SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Aim

The aim of this study was to identify clinical and radiological predictors of activities of daily living (ADL) outcomes in children with cerebellar atrophy.

Method

Over a period of 5 years, we evaluated 44 participants (25 males, 19 females) children with confirmed cerebellar atrophy using magnetic resonance imaging (MRI). The median age at the time of assessment 9 years; range 16mo–18y. Participants were grouped according to whether the cerebellar atrophy was isolated or associated with other radiological abnormalities. Severity of cerebellar atrophy was graded using qualitative and quantitative scoring systems. A standardized ADL assessment was used to characterize functional outcomes. The characteristics of the participants were analysed using descriptive statistics.

Results

The mean age at symptom onset was 20 months (range birth–10y). The group with isolated cerebellar atrophy had better outcomes than the group with cerebellar atrophy associated with other radiological abnormalities, with a mean total ADL score difference of 8.0 points (95% confidence interval 1.8–14.2 points, p=0.01). Age at onset of cerebellar atrophy before 2 years of age, progression of cerebellar atrophy on magnetic resonance imaging, presence of seizures, and decreased size of transverse cerebellar hemisphere diameter were all associated with worse outcomes.

Interpretation

We present a prospective study of clinical and radiological predictors of ADL outcome in children with cerebellar atrophy. This information may be useful in the diagnosis and future management of this complex group of disorders.

Abbreviations
ADL

Activities of daily living

CA+

Cerebellar atrophy plus

FLAIR

Fluid-attenuated inversion recovery

ICA

Isolated cerebellar atrophy

Childhood-onset cerebellar atrophy is associated with a heterogeneous group of disorders.[1, 2] Cerebellar atrophy is defined as a progressive loss of cerebellar tissue, but with initially normal structures, in a posterior fossa of normal size.[3] The number of vermian and cerebellar lobules is normal, but arborization of the lobes of the vermis is accentuated as a result of enlarged interfoliate fissures and sulci. Non-genetic causes of cerebellar atrophy include cerebellar infections; immune-mediated processes such as systemic lupus erythematosus, multiple sclerosis, and paraneoplastic syndromes; toxins; some antiepileptic medications; tumours; and vascular events.[4, 5] Genetic causes of cerebellar atrophy include chromosomal, single gene, mitochondrial, and other metabolic disorders.[2, 3, 5] The most common genetic disorders associated with cerebellar atrophy are presented in Table 1.

Table 1. Common genetic causes of childhood-onset cerebellar atrophy
OnsetDisease
0–1y (infancy)Pontocerebellar hypoplasia types 1 and 2
Infantile neuroaxonal dystrophy
Pelizaeus–Merzbacher disease
Congenital disorders of glycosylation
Infantile-onset spinocerebellar ataxia
Cockayne syndrome
Neuronal ceroid lipofuscinosis (infantile)
Spinocerebellar ataxia, autosomal recessive with axonal neuropathy
Mevalonate kinase deficiency
3-Methylglutaconicaciduria
Salla disease
Progressive encephalopathy with oedema, hypsarrhythmia, and optic atrophy
1–2y (late infancy)Hypomyelination with atrophy of the basal ganglia and cerebellum
Ataxia telangiectasia
Marinesco–Sjögren syndrome
Neuronal ceroid lipofuscinosis (late infantile)
2–18y (early and late childhood)Ataxia with oculomotor apraxia 1
Ataxia with oculomotor apraxia 2
Ataxia-telangiectasia-like disorder
Spastic ataxia of Charlevoix–Saguenay
Cerebrotendinous xanthomatosis
Myoclonic epilepsy of Unverricht–Lundborg disease
Coenzyme Q10 deficiency
Episodic ataxia 1
CACNA1A-related disorders
Leukoencephalopathy with ataxia, hypodontia, and hypomyelination
Juvenile-onset dentatorubral pallidoluysian atrophy
Neuronal ceroid lipofuscinosis (late infantile/juvenile)
Late-onset GM2 gangliosidosis
VariableMitochondrial
Spinocerebellar ataxia (s)
Niemann–Pick disease type C
Adrenoleukodystrophy

The identification of specific aetiologies of cerebellar atrophy is an essential step in counselling families regarding potential outcomes and prognosis for their children. Accurate genetic diagnosis of these disorders is complicated by both clinical and genetic heterogeneity, however, and many patients never receive a genetic diagnosis.[2, 6] In a recent retrospective study of 300 children with cerebellar atrophy, a diagnosis was established in less than half (47%).[6] As the diagnosis depends on whether the cerebellar atrophy is isolated or associated with additional signs such as hypomyelination, progressive white matter abnormalities, basal ganglia changes, or cerebellar white matter abnormalities,[3, 6] we wished to determine whether functional outcomes would be different between these two neuroradiologically defined groups.

Clinical trials of potential therapies for adult patients with cerebellar disorders are rapidly emerging, and highlight the need for accurate outcome measures, including an assessment of activities of daily living (ADL).[7, 8] Studies assessing the neurodevelopmental and functional outcome in children with ataxias associated with cerebellar atrophy are lacking, and there are no prospective studies which assess functional outcomes in children with cerebellar atrophy. We present the first prospective study of clinical and radiological predictors of ADL outcome in children with cerebellar atrophy, and describe the prevalence of disability related to ADL in this patient population.

Method

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Study population

Participants were identified during an institutional research ethics board-approved study of children with cerebellar atrophy over a 5-year period (2007–2012). They were selected consecutively upon presentation to the Neurogenetics Clinic, Hospital for Sick Children, Toronto, Canada, with informed parental consent. Participants with cerebellar atrophy due to confirmed treated posterior fossa tumours, posterior fossa surgery, cerebellar haemorrhage, or ischaemia (including perinatal ischaemic insult) and non-genetic aetiologies of cerebellar atrophy were excluded. Participants with posterior fossa malformations including Dandy–Walker and Chiari malformations type 1 and 2, Joubert syndrome, rhomboencephalosynapsis, and mega cisterna magna were also excluded.

A total of 44 participants, (25 males, 19 females) were evaluated by a neurogeneticist (GY). Detailed clinical data including sex, age at onset of symptoms, mode of presentation, and family history were collected. A standardized neurological examination was performed at each visit. All patients had a standardized panel of laboratory investigations including metabolic studies (pyruvate, lactate, peroxisomal and lysosomal enzyme activity, lipid profiles, alpha-fetoprotein, amino acids, organic acids, and transferrin isoelectric focusing), immunological studies, and cytogenetic tests. Molecular genetic tests, electromyography, nerve conduction studies, echocardiography, and muscle and skin biopsies were carried out when deemed necessary, according to clinical symptoms.

Magnetic resonance imaging studies

Brain magnetic resonance imaging (MRI; 1.5T) was performed at least once on each patient. MRI was performed 73 times on 44 participants (1.5T). All examinations included sagittal T1-weighted and axial/coronal T2-weighted images; fluid-attenuated inversion recovery (FLAIR) sequences, diffusion-weighted imaging, and magnetic resonance spectroscopy were reviewed when available. The median age at first MRI was 5 years 10 months (range 1d–18y 11mo). Follow-up MRI was carried out in 28 participants, with the time from initial MRI ranging from 3 to 7 years.

All MRI findings and neuroradiology reports were reviewed by an experienced neuroradiologist (SB). The presence of atrophy involving the cerebellar vermis or hemispheres was noted. Biometric measurements of the brainstem were performed with specific anteroposterior measurements of the mesencephalon, pons, medulla, and clava. The craniocaudal dimension of the vermis, or maximum vermian height, was measured on the midline sagittal image as per the published criteria for cerebello vermian biometry.[9] The maximum transverