- Email: [email protected]
Natural history of canine paroxysmal movement disorders in Labrador retrievers and Jack Russell terriers
The Veterinary Journal 213 (2016) 33–37
Contents lists available at ScienceDirect
The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l
Natural history of canine paroxysmal movement disorders in Labrador retrievers and Jack Russell terriers Mark Lowrie a,*, Laurent Garosi b a b
Dovecote Veterinary Hospital, 5 Delven Lane, Castle Donington, Derby DE74 2LJ, UK Davies Veterinary Specialists, Manor Farm Business Park, Higham Gobion, Hitchin SG5 3HR, UK
A R T I C L E
I N F O
Article history: Accepted 3 March 2016 Keywords: Seizure Dog Dyskinesia Treatment Progression
A B S T R A C T
Delineation of the typical disease progression in canine paroxysmal dyskinesia (PD) may assist in evaluating therapeutic agents during clinical trials. Our objective was to establish the natural disease course in a group of dogs diagnosed with PD that received no medication. Fifty-nine dogs (36 Labradors, 23 JRTs) with clinically confirmed PD and a follow-up of ≥3 years were retrospectively reviewed. Dogs with PD had a young onset, were triggered by startle or sudden movements, and had a male bias (75%) with the majority being entire sample population. Twenty-one dogs (36%) had at least one event comprising cluster episodes. Episode duration and frequency varied dramatically, even within an individual. Median follow-up was 7 years. No concurrent disease was identified in any dog that was investigated. The natural history was self-limiting with 32% entering remission and an improvement in 75%. Episodes reduced in terms of frequency and duration in Labradors and JRTs respectively. Remission was lower in dogs with cluster episodes than those without. These findings suggest that the diagnostic yield of advanced neuroimaging techniques in dogs with video footage and historical data supporting PD, without neurological deficits, is low. The presence of cluster episodes is of predictive value for the prognosis of canine PD. Future research should be cautious in reporting treatment response for PD without first considering the spontaneous remission rate and improvements in untreated dogs documented in this study. Crown Copyright © 2016 Published by Elsevier Ltd. All rights reserved.
Introduction Paroxysmal dyskinesias (PD) are increasingly described although they remain poorly characterised in veterinary literature. They are recognised as a group of hyperkinetic paroxysmal movement disorders whose main feature is involuntary sustained muscle contraction. In PD, recurrent episodes are assumed to be the result of a molecular or structural abnormality and treatment, when indicated, is targeted at controlling episode frequency and severity. Known causes of PD in dogs are thus far limited to genetic (Forman et al., 2012; Gill et al., 2012; O’Brien et al., 2015), drug-induced (Kube et al., 2006; Mitek et al., 2013), and dietary factors (Lowrie et al., 2015). The first genetically mapped paroxysmal movement disorder was episodic falling syndrome in the Cavalier King Charles spaniel (Forman et al., 2012; Gill et al., 2012). Phenobarbital was found to elicit dyskinesia in an epileptic Chow (Kube et al., 2006). Border terriers suffer from canine epileptoid cramping syndrome (Black et al., 2013) which has been found to be a manifestation of gluten sensitivity (Lowrie et al., 2015). Although PD is most commonly of idiopathic or familial aetiology, other causes in people exist; for example, secondary or
* Corresponding author. Tel.: +44 1332 810395. E-mail address: [email protected] (M. Lowrie). http://dx.doi.org/10.1016/j.tvjl.2016.03.007 1090-0233/Crown Copyright © 2016 Published by Elsevier Ltd. All rights reserved.
symptomatic dyskinesia is reported, resulting from structural central nervous system lesions such as multiple sclerosis, head trauma, cerebral palsy, cerebrovascular accidents or encephalitis (Bax et al., 2005; Lotze and Jankovic, 2003; Dale et al., 2009). A current classification of human PD is one proposed by Demirkiran and Jankovic (1995), based solely on precipitating factors of the episodes and not phenomenology. This idea was born out of the observation that each type of PD can manifest with dystonia, chorea, athetosis, or a combination of abnormal movements. The classification distinguishes four categories: 1. paroxysmal kinesigenic dyskinesia (PKD) – incited by sudden movements. 2. paroxysmal nonkinesigenic dyskinesia (PNKD) – occurring spontaneously at rest 3. paroxysmal exertion-induced dyskinesia (PED) – precipitated by fatigue 4. paroxysmal hypnogenic dyskinesia (PHD) – attacks occur during sleep Various therapies are reported in the management of these conditions in people. Phenobarbital (Harcourt-Brown, 2008), acetazolamide (Gill et al., 2012; O’Brien et al., 2015) and clonazepam (Garosi et al., 2002) have been prescribed with varying success in
34
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
Table 1 Definitions of clinical signs that incorporate the term ‘dyskinesia’. Chorea Athetosis
Choreoathetosis Ballism Dystonia
An abrupt, unsustained contraction of different muscle groups A prolonged, slow contraction of the trunk muscles resulting in bending and writhing of the body and precluding maintenance of a stable posture Involuntary movements that have characteristics of both chorea and athetosis An abrupt contraction of the limb muscles which results in flailing movement of the limb and is often unilateral A sustained involuntary contraction of a group of muscles producing abnormal postures
dogs. However, future therapies should be given with an understanding of the natural progression of this condition in order to establish their efficacy. The natural course of PD is unknown. Delineation of the typical disease progression may assist in the evaluations of therapeutic agents during clinical trials. The following analysis was designed using the largest reported sample of dogs affected with PD to date. We performed this retrospective study to determine the prevalence of positive diagnostic findings in dogs with PD and to determine the natural course of this disease in two breeds of dog seen commonly with PD at our clinic; the Labrador and Jack Russell terrier (JRT). Materials and methods The study was designed as a retrospective hospital based study with followup. Dogs included in this study were client-owned clinical cases diagnosed with PD evaluated between January 2005 and January 2015 that had received no medication for PD, and with a follow-up of at least 3 years. Criteria were used to limit the