Article Text

Download PDFPDF

Research paper
The impact and prognosis for dystonia in childhood including dystonic cerebral palsy: a clinical and demographic tertiary cohort study
  1. Jean-Pierre Lin1,
  2. Daniel E Lumsden1,2,
  3. Hortensia Gimeno1,
  4. Margaret Kaminska1
  1. 1Complex Motor Disorders Service, Evelina London Children's Hospital, Guy's & St Thomas’ NHS Foundation Trust, London, UK
  2. 2Rayne Institute, King's College London, London, UK
  1. Correspondence to Dr Jean-Pierre Lin, Consultant Paediatric Neurologist, Complex Motor Disorders Services, Evelina Children's Hospital, Guy's and St Thomas’ NHS Foundation Trust, Westminster Bridge Road, London, SE1 7EH, UK; jean-pierre.lin{at}gstt.nhs.uk and jeanpierrelin{at}icloud.com

Abstract

Introduction and methods The impact of dystonia in childhood is poorly understood. We report our experience of referrals between 2005 and 2012.

Results Of 294/315 assessable children, 15/294 had pure spasticity, leaving 279/294 with dystonia classified as primary (30/279: 10.7%); primary-plus (19/279: 6.8%) and secondary (230/279: 82.4%) dystonia, including heredodegenerative dystonia (29/279: 10.3%); 150/279 (53.7%) with cerebral palsy and 51/279 (18.2%) acquired brain injury. Definitive diagnoses were available in 222/294 (79.6%), but lower in primary/primary-plus compared with secondary groups (11/49 vs 211/230: Fisher's exact test p<0.0001). Spasticity comorbidity was present in 79/230 (34.3%) children.

Median age (interquartile years) at referral was 9.75 (6.58–13), not significantly differing by aetiology (Kruskal–Wallis test p>0.05); dystonia-onset age was 3 (0.5–7.0) for primary/primary-plus and 0.25 (0.08–0.8) in the secondary/CP groups. Dystonia duration at referral was 4.75 years (3.0–10.33) for primary/primary-plus groups and 7.83 (5.4–11) in the secondary group. The mean (interquartile range) proportion of life lived with dystonia, derived as dystonia duration normalised to age was 0.68 (0.31–0.96); 0.59 (0.35–0.8); 0.75 (0.62–0.95)and 0.9 (0.92–0.99) for primary, primary-plus, heredodegenerative and secondary-static dystonias respectively.

Only 91/279 (32.6%) experienced a period of normal motor development. Carers perceived dystonia deterioration in 168/279 (60.2%), stabilisation in 88/279 (31.5%) and improvement in 23/279 (8.2%). Dystonia occurred in 26/225 (11.6%) siblings: 14/26 secondary and 5/26 heredodegenerative dystonia. Comorbidities were identified in 176/279 (63.1%) cases.

Gross Motor Function Classification System (GMFCS) levels I–III were commoner in primary/primary-plus (37/49: 75%) compared with secondary/CP (29/230: 13%) cases, χ2 p<0.0001).

Discussion In this selective cohort, childhood dystonia is severe, presenting early before worsening without remission. Secondary dystonias spend a higher proportion of life living with dystonia and lower functional capacity. Despite referral bias, services offering neurosurgical interventions and health service planning agencies should understand the context and predicament of life with childhood dystonia.

  • Dystonia
  • Paediatric Neurology
  • Spasticity
  • Cerebral Palsy
  • Movement Disorders

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Movement disorders including dystonia, chorea and hemiballismus represent some of the most challenging clinical problems. They reflect imbalances between desired and competing motor patterns thought to arise from the physiological interaction of cerebellothalamocortical basal ganglia circuits: summarised as ‘a failure to inhibit unwanted movements (eg, chorea, dystonia and tics) through abnormal activation patterns of groups of striatal neurons.1 Dystonia has many causes and often no structural lesion can be found to explain very severe dystonia. It is widely recognised by clinicians managing childhood dystonic movement disorders that the causes and associated features differ from those seen in adulthood.2 The clinical features of childhood dystonia are likely to be linked to disturbances in the recently described segregated and integrative connectivity patterns in the human basal ganglia.3 This model co-locates in each basal ganglia structure posterior sensorimotor functions, middle cognitive associative functions and anterior limbic functions. Likewise, our understanding of how we develop habitual as opposed to goal-directed movements is likely to be as important for dystonia in adults and children as it is for Parkinson's disease.4

The impact of dystonia on the important phases of childhood is not well described, and there are no clear descriptions of the natural history of dystonia in children to inform carers or health professionals about prognosis for motor and non-motor features.

