SEARCH

SEARCH BY CITATION

Keywords:

  • Basal nuclei;
  • Channelopathy;
  • Movement disorder;
  • Muscle membrane;
  • Seizure

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

Background: Paroxysmal dyskinesias are episodes of abnormal, involuntary movement or muscle tone, distinguished from seizures by the character of the episode and lack of seizure activity on ictal EEG.

Hypothesis: Paroxysmal dyskinesia is an inherited, autosomal recessive disorder in Chinook dogs.

Animals: Families of Chinook dogs with paroxysmal dyskinesia.

Methods: Pedigrees and medical histories were reviewed for 299 Chinook dogs. A family of 51 dogs was used for analysis. Episodes were classified as seizures, paroxysmal dyskinesia, or unknown, and segregation analysis was performed.

Results: Paroxysmal dyskinesia was identified in 16 of 51 dogs and characterized by an inability to stand or ambulate, head tremors, and involuntary flexion of 1 or multiple limbs, without autonomic signs or loss of consciousness. Episode duration varied from minutes to an hour. Inter-ictal EEGs recorded on 2 dogs with dyskinesia were normal. Three dogs with dyskinesia also had generalized tonic-clonic seizures. One of 51 dogs had episodes of undetermined type. Phenotype was unknown for 6 of 51 dogs, and 28 dogs were unaffected. Segregation was consistent with an autosomal recessive trait.

Conclusions and Clinical Importance: This movement disorder is prevalent in the Chinook breed, and consistent with a partially penetrant autosomal recessive or polygenic trait. Insufficient evidence exists for definitive localization; episodes may be of basal nuclear origin, but atypical seizures and muscle membrane disorders remain possible etiologies. The generalized seizures may be a variant phenotype of the same mutation that results in dyskinesia, or the 2 syndromes may be independent.

Abbreviations:
CSF

cerebrospinal fluid

EA-1, EA-2

episodic ataxia-1, -2

MRI

magnetic resonance imaging

PDC

paroxysmal dystonic choreoathetosis

PED

paroxysmal exertion-induced dystonia

PHD

paroxysmal hypnogenic dyskinesia

PKC

paroxysmal kinesigenic choreoathetosis

PKD

paroxysmal kinesigenic dyskinesia

PNKD

paroxysmal nonkinesigenic dyskinesia

Paroxysmal dyskinesias are episodes of abnormal involuntary hyperkinetic movement or muscle tone, distinguished from seizures by the appearance of the observed movements, lack of altered consciousness, lack of autonomic signs, and (with very few exceptions) lack of seizure activity on ictal EEG recordings.1 The term dyskinesia specifically implies difficulty performing a voluntary movement. Often this is due to involuntary movements of the body that prohibit normal voluntary movement. These involuntary movements may include any of the following: dystonia, ballismus, hemiballismus, chorea, and athetosis (definitions, Table 1). The involuntary movements do not, however, include clonic movements or automatisms typically observed in generalized seizures.

Table 1.   Terminology used to describe movements in dyskinesias in human medicine.
DystoniaAbnormal tonicity in the tissues, resulting in impairment of voluntary movement. This abnormal tone may be either hypertonicity or hypotonicity
ChoreaIrregular, spasmodic, involuntary movements of the limbs or facial muscles. Chorea is often accompanied by hypotonia
AthetosisA constant succession of slow, writhing, involuntary movements of flexion, extension, pronation, and supination of the fingers and hands, and sometimes toes and feet. Movements may have characteristics of both chorea and athetosis, and in such cases are described as choreoathetosis
Ballism / HemiballismInvoluntary movement of the proximal limb musculature, characterized by jerking, flinging movements of the limb. Signs are commonly unilateral, in which they are described as hemiballismus

Paroxysmal dyskinesias are not well characterized in veterinary medicine, and very few reports of dyskinesias exist in the veterinary literature.2–5 Historically, several classification schemes have been presented in the human literature to differentiate among the forms of paroxysmal dyskinesias. Recently, a modified classification scheme has been proposed for human dyskinesias that is more detailed, and takes into account not only duration of attack and precipitating factors, but also phenomenology and etiology.6

The specific classifications of paroxysmal dyskinesias are derived from the combination and character of described movements, as well as duration, frequency, precipitating factors (eg, sudden movements, prolonged exercise, stress, fatigue, coffee, heat, or cold), and pathophysiology (eg, ion channel disorder, basal nuclei disorder, or symptomatic in association with other neurologic or metabolic disorders).6 Based on these factors, paroxysmal dyskinesias may be classified as one of the following: (1) paroxysmal kinesigenic choreoathetosis/dyskinesias (PKC/PKD), (2) paroxysmal dystonic choreoathetosis/paroxysmal nonkinesigenic dyskinesias (PDC/ PNKD), (3) paroxysmal exertion-induced dystonia (PED), (4) paroxysmal hypnogenic dyskinesia (PHD), or (5) episodic ataxia (EA-1 or EA-2).6 Dyskinesias within any of these specific classifications may be further distinguished by etiology, namely familial, sporadic, or secondary to underlying metabolic disorders.

Although the clinical features of dyskinesias are becoming more clearly recognized in human medicine, the pathophysiology and neuroanatomic origin of paroxysmal dyskinesias is still not fully understood. Knowledge of the pathophysiology of dyskinesia in dogs is even more limited, yet dyskinesias are becoming increasingly recognized in dogs. Phenotypic characterization, genome scanning, and identification of specific genetic mutations responsible for dyskinesias are necessary to advance our understanding of these disorders.

We identified a familial paroxysmal dyskinesia prevalent in Chinook dogs, and performed pedigree analysis to characterize the mode of inheritance. To the authors' knowledge, this is the first description of a familial paroxysmal dyskinesia reported in the veterinary literature.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

The original proband in a family of Chinook dogs with dyskinesia was identified in 1998 by owner submission to the internet epilepsy database (http://www.canine-epilepsy.net) at the University of Missouri–Columbia, College of Veterinary Medicine. Additional cases were recruited from 1998 to 2006 via the internet database and solicitation through Chinook breed clubs and owner associations. Pedigrees, medical history information, litter information, and EDTA-preserved blood samples for DNA isolation were collected from all Chinook dogs (affected and unaffected) participating in the study. Detailed surveys, owner interview via phone conversation, and, where available, review of video recordings of episodes were performed to characterize the nature of episodes as described below (see “Classification of phenotype”), and to classify them as generalized seizures, paroxysmal dyskinesia, or both. Dogs in which episodes were determined to be idiopathic (eg, no evidence of concurrent or underlying disease on physical examination, neurologic examination and minimum database), but for which the character was not clearly consistent with paroxysmal dyskinesia or generalized tonic-clonic seizure, were reported as having episodes of uncharacterized type. Dogs with episodes of uncharacterized type, with unknown status, unknown parentage, and dogs that died before adulthood were excluded from pedigree analysis, and are indicated as such in Results.

Classification of Phenotype

Clinical features considered in the characterization of phenotype were the presence or absence of generalized tonic-clonic motor activity, conscious awareness, autonomic signs, and specific limb movements (eg, dystonia, choreoathetosis, ballism). See Table 1 for more detailed descriptions and terminology. Generalized seizures were defined as episodes of brief duration (<5 minutes) affecting the entire body, in which the animal exhibited tonic-clonic motor activity and loss of consciousness, with or without autonomic signs. Episodes were only classified as paroxysmal dyskinesia if the dog maintained conscious awareness, lacked autonomic signs, and if there was no generalized tonic-clonic component to the episode. Episodes of focal motor activity in the presence of autonomic signs, with or without mental impairment, would have been classified as a simple or complex partial seizure as appropriate, but such episodes were not reported in this study population.

Pedigree Analysis

Segregation analysis was performed on the pedigrees and phenotypes. Likelihood ratio tests were performed for those families for which the status of littermates was known to determine if the proportion of affected offspring was consistent with a fully penetrant autosomal dominant, autosomal recessive, or sex-linked mode of inheritance. The Davie singles method7 was used to correct for ascertainment bias in the autosomal recessive inheritance model, and confidence intervals were calculated. Correction for ascertainment bias was necessary because of the greater likelihood of identifying litters in which there are affected individuals. More specifically, the Davie method of ascertainment bias correction accounts for independent multiple ascertainments within litters. A chi-squared analysisa based on the log-likelihoods for the pedigree and phenotypes under each mode of inheritance was used to determine the most likely mode of inheritance. A P≤ .05 allowed exclusion of the mode of inheritance tested in the model, and P > .05 failed to exclude the mode of inheritance.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

Pedigrees and medical histories were reviewed for 299 Chinook dogs participating in the University of Missouri Epilepsy Study. A family of 189 dogs was selected for visual scrutiny of the pedigree, and descriptive characterization of the phenotype (Fig 1). Paroxysmal dyskinesia was identified in 35 of 189 dogs (17 males, 18 females) and characterized by an inability to stand or ambulate, head tremor, and dystonia or dyskinesia of 1 or multiple limbs, without loss of consciousness or autonomic signs (Fig 2 and supporting information Video S1). Based on evaluations of videotaped episodes (10 of 35 affected dogs), information from owner surveys (all dogs) and verbal interviews of dog owners (all dogs), the phenomenology was found to be very similar among dogs. The following description captures the summarized information provided on questionnaires, supplemented with information acquired during detailed conversations with owners and visual evaluation of videotapes.

image

Figure 1.  Pedigree of the family of 189 Chinook dogs used in the descriptive study. The 11 modern founders of the breed occur among the oldest generation represented in this pedigree. Pedigree symbols are standard for male (square) and female (circle). Phenotypes are represented as follows: inline image Normal (unaffected), inline image Paroxysmal dyskinesia only, inline image Generalized tonic-clonic seizures only, inline image Paroxysmal dyskinesia and generalized tonic-clonic seizures, inline image Episodes of uncharacterized type, inline image Individual of unknown status, inline image Deceased before adulthood.

Download figure to PowerPoint

image

Figure 2.  Still photo of a Chinook dog having an episode of dyskinesia. See also Supplemental Video S1, of a dog during an episode of dyskinesia. The episode in the video is characterized primarily by flexion of the limbs and mild tremors. The dog is clearly alert and responsive to the people around him, but unable to rise. At the end, there are no post-ictal behavioral changes. The total duration of the episode in the video was 5 minutes.

Download figure to PowerPoint

The movements observed during Chinook paroxysmal dyskinesia episodes included dystonia (eg, flexion of a limb or limbs), chorea (eg, repetitive, small-range movements of a limb or limbs), and ballismus (eg, kicking or flailing of a limb or limbs), but athetotic movements were not described or observed in these dogs. Episodes affected all 4 limbs or were unilateral (the affected side may vary between episodes). Some dogs exhibited head tremors as a component of the episode. Urination, defecation, and other autonomic signs were consistently absent. The duration of episodes ranged from minutes to an hour, and episodes occurred as often as several per day or as infrequently as several months or years between attacks. Episodes did not appear to be induced by sudden movements. Dogs were reported to be either normal or lethargic after an episode, but were mentally appropriate throughout. Blindness was never reported as a component of the episodes or post-ictal period. Age of onset ranged from 2 to 6 months up to 5 years of age. One owner reported an age of onset >5 years of age. Most affected dogs had their 1st episode before 3 years of age.

Intra-ictal EEGs were not acquired because of logistical limitations, but inter-ictal EEGs were recorded for a duration of 14–18 minutes on 2 dogs with paroxysmal dyskinesia, and were normal. Eight-channel EEG studies were acquired with 28-G, platinum-coated needle electrodes in standard configuration according to the Redding montage.8 Muscle fasiculations were identified, isolated, and eliminated by injecting 0.1–0.2 mL of 2% lidocaine around the electrode(s) that showed the interference. In both cases, EEG studies were acquired after anesthesia for magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) collection. Animals had been preanesthetized with buprenorphine (0.01 mg/kg IM), induced with propofol (3–6 mg/kg IV to effect), and maintained on inhalant isoflurane in oxygen. After isoflurane was discontinued, 100% oxygen was administered for 10 minutes, after which time animals were allowed to breathe room air and the EEG studies commenced. The EEG recordings continued until the animals no longer tolerated continued recording (approximately 14–18 minutes of EEG recording). During this time, no epileptic activity was identified in either animal.

Both of the dogs that had EEG studies also had MRI brain scans and CSF analysis, pre- and postexercise blood lactate and pyruvate concentrations, CBCs, serum biochemical profiles (one of which was acquired immediately post-ictally), and thyroid testing. One dog also had a urinalysis performed immediately post-ictally. All results were within reference ranges.

In addition to the 2 cases with extensive diagnostic workups presented, the following diagnostic tests were performed. Ten dogs had CBCs and serum biochemical profiles, 2 of which were acquired intra-ictally and 2 of which were acquired and analyzed 1 day post-ictally (all results were within reference range). No evidence of acid-base or electrolyte imbalances was identified on these profiles. One dog had a urinalysis performed, which was within reference range. One dog had a blood lead concentration measured, which was within reference range. Six dogs had thyroid tests reported, which were within reference range. One additional dog was diagnosed with hypothyroidism several years after the episodes occurred, and received thyroid hormone supplementation. Other information regarding episode occurrence in association with the diagnosis and treatment of hypothyroidism in this dog was not available. Diagnostic information was not available for the remaining dogs in the study.