phenotype of cases included by restricting selection to Labradors and JRTs only. These breeds that were chosen with PD have not been previously characterised in these breeds, though they commonly present with PD in our clinic. Analysis of each breed was performed separately. Inclusion required video evidence of at least one typical episode of PD at the time of or following investigation and its assessment by a Boardcertified neurologist. Dogs that had received treatment for PD were excluded from analysis. All dogs had to have at least a 6-month history of PD. Using phenomenological classification and in conjunction with the history, a diagnosis of PD was suspected in accordance with previous diagnostic guidelines (Packer et al., 2010; Black et al., 2013; Lowrie et al., 2015). Briefly, a diagnosis of PD was suspected when dogs had episodes in the absence of mentation alterations, autonomic signs, abnormal inter-ictal signs and post-ictal behaviour. Furthermore, the core movement had to be dyskinesia; that is, involving movement of the limb(s) (i.e. chorea, athetosis, dystonia or ballism; see Table 1 for definitions). Video examples are available of Labradors and JRTs having typical episodes (see Appendix: Supplementary Video S1 and S2). Clinical data were extracted, including signalment, duration and frequency of clinical signs (including the length of observation period over which these data were based), triggers (i.e. owners were asked if anything they had identified could trigger an episode) and survival data. Clinicopathologic data and diagnostic imaging results were also recorded where performed to identify concurrent disease processes. Follow-up was assessed via telephone interview with the owners of the dogs and/or referring veterinarians. The information obtained included current episode frequency and duration, presence or absence of cluster episodes, potential episode triggers, whether the dog was currently alive or dead, and, if applicable, the date and presumed cause of death. Two observational periods were chosen to assess episode frequency and duration; presentation and follow-up. The first observational period was calculated as the number of episodes per month in the 6 months prior to presentation at our clinic. Episode frequency at follow-up was calculated as the number of episodes per month in the 2 years preceding follow-up or in the 2 years prior to death and provided the second observational period. Episode duration in these observational periods was assigned to one of the following categories; <2 min, 2–5 min, 5–10 min, 10– 30 min, 30–60 min, 60–120 min, and >120 min. Cluster episodes were defined as more than one episode in a week. Remission of PD was defined as no episodes for ≥2 years. Remission was defined as either ‘early’ (i.e. within 24 months of onset), or ‘late’ (i.e. ≥24 months following onset).
Results Fifty-nine dogs met the inclusion criteria and were included in the analysis incorporating 36 Labradors and 23 JRTs. The majority (53/59) of owners kept a diary of each observed episode.
Signalment The median age of the 36 Labradors at episode onset was 2 years 3 months (range, 9 months to 10 years 8 months). Seven Labradors were female (19%; 2/7 neutered) and 29 were male (81%; 9/29 neutered). In contrast, the median age of onset in JRTs was 4 years 8 months (range, 1 year to 8 years) with 8/23 being female (35%; 5/8 neutered) and 15/23 being male (65%; 4/15 neutered). Triggers Episodes began most commonly following extremes of temperature in JRTs (83%; 19/23 dogs) and after sudden movements or being startled (58%; 21/36 dogs) or with excitement (25%; 9/36 dogs) in Labradors. The episodes always occurred at home and never happened when sleeping or during exercise. No diurnal rhythm was observed in any dog and many owners described stress or variation in daily routine as an inciting factor in both breeds. Frequency and duration of episodes The median duration of clinical signs in all dogs at presentation was 9 months (range 6 to 23 months); 10 months in Labradors (range 7 to 23 months) and 8 months in JRTs (range 6 to 21 months). The reported median frequency at presentation was one episode every 3 weeks in Labradors (range, one every 6 months to 12/ month) and one per month in JRTs (range, one every 6 months to two per month). Nineteen Labradors (19/44; 43%) and seven JRTs (2/23; 9%) had at least one event comprising cluster episodes. On average, duration of episodes in both Labradors and JRTs was similar (Figs. 1,2) although episodes varied dramatically in this respect, even within an individual. Results of investigations All dogs had unremarkable haematology and biochemistry profiles. Creatine kinase was increased in three dogs (all Labradors). Further testing in some dogs, which did not reveal any abnormalities, included dynamic bile acid testing (n = 25); urinalysis (n = 19); serum ammonia (n = 14); thyroid status (n = 8); ACTH stimulation test (n = 3); pre- and post-exercise plasma lactate and pyruvate concentrations (n = 2); acetylcholine receptor antibody serology (n = 2); ionised calcium (n = 1); echocardiogram (n = 12); halter monitoring during an episode (n = 1); cardiac (n = 2) and abdominal ultrasound (n = 14); radiographs of the chest (n = 17), abdomen (n = 2), stifles (n = 4), hips (n = 6) and spine (n = 1); urinary organic acids (n = 3); Neospora (n = 6) and Toxoplasma (n = 5) serology; electrophysiology (motor nerve conduction and electromyography: n = 5); CSF examination (n = 48); and MRI of the brain (n = 48), lumbosacral spine (n = 3), thoracic spine (n = 3), cervical spine (n = 3), and lumbar spine (n = 3). Follow-up Median follow-up time (from first observed episode to date of death or follow-up) in all dogs was 7 years. The median follow-up time was 6 years 8 months (range, 3 years to 12 years 2 months) in Labradors and 9 years (range, 3 years 4 months to 14 years 10 months) in JRTs. The frequency and duration of episodes was reported to have decreased in 25/36 (86%) Labradors, progressed in five Labradors and remained static in six Labradors at follow-up (Fig. 1). Owners of JRTs reported an improvement in duration and frequency in 19/ 23 (57%) dogs, a deterioration in 2/23 and a static course in 2/23 dogs (Fig. 2). Median frequency at follow-up was one episode every
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
35
Fig. 1. A chart comparing the duration of episodes at presentation and follow-up of paroxysmal dyskinesia in Labradors.
Fig. 2. A chart comparing the duration of episodes at presentation and follow-up of paroxysmal dyskinesia in Jack Russell terriers.
4 months in Labradors (range, 0–1 per month) and one every 6 months in JRTs (range, 0–1 every 6 weeks). Overall the episodes were reduced in terms of frequency and duration in both Labradors and JRTs. Fourteen Labradors (39%) met the criteria of spontaneous remission of which seven dogs had entered early remission while seven further dogs had entered late remission. The median time of remission in these Labradors was 5 years 8 months (range, 11 years 7 months to 14 years). Of the 22 Labradors failing to achieve remission, 18 dogs had a history of cluster episodes (i.e. representing 95% of the Labradors observed to have cluster episodes). Five JRTs (22%) were classified as entering spontaneous remission. All five were classified as having late remission. Median time in remission for these JRTs was 8 years (range, 7 years to 13 years 6 months). Both JRTs with cluster episodes failed to enter remission. Episode remission was lower in dogs with cluster episodes than those without. Fifty-seven of the 59 dogs that participated in the study were alive at the time of follow-up. The two deceased dogs were: a Labrador euthanased with lymphoma at 12 years 2 months, 10 years and 4 months after the onset of PD (this dog had been classified as being in late remission), and a Labrador euthanased as a result of severe arthritis at 11 years 8 months. All JRTs were alive at the time of follow-up. No episode was fatal in any dog. Dogs that were diagnosed with PD and had less than 3 years of follow-up were identified and none died or were euthanased in this time frame.
Discussion The main results of this retrospective, long-term populationbased study of 59 dogs followed since their first episode of PD, were: one; at the end of the median follow-up of 7 years 6 months, 39% of 36 Labradors were in spontaneous remission, compared to 22% of 23 JRTs. Secondly; among those in remission, 37% of dogs achieved uninterrupted episode freedom within 2 years of onset of the disease. The remaining 63% achieved spontaneous remission after 2 years. Thirdly; dogs that did not achieve remission were more likely to have experienced cluster episodes. There has been considerable confusion and disagreement amongst veterinarians pertaining to the diagnosis of PD and their distinction from epilepsy. There is strong evidence to suggest that similar pathogenic mechanisms may be involved in generating epileptic seizures and PD (Crompton and Berkovic, 2009). However, despite this common ancestry, their phenotype and clinical approach is distinct. Although simple partial seizures may manifest similarly and potentially be misdiagnosed as PD in that focal motor activity can occur while consciousness is maintained, the lack of autonomic signs, generalised nature of the dyskinesia, and long duration of episodes in some dogs make simple partial seizures unlikely. PD, with video correlation, normal inter-ictal findings and normal neurological examination, is unlikely to yield abnormal findings when performing screening diagnostic procedures. The aetiology of PD is uncertain but diagnostic screening for a structural cause has a low yield and this study may aid clinicians to define the most
36
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
cost effective approach in PD by avoiding further screening diagnostic tests when not indicated by unexpected abnormalities on history or clinical examination. The diagnosis of PD is therefore speculative in many cases. In people, it is the core feature of the movement disorder that should be considered to achieve a diagnosis. The reason for this is that many of these movement disorders may occur concurrently. In examining cohorts of Labradors and JRTs we accept the possibility that some of these dogs may have presented with more than just dyskinesia; for example myokymia and neuromyotonia, in the case of JRTs (Bhatti et al., 2011). However, when the dyskinesia predominated and involved all four limbs, this was sufficient for a diagnosis of PD in this study when the other inclusion criteria were fulfilled. Myokymia and neuromyotonia was excluded on clinical grounds as twitches remain within the affected body segment and movement is not observed. The clinical presentation of affected Labradors and JRTs was remarkably stereotyped fulfilling many of the criteria of PNKD in people (Jankovic and Demirkiran, 2002). In people, these signs can start from childhood but may develop at maturity. Episodes can last from minutes to hours with a notable predominance of males (Bruno et al., 2007). The frequency of attacks varies between individuals, ranging from daily attacks to only a few in a lifetime. Episodes may also be induced by a sudden change in position, for example from a sitting to standing position, or by being startled (Bruno et al., 2007). Neurological examination is normal between attacks. Episode frequency in people with PNKD wanes over time decreasing considerably or abating in adulthood, as found in 75% of dogs in our study. Other features consistent with PNKD include the lack of response to antiepileptic medication, the relatively low frequency of episodes and their relatively long duration. An autosomal dominant family history is present in about one in four cases in people (Jankovic and Demirkiran, 2002) but genetic studies have not yet been performed in Labradors. A dominant mode of inheritance seems unlikely given the limited pedigree data obtained from Labradors in this study (data not shown). A genetic association in JRTs is possible but was not investigated here due to the difficulties in identifying familial links in such a heterogeneous gene pool. The clinical variation observed between individuals in this study may represent distinct phenotypes within the breeds assessed or simply a wide spectrum of severity with a single aetiology. One challenge encountered in this study was the ability to construct a meaningful measure of episode frequency given that a dog may have six episodes in short succession followed by a period of 12 month quiescence, compared to a dog that has one episode every 2 months. The average episode frequency for these dogs would be classed as one episode every 2 weeks. However, both dogs clearly have a different phenotypic severity. This same idea also highlights the difficulty in interpreting treatment success as the nature of this condition is that it waxes and wanes and a short-term improvement following instigation of treatment could be falsely attributed to medication rather than disease repose. This was the reason for introducing the concept of cluster episodes to try and distinguish these two presentations. In dogs, the first genetically mapped PD was that associated with EFS in Cavalier King Charles spaniels (Forman et al., 2012; Gill et al., 2012). The mutation was characterised as a deletion affecting the brevican gene (BCAN) which encodes a brain-specific component of the extracellular matrix proteoglycan complex. It is thought to be involved in homeostasis and mutations of this protein result in a disruption of axonal conduction and synaptic stability. An interesting observation in EFS is that the condition is self-limiting in some cases (Forman et al., 2012). It is suggested that compensatory pathways involving up regulation of other proteoglycans may occur and account for this rectification. Dogs in our study also seemed to show this improvement with age so a similar compensatory mechanism may be considered.