A complicating factor is the impact of dystonia on the developing brain, the growing musculoskeletal system, functional adaptation to increasing demands and the extent to which early interventions can improve the prognosis for functionally adaptive development.

A clearer understanding of the predicament of children living with dystonia is essential to evaluate the efficacy of therapeutic strategies for relieving dystonia in childhood and in particular to understand whether there are any crucial ‘windows of opportunity’ for such interventions.

A significant limitation to our understanding is that dystonia in childhood may be under-recognised by paediatricians and allied health professionals working with children5 despite definitions by the Taskforce on Childhood Motor Disorders as “….a movement disorder in which involuntary sustained or intermittent muscle contractions cause twisting and repetitive movements, abnormal postures, or both”.6 This is also consistent with the most recently proposed consensus definition for dystonia in adulthood.7 Dystonia in childhood may coexist with spasticity8 ,9 in symptomatic or ‘secondary’ disorders,10 but the data on frequency of this association are again limited. With advances in neurostimulator technology, deep brain stimulation (DBS) may now be possible in younger children of 3–4 years of age or weighing 10 kg.11 ,12 Given the apparently increased response to DBS seen with shorter disease duration,12–14 there is increasing interest in exploring earlier intervention with neuromodulation. On the other hand, understanding what childhood disorders share in common or the physical and temporal characteristics that make them different from each other could help explain differences in outcome.

It is therefore essential that the characteristics and operational definitions relating to childhood dystonia are better understood for accurate ascertainment of numbers of children affected; to determine similarities and differences between childhood and adult features and also for future interventional study design such as neuromodulation with DBS and other therapeutic modalities.

Methods

Data were collected retrospectively from the clinical records of children and young people (CAYP) referred to our supraregional Complex Motor Disorder Service (CMDS) for possible DBS, Intrathecal Baclofen infusion pump (ITB) or other medical interventions to relieve hypertonicity between July 2005 and January 2012. The CMDS receives referrals from tertiary referral centres throughout England, Wales, Scotland, Northern Ireland and also the Republic of Ireland.

A pro-forma procedure was used to record clinical characteristics. Data were collected regarding clinical and demographic features at initial assessment by the CMDS, including age of onset and duration of dystonic symptoms, aetiological classification of dystonia,15 absence or presence of spasticity as defined by the Taskforce for Childhood Motor Disorders8 and the operational definitions described in Lin16 and Lin17 . Particular note was made of the underlying diagnosis, presence of other motor signs and other coincident medical conditions. A primary dystonia was classified in the absence of a structural abnormality on MRI or of a demonstrable neurometabolic disorder. Neurotransmitter screening and L-DOPA/dopamine agonist trials were performed where indicated but had usually been performed by the referring centres. The commonest genetic testing included CGH arrays, DYT1, DYT11, DYT6, TITF1 and GCH1. Further screening was performed where appropriate. Where dystonia parkinsonism was clinically detected, extensive Parkinson mutation screening was often performed.

The functional classification was measured by the Gross Motor Function Classification System (GMFCS) 18 and is described in box 1. Parental perception of whether dystonia was improving, static or worsening over time, the presence of siblings affected with dystonia and whether the child had experienced a period of normal motor development were also recorded.

Box 1 Gross Motor Function Classification System

GMFCS Level I

Walks without restrictions

Athletic skills: can run and jump but speed, balance and coordination reduced

Climbs stairs without handrail

GMFCS Level II

Walks unassisted

Limitations walking outdoors on uneven surfaces, inclines and crowds

Negotiates stairs using handrail

GMFCS Level III

Walks with assistive mobility devices indoors and outdoors on level surfaces

May propel a manual wheelchair

Requires assistance for long distances or uneven surfaces

Climbs stairs using handrail

Has floor mobility

Sits independently

GMFCS Level IV

Walking indoors severely limited even with assistive devices

May steer powered wheelchair

Standing transfers may be possible with or without assistance

Supported sitting only

GMFCS Level V

No independent mobility

Cannot sit or stand independently: poor head control

May use powered mobility

After Palisano et al18

Results

Sufficient clinical details were available on 294/315 referrals (93.3%). Fifteen children were excluded as they were found to have pure spasticity on clinical examination. The remaining 279/294 (figure 1) were classified as primary (30/279: 10.8%); primary-plus (19/279: 6.8%) and secondary (230/279: 82.4%) dystonia, respectively, including heredodegenerative dystonia (29/279: 10.4%) leaving 201/279 (72%) with secondary (non-progressive) dystonia comprising either cerebral palsy (150/279: 53.3%) or acquired brain injury (51/279: 18.2%) of multiple aetiology (figure 1).

Figure 1

Clinical demographics of cohort, breaking down aetiological classification and Gross Motor Function Classification System.

Definitive aetiological diagnoses (figure 2) had been made at the point of referral to our service across the cohort in 222 cases (79.6%). The proportion of children with an aetiological diagnosis was lower in the combined primary-primary-plus group (DYT1=4; idiopathic=26; dystonia-myoclonus DYT11-positive=5; dystonia-myoclonus syndrome=7; Benign Hereditary Chorea TITF1 mutation positive=2; dystonia parkinsonism=2; other mixed=3) compared with the secondary group (11/49 compared with 211/230, Fisher's exact test p<0.0001). Coexistent spasticity was present in 79/230 (34.3%) of children after excluding primary/primary-plus dystonias.

Figure 2

Diagnostic cause of dystonia across the group. Categories were thresholded at 1% of the total number of children in the cohort. All diagnosis with <1% of the total cohort were combined in the ‘other’ group. Vertical red arrows indicate those cases that would collectively meet a diagnosis of cerebral palsy. These have been further subdivided into cerebral palsy caused by hypoxic ischaemic encephalopathy (HIE) at term, as a consequence of premature delivery, as a consequence of kernicterus and miscellaneous other causes. HIE, hypoxic ischaemic encephalopathy; CP, cerebral palsy; PKAN, pantothenate kinase associated neurodegeneration; CVA, cerebrovascular accident; TBI, traumatic brain injury; RTA, road traffic accident.

The median age (interquartile years) at referral was 9.75 (6.58–13), with no statistical significant difference between aetiological classifications (Kruskal–Wallis test p>0.05); dystonia-onset age was 3 (0.5–7.0) years for primary/primary-plus and 0.25 (0.08–0.8) years in the secondary/cerebral palsy (CP) groups. A significant difference in duration of dystonia at the time of referral was also found between the primary/primary-plus and secondary dystonia/CP groups, median (interquartile) dystonia duration at referral being 4.75 years (3.0–10.33) in the primary/primary-plus group and 7.83 (5.4–11) in the secondary/CP group. The mean (interquartile range) proportion of life lived with dystonia, derived as dystonia duration normalised to age was 0.68 (0.31–0.96); 0.59 (0.35–0.8); 0.75 (0.62–0.95)and 0.9 (0.92-0.99) for primary, primary-plus, heredodegenerative and secondary-static dystonias respectively.

Only 91/279 (32.6%) of children with dystonia were reported to have had a period of normal motor development (figure 3A), a factor that may have important prognostic significance.

Figure 3

(A) Proportion of children experiencing normal motor development prior to onset of dystonia, expressed as percentage of total number of children within each aetiological classification. (B) Carer perception of change in dystonia over time (improvement, no change or deterioration) expressed as a percentage of total number of children within each aetiological classification.

Parental perception of change in dystonia over time (figure 3B) was that deterioration had occurred in 168/279 (60.2%), dystonia was stable in 88/279 (31.5%), with perceived improvement in only 23/279 (8.2%).

Worsening of dystonia over time was the most common perception across all aetiological groups: 20/30 (66%) of the primary dystonia group, 14/19 (73%) of the primary-plus group and 134/230 (58%) of the secondary dystonia/CP group. Dystonia in childhood worsened regardless of aetiology (figure 3B).

The majority of children had siblings (225/279, 80.6%), of whom 26/225 (11.6%) also had documented dystonic symptoms. The distribution of aetiological classification of cases with affected siblings was primary (5/26), primary-plus (2/26) and secondary (19/26), of which 5/19 were attributable to heredodegenerative causes. Additional medical problems not directly related to dystonia were identified in the majority of CAYP 176/279 (63.1%).

The distribution of GMFCS levels across the cohort was dominated by marked severity as follows (see box 1 for definitions): level I=26, level II=26, level III=14, level IV=40 and level V=170. A difference in functional classification was found across aetiological groups. Dichotomising children into higher functioning (GMFCS levels I–III) and lower functioning (GMFCS levels IV–V), a higher proportion of higher functioning was seen in the primary/primary-plus compared with secondary dystonia groups (37/49 or 75% in GMFCS I–III compared with 29/230 or 13% in GMFCS I–III for the secondary cases, χ2 p<0.0001).

Discussion

With increasing interest in early intervention in the management of dystonia, a full understanding of the characteristics of the dystonia in childhood is essential. Our study reflects a UK-wide pattern of referral of children with dystonia for possible DBS or ITB neurosurgical intervention. This study reports the overall characteristics of all these children with a view to providing an overview of the predicament of living with childhood dystonia in the early 21st century. These referrals were motivated by clinicians and families seeking improved medical and surgical solutions to improve comfort through dystonia management.

Notwithstanding the evident referral bias inherent in this selected cohort study, two pieces of received wisdom regarding childhood dystonia are confirmed:

  1. Secondary dystonia (230/279) is far more common than primary dystonia (30/279) or primary-plus dystonia (19/279). Dystonic CP being the largest grouping within the secondary dystonias (150/230).

  2. Spasticity coexists with dystonia in a third of cases (79/230).

Our understanding of the pathophysiology of dystonia, including concepts of abnormal inhibition, abnormalities of sensorimotor integration and abnormal plasticity, has largely been informed by the findings of studies in the adult population with primary or focal dystonias.19 It is far from clear to what extent findings relating to adult pathophysiology can be extrapolated to childhood dystonia, and emerging evidence would suggest that the cardinal features may not be shared by the secondary dystonia/CP children.20

A number of other observations emerge from our current study. Regardless of dystonia aetiology or duration, two-thirds of families in this cohort reported a worsening in dystonia over time, with few families, only 8%, perceiving an improvement in dystonia symptoms (figure 3B). Children with secondary dystonia/CP were more likely to have a confirmed aetiological diagnosis compared with primary/primary-plus dystonia. The primary dystonias are generally considered to be genetic disorders. A lack of a confirmed causative mutation limits genetic counselling for families and introduces greater clinical uncertainty regarding the therapeutic prognosis. In some childhood epilepsies, a positive genetic diagnosis was felt by families to help with accepting the neurological condition, facilitating access to therapies and to respite care,21 highlighting the potential importance of such a diagnosis for children with neurological conditions.

The majority of children described in this sample often presented with other, non-dystonia-related medical conditions, of which seizures in infancy were commonly experienced, for example, in perinatal hypoxic-ischaemic encephalopathy. The relative influence of these other medical concerns and the potential for antidystonic therapies to exacerbate them further add to the complexity in management and clinical heterogeneity in the cohort.

In a significant minority of cases, 11–12% of siblings exhibited dystonia not confined to primary dystonias, which may be worthy of detailed molecular genetic exploration. Involvement of other children and family burden of care is of particular importance when planning interventions that may be delivered at a distance from the family home, for example, DBS surgery. Provision of care for affected siblings while a family member undergoes such a procedure is likely to create more practical difficulties for the family.

The systematic occurrence of a previously normal motor development across all aetiological categories including a fifth of children with secondary dystonia/CP was noteworthy given the potential prognostic value for the impact of intervention. Previous normal development might be construed as ‘money in the bank’, that is, a potentially accessible developmental resource if the dystonia can be moderated. This contrasts with a history of normal development in two-thirds of children with primary, primary-plus and heredogenerative dystonias (figure 3A).

Another key observation from our sample was the greater severity of functional impairment in the children with secondary compared with primary/primary-plus dystonia. There was clearly a trend towards referring the severest cases that are also the most difficult to manage, hence the very high proportion of GMFCS level IV and V cases.

It is impossible to determine the precise contribution of coincident spasticity or of other medical conditions to impairment in gross motor function. Our experience with children from this cohort undergoing selection for DBS surgery indicates a greater severity of dystonia as defined by the GMFCS and higher baseline Burke–Fahn–Marsden Dystonia Rating Scale22 scores in CAYP with secondary dystonia/CP.12

Age at referral to our service did not differ by aetiological group, but age of onset did. Consequently, duration of dystonia differed at the time of referral. Dystonia duration at the time of DBS surgery appears to inversely correlate with response following DBS,12–14 leading to an important therapeutic objective of earlier intervention. Since secondary dystonia/CP onset occurs at approximately half the age at which primary dystonias present, early intervention necessarily implies intervening at much younger ages in secondary dystonia/CP. In this cohort, the secondary/CP dystonia cases had spent twice as long living with dystonias as their primary/primary-plus counterparts. There is also limited evidence to suggest that the effects of certain medication such as trihexyphenidyl are better tolerated in younger children.2 ,23 Taken together, these findings suggest that early referral to specialist services for childhood dystonia management at a younger age should be encouraged. Appropriate measures of change, health-related quality of life and care burden for children with dystonia are required to measure meaningful change after intervention and to inform study sample size. Reported improvements following DBS in preliminary studies using the Melbourne Assessment of Upper Limb Function (MAULF) 24 and The Canadian Occupational Performance Measure may provide more useful measures of functional change in children than a change in dystonia alone.25–27

Trials of early intervention before the child has experienced greater than 75% of life living with dystonia have been advocated.7 Such studies are required to determine whether the downstream consequences of dystonia in childhood can be alleviated and may contribute to our understanding of the early impact of primary and secondary dystonia in children. Notwithstanding the likely under-recognition and reporting of dystonia CP, this study also underlines the relevance and contribution of dystonia pathophysiology to the disability of cerebral palsy.5 ,7 ,16 ,17 ,28 A direct focus of the clinical strategy on reducing the proportion of life lived with dystonia may prove beneficial to all children growing up with dystonia, irrespective of aetiology, by relieving pain, increasing independence and preventing deformity.

Limitations

This study has all the limitations of a cohort study reflecting referral of the worst dystonias of childhood and probably excludes the mild focal and generalised dystonias. But the referral pattern is probably typical of the range and type of challenging dystonic children seen in most regional clinics.

More work is clearly required to understand differences between the developmental plasticity of typically developing children and children living with dystonia. For instance: do children with dystonia have ‘too much plasticity’ or on the contrary ‘not enough plasticity’? The functional burden of dystonia, likelihood of worsening and lack of likely spontaneous remission in this selected cohort further serves to emphasise the importance of increasing interventional research efforts in the paediatric dystonia population to reduce comorbidities and improve the quality of life for children with dystonia. More work is required to recognise the severity of dystonia in children,29 obtain true population data on dystonia in children and especially to determine the quality of life of children with dystonia and the burden of care to parents and carers. Improved understanding and management plans for dystonia emergencies such as ‘brittle dystonia’ and status dystonicus30 in children will be helped by an overall understanding of the demographic impact of dystonia in children.

Dystonia severity, timing of onset, duration, proportion of life lived with dystonia, nature and timing of intervention and time at follow-up are all factors likely to influence assessment of outcome. Defining what constitutes meaningful change in childhood dystonia remains a challenge before formal multicentre clinical trials can begin to demonstrate ‘what works’ in these complex movement disorders.

Acknowledgments

We thank the families and children and their referring clinicians within the UK and Ireland. This work was supported by a generous grant from the Guy's and St Thomas’ Charitable Trust number G060708 and a Dystonia Society UK grant 01/2011.

References

Footnotes

  • Contributors J-PL conceived, organised, supported, executed, wrote and reviewed the study. DEL and HG organised, supported, executed, wrote and reviewed the study. MK supported, executed, wrote and reviewed the study.

  • Funding DEL is partly supported by a Dystonia Society UK grant 01/2011, and HG has also been supported by a Dystonia Society UK grant 07/2013. J-PL has held grants from the Guy's and St Thomas Charity (grant G060708), the Dystonia Society UK (grant 01/2011 and 07/2013) and Action Medical Research (grant GN2097). All authors have received unrestricted educational support from Medtronic Ltd. J-PL has acted as a consultant for Medtronic Ltd.

  • Competing interests The Complex Motor Disorders Service has benefited from an unrestricted educational grant by Medtronic to present work at international conferences.

  • Provenance and peer review Not commissioned; externally peer reviewed.