but this increase was not significant (P=0.62). Thus, the difference‐in‐difference represented an approximately 3.5% increase in activity from the average activity point score for the treatment group. In other words, when owners were asked to increase their dog's activity by 20%, an increase of 3.5% was actually performed.

FIG 1.

FIG 1

Open in a new tab

Difference‐in‐difference plot displaying predicted average activity points during the baseline activity period (months 1 to 3) and during the treatment period (months 4 to 6, dotted line) for the treatment (red) versus control (blue) groups. This plot is controlling for age, sex and neuter/spay status. Compared with control dogs, dogs in the treatment group, on average, had an estimated increase in activity of 219 activity points between the baseline activity period and the prescribed increase in activity period, although this was not statistically significant (SE, 164; CI=−103 to 541; P=0.18)

Control dogs actually had a decrease in activity during months 4 to 6 (mean, 7056 activity points; 95% CI=5997 to 8116), relative to their baseline activity level (mean 7220 activity points; 95% CI=6161 to 8280). These estimates correspond to a decrease of 164 activity points that was not significant (95% CI=−401 to 73, P=0.17).

Seizure frequency

Using an intention‐to‐treat analysis, the difference‐in‐difference was evaluated by comparing the change in seizure frequency between the control and treatment groups. Compared with the control group, the number of seizures increased significantly by 0.38 (CI=0.09 to 0.68) seizures per month in the treatment group (P=0.01), suggesting a slight increase in seizure frequency among epileptic dogs assigned to 3 months of increased activity (Fig 2). The sensitivity analysis provided similar results (P=0.02). The full results from the model are given in Table 2. Among the treatment and control groups, the incidence of seizures per month decreased over the study period. In the treatment group, number of seizures per month decreased from 1.12 (95% CI=0.58 to 2.16) in the baseline period to 1.06 (95% CI=0.55 to 2.04) in the prescription activity period (P=0.61). The sensitivity analysis was similar (P=0.66). Among control dogs, the incidence of seizures per month decreased from 1.12 (SE, 0.38, CI=0.58 to 2.17) in the baseline period to 0.74 (CI=0.37 to 1.41) over months 4 to 6 (P=0.0004).

FIG 2.

FIG 2

Open in a new tab

Average number of seizures over 3‐month period (left) and average number of seizure days per month (right) for the treatment versus control group. Compared with the control group, the treatment group's number of seizures increased significantly by 0.34 seizures per month (P=0.012). There was no significant difference in the change in the seizure day incidence between the treatment and control groups (P=0.48). The red line corresponds to the treatment group; the blue line corresponds to the control group

Table 2.

Results of sleep score, number of seizures per month and the number of seizure days per month for multiple variables

Outcome Sleep scores Seizures Seizure days
Model type Linear Reg Poisson Reg Poisson Reg
Parameter (1) (2) (3)
Constant 0.847 (0.04) 0.217 (0.742) −0.170 (0.654)
Difference in difference terms
Treatment −0.006 (0.017) −0.006 (0.35) 0.060 (0.308)
Treatment period −0.008 (0.004) −0.437*** (0.123) −0.112 (0.146)
(Treatment)×(treatment period) 0.012* (0.005) 0.381* (0.151) 0.127 (0.18)
Baseline covariates
Age (years) −0.004 (0.005) 0.009 (0.088) 0.016 (0.076)
Sex, male 0.012 (0.018) −0.306 (0.362) −0.250 (0.311)
Neutered or spayed 0.028 (0.029) 0.030 (0.548) 0.048 (0.472)
Medications (dose in 1000×mg/kg/day):
Phenobarbital −0.002 (2.708)
Potassium bromide −0.995* (0.455)
Zonisamide 0.834 (0.974)
Levetiracetam (intermediate release) −0.196 (0.138)
Levetiracetam (extended release) 0.114 (0.231)
Gabapentin −0.370 (2.15)
Open in a new tab

Reg, Regression

All coefficients are given on the linear scale with standard errors in parenthesis

Significance values are given as: *P<0.05; **P<0.01; ***P<0.001

The number of seizure days per month was also compared between the treatment and control groups. There was no significant difference in the change in the seizure day incidence between the control and treatment groups (P=0.48, Fig 2). Among treatment dogs, seizure days per month increased from 0.81 (CI=0.45 to 1.45) in the baseline period to 0.825 (CI=0.47 to 1.46) in the activity prescription period (P=0.90). Among control dogs, the incidence rate of seizure days per month decreased from 0.765 (CI=0.43 to 1.37) in the initial baseline period to 0.68 (CI=0.38 to 1.22) in the treatment period (P=0.444).

Partial compliers

Given that few participants complied with the prescribed 20% increase in activity (n=5), a subgroup analysis was performed to compare seizure frequency in dogs with at least a 10% increase in activity during months 4 to 6 (termed “partial compliers”; n=11) compared with control activity dogs. The results from the partial compliers, supports the initial analysis, in that the number of seizures per month were significantly increased in the partial complier group compared with the control group (Fig 3). Additionally, the number of seizure days per month was also significantly increased in the partial complier group compared with the control group (Fig 3). Descriptive data of the partial compliers, treatment group and control groups are shown in Table 3.

FIG 3.

FIG 3

Open in a new tab

Average number of seizures per month (left) and seizure days per month incidence (right) for the partial compliers, defined as dogs in the treatment group who had their activity increased by at least 10%, versus the control group. The partial compliers' number of seizures increased by 0.51 seizures per month, whereas the control group's number of seizures decreased by 0.54 seizures per month. A similar trend was seen in seizure days, where partial compliers increased by 0.49 per month and the control group decreased by 0.05 per month. The orange line corresponds to the partial compliers of the treatment group; the blue line corresponds to the control group

Table 3.

Description of seizures per month and seizure days per month per group

Partial compliers Treatment group Control group
Seizure incidence, treatment period 1.576 1.818 1.354
Seizure incidence, baseline period 1.061 1.829 1.896
Average seizure days, treatment period 1.212 1.279 1.104
Average seizure days, baseline period 0.727 1.234 1.156
Number of dogs 11 37 32
Open in a new tab

Seizure incidence are given as number of seizures per dog, per month

Sleep score

Mean sleep scores were higher for treatment dogs in the 4‐ to 6‐month exercise assignment period compared with control dogs when controlling for age, sex, spay/neuter status and seizure medication dosages. On average, dogs in the exercise treatment group had a statistically significant increase in sleep scores of 1.2% (i.e. better sleep quality) compared with the control group (95% CI for the difference=0.2% to 2.3%, P=0.02, Fig 4). The sensitivity analysis results were similar (P=0.03). In addition to this association, higher dosages of potassium bromide were associated with lower sleep scores (i.e. worse sleep quality) (P=0.03, Table 2). No other AED dosages were positively or negatively associated with sleep score in this population.

FIG 4.

FIG 4

Open in a new tab

Difference‐in‐difference plot displaying predicted average sleep score during the baseline activity period (months 1 to 3) and during the treatment period (months 4 to 6) for the treatment (red) versus control (blue) group. This plot controls for age, sex, neuter/spay status and seizure medication doses. Sleep scores improved by 1.2% on average (i.e. better sleep quality) for dogs in the treatment group versus the control group when age, sex, spay/neuter status and seizure medication dosages were controlled for (CI=0.2 to 2.3%, SE, 0.5%, P=0.0198)

DISCUSSION

Idiopathic epilepsy is the most common chronic neurological condition in dogs (Packer et al. 2014). While AEDs are the mainstay of treatment, their adverse effects in dogs contribute to a decreased perceived quality of life by their owners; in addition, 20 to 30% of epileptic dogs are poorly controlled despite appropriate AED therapy (Chang et al. 2006, Packer et al. 2014, Wessmann et al. 2016). Consequently, there is a strong need to identify adjunctive treatment options to decrease seizure frequency in epileptic dogs. The current study is the first prospective, randomised, placebo‐controlled clinical trial to study the effect of increased physical activity, as measured by a CAMD, on seizure frequency, as well as sleep quality, in dogs with idiopathic epilepsy receiving AED therapy.