It was important that we had a representative measure of frequency and duration prior to presentation and at follow-up. The nature of this study was that dogs were examined at different time intervals following onset and so a comparison between dogs was difficult. Clinical frequency at presentation in this study was based on the 6 months prior to presentation whereas follow-up was based on the 2 years prior to follow-up in each individual. This was chosen to allow a comparison to be drawn and for future investigations to use similar indices to enable comparison between studies. Our results are liable to recall bias. By asking owners the questions at two different time points we may find their perception of the problem and hence their answers will be influenced by their level of concern. Data on remission should not be subject to recall bias but results of improvement in PD may be amplified. Information on triggers was obtained by asking owners if they had observed anything that may induce an episode. It was surprising to see the level of unanimity obtained in the responses. However, it is important to emphasise that a failure to report a certain trigger does not necessarily exclude it as owners may not have observed certain triggers in their dog’s condition. In this respect, the triggers reported may have an increased incidence than is reported here. Even though the signs that occur with PD are fairly well characterised, progression of the disease remains uncertain. Clarification of the disease course is useful in determining the efficacy of therapeutic agents that are designed to control the signs and helpful when counselling owners with affected dogs. The current cohort, the largest reported sample of dogs affected with PD, helps to clarify the typical course of PD. The progression of PD reported herein provides the clinician with the general disease progression as observed by owners with dogs affected with the condition. This information can be used when counselling such owners regarding the expected progression of PD and in monitoring response to therapeutic agents. Conclusion The natural history of canine PD varies. In some, the episodes are static or slowly progressive with increased severity and frequency. The overall tendency is for the episodes to stabilise in the first few years following onset, at which point they may improve in the majority with clinical remission in the minority. Treatment trials for canine PD should consider the natural history of this disease in untreated dogs before misattributing remission to specific treatment effects. Conflict of interest statement None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of the paper. Acknowledgements The authors thank all owners for their time in participating to this study. Appendix: Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.tvjl.2016.03.007. References Bax, M., Goldstein, M., Rosenbaum, P., Leviton, A., Paneth, N., Dan, B., Jacobsson, B., Damiano, D., Executive Committee for the Definition of Cerebral Palsy, 2005. Proposed definition and classification of cerebral palsy, April 2005. Developmental Medicine and Child Neurology 47, 571–576. Bhatti, S.F., Vanhaesebrouck, A.E., Van Soens, I., Martlé, V.A., Polis, I.E., Rusbridge, C., Van Ham, L.M., 2011. Myokymia and neuromyotonia in 37 Jack Russell terriers. The Veterinary Journal 189, 284–288.
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
Black, V., Garosi, L., Lowrie, M., Harvey, R.J., Gale, J., 2013. Phenotypic characterisation of canine epileptoid cramping syndrome in the Border terrier. The Journal of Small Animal Practice 55, 102–107. Bruno, M.K., Lee, H.Y., Auburger, G.W., Friedman, A., Nielsen, J.E., Lang, A.E., Bertini, E., Van Bogaert, P., Averyanov, Y., Hallett, M., et al., 2007. Genotype-phenotype correlation of paroxysmal nonkinesigenic dyskinesia. Neurology 68, 1782– 1789. Crompton, D.E., Berkovic, S.F., 2009. The borderland of epilepsy: Clinical and molecular features of phenomena that mimic epileptic seizures. The Lancet. Neurology 8, 370–381. Dale, R.C., Irani, S.R., Brilot, F., Pillai, S., Webster, R., Gill, D., Lang, B., Vincent, A., 2009. N-methyl-D-aspartate receptor antibodies in pediatric dyskinetic encephalitis lethargica. Annals of Neurology 66, 704–709. Demirkiran, M., Jankovic, J., 1995. Paroxysmal dyskinesias: Clinical features and classification. Annals of Neurology 38, 571–579. Forman, O.P., Penderis, J., Hartley, C., Hayward, L.J., Ricketts, S.L., Mellersh, C.S., 2012. Parallel mapping and simultaneous sequencing reveals deletions in BCAN and FAM83H associated with discrete inherited disorders in a domestic dog breed. PLoS Genetics doi:10.1371/journal.pgen.1002462. Garosi, L.S., Platt, S.R., Shelton, G.D., 2002. Hypertonicity in Cavalier King Charles spaniels. Journal of Veterinary Internal Medicine 16, 330. Gill, J.L., Tsai, K.L., Krey, C., Noorai, R.E., Vanbellinghen, J.F., Garosi, L.S., Shelton, G.D., Clark, L.A., Harvey, R.J., 2012. A canine BCAN microdeletion associated with episodic falling syndrome. Neurobiology of Disease 45, 130–136.
37
Harcourt-Brown, T., 2008. Anticonvulsant responsive, episodic movement disorder in a German shorthaired pointer. The Journal of Small Animal Practice 49, 405–407. Jankovic, J., Demirkiran, M., 2002. Classification of paroxysmal dyskinesias and ataxias. Advances in Neurology 89, 387–400. Kube, S.A., Vernau, K.M., LeCouteur, R.A., 2006. Dyskinesia associated with oral phenobarbital administration in a dog. Journal of Veterinary Internal Medicine 20, 1238–1240. Lotze, T., Jankovic, J., 2003. Paroxysmal kinesigenic dyskinesias. Seminars in Pediatric Neurology 10, 68–79. Lowrie, M., Garden, O.A., Hadjivassiliou, M., Harvey, R.J., Sanders, D.S., Powell, R., Garosi, L., 2015. The clinical and serological effect of a gluten-free diet in Border terriers with epileptoid cramping syndrome. Journal of Veterinary Internal Medicine 29, 1564–1568. Mitek, A.E., Clark-Price, S.C., Boesch, J.M., 2013. Severe propofol-associated dystonia in a dog. Canadian Veterinary Journal 53, 471–474. O’Brien, D., Kolicheski, A., Packer, R., Thomovsky, S., Taylor, J., Schnabel, R., Berg, J., Vasquez, L., Johnson, G., 2015. Paroxysmal non-kinesogenic dyskinesia in soft coated wheaten terriers associated with a missense mutation in PIGN and responds to acetazolamide therapy. Journal of Veterinary Internal Medicine doi:10.1111/jvim.13002. Packer, R.A., Patterson, E.E., Taylor, J.F., Coates, J.R., Schnabel, R.D., O’Brien, D.P., 2010. Characterization and mode of inheritance of a paroxysmal dyskinesia in Chinook dogs. Journal of Veterinary Internal Medicine 24, 1305–1313.
Contents lists available at ScienceDirect
The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l
Natural history of canine paroxysmal movement disorders in Labrador retrievers and Jack Russell terriers Mark Lowrie a,*, Laurent Garosi b a b
Dovecote Veterinary Hospital, 5 Delven Lane, Castle Donington, Derby DE74 2LJ, UK Davies Veterinary Specialists, Manor Farm Business Park, Higham Gobion, Hitchin SG5 3HR, UK
A R T I C L E
I N F O
Article history: Accepted 3 March 2016 Keywords: Seizure Dog Dyskinesia Treatment Progression
A B S T R A C T
Delineation of the typical disease progression in canine paroxysmal dyskinesia (PD) may assist in evaluating therapeutic agents during clinical trials. Our objective was to establish the natural disease course in a group of dogs diagnosed with PD that received no medication. Fifty-nine dogs (36 Labradors, 23 JRTs) with clinically confirmed PD and a follow-up of ≥3 years were retrospectively reviewed. Dogs with PD had a young onset, were triggered by startle or sudden movements, and had a male bias (75%) with the majority being entire sample population. Twenty-one dogs (36%) had at least one event comprising cluster episodes. Episode duration and frequency varied dramatically, even within an individual. Median follow-up was 7 years. No concurrent disease was identified in any dog that was investigated. The natural history was self-limiting with 32% entering remission and an improvement in 75%. Episodes reduced in terms of frequency and duration in Labradors and JRTs respectively. Remission was lower in dogs with cluster episodes than those without. These findings suggest that the diagnostic yield of advanced neuroimaging techniques in dogs with video footage and historical data supporting PD, without neurological deficits, is low. The presence of cluster episodes is of predictive value for the prognosis of canine PD. Future research should be cautious in reporting treatment response for PD without first considering the spontaneous remission rate and improvements in untreated dogs documented in this study. Crown Copyright © 2016 Published by Elsevier Ltd. All rights reserved.
Introduction Paroxysmal dyskinesias (PD) are increasingly described although they remain poorly characterised in veterinary literature. They are recognised as a group of hyperkinetic paroxysmal movement disorders whose main feature is involuntary sustained muscle contraction. In PD, recurrent episodes are assumed to be the result of a molecular or structural abnormality and treatment, when indicated, is targeted at controlling episode frequency and severity. Known causes of PD in dogs are thus far limited to genetic (Forman et al., 2012; Gill et al., 2012; O’Brien et al., 2015), drug-induced (Kube et al., 2006; Mitek et al., 2013), and dietary factors (Lowrie et al., 2015). The first genetically mapped paroxysmal movement disorder was episodic falling syndrome in the Cavalier King Charles spaniel (Forman et al., 2012; Gill et al., 2012). Phenobarbital was found to elicit dyskinesia in an epileptic Chow (Kube et al., 2006). Border terriers suffer from canine epileptoid cramping syndrome (Black et al., 2013) which has been found to be a manifestation of gluten sensitivity (Lowrie et al., 2015). Although PD is most commonly of idiopathic or familial aetiology, other causes in people exist; for example, secondary or
* Corresponding author. Tel.: +44 1332 810395. E-mail address: [email protected] (M. Lowrie). http://dx.doi.org/10.1016/j.tvjl.2016.03.007 1090-0233/Crown Copyright © 2016 Published by Elsevier Ltd. All rights reserved.
symptomatic dyskinesia is reported, resulting from structural central nervous system lesions such as multiple sclerosis, head trauma, cerebral palsy, cerebrovascular accidents or encephalitis (Bax et al., 2005; Lotze and Jankovic, 2003; Dale et al., 2009). A current classification of human PD is one proposed by Demirkiran and Jankovic (1995), based solely on precipitating factors of the episodes and not phenomenology. This idea was born out of the observation that each type of PD can manifest with dystonia, chorea, athetosis, or a combination of abnormal movements. The classification distinguishes four categories: 1. paroxysmal kinesigenic dyskinesia (PKD) – incited by sudden movements. 2. paroxysmal nonkinesigenic dyskinesia (PNKD) – occurring spontaneously at rest 3. paroxysmal exertion-induced dyskinesia (PED) – precipitated by fatigue 4. paroxysmal hypnogenic dyskinesia (PHD) – attacks occur during sleep Various therapies are reported in the management of these conditions in people. Phenobarbital (Harcourt-Brown, 2008), acetazolamide (Gill et al., 2012; O’Brien et al., 2015) and clonazepam (Garosi et al., 2002) have been prescribed with varying success in
34
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
Table 1 Definitions of clinical signs that incorporate the term ‘dyskinesia’. Chorea Athetosis
Choreoathetosis Ballism Dystonia
An abrupt, unsustained contraction of different muscle groups A prolonged, slow contraction of the trunk muscles resulting in bending and writhing of the body and precluding maintenance of a stable posture Involuntary movements that have characteristics of both chorea and athetosis An abrupt contraction of the limb muscles which results in flailing movement of the limb and is often unilateral A sustained involuntary contraction of a group of muscles producing abnormal postures
dogs. However, future therapies should be given with an understanding of the natural progression of this condition in order to establish their efficacy. The natural course of PD is unknown. Delineation of the typical disease progression may assist in the evaluations of therapeutic agents during clinical trials. The following analysis was designed using the largest reported sample of dogs affected with PD to date. We performed this retrospective study to determine the prevalence of positive diagnostic findings in dogs with PD and to determine the natural course of this disease in two breeds of dog seen commonly with PD at our clinic; the Labrador and Jack Russell terrier (JRT). Materials and methods The study was designed as a retrospective hospital based study with followup. Dogs included in this study were client-owned clinical cases diagnosed with PD evaluated between January 2005 and January 2015 that had received no medication for PD, and with a follow-up of at least 3 years. Criteria were used to limit the phenotype of cases included by restricting selection to Labradors and JRTs only. These breeds that were chosen with PD have not been previously characterised in these breeds, though they commonly present with PD in our clinic. Analysis of each breed was performed separately. Inclusion required video evidence of at least one typical episode of PD at the time of or following investigation and its assessment by a Boardcertified neurologist. Dogs that had received treatment for PD were excluded from analysis. All dogs had to have at least a 6-month history of PD. Using phenomenological classification and in conjunction with the history, a diagnosis of PD was suspected in accordance with previous diagnostic guidelines (Packer et al., 2010; Black et al., 2013; Lowrie et al., 2015). Briefly, a diagnosis of PD was suspected when dogs had episodes in the absence of mentation alterations, autonomic signs, abnormal inter-ictal signs and post-ictal behaviour. Furthermore, the core movement had to be dyskinesia; that is, involving movement of the limb(s) (i.e. chorea, athetosis, dystonia or ballism; see Table 1 for definitions). Video examples are available of Labradors and JRTs having typical episodes (see Appendix: Supplementary Video S1 and S2). Clinical data were extracted, including signalment, duration and frequency of clinical signs (including the length of observation period over which these data were based), triggers (i.e. owners were asked if anything they had identified could trigger an episode) and survival data. Clinicopathologic data and diagnostic imaging results were also recorded where performed to identify concurrent disease processes. Follow-up was assessed via telephone interview with the owners of the dogs and/or referring veterinarians. The information obtained included current episode frequency and duration, presence or absence of cluster episodes, potential episode triggers, whether the dog was currently alive or dead, and, if applicable, the date and presumed cause of death. Two observational periods were chosen to assess episode frequency and duration; presentation and follow-up. The first observational period was calculated as the number of episodes per month in the 6 months prior to presentation at our clinic. Episode frequency at follow-up was calculated as the number of episodes per month in the 2 years preceding follow-up or in the 2 years prior to death and provided the second observational period. Episode duration in these observational periods was assigned to one of the following categories; <2 min, 2–5 min, 5–10 min, 10– 30 min, 30–60 min, 60–120 min, and >120 min. Cluster episodes were defined as more than one episode in a week. Remission of PD was defined as no episodes for ≥2 years. Remission was defined as either ‘early’ (i.e. within 24 months of onset), or ‘late’ (i.e. ≥24 months following onset).
Results Fifty-nine dogs met the inclusion criteria and were included in the analysis incorporating 36 Labradors and 23 JRTs. The majority (53/59) of owners kept a diary of each observed episode.
Signalment The median age of the 36 Labradors at episode onset was 2 years 3 months (range, 9 months to 10 years 8 months). Seven Labradors were female (19%; 2/7 neutered) and 29 were male (81%; 9/29 neutered). In contrast, the median age of onset in JRTs was 4 years 8 months (range, 1 year to 8 years) with 8/23 being female (35%; 5/8 neutered) and 15/23 being male (65%; 4/15 neutered). Triggers Episodes began most commonly following extremes of temperature in JRTs (83%; 19/23 dogs) and after sudden movements or being startled (58%; 21/36 dogs) or with excitement (25%; 9/36 dogs) in Labradors. The episodes always occurred at home and never happened when sleeping or during exercise. No diurnal rhythm was observed in any dog and many owners described stress or variation in daily routine as an inciting factor in both breeds. Frequency and duration of episodes The median duration of clinical signs in all dogs at presentation was 9 months (range 6 to 23 months); 10 months in Labradors (range 7 to 23 months) and 8 months in JRTs (range 6 to 21 months). The reported median frequency at presentation was one episode every 3 weeks in Labradors (range, one every 6 months to 12/ month) and one per month in JRTs (range, one every 6 months to two per month). Nineteen Labradors (19/44; 43%) and seven JRTs (2/23; 9%) had at least one event comprising cluster episodes. On average, duration of episodes in both Labradors and JRTs was similar (Figs. 1,2) although episodes varied dramatically in this respect, even within an individual. Results of investigations All dogs had unremarkable haematology and biochemistry profiles. Creatine kinase was increased in three dogs (all Labradors). Further testing in some dogs, which did not reveal any abnormalities, included dynamic bile acid testing (n = 25); urinalysis (n = 19); serum ammonia (n = 14); thyroid status (n = 8); ACTH stimulation test (n = 3); pre- and post-exercise plasma lactate and pyruvate concentrations (n = 2); acetylcholine receptor antibody serology (n = 2); ionised calcium (n = 1); echocardiogram (n = 12); halter monitoring during an episode (n = 1); cardiac (n = 2) and abdominal ultrasound (n = 14); radiographs of the chest (n = 17), abdomen (n = 2), stifles (n = 4), hips (n = 6) and spine (n = 1); urinary organic acids (n = 3); Neospora (n = 6) and Toxoplasma (n = 5) serology; electrophysiology (motor nerve conduction and electromyography: n = 5); CSF examination (n = 48); and MRI of the brain (n = 48), lumbosacral spine (n = 3), thoracic spine (n = 3), cervical spine (n = 3), and lumbar spine (n = 3). Follow-up Median follow-up time (from first observed episode to date of death or follow-up) in all dogs was 7 years. The median follow-up time was 6 years 8 months (range, 3 years to 12 years 2 months) in Labradors and 9 years (range, 3 years 4 months to 14 years 10 months) in JRTs. The frequency and duration of episodes was reported to have decreased in 25/36 (86%) Labradors, progressed in five Labradors and remained static in six Labradors at follow-up (Fig. 1). Owners of JRTs reported an improvement in duration and frequency in 19/ 23 (57%) dogs, a deterioration in 2/23 and a static course in 2/23 dogs (Fig. 2). Median frequency at follow-up was one episode every
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
35
Fig. 1. A chart comparing the duration of episodes at presentation and follow-up of paroxysmal dyskinesia in Labradors.
Fig. 2. A chart comparing the duration of episodes at presentation and follow-up of paroxysmal dyskinesia in Jack Russell terriers.