All dogs had physical and neurologic examinations performed by a veterinarian, one of which was performed intra-ictally, with no deficits or abnormalities identified other than the dyskinesia; 4 of these neurologic examinations were performed by board-certified veterinary neurologists, with the remainder performed by general practitioners.

Pedigree Analysis

A family of 51 dogs, for which sufficient parental information was known for both parents, was used for segregation analysis (Fig 3). Paroxysmal dyskinesia was identified in 16 of the 51 dogs (8 males, 8 females). Three of the 16 dogs with paroxysmal dyskinesia also had generalized tonic-clonic seizures. Twenty-eight of the 51 dogs were unaffected. Exclusions to the analysis included the following 7 dogs: 1 of 51 dogs had episodes of undetermined type, and phenotype was unknown for 6 of 51 dogs: 4 because of neonatal death and 2 because of missing or incomplete case information.

image

Figure 3.  Pedigree of the family of 51 Chinook dogs used in the segregation analysis. Arrows indicate probands. The dotted lines show the relationship with 2 grandparents for the purpose of illustrating familial relationships among litters; these 2 individuals (noted on the pedigree as “grandparent”) were not included in the analysis. This family of 51 individuals is derived from the right half of the full pedigree presented in Fig 1. The basic configuration of the full pedigree is preserved, to allow for visual comparison between the pedigrees. Pedigree symbols are standard for male (square) and female (circle). Phenotypes are represented as follows: inline image Normal (unaffected), inline image Paroxysmal dyskinesia only, inline image Generalized tonic-clonic seizures only, inline image Paroxysmal dyskinesia and generalized tonic-clonic seizures, inline image Episodes of uncharacterized type, inline image Individual of unknown status, inline image Deceased before adulthood.

Download figure to PowerPoint

Among the 189 individuals in the larger family pedigree, sufficient medical information was known about the phenotype of at least 1 parent in 15 litters with affected offspring. Among these 15 litters in which at least 1 affected offspring was produced, 4 of 18 (22%) of the known parents were themselves affected. Among the family pedigree of 51 individuals, sufficient medical information was known about both parents in all 7 litters in which at least 1 affected offspring was produced. For these 7 litters, 2 of 10 (20%) parents were themselves affected.

The proportion of affected males and females in the pedigree was not significantly different. For the larger family pedigree of 189 individuals, 17 males and 18 females were affected (χ2= 0.02, df= 1, P= .89), and for the smaller family of 51 individuals used in the segregation analysis, 8 males and 8 females were affected (χ2= 0, df= 1, P= 1).

Sufficient medical information was known for a family of 51 individuals for use in the segregation analysis. Of these, 7 offspring were excluded from the analyses because of the following reasons: 4 died before adulthood, 1 had episodes that could not be clearly characterized, and 2 had unknown phenotype status. Therefore, 35 offspring were included in our analysis. A likelihood ratio test was performed to determine if our observed number of affected offspring was concordant with the genotype of parents, assuming a fully penetrant autosomal recessive mode of inheritance. Ascertainment bias was corrected by the Davie singles method7 to account for the increased likelihood of identifying litters in which at least 1 offspring was affected (Tables 2 and 3).

Table 2.   Expected, observed, and corrected family data used in the segregation analysis for offspring of carrier-carrier (Aa×Aa) parents. Assessment of ascertainment bias is presented below the table.7
Family/ LitterParentage (female to male)Raw # Offspring in Litter# Deceased or Unknown (excluded)TR% AffectedProbandsJQ
  1. p, probability of an individual being affected; , corrected probability of an individual being affected; T, total number of siblings (excluding unknown status and those deceased as juvenile/neonate); R, number of affected siblings; J, number of families with exactly 1 proband; Q, number of families with exactly 2 probands. Families are identified as A-F according to parentage, with litters numbered 1–7 corresponding to the pedigree in Figure 3.

A/1Aa×Aa413266000
B/2Aa×Aa312150110
E/6Aa×Aa918225110
Total   T= 13R= 5382J= 2Q= 0
Observed segregation frequency =R/T× 100 = 38%.
Corrected segregation frequency () = [RJ] / [TJ] = 0.27 × 100 = 27%.
Var = [(RJ)(TR)] / [(TJ)3] + [2Q (TR)2] / [(TJ)4] = 0.025.
95% confidence interval =± 1.96 (SE) = 0.27 ± 0.31.
95% confidence interval range, 0–0.58.
Table 3.   Expected, observed, and corrected family data used in the segregation analysis for offspring of carrier-affected (Aa×aa) parents. Assessment of ascertainment bias is presented below the table.7
Family/ LitterParentage (female to male)Raw # Offspring in Litter# Deceased or Unknown (excluded)TR% AffectedProbandsJQ
  1. p, probability of an individual being affected; , corrected probability of an individual being affected; T, total number of siblings (excluding unknown status and those deceased as juvenile/neonate); R, number of affected siblings; J, number of families with exactly 1 proband; Q, number of families with exactly 2 probands. Families are identified as A-F according to parentage, with litters numbered 1–7 corresponding to the pedigree in Figure 3.

C/3Aa×aa927457110
C/4Aa×aa514125110
D/5Aa×aa4044100110
F/7aa×Aa817114110
Total   T= 22R= 10454J= 4Q= 0
Observed segregation frequency =R / T× 100 = 45%.
Corrected segregation frequency ) = [R − J] / [T − J] = 0.33 × 100%= 33%.
Var = [(R − J)(T − R)] / [(T − J)3] + [2Q (T − R)2] / [(T − J)4] = 0.012.
95% confidence interval =± 1.96 (SE) = 0.33 ± 0.22.
95% confidence interval range, 0.11–0.55.

Without regard to parental phenotype, 15 of 35 (42.8%) offspring were affected. Of these, 13 offspring were the result of presumed carrier-carrier matings (ie, unaffected [Aa] × unaffected [Aa] parents), and 22 offspring were the result of presumed carrier-homozygous recessive matings (ie, unaffected [Aa] × affected [aa] parents). The expected proportion of affected individuals in the carrier-carrier mating was 25%, and the expected proportion of affected individuals in the carrier-homozygous recessive mating was 50%. Chi-squared analyses comparing the observed to expected proportions of affected offspring in our population, with and without correction for ascertainment bias, are presented in Table 4. Based on the χ2 values obtained for all analyses, we cannot reject the null hypothesis that the paroxysmal dyskinesia is a fully penetrant autosomal recessive trait.

Table 4.   Chi-square analysis data and results for the autosomal recessive model of inheritance.
 Actual (observed)ExpectedDeviation from Expected[Deviation2] / Expected
  1. Results with and without correction for ascertainment bias are presented.

Raw analysis
 Phenotype: Aa×Aa mating
  Affected53.251.750.94
  Unaffected89.75−1.750.31
 131301.26
    χ2= 1.26, df= 1, P= .26; therefore cannot reject recessive mode of inheritance
 Phenotype: Aa×aa mating
  Affected1011−10.09
  Unaffected121110.09
 222200.18
    χ2= 0.18, df= 1, P = .67; therefore cannot reject recessive mode of inheritance
Corrected for ascertainment bias
 Phenotype: Aa×Aa mating
  Affected3.513.250.260.02
  Unaffected9.499.75−0.260.01
 131300.03
    χ2= 0.03, df= 1, P= .86; therefore cannot reject recessive mode of inheritance
 Phenotype: Aa×aa mating
  Affected7.2611−3.741.27
  Unaffected14.74113.741.27
 222202.54
    χ2= 2.54, df= 1, P= .11; therefore cannot reject recessive mode of inheritance

The chi-squared results are further supported by evaluation of their associated 95% confidence intervals. Based on the variance of 0.025 (variance of , Table 2), the 95% confidence interval for a carrier-carrier mating was 0–0.58, which included the expected frequency of 0.25 for autosomal recessive inheritance (Table 2). Based on the variance of 0.012 (variance of , Table 3), the 95% confidence interval for the proportion of affected progeny from a carrier-affected mating was 0.11–0.55, which included the expected frequency of 0.50 for autosomal recessive inheritance (Table 3).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

The Chinook breed was begun in the early 1900's by explorer Arthur Treadwell Walden using crosses between other northern breeds to create a sturdy, intelligent sled dog. His Chinooks are said to have accompanied Admiral Byrd and him on expeditions to Antarctica. The breed grew, but later declined and by 1981 only 11 Chinooks (4 males and 7 females) remained to form the modern foundation of the breed (J.S. Bell and G.S. Johnson, unpublished data). These modern founders can be traced back to 4 common ancestors, which included only 1 female (J.S. Bell and G.S. Johnson, unpublished data). Genetic bottlenecks, such as that which occurred in the recent history of the Chinooks, predispose populations to undesirable recessive traits, due to the fact that each founder carries undesirable alleles for several traits, enhancing the likelihood that at least 1 undesirable phenotype will arise due to drift alone.

Fully penetrant autosomal dominant, mitochondrial, and X-linked dominant and recessive modes of inheritance are excluded by the pedigree and phenotype data (Fig 3). Litters B2 and E6 provide examples of this exclusion, in that an affected female is born to 2 unaffected parents. Relatively few (20–22%) parents of affected individuals were themselves affected. The percentage of affected parents in families with affected progeny is sufficiently low to suggest that a partially penetrant autosomal dominant mode of inheritance is unlikely. Affected individuals are evenly distributed among males and females within the pedigree, making a sex-linked mode of inheritance unlikely. In addition to the lack of sex bias, in several families affected females were born to unaffected males, which precludes a fully penetrant X-linked mode of inheritance. Based on these pedigree and phenotype data, and results of the chi-squared analyses, the mode of inheritance of paroxysmal dyskinesia in the Chinook breed is consistent with an autosomal recessive trait.

Ascertainment bias often is a concern in pedigree analysis due to the over-sampling of litters with affected individuals, and lack of sampling of litters in which there are no affected individuals. There are a number of methods available for correction of ascertainment bias, but the Davie singles method7 is more robust in that it allows for the independent ascertainment of multiple individuals within a litter. After applying this correction, our results remain consistent with an autosomal recessive mode of inheritance.

A common problem in pedigree analyses is inaccurate phenotyping. This is particularly relevant for situations in which phenotypes are episodic or transient in nature, and mild in effect. It is possible that our analyses may be inaccurate due to difficulties associated with accurate phenotype classification. A number of dogs classified as normal may actually be affected with paroxysmal dyskinesia due to the sporadic and mild nature of these episodes. These episodes may have been missed by dog owners, or perhaps some of the younger dogs included in the study have yet to express the phenotype. We suspect that such phenotyping errors are inevitable, and exist to some degree in these data. Given the high P values in our chi-squared analyses however, phenotyping errors in young dogs are unlikely to result in important alterations in results or interpretation.

Adding to the complexity of phenotyping in this study is the occurrence of generalized tonic-clonic seizures in some of the dogs. Among the larger pedigree of 189 dogs, 4 dogs had generalized tonic-clonic seizures, and 3 dogs had both generalized seizures and paroxysmal dyskinesia. This may be a variant phenotype expressed by dogs affected with paroxysmal dyskinesia, or alternatively this may be a concurrent but genetically unrelated phenotype. There are reports in the human literature of dyskinesias concurrent with generalized seizures as phenotypic expressions of the affected genotype.9–12 Because of the sporadic occurrence of such individuals in this pedigree and the predominance of the dyskinesia as a phenotype, we elected to consider the 4 dogs with generalized seizures as unaffected for paroxysmal dyskinesia in our analysis. Future genetic analyses should consider each phenotype separately in genome scans, and include larger samples of individuals expressing both or each phenotype to establish the correct relationship between dyskinesia and generalized tonic-clonic seizures in Chinook dogs.

As with many types of epileptic seizure disorders, it is possible that there is an epistatic interaction between loci that contribute to the expressed phenotype of this disorder. In a genetic study of the role of a DYT1 mutation in human dystonia, the authors found that some individuals with the putative disease genotype do not express the phenotype of dystonia.13 These authors theorize that this may be due to the need for secondary environmental insults or other predisposing genetic factors that promote the appearance and clinical manifestation of dystonia. Although not symptomatically apparent, there is electrophysiologic evidence of phenotypically normal individuals with the disease genotype being affected, as in patients with clinical dystonia.13

Phenotypically, the paroxysmal dyskinesia observed in this family of Chinooks appears consistent with a PDC/PNKD according to the modified classification scheme used for human paroxysmal dyskinesias, but there is inevitable overlap in the categories of classification.6 The lack of exercise-induced episodes makes PKC/PKD and PED less likely, the short duration of episodes and presence of dystonic movements are not consistent with the episodic ataxias (EA-1/EA-2), and episodes do not occur during sleep as with PHD. As reported for human cases,1,6,14 stress and fatigue have been suspected triggers according to Chinook breeders and owners, but a consistent association was not identified in this study. Unlike in human patients, it is not possible to evaluate coffee, tea, or alcohol consumption as precipitating factors. Although smoking has been identified as a precipitating factor in human cases of PDC/PNKD,6 smoking by the owners was not evaluated in the Chinook households. Other triggers (eg, cold, potassium) were not consistently reported by Chinook owners or breeders. Challenges to induce episodes by diet or cold exposure have not been performed, but most dogs of the Chinook breed live in the New England region with cold winter temperatures. This environment has not yielded an association of episodes with environmental temperature. Some owners notice episodes at hot temperatures whereas others at cold temperatures.

Based on the overall phenomenology, the paroxysmal dyskinesia described in Chinooks is unlikely to be a simple partial seizure of cerebrocortical origin. Although simple partial seizures may manifest similarly and potentially be misdiagnosed as paroxysmal dyskinesia in that focal motor activity can occur while consciousness is maintained, the lack of autonomic signs, generalized nature of the dyskinesia, long duration of episodes in some dogs, and normal inter-ictal EEG recordings in 2 dogs make simple partial seizures unlikely. Normal EEGs are reported in human cases of PNKD.1 Notwithstanding the limited sample size, an epileptic disorder could be present, but epileptiform activity may not be evident during the inter-ictal period, or the activity could be sufficiently deep that the surface electrodes of the EEG were unable to detect the pattern. This possibility could be further supported by the reported lethargy in some dogs after an episode. On the other hand, lethargy could be either cerebral in origin or simply from muscle fatigue, among other origins. The generalized tonic-clonic seizures reported in some individuals could be an unrelated problem, or an alternative manifestation of the disease. In several mutations that cause an episodic dyskinesia in humans, an epilepsy phenotype is reported in some family members.9,15,16 EEG studies in a larger number of canine paroxysmal dyskinesia cases (specifically intra-ictally acquired) would be necessary to reliably confirm or refute the presence of epileptic EEG activity.

The head tremor noted during some episodes may be consistent with dystonic head tremor, in which the tremor is associated only with those body regions affected by dystonia.17,18 The lack of other tremor movements and infrequent nature of these episodes make a primary tremor disorder unlikely. Other differential diagnoses that could present similarly include toxins, calcium-related tetanic episodes, metabolic myopathies, and myotonic syndromes. Toxins are unlikely to be episodic and affect so many dogs in various living environments. Furthermore, the lack of abnormalities on serum biochemical profiles and the identifiable pattern of inheritance make toxic exposure an unlikely etiology in these cases. Calcium-related (tetanic) myopathies also are unlikely, due to lack of abnormalities in both intra-ictal and inter-ictal serum calcium concentrations and the infrequent and relatively short duration of episodes. Metabolic myopathies (including hypothyroid-related myopathies) are a large group of disorders with varying clinical presentations. Metabolic myopathies are unlikely given the normal intra-ictal serum biochemical profiles, electrolyte concentrations, thyroid hormone concentrations, normal pre- and postexercise lactate and pyruvate concentrations, and the infrequent occurrence of episodes. No association has been identified between the occurrence of episodes and anesthesia or exercise, making malignant hyperthermia an unlikely etiology. Disorders such as myotonia are more difficult to exclude definitively, although there was no evidence of muscle dimpling or myokymic muscle movements in these dogs, and the clinical picture was not consistent with a typical myotonia; episodes were infrequent and animals did not exhibit the collapse typical of myotonia. However, intra-ictal electromyographic studies would be essential to definitively exclude myotonic disorders.

Inconsistencies in diagnostic evaluations are a limitation of this study. The aim of this study was primarily phenotype characterization and genetic analysis. As such, complete diagnostic evaluations were available only on 2 dogs. Given the highly consistent nature of the observed phenotype, the strength of this study is derived from the numbers of affected cases of related individuals, rather than specific diagnostic findings. Sporadic or acquired disease etiologies are unlikely in this population given the high incidence of identical phenotypes in related individuals. As a result, diagnostic information serves primarily to elucidate the underlying pathophysiology responsible for the phenotype, and as such even a limited number of complete evaluations should be representative of the phenotype in the general study population. Of note, genetic screening was not performed in these cases because there is no genetic disease yet identified in veterinary patients that fits this phenotype or breed.

The neuroanatomic origin of movement disorders such as paroxysmal dyskinesia is not specifically known. In general, movement disorders may have origins in the cerebrocortical neurons (eg, seizures and epilepsies), extrapyramidal system (eg, basal nuclei), or peripheral nervous system (eg, muscle membrane, peripheral nerve). Evidence for basal nuclei origin of the paroxysmal dyskinesias in humans comes from positron-emission tomography scans during an episode and response of some dyskinesias to dopaminergic drugs.1 With the increase in genetic discoveries, a trend is emerging in which paroxysmal dyskinesias appear to be a result of mutations in various electrolyte channels,9,19–22 although other mutations also have been identified (eg, a glutamate transporter EAAT110 and the DYT1 gene13). Such channelopathies could explain the cooccurrence of epilepsy and paroxysmal dyskinesia, given that such electrolyte channels exist in cerebrocortical neurons as well as basal ganglia.12

The term channelopathy refers to disorders of ion channel function, which may affect membrane potentials of muscles and peripheral nerves, or of neurons in the central nervous system. Channelopathies may occur on a biochemical level, biophysical level, or both as an etiology for seizures or movement disorders, or alternatively the term may be used to describe a specific subset of muscle membrane diseases (eg, hyperkalemic periodic paralysis). Clinical presentation of the paroxysmal dyskinesia described here is not typical of the muscle membrane channelopathies such as hyperkalemic periodic paralysis, and one of the Chinooks had ictal electrolyte concentrations evaluated, which were normal. Electrophysiologic studies would be necessary to confirm this clinical impression; but electromyographic and nerve conduction studies, as well as muscle and nerve biopsies, were not performed in these dogs, because there was no evidence of a primary muscle disease on clinical evaluation. Although this information would be of great interest academically, episodes were short and discrete, occurred on an irregular basis, and lacked evidence of collapse, weakness, and muscle atrophy, which is inconsistent with a primary muscle disorder. Intra-ictal electromyographic studies would be essential to conclusively exclude a myotonic disorder, but timing studies to coincide with episodes would be difficult.

The lack of electrolyte abnormalities, acid-base derangements (eg, TCO2 and anion gap concentrations on serum biochemical profiles), and metabolic abnormalities (eg, liver enzyme activities, thyroid hormone concentrations, serum glucose concentrations) on serum biochemical profiles, which in some cases were performed intra-ictally or immediately post-ictally, makes underlying electrolyte imbalances or metabolic etiologies less likely.

Although all previous reports of dyskinesias in the veterinary literature describe movement disorders that resemble paroxysmal dyskinesias of some form, the paroxysmal dyskinesia reported here exhibits some differences from previously reported dyskinesias. The movement disorder in 2 litters of Boxer pups was described as episodes of brief duration and high frequency of events per day, which differs from the Chinook dogs described here.3,6 Furthermore, the unilateral focal movements described in the Boxer pups also suggests a different classification type of movement disorder than the Chinook dogs.3 PKC/PKD was reported in a German Shorthaired Pointer.4 As described in human PKC/PKD, the episodes in this dog appeared to be triggered by activity.6 Paroxysmal dyskinesias with dystonic and choreoathetotic movements also have been described in the Bichon Frise2 but the unusual feature in the Bichon Frise is the tendency for the dyskinesia movements to affect 1 limb, or in some cases to progress sequentially to other limbs. Fine to severe twitches of the facial, neck, and shoulder muscles accompanied by a normal EEG were reported in an epileptic Chow Chow and responded to discontinuation of phenobarbital therapy.5

As reported in humans with PDC/PNKD, antiepileptic drugs generally have not been effective.6,14 Treatment with clonazepam, acetazolamide, carbamazepine, and other drugs has been attempted, but has been met with variable success.1,6,14,23 Although there is a physiological basis for the use of carbonic anhydrase inhibitors, evidence-based data are limited in both the veterinary and human literature.

In conclusion, we report a familial paroxysmal dyskinesia in the Chinook dog breed, which is similar to the PDC/PNKD reported in humans. Pedigree analysis is consistent with an autosomal recessive mode of inheritance. Genome-wide association analysis currently is underway to identify the chromosomal location of the mutation underlying the phenotype expressed by individuals in this family. By genetic analysis, the underlying etiology for this movement disorder may be determined.

Footnotes

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

a Microsoft Excel 2008 for Mac, ver. 12.2.4, Microsoft Corporation, Redmond, WA

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

We thank Liz Hansen for her valuable technical support in this study, as well as the Chinook Owner's Association and the individual Chinook owners and breeders who participated in data collection for this study. We also thank Dr Jay McDonnell for performing EEGs; and Tomoyuki Awano, DVM, PhD, and Gary S. Johnson, DVM, PhD, for their assistance with DNA and pedigree collection.

Disclaimer: Supporting information is published as submitted and not corrected or checked for scientific content, typographical errors, or functionality. The responsibility for scientific accuracy remains entirely with the authors.

Funding: This work was funded in part by the AKC Canine Health Foundation grants 847-A and 762.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgments
  8. References
  9. Supporting Information

Video S1. Video of a dog during an episode of dyskinesia. The episode in the video is characterized primarily by flexion of the limbs and mild tremors. The dog is clearly alert and responsive to the people around him, but unable to rise. At the end, there are no post-ictal behavioral changes. The total duration of the episode in the video was 5 minutes.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
JVIM_629_sm_suppinfo.mov2362KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.