CAMD serves as an innovative, objective method to assess physical activity in dogs. These devices not only measure physical activity in healthy dogs with reasonable accuracy (Chan et al. 2005) but have been used to assess response to treatment in pruritic and osteoarthritic dogs (Brown et al. 2010, Wernimont et al. 2018). In epileptic dogs receiving AED therapy, CAMD has been used to measure baseline activity level (Barry et al. 2021) and has been evaluated for use as a method to detect seizure activity (Muñana et al. 2020). Our study is the first to use this device to objectively monitor activity changes in epileptic dogs in response to a prescribed activity treatment. The FitBark™ was chosen as the CAMD for this study, as it has been used to evaluate the activity levels of dogs of many breeds, ages, sex and geographic locations, as well as to study the differences in daily activity in dogs without medical conditions and dogs with osteoarthritis, allergies and obesity (WWW Document 2018a).

Using this device, we found that dogs in the treatment group had their activity increased during the exercise prescription period whereas dogs in the control group had a decrease in activity levels, relative to the baseline period, although the difference was not statistically significant. Humans using pedometers decrease their step counts in the winter months and may decrease their activity over time as the novelty of the pedometer fades (Hamilton et al. 2008, Ho et al. 2013). Consequently, our finding of decreasing activity in the control group may be due to the diminishing novelty of the device, as well as more dogs in both the treatment and control groups starting the study in summer months. The enrolment patterns were consistent between treatment and control groups (Pearson chi‐square test, P=0.735), and so we would expect the treatment group to be exposed to these same factors.

Contrary to our hypothesis, we also found that the seizure frequency increased in the treatment group compared with the control group during the exercise prescription period, suggesting that the prescribed activity increase may have contributed to increased seizure frequency. However, few owners complied with the prescribed 20% increase in activity for their dogs; consequently, a subgroup analysis was performed to investigate seizure frequency among dogs with at least a 10% increase in activity, supporting the possibility that exercise may lead to a small, but significant, increase in overall seizure frequency. The difference in the number of seizure days per month was not significant in the intent‐to‐treat model, suggesting that dogs in the treatment group experienced more cluster seizures per day.

Research regarding the relationship between seizures and physical activity in humans is limited and has yielded mixed results. Several studies report a decrease in seizure frequency with exercise intervention. In one study, women with pharmacologically intractable epilepsy had fewer seizures when participating in aerobic dancing for 15 weeks (Eriksen et al. 1994), and in two other studies, yoga training was associated with reduced seizures in patients with refractory epilepsy (Lundgren et al. 2008, Sathyaprabha et al. 2008). However, other research has found no effect of either 4‐ or 12‐week physical training programs on seizure frequency (Nakken et al. 1990, McAuley et al. 2001). In contrast to this research, some epileptic patients report seizures in association with exercise instead. While data is limited, 2% to 10% of epileptic patients may have exercise‐induced seizures, defined by seizures occurring in more than 50% of training sessions (Bjorholt et al. 1990, Nakken 1999, Carrizosa‐moog et al. 2018). In addition, in a survey of 204 adults with epilepsy, approximately 10% reported seizures in connection with exercise and in a survey of 193 people with epilepsy, 18% of respondents reported having seizures before, during or after exercise (Nakken 1999, Ablah et al. 2009).

Our study adds to this body of research with the finding that in dogs with idiopathic epilepsy receiving AED therapy, increased activity may increase seizure frequency. In humans, a variety of mechanisms have been proposed to explain seizures associated with physical activity. Hyperhydration, fatigue physical and mental stress, hyperthermia and hypoglycaemia secondary to physical exercise are factors that have been speculated to explain exercise‐associated seizures; however, research on this topic is limited (Arida et al. 2008, Carrizosa‐moog et al. 2018). Seizure frequency may also be influenced by the type or intensity of exercise. People with epilepsy who have exercise‐associated seizures often experience the seizures with high‐intensity exercise, such as ball games, jogging or hiking (Nakken 1999; Ablah et al. 2009). In contrast, several of the studies that report a decrease in seizures with exercise assigned low‐intensity physical activity, such as yoga (Lundgren et al. 2008; Sathyaprabha et al. 2008). Our study did not dictate the type or intensity of exercise that dogs engaged in; consequently, future studies evaluating how exercise type and intensity relate to seizures in dogs is an important next step.

Interestingly, our study found that dogs assigned to an increase in activity over 3 months had a significant increase in sleep scores relative to control dogs. Higher sleep scores suggest dogs in the treatment group were less restless, and while the accelerometer cannot definitively determine that dogs were sleeping and not resting, it offers a more objective assessment of sleep than owner observation alone. Owners, i.e., may not be able accurately assess how long their dog slept overnight or when they are not present (Kinsman et al. 2020). It is also more feasible than an EEG, which is used frequently in humans to evaluate sleep but infrequently in veterinary patients due to limited availability and patient cooperation.

Epilepsy and sleep have a complex relationship, with sleep disturbances being more prevalent among human epileptics compared with age and gender‐matched controls (De Weerd et al. 2004). These disturbances can manifest as insomnia, sleep apnoea, parasomnia or rapid eye movement behavioural disorders (Ismayilova et al. 2015) and are associated with a significant decrease in quality of life (Gutter et al. 2019; De Weerd et al. 2004). In healthy adults as well as those with chronic health conditions associated with sleep disturbances, such as rheumatoid arthritis and cardiovascular disease, exercise has been shown to improve sleep quality or reduce sleep disturbances (Yamamoto et al. 2007, Durcan et al. 2014, Dolezal et al. 2017). Its effect on sleep in people with epilepsy, though, is unclear. One study assigned women with refractory epilepsy to an exercise program and found that they reported fewer sleeping problems (Eriksen et al. 1994). However, another study assigned children with epilepsy to an exercise intervention and found that while subjective sleep quality as measured by parent questionnaires improved, the children's physical activity, sleep efficiency and time asleep did not increase (Do et al. 2020). Our study is the first to examine this relationship in dogs with idiopathic epilepsy and found that dogs assigned to an exercise treatment had improved sleep scores. In healthy humans, a multitude of mechanisms involving changes in body temperature, vagal modulation, hormone levels and mood have been proposed to explain exercise's beneficial effect on sleep; however, the main mechanism remains a topic of debate (Uchida et al. 2012, Stutz et al. 2019).

Our study had several limitations, particularly low compliance of owners actually increasing their dog's activity level. Only five owners complied with the 20% prescribed increase in activity, with 11 increasing their activity by at least 10%. However, in analysing this group of partial compliers separately, the relationship between increased activity and seizure frequency became more significant, strengthening the hypothesis that exercise may have actually increased seizure frequency. The post hoc partial complier analysis is limiting since it does not capture the intention to treat effect and the sample size of compliers is low. CAMD data also relied on owners charging the device and keeping it on their dogs. This, combined with technical issues, led to occasional gaps in data collection. Since the CAMD is not waterproof, it needed to be removed during swimming, which may have underestimated a dog's activity level. Owners’ reports were also relied on for seizure monitoring and tracking. While EEG is the gold standard of seizure detection, in veterinary medicine, use of this method is limited due to patient cooperation and the need for extended monitoring (Uriarte & Saiz 2016).

In conclusion, our study found that seizure frequency slightly, but significantly, increased during periods of prescribed physical activity in dogs with idiopathic epilepsy receiving AED therapy, compared with dogs without an increase in physical activity over a 3‐month period. Dogs with increased activity also had improved sleep scores during this period. Further studies are needed to understand the types of activities possibly associated with seizures in epileptic dogs, as well as the relationship between sleep quality and seizure activity in dogs. Our results highlight the importance of having owners track seizures in relation to their dog's daily level of physical activity.

Funding

This study was funded by the University of Wisconsin – Madison, School of Veterinary Medicine, Companion Animal Grant (233 AAG5571 872100 4).

Conflict of Interest

None of the authors of this article has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper.

Author contributions

Kylie Grady: Data curation (equal); formal analysis (supporting); investigation (equal); methodology (supporting); project administration (supporting); writing – original draft (lead); writing – review and editing (supporting). Starr Cameron: Conceptualization (equal); data curation (equal); formal analysis (supporting); funding acquisition (equal); investigation (lead); methodology (lead); project administration (lead); software (equal); supervision (lead); writing – original draft (supporting); writing – review and editing (lead). Sean P. Kent: Conceptualization (supporting); data curation (supporting); formal analysis (lead); methodology (equal); software (lead); validation (lead); writing – original draft (supporting); writing – review and editing (supporting). Heidi Barnes Heller: Conceptualization (equal); funding acquisition (equal); methodology (supporting); writing – review and editing (supporting). Megan Marie Barry: Data curation (equal); funding acquisition (supporting); investigation (equal); methodology (supporting); writing – original draft (supporting); writing – review and editing (supporting).

References

  1. Abadie, A. (2005) Semiparametric difference‐in‐differences estimators. Review of Economic Studies 72, 1‐19 [Google Scholar]
  2. Ablah, E. , Haug, A. , Konda, K. , et al. (2009) Exercise and epilepsy: a survey of Midwest epilepsy patients. Epilepsy and Behavior 14, 162‐166 [DOI] [PubMed] [Google Scholar]
  3. Arida, R. M. , Cavalheiro, E. A. , Da Silva, A. C. , et al. (2008) Physical activity and epilepsy: proven and predicted benefits. Sports Medicine 38, 607‐615 [DOI] [PubMed] [Google Scholar]
  4. Arida, R. M. , Scorza, F. A. , Gomes da Silva, S. , et al. (2010) The potential role of physical exercise in the treatment of epilepsy. Epilepsy and Behavior 17, 432‐435 [DOI] [PubMed] [Google Scholar]
  5. Barry, M. , Cameron, S. , Kent, B. , et al. (2021) Daytime and noctural activity in treated dogs with idiopathic epilepsy compared to matched unaffected controls. Journal of Veterinary Internal Medicine 35, 1826‐1833 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Berendt, M. , Gredal, H. & Alving, J. (2004) Characteristics and phenomenology of epileptic partial seizures in dogs: similarities with human seizure semiology. Epilepsy Research 61, 167‐173 [DOI] [PubMed] [Google Scholar]
  7. Bjorholt, P. G. , Nakken, K. , Rohme, K. , et al. (1990) Leisure time habits and physical fitness in adults with epilepsy. Epilepsia 31, 83‐87 [DOI] [PubMed] [Google Scholar]
  8. Brown, D. C. , Boston, R. C. & Farrar, J. T. (2010) Use of an activity monitor to detect response to treatment in dogs with osteoarthritis. Journal of the American Veterinary Medical Association 237, 66‐70 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Campos, D. V. , Lopim, G. M. , de Almeida, V. S. , et al. (2016) Effects of different physical exercise programs on susceptibility to pilocarpine‐induced seizures in female rats. Epilepsy & Behavior 64, 262‐267 [DOI] [PubMed] [Google Scholar]
  10. Campos, D. V. , Lopim, G. M. , da Silva, D. A. , et al. (2017) Epilepsy and exercise: an experimental study in female rats. Physiology & Behavior 171, 120‐126 [DOI] [PubMed] [Google Scholar]
  11. Carrizosa‐moog, J. , Ladino, L. D. , Benjumea‐Cuartas, V. , et al. (2018) Epilepsy, physical activity and sports: a narrative review. The Canadian Journal of Neurological Sciences 45, 624‐632 [DOI] [PubMed] [Google Scholar]
  12. Chan, C. B. , Spierenburg, M. , Ihle, S. L. , et al. (2005) Use of pedometers to measure physical activity in dogs. Journal of the American Veterinary Medical Association 226, 2010‐2015 [DOI] [PubMed] [Google Scholar]
  13. Chandler, K. (2006) Canine epilepsy: what can we learn from human seizure disorders? The Veterinary Journal 172, 207‐217 [DOI] [PubMed] [Google Scholar]
  14. Chang, Y. , Mellor, D. & Anderson, T. (2006) Idiopathic epilepsy in dogs: owners’ perspectives on management with phenobarbitone and/or potassium bromide. Journal of Small Animal Practice 47, 574‐581 [DOI] [PubMed] [Google Scholar]
  15. De Weerd, A. , De Haas, S. , Otte, A. , et al. (2004) Subjective sleep disturbance in patients with partial epilepsy: a questionnaire‐based study on prevalence and impact on quality of life. Epilepsia 45, 1397‐1404 [DOI] [PubMed] [Google Scholar]
  16. Do, J. , Webster, R. J. , Longmuir, P. E. , et al. (2020) Physically active children with epilepsy have good objective sleep duration and efficiency despite subjective reports of fatigue and sleep problems. Epilepsy & Behavior 104, 106853 [DOI] [PubMed] [Google Scholar]
  17. Dolezal, B. A. , Neufeld, E. V. , Boland, D. M. , et al. (2017) Interrelationship between sleep and exercise: a systematic review. Advances in Preventive Medicine 2017, 1‐14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Durcan, L. , Wilson, F. & Cunnane, G. (2014) The effect of exercise on sleep and fatigue in rheumatoid arthritis: a randomized controlled study. Journal of Rheumatology 41, 1966‐1973 [DOI] [PubMed] [Google Scholar]
  19. Eriksen, H. R. , Ellertsen, B. , Gronningsaeter, H. , et al. (1994) Physical exercise in women with intractable epilepsy. Epilepsia 35, 1256‐1264 [DOI] [PubMed] [Google Scholar]
  20. Esquivel, E. , Chaussain, M. , Plouin, G. , et al. (1991) Physical exercise and voluntary hyperventilation in childhood absence epilepsy. Electroencephalography and Clinical Neurophysiology 79, 127‐132 [DOI] [PubMed] [Google Scholar]
  21. Gutter, T. , Callenbach, P. M. C. , Brouwer, O. F. , et al. (2019) Prevalence of sleep disturbances in people with epilepsy and the impact on quality of life: a survey in secondary care. Seizure: European. Journal of Epilepsy 69, 298‐303 [DOI] [PubMed] [Google Scholar]
  22. Häfele, C. A. , Freitas, M. P. , Cozzensa da Silva, M. , et al. (2017) Are physical activity levels associated with better health outcomes in people with epilepsy? Epilepsy & Behavior 72, 28‐34 [DOI] [PubMed] [Google Scholar]
  23. Hamilton, S. L. , Clemes, S. A. & Griffiths, P. L. (2008) UK adults exhibit higher step counts in summer compared to winter months. Annals of Human Biology 35, 154‐169 [DOI] [PubMed] [Google Scholar]
  24. Ho, V. , Simmons, R. K. , Ridgway, C. L. , et al. (2013) Is wearing a pedometer associated with higher physical activity among adolescents? Preventive Medicine 56, 273‐277 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ismayilova, V. , Demir, A. U. & Tezer, F. I. (2015) Subjective sleep disturbance in epilepsy patients at an outpatient clinic: a questionnaire‐based study on prevalence. Epilepsy Research 115, 119‐125 [DOI] [PubMed] [Google Scholar]
  26. Kataria, L. & Vaughn, B. V. (2016) Sleep and epilepsy. Sleep Medicine Clinics 11, 25‐38 [DOI] [PubMed] [Google Scholar]
  27. Kinsman, R. , Owczarczak‐Garstecka, S. , Casey, R. , et al. (2020) Sleep duration and behaviours: a descriptive analysis of a cohort of dogs up to 12 months of age. Animals 10, 1‐14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lundgren, T. , Dahl, J. , Yardi, N. , et al. (2008) Acceptance and commitment therapy and yoga for drug‐refractory epilepsy: a randomized controlled trial. Epilepsy and Behavior 13, 102‐108. 10.1016/j.yebeh.2008.02.009 [DOI] [PubMed] [Google Scholar]
  29. McAuley, J. W. , Long, L. , Heise, J. , et al. (2001) A prospective evaluation of the effects of a 12‐week outpatient exercise program on clinical and behavioral outcomes in patients with epilepsy. Epilepsy & Behavior 2, 592‐600 [DOI] [PubMed] [Google Scholar]
  30. Muñana, K. R. , Nettifee, J. A. , Griffith, E. H. , et al. (2020) Evaluation of a collar‐mounted accelerometer for detecting seizure activity in dogs. Journal of Veterinary Internal Medicine 34, 1239‐1247 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nakken, K. O. (1999) Physical exercise in outpatients with epilepsy. Epilepsia 40, 643‐651 [DOI] [PubMed] [Google Scholar]
  32. Nakken, K. , Bjerholt, P. G. , Johannessen, S. I. , et al. (1990) Effect of physical training on aerobic capacity, seizure occurrence, and serum level of antiepileptic drugs in adults with epilepsy. Epilepsia 31, 88‐94 [DOI] [PubMed] [Google Scholar]
  33. Nakken, K. , Loyning, A. , Loyning, T. , et al. (1997) Does physical exercise influence the occurrence of epileptiform EEG discharges in children? Epilepsia 38, 279‐284 [DOI] [PubMed] [Google Scholar]
  34. Packer, R. M. A. , Shihab, N. K. , Torres, B. B. J. , et al. (2014) Clinical risk factors associated with anti‐epileptic drug responsiveness in canine epilepsy. PLoS One 9, e106026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Packer, R. M. A. , Shihab, N. K. , Torres, B. B. J. , et al. (2015) Responses to successive anti‐epileptic drugs in canine idiopathic epilepsy. Veterinary Record. 176, 203 [DOI] [PubMed] [Google Scholar]
  36. Packer, R. M. A. , Hobbs, S. L. & Blackwell, E. J. (2019) Behavioral interventions as an adjunctive treatment for canine epilepsy: a missing part of the epilepsy management toolkit? Frontiers in Veterinary Science 6, 571‐579 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Peixinho‐Pena, L. F. , Fernandes, J. , de Almeida, A. , et al. (2012) A strength exercise program in rats with epilepsy is protective against seizures. Epilepsy & Behavior 25, 323‐328 [DOI] [PubMed] [Google Scholar]
  38. Podell, M. , Fenner, W. R. & Powers, J. D. (1995) Seizure classification in dogs from a nonreferral‐based population. Journal of the American Veterinary Medical Association 206, 1721‐1728 [PubMed] [Google Scholar]
  39. R Core Team . (2020) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria
  40. Sathyaprabha, T. N. , Satishchandra, P. , Pradhan, C. , et al. (2008) Modulation of cardiac autonomic balance with adjuvant yoga therapy in patients with refractory epilepsy. Epilepsy & Behavior 12, 245‐252 [DOI] [PubMed] [Google Scholar]
  41. Stutz, J. , Eiholzer, R. & Spengler, C. M. (2019) Effects of evening exercise on sleep in healthy participants: a systematic review and meta‐analysis. Sports Medicine 49, 269‐287. 10.1007/s40279-018-1015-0 [DOI] [PubMed] [Google Scholar]
  42. Uchida, S. , Shioda, K. , Morita, Y. , et al. (2012) Exercise effects on sleep physiology. Frontiers in Neurology 3, 1‐5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Uriarte, A. & Saiz, I. M. (2016) Canine versus human epilepsy: are we up to date ? Journal of Small Animal Practice 57, 115‐121 [DOI] [PubMed] [Google Scholar]
  44. Wernimont, S. M. , Thompson, R. J. , Mickelsen, S. L. , et al. (2018) Use of accelerometer activity monitors to detect changes in pruritic behaviors: interim clinical data on 6 dogs. Sensors 18, 249 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wessmann, A. , Volk, H. A. , Packer, R. M. A. , et al. (2016) Quality‐of‐life aspects in idiopathic epilepsy in dogs. Veterinary Record. 179, 229 [DOI] [PubMed] [Google Scholar]
  46. WWW Document . (2018a) Fitbark for research. https://www.fitbark.com/research/ Accessed October 2018.
  47. WWW Document . (2018b) What are BarkPoints? Are they steps? https://www.fitbark.com/articles/what‐are‐barkpoints/. Accessed October 2018.
  48. Yamamoto, U. , Mohri, M. , Shimada, K. , et al. (2007) Six‐month aerobic exercise training ameliorates central sleep apnea in patients with chronic heart failure. Journal of Cardiac Failure 13, 825‐829 [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Small Animal Practice are provided here courtesy of Wiley

RESOURCES