4 months in Labradors (range, 0–1 per month) and one every 6 months in JRTs (range, 0–1 every 6 weeks). Overall the episodes were reduced in terms of frequency and duration in both Labradors and JRTs. Fourteen Labradors (39%) met the criteria of spontaneous remission of which seven dogs had entered early remission while seven further dogs had entered late remission. The median time of remission in these Labradors was 5 years 8 months (range, 11 years 7 months to 14 years). Of the 22 Labradors failing to achieve remission, 18 dogs had a history of cluster episodes (i.e. representing 95% of the Labradors observed to have cluster episodes). Five JRTs (22%) were classified as entering spontaneous remission. All five were classified as having late remission. Median time in remission for these JRTs was 8 years (range, 7 years to 13 years 6 months). Both JRTs with cluster episodes failed to enter remission. Episode remission was lower in dogs with cluster episodes than those without. Fifty-seven of the 59 dogs that participated in the study were alive at the time of follow-up. The two deceased dogs were: a Labrador euthanased with lymphoma at 12 years 2 months, 10 years and 4 months after the onset of PD (this dog had been classified as being in late remission), and a Labrador euthanased as a result of severe arthritis at 11 years 8 months. All JRTs were alive at the time of follow-up. No episode was fatal in any dog. Dogs that were diagnosed with PD and had less than 3 years of follow-up were identified and none died or were euthanased in this time frame.
Discussion The main results of this retrospective, long-term populationbased study of 59 dogs followed since their first episode of PD, were: one; at the end of the median follow-up of 7 years 6 months, 39% of 36 Labradors were in spontaneous remission, compared to 22% of 23 JRTs. Secondly; among those in remission, 37% of dogs achieved uninterrupted episode freedom within 2 years of onset of the disease. The remaining 63% achieved spontaneous remission after 2 years. Thirdly; dogs that did not achieve remission were more likely to have experienced cluster episodes. There has been considerable confusion and disagreement amongst veterinarians pertaining to the diagnosis of PD and their distinction from epilepsy. There is strong evidence to suggest that similar pathogenic mechanisms may be involved in generating epileptic seizures and PD (Crompton and Berkovic, 2009). However, despite this common ancestry, their phenotype and clinical approach is distinct. Although simple partial seizures may manifest similarly and potentially be misdiagnosed as PD in that focal motor activity can occur while consciousness is maintained, the lack of autonomic signs, generalised nature of the dyskinesia, and long duration of episodes in some dogs make simple partial seizures unlikely. PD, with video correlation, normal inter-ictal findings and normal neurological examination, is unlikely to yield abnormal findings when performing screening diagnostic procedures. The aetiology of PD is uncertain but diagnostic screening for a structural cause has a low yield and this study may aid clinicians to define the most
36
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
cost effective approach in PD by avoiding further screening diagnostic tests when not indicated by unexpected abnormalities on history or clinical examination. The diagnosis of PD is therefore speculative in many cases. In people, it is the core feature of the movement disorder that should be considered to achieve a diagnosis. The reason for this is that many of these movement disorders may occur concurrently. In examining cohorts of Labradors and JRTs we accept the possibility that some of these dogs may have presented with more than just dyskinesia; for example myokymia and neuromyotonia, in the case of JRTs (Bhatti et al., 2011). However, when the dyskinesia predominated and involved all four limbs, this was sufficient for a diagnosis of PD in this study when the other inclusion criteria were fulfilled. Myokymia and neuromyotonia was excluded on clinical grounds as twitches remain within the affected body segment and movement is not observed. The clinical presentation of affected Labradors and JRTs was remarkably stereotyped fulfilling many of the criteria of PNKD in people (Jankovic and Demirkiran, 2002). In people, these signs can start from childhood but may develop at maturity. Episodes can last from minutes to hours with a notable predominance of males (Bruno et al., 2007). The frequency of attacks varies between individuals, ranging from daily attacks to only a few in a lifetime. Episodes may also be induced by a sudden change in position, for example from a sitting to standing position, or by being startled (Bruno et al., 2007). Neurological examination is normal between attacks. Episode frequency in people with PNKD wanes over time decreasing considerably or abating in adulthood, as found in 75% of dogs in our study. Other features consistent with PNKD include the lack of response to antiepileptic medication, the relatively low frequency of episodes and their relatively long duration. An autosomal dominant family history is present in about one in four cases in people (Jankovic and Demirkiran, 2002) but genetic studies have not yet been performed in Labradors. A dominant mode of inheritance seems unlikely given the limited pedigree data obtained from Labradors in this study (data not shown). A genetic association in JRTs is possible but was not investigated here due to the difficulties in identifying familial links in such a heterogeneous gene pool. The clinical variation observed between individuals in this study may represent distinct phenotypes within the breeds assessed or simply a wide spectrum of severity with a single aetiology. One challenge encountered in this study was the ability to construct a meaningful measure of episode frequency given that a dog may have six episodes in short succession followed by a period of 12 month quiescence, compared to a dog that has one episode every 2 months. The average episode frequency for these dogs would be classed as one episode every 2 weeks. However, both dogs clearly have a different phenotypic severity. This same idea also highlights the difficulty in interpreting treatment success as the nature of this condition is that it waxes and wanes and a short-term improvement following instigation of treatment could be falsely attributed to medication rather than disease repose. This was the reason for introducing the concept of cluster episodes to try and distinguish these two presentations. In dogs, the first genetically mapped PD was that associated with EFS in Cavalier King Charles spaniels (Forman et al., 2012; Gill et al., 2012). The mutation was characterised as a deletion affecting the brevican gene (BCAN) which encodes a brain-specific component of the extracellular matrix proteoglycan complex. It is thought to be involved in homeostasis and mutations of this protein result in a disruption of axonal conduction and synaptic stability. An interesting observation in EFS is that the condition is self-limiting in some cases (Forman et al., 2012). It is suggested that compensatory pathways involving up regulation of other proteoglycans may occur and account for this rectification. Dogs in our study also seemed to show this improvement with age so a similar compensatory mechanism may be considered.
It was important that we had a representative measure of frequency and duration prior to presentation and at follow-up. The nature of this study was that dogs were examined at different time intervals following onset and so a comparison between dogs was difficult. Clinical frequency at presentation in this study was based on the 6 months prior to presentation whereas follow-up was based on the 2 years prior to follow-up in each individual. This was chosen to allow a comparison to be drawn and for future investigations to use similar indices to enable comparison between studies. Our results are liable to recall bias. By asking owners the questions at two different time points we may find their perception of the problem and hence their answers will be influenced by their level of concern. Data on remission should not be subject to recall bias but results of improvement in PD may be amplified. Information on triggers was obtained by asking owners if they had observed anything that may induce an episode. It was surprising to see the level of unanimity obtained in the responses. However, it is important to emphasise that a failure to report a certain trigger does not necessarily exclude it as owners may not have observed certain triggers in their dog’s condition. In this respect, the triggers reported may have an increased incidence than is reported here. Even though the signs that occur with PD are fairly well characterised, progression of the disease remains uncertain. Clarification of the disease course is useful in determining the efficacy of therapeutic agents that are designed to control the signs and helpful when counselling owners with affected dogs. The current cohort, the largest reported sample of dogs affected with PD, helps to clarify the typical course of PD. The progression of PD reported herein provides the clinician with the general disease progression as observed by owners with dogs affected with the condition. This information can be used when counselling such owners regarding the expected progression of PD and in monitoring response to therapeutic agents. Conclusion The natural history of canine PD varies. In some, the episodes are static or slowly progressive with increased severity and frequency. The overall tendency is for the episodes to stabilise in the first few years following onset, at which point they may improve in the majority with clinical remission in the minority. Treatment trials for canine PD should consider the natural history of this disease in untreated dogs before misattributing remission to specific treatment effects. Conflict of interest statement None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of the paper. Acknowledgements The authors thank all owners for their time in participating to this study. Appendix: Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.tvjl.2016.03.007. References Bax, M., Goldstein, M., Rosenbaum, P., Leviton, A., Paneth, N., Dan, B., Jacobsson, B., Damiano, D., Executive Committee for the Definition of Cerebral Palsy, 2005. Proposed definition and classification of cerebral palsy, April 2005. Developmental Medicine and Child Neurology 47, 571–576. Bhatti, S.F., Vanhaesebrouck, A.E., Van Soens, I., Martlé, V.A., Polis, I.E., Rusbridge, C., Van Ham, L.M., 2011. Myokymia and neuromyotonia in 37 Jack Russell terriers. The Veterinary Journal 189, 284–288.
M. Lowrie, L. Garosi / The Veterinary Journal 213 (2016) 33–37
Black, V., Garosi, L., Lowrie, M., Harvey, R.J., Gale, J., 2013. Phenotypic characterisation of canine epileptoid cramping syndrome in the Border terrier. The Journal of Small Animal Practice 55, 102–107. Bruno, M.K., Lee, H.Y., Auburger, G.W., Friedman, A., Nielsen, J.E., Lang, A.E., Bertini, E., Van Bogaert, P., Averyanov, Y., Hallett, M., et al., 2007. Genotype-phenotype correlation of paroxysmal nonkinesigenic dyskinesia. Neurology 68, 1782– 1789. Crompton, D.E., Berkovic, S.F., 2009. The borderland of epilepsy: Clinical and molecular features of phenomena that mimic epileptic seizures. The Lancet. Neurology 8, 370–381. Dale, R.C., Irani, S.R., Brilot, F., Pillai, S., Webster, R., Gill, D., Lang, B., Vincent, A., 2009. N-methyl-D-aspartate receptor antibodies in pediatric dyskinetic encephalitis lethargica. Annals of Neurology 66, 704–709. Demirkiran, M., Jankovic, J., 1995. Paroxysmal dyskinesias: Clinical features and classification. Annals of Neurology 38, 571–579. Forman, O.P., Penderis, J., Hartley, C., Hayward, L.J., Ricketts, S.L., Mellersh, C.S., 2012. Parallel mapping and simultaneous sequencing reveals deletions in BCAN and FAM83H associated with discrete inherited disorders in a domestic dog breed. PLoS Genetics doi:10.1371/journal.pgen.1002462. Garosi, L.S., Platt, S.R., Shelton, G.D., 2002. Hypertonicity in Cavalier King Charles spaniels. Journal of Veterinary Internal Medicine 16, 330. Gill, J.L., Tsai, K.L., Krey, C., Noorai, R.E., Vanbellinghen, J.F., Garosi, L.S., Shelton, G.D., Clark, L.A., Harvey, R.J., 2012. A canine BCAN microdeletion associated with episodic falling syndrome. Neurobiology of Disease 45, 130–136.
37
Harcourt-Brown, T., 2008. Anticonvulsant responsive, episodic movement disorder in a German shorthaired pointer. The Journal of Small Animal Practice 49, 405–407. Jankovic, J., Demirkiran, M., 2002. Classification of paroxysmal dyskinesias and ataxias. Advances in Neurology 89, 387–400. Kube, S.A., Vernau, K.M., LeCouteur, R.A., 2006. Dyskinesia associated with oral phenobarbital administration in a dog. Journal of Veterinary Internal Medicine 20, 1238–1240. Lotze, T., Jankovic, J., 2003. Paroxysmal kinesigenic dyskinesias. Seminars in Pediatric Neurology 10, 68–79. Lowrie, M., Garden, O.A., Hadjivassiliou, M., Harvey, R.J., Sanders, D.S., Powell, R., Garosi, L., 2015. The clinical and serological effect of a gluten-free diet in Border terriers with epileptoid cramping syndrome. Journal of Veterinary Internal Medicine 29, 1564–1568. Mitek, A.E., Clark-Price, S.C., Boesch, J.M., 2013. Severe propofol-associated dystonia in a dog. Canadian Veterinary Journal 53, 471–474. O’Brien, D., Kolicheski, A., Packer, R., Thomovsky, S., Taylor, J., Schnabel, R., Berg, J., Vasquez, L., Johnson, G., 2015. Paroxysmal non-kinesogenic dyskinesia in soft coated wheaten terriers associated with a missense mutation in PIGN and responds to acetazolamide therapy. Journal of Veterinary Internal Medicine doi:10.1111/jvim.13002. Packer, R.A., Patterson, E.E., Taylor, J.F., Coates, J.R., Schnabel, R.D., O’Brien, D.P., 2010. Characterization and mode of inheritance of a paroxysmal dyskinesia in Chinook dogs. Journal of Veterinary Internal Medicine 24, 1305–1313.