Epilepsy in the Irish Wolfhound
05/25/2005

Epilepsy in the Irish Wolfhound

Margret L. Casal Dr med. vet, PhD¹, Richard Munuve BVM¹, 
M. Anne Janis MS², Petra Werner Dr med vet¹, Paula Henthorn PhD¹

From the ¹Section of Medical Genetics, School of Veterinary Medicine, University of Pennsylvania and the ²Irish Wolfhound Seizure Study, Inc.

Acknowledgements: The authors thank the Jacky Karpinski fund and the Irish Wolfhound Seizure Study fundraisers for financial assistance and the owners and breeders for submitting the information needed for this retrospective study.  The Irish Wolfhound Seizure Study, Inc. is a non-profit organization.

Abstract

During the last fifteen years, an increase in the proportion of Irish Wolfhounds with seizure disorders has been observed.Clinical data and pedigrees from closely related Irish wolfhounds were collected retrospectively and analyzed.  Idiopathic epilepsy was diagnosed, by exclusion of other causes for seizures, in 146 of 796 Irish wolfhounds (18.3%) from 115 litters.  Males were more commonly affected than females (61.6% versus 38.4%), and life expectancy of affected dogs was decreased by two years when compared to the average Irish wolfhound population.  The heritability index for the affected dogs, their littermates and unaffected parents was 0.87. The first seizure occurred by age 3 years in 73% of all dogs and there appeared to be a difference in the inheritance pattern between dogs with “early” and “late” onset seizures.  In the group with “early onset” seizures, heritability increased slightly (0.88), and the gender bias was no longer significant. An autosomal recessive mode of inheritance, with incomplete penetrance and a possible sex-predilection, most simply explains the segregation of epileptic dogs in the pedigrees examined.

Introduction

A seizure is defined as “an occasional sudden, excessive, rapid, and local discharge of gray matter” and may be generalized or partial ¹. Idiopathic epilepsy is the predominant diagnosis in dogs with generalized, recurring seizures.  Occasionally, some dogs with idiopathic epilepsy may only experience focal seizures or may even retain consciousness during the seizure. In the overall dog population 0.5 – 5.7% ²have seizures.

Idiopathic epilepsy has been described in a variety of dog breeds and has been suspected or shown to be a genetic disease ².  Various modes of inheritance of epilepsy have been deduced in different breeds, including simple autosomal recessive ³ with a sex-modifier ^{4, 5}, polygenic recessive ⁶ with a sex-predilection ⁷, polygenic with a large contribution by a single autosomal recessive gene ⁸, and a multi-locus model ⁹.

Over the years a variety of medications have become available for the treatment of idiopathic epilepsy in humans but only a handful have been tested and approved for use in the dog ¹⁰.  Unfortunately therapy is not always successful, which often leads to euthanasia of affected dogs ¹¹.  To reduce the number of losses, the disease should be eliminated from the breed population, which can be achieved with carefully designed breeding programs.  However, in order to develop such programs the mode of inheritance needs to be understood and tests to determine carrier status must be developed. 

The purposes of this study were to document idiopathic epilepsy and define the mode of inheritance in the Irish wolfhound.  Information on affected and related, non-affected Irish wolfhounds collected by questionnaire and pedigree analysis are presented here.

Materials and Methods

Animals

Medical records and pedigree information from 120 litters (N=706) of Irish wolfhounds and their parents were obtained from veterinarians and owners by detailed questionnaire including requests for relevant laboratory reports.  The diagnosis of idiopathic epilepsy in the 146 affected dogs was based on a history of more than two seizures in the absence of other medical problems, as defined by normal physical and neurological examinations, as well as normal hematological, serum biochemical, and urine analyses.  Additional information collected included date of birth, gender, age of first observed seizure, number of known seizures, age at death, and cause of death. Owners of epileptic dogs also provided information on the seizure characteristics, such as observation of partial or generalized motor problems, urination, defecation, salivation, collapse and relationship of seizure onset to the time of day, feeding, walking, sleeping or estrus cycle. Necropsies were performed on 11 affected dogs at the Veterinary School of the University of Pennsylvania, Department of Pathobiology and one at the Virginia-Maryland Regional College of Veterinary Medicine, Department of Biomedical Sciences and Pathobiology.  Dogs were considered phenotypically unaffected if no seizure had been observed during the dog’s lifetime.  Seizure-free dogs that died before age four were not included in the data analysis and 37 dogs that had seizures in the presence of seizure-associated conditions or diseases, such as hyperthermia, portosystemic shunts, head trauma, or brain/neurological lesions, such as neoplasia and ehrlichiosis, were excluded from the study. 

Genetic analysis

Pedigrees were examined from dogs with clear identifiers such as date of birth, age at death, gender, and known health status of littermates.  Dogs were considered phenotypically unaffected if no seizure had been seen in absence of an underlying brain/neurological lesion by four years of age or by the time of their death.  The vertical pedigrees were drawn and scrutinized to see whether the phenotypes and the assumed genotypes in each generation were compatible with expectations under assumptions of the simple Mendelian modes of inheritance.  Segregation analysis was performed to correct for incomplete ascertainment ¹² and heritability was calculated using standard methods ¹³  

Results

Clinical Findings

Idiopathic epilepsy was diagnosed in 146 (18.3%) Irish wolfhounds from 120 related litters containing 796 dogs.  For females, the ages of their first observed seizure ranged from 6 months to 84 months (mean=27.6±16.8).  The incidence of females with onset of seizures between 3 and 4 years of age was very low, with the majority (82%) of first observed seizures occurring before 3 years of age (hereafter to be referred to as  “early” onset seizures versus “late” onset seizures beginning after age 3).  The age of onset for males ranged from 6 months to 107 months (mean=33.9±18.5; Figure 1).  Only 67% of males experienced their first seizure before the age of 3 years (“early onset”), whereas the onset of seizures occurred by 4 years of age in 83% of all male dogs.  Overall, significantly more dogs had their first seizure before 3 years of life than after three years (p

(see Figure 1 at left:  Number of dogs at various ages at first seizure.)  Statistically significant differences in the age of onset of seizures (p

The ratio of affected male dogs (90) to female (56) was 1:0.62, differing significantly from the 1:1.08 male to female ratio for the general Irish wolfhound population during the same time period (p=0.001; an average of 2502 dogs were born yearly).  In the “early onset” group the ratio of affected males to females (60:46) was not significantly different (p=0.27) from the general wolfhound population.

The coefficient of inbreeding was calculated for 5 and 10 generations and was examined with respect to the age of onset of disease in all dogs as well as in males and females separately.  There was no significant correlation between age at first seizure and the coefficient of inbreeding, regardless of gender, or in all dogs considered together (R² = 0.02-0.06).  However, in 23 litters with multiple affected dogs, the onset of seizures in littermates occurred within 12 months of each other with the exception of seven litters in which one dog of each litter had its first seizure later in life. In the single instance where affected full siblings were born in separate litters (2 affected dogs in each of two litters) , the age of first seizure covered a range of 3 years (18, 31, 39, and 54 months,). 

The prodromal signs observed by owners or breeders in affected dogs included: teeth chattering, anxiety, mental detachment (”spaciness”), standing with a blank stare, wandering aimlessly, increased attention seeking, and sitting or lying down immediately before a seizure.  Interestingly, some owners observed “aberrant” behavior as puppies or months before the first seizure.  Reported aberrant behaviors included the dog submerging its entire head when drinking, sound and touch sensitivity, shyness of doors or other barriers, exaggerated flight distance, and shyness to an approaching perceived threat.  Affected males were often reported to be hyperactive and hard to control while females were commonly reported as being withdrawn and unable to relax.

The clinical signs were generalized seizures, as defined by uncontrolled muscular activity.  All but four of the affected animals in this study experienced recurring grand mal seizures, with seizure duration varying from 30 seconds to 4 minutes.  Six of the affected dogs began with mild generalized seizures that progressed to grand mal seizures over the course of their disease.  Two animals were reported as having petit mal seizures and another two exhibited the “fly biting” type of seizure.  In most dogs (38.4%) seizure episodes occurred during sleep regardless of time of day.  In other dogs (13.7%), seizure episodes were reported to occur only while awake and active.  In 22.6% of the affected dogs, the seizures occurred both during sleep and while awake.  For 37 (25.3%) affected dogs this information was not recorded by the owner.  The duration of each seizure in a particular dog was variable and was not predictive for the following seizure.  Also, types of seizures were variable within the litters with multiple affected dogs.  One affected dog had cluster seizures from the onset of disease, while the condition advanced from single grand mal seizures to cluster seizures in 29 affected Irish wolfhounds.  While the “late onset” seizures were mainly classified as grand mal type seizures, they subjectively appeared to be milder and less frequent than seizures in the “early onset” dogs.  In general, cluster seizures did occur in the “late onset” population of dogs but they were less common and the dogs required less medical treatment.  Owners of multiple seizuring dogs described the “late onset” seizure dogs as having a better quality of life. 

The postictal phase was characterized by wandering aimlessly, disorientation and confusion, running into stationary objects, and increased hunger.  Almost all of the dogs with cluster seizures showed a prolonged recovery phase lasting more than 36 hours.  A change in general behavior was noted in 13 dogs after the onset of disease.  Recorded changes included sensitivity to noise (4 dogs) and light (1 dog), loss of vision (3 dogs), and shyness toward strangers (5 dogs).

There was no significant difference in longevity after the first seizure between affected males and females at p14.  Of the 126 affected dogs that have died, 76 deaths (60.3%) were directly attributed to the seizure disorder, with the cause of death reported as death during seizure (16.1% of all seizure related deaths) or euthanasia because of uncontrolled seizures (65.8%), hepatotoxicity from phenobarbital treatment (6.6%), aspiration during seizures leading to fatal pneumonia (6.6%), or euthanasia because of non-recovery after a seizure (3.9%).  Fifty affected dogs died of unrelated causes and 20 affected dogs were still alive at the time of writing.  Detailed necropsies were performed on 12 of the affected Irish wolfhounds that were included in the study.  No neurological or systemic lesions were found in these 12 dogs that could explain the recurrent seizures. 

 
(see Figure 2 at left:  Life expectancy plotted against age at first seizure (± SEM).)  There was no statistically significant difference between any of the groups (p

Drug Therapy

A relatively large number of the affected dogs (36.6%) were not receiving any medication.  Conventional maintenance therapy of phenobarbital, potassium bromide, clorazepate, dilantin or diazepam was used alone or in combination for the majority of affected dogs.  Phenobarbital was by far the most commonly used drug for amelioration of the seizures, with 30.8% of all affected dogs receiving this drug.  In 13.7% of the animals, phenobarbital was used in combination or alternately with potassium bromide.  Potassium bromide alone was used in only 4 dogs (2.7%).  Gabapentins were used experimentally in 3 animals.  Two animals were treated with a combination of conventional and alternative therapies.  Four individuals were treated with homeopathic therapy alone.  None of the treatments eliminated the seizure activity but treatment was reported by owners to reduce frequency and severity of the seizures in the affected dogs.  Interestingly, there was no discernable difference in the frequency or severity of seizures with the use of homeopathy as compared to conventional maintenance therapy.  Side effects were reported from potassium bromide (ataxia; 1 dog) and phenobarbital (hepatotoxicity and motor dysfunction; 5 dogs).  Response to therapy within litters with multiple affected dogs was not consistent, as each dog reacted differently to the same drugs given. 

Genetic Analysis

Affected dogs were present in 115 litters containing 773 dogs.  Five litters (23 dogs) without affected dogs were born to matings between parents that had both been known to produce affected offspring.  A representative portion of the larger pedigree illustrating the inheritance of idiopathic epilepsy is shown in Figure 3.  The average inbreeding coefficient (calculated over ten generations) for all the dogs entered into the study was 0.156 with a range of 0.0353 – 0.35.  The major observation made from examining the pedigree was that the vast majority of affected dogs were born to healthy parents (143 affected dogs from 112 litters).  In keeping with this observation the trait was observed to skip generations.  In 37 of the litters born to unaffected parents, one of the unaffected parents was related to other affected dogs.  Only 3 of the affected dogs had an affected parent, and breedings between an affected and an unaffected parent could produce either all unaffected offspring or a mix of affected and unaffected offspring in the same litter.  The segregation ratio and the heritability (h²) were calculated from litters in which both parents never had a history of seizures and in which medical information was available on all littermates.  Ascertainment bias was corrected for using the singles method, as adopted by Nicholas 1987 ¹³.  The analysis for all dogs resulted in an h² of 0.87 and an estimated segregation ratio of 0.083 ± 0.0001, which is significantly lower than would be expected from a fully penetrant autosomal recessive trait.   Because of the potential difference between “early” and “late” onset seizures, the values were recalculated using only affected dogs that had their first seizure before 36 months.  The h2 for the “early onset” group increased only slightly to 0.88 and the estimated segregation ratio increased to 0.095 ± 0.0002. 

(see Figure 3 at left:  Small portion of the much larger pedigree of Irish wolfhounds with idiopathic epilepsy.)  Squares and circles represent males and females, respectively.  Open and filled-in symbols designate normal dogs and affected wolfhounds, respectively.  While not always indicated, all dogs in this pedigree have common ancestors. 

Discussion

Idiopathic epilepsy has been described in a variety of dog breeds along with different apparent or suspected modes of inheritance.  A familial predisposition had been recognized early on in keeshonds, terriers, corgis, retrievers, pointers, setters, spaniels, collies, Alsatian dogs, boxers, Pekinese^{15, 16}, Tervueren shepherds ¹⁷, dachshunds ^{15,16, 18},  and in the Horak laboratory dog ¹⁹.  Subsequently, epilepsy has been the subject of investigation in many institutions.  In the German shepherd and in the keeshond, idiopathic epilepsy was thought to have a simple autosomal recessive basis ^{3, 20-22}.  Epilepsy in beagles was thought to be homologous to Lafora’s disease in man based on the similar findings from histopathological examination ²³ and the mode of inheritance was described as autosomal recessive with a sex-linked “suppressor” ^{4, 5}.  For epilepsy in the golden retriever, a multifactorial etiology, including genetic and environmental factors has been described ⁹.  Larger studies were performed in the Labrador retriever, Bernese mountain dog, and the Belgian tervueren.  Pedigrees from 44 families with 55 epileptic Labrador retrievers were analyzed and the results were suggestive of a polygenic, recessive mode of inheritance ⁶.  Analysis of pedigrees from 4005 Bernese mountain dogs with 50 affected dogs revealed a polygenic, recessive mode of inheritance with a predisposition for males ⁷.  The Tervueren has been described as the dog breed with one of the highest incidences of seizures ^{17, 24}. It is, therefore, no surprise that a number of genetic studies have been performed in this breed, with results proposing a single major gene with a large effect on seizures ^{8, 24-26}.  This is the first report of hereditary epilepsy in the Irish wolfhound breed. 

In our study, an 18.3% incidence of idiopathic epilepsy was found among related Irish wolfhounds with a high heritability of 0.87 for the trait. These results suggest the presence of a major gene, as the heritability was higher than 0.5.  The heritability did not change significantly if only dogs with “early onset” seizures were used in the calculations.  In a similarly performed study a heritability of 0.77 was estimated in the Belgian Tervueren and 0.76 for the Belgian sheepdog ⁸.  In the Irish wolfhounds, significantly more males were affected than females.  This is in contrast to all but two other studies in which no sex predilection was found ^{3, 6, 8, 9}.   The results in the Irish wolfhounds are similar to the findings in the beagle and the Bernese mountain dog ^{4, 5, 7}, where the involvement of a sex-modifier gene was hypothesized.  However, when only those dogs with onset of seizure activity before 3 years of age were taken into account, there was not a significant difference between the percentage of affected females and males. 

The age at first seizure ranged from 6 months to almost nine years of age in the affected Irish wolfhounds, with the majority (73.8%) of first seizures occurring at or before three years of age.  These results are similar to what has been found in the general canine population ^{2, 27} and in specific breeds ^{6, 7, 9}.  In contrast, a study in Danish Labrador retrievers showed that the majority of seizures occurring before 4 years of age were only partial seizures and an equal number of generalized seizures occurred before and after four years of age ²⁸.  In beagles, the age of onset ranged from 11 to 70 months of age with 75% of the seizures occurring before the age of 40 months, which is at a slightly older age than in the Irish wolfhounds ⁴.  However, the number of beagles examined was much smaller than the Irish wolfhounds in this study (29 versus 146).  In the Irish wolfhounds, there was no correlation between the type or onset of seizure and the coefficient of inbreeding, similar to the findings in the Labrador and the Bernese mountain dog ^{6, 7}.

Most of the Irish wolfhounds had clinical signs and grand mal type seizures, consistent with findings in almost all other studied dog breeds ^{2, 4, 7, 9, 27}.   However, grand mal seizures advanced to cluster seizures in 29 (19.9%) of the affected Irish wolfhounds, which is a large number considering that only 27% of all dogs referred for the treatment of cluster seizures suffer from (primary) idiopathic epilepsy ²⁹.  It is possible that the Irish wolfhounds have a more severe form of idiopathic epilepsy than has been previously reported in other breeds.

Similar to affected dogs of other breeds, none of the therapies eliminated the seizures completely, but the frequency and severity of the clinical signs were reduced in most Irish wolfhounds.  Six dogs in the study exhibited side effects from their therapy which resulted in death in five dogs; a finding which has been described previously ³⁰.  Poor response to treatment has been attributed to inadequate drug therapy ^{11, 27, 29} and unresponsiveness to therapy ⁸. 

Irish wolfhounds with idiopathic epilepsy had a life expectancy that was decreased by two years in comparison to the general Irish wolfhound population14.  Of the affected dogs that died, 62% succumbed to disorders of quality of life issues directly attributable to idiopathic epilepsy.  These losses and the decreased life expectancy underscore the need for elimination of the disease from the breed by implementing a selective breeding program.

Pedigree examination revealed several features supporting an autosomal recessive mode of inheritance.  A simple dominant mode of inheritance can be excluded because the vast majority of affected dogs were born to clinically normal parents.  An X-linked mode of inheritance must be entertained because the incidence of seizures in males was significantly higher than in females.  However, almost all of the affected females were born to clinically normal parents; in an X-linked trait one would expect affected females to be born from affected males bred to carrier females.   The estimated segregation ratio is not consistent with a fully penetrant simple autosomal recessive mode of inheritance.  It was therefore concluded that seizures in the Irish wolfhounds are inherited as a non-fully penetrant autosomal recessive trait with a sex predilection for males in all of the affected dogs and as a non-fully penetrant autosomal recessive trait without a sex predilection in the “early onset” group. 

Genetic studies in humans and mice with idiopathic epilepsy have revealed a number of causative genes for specific forms of epilepsy ³¹.  Mutations for monogenic epilepsies have been found primarily in genes coding for ion channels ³² as well as in two different proteins that may be involved in neurogenesis ³³.  More recently, the molecular basis for autosomal recessive progressive myoclonic epilepsy (Lafora disease) in the miniature wirehaired dachshund has been identified ³⁴.  However, this well- defined form of epilepsy (not idiopathic), which is characterized by myoclonic type seizures with rapid, progressive mental deterioration and polyglucosan intracellular inclusions ³⁵, is clearly distinct from the form or forms of epilepsy observed in Irish wolfhounds and other breeds.  In considering the genetic bases for epilepsy in various dog breeds, including Irish wolfhounds it is important to consider that most forms of idiopathic generalized epilepsy in humans are inherited as complex genetic traits ³⁶.  A more recent study in humans has demonstrated a major susceptibility gene with modifiers that influence the clinical appearance of the seizure ³⁷.  We suspect that a similar mechanism is responsible for the phenotypic appearance of idiopathic epilepsy in the Irish wolfhound.  Given the high heritability of epilepsy in this breed, linkage analysis studies to determine the genes involved are feasible. 

Conclusion

The high incidence of seizures in the Irish wolfhound along with the decreased longevity in affected animals and unsatisfactory treatment modalities underscore the need for a breeding program designed to eliminate the disease from the breed.  Documentation of the inheritance pattern of the disease is important not only in designing such a program, but also in performing additional genetic studies that can lead to the development DNA-based genetic tests for the identification of disease predisposing alleles, which can greatly enhance the efficiency of such a program.

© 2005 Margret Casal, Dr med vet, PhD, et.al.
All rights reserved.   Reprint requests to be directed to iwstudy@earthlink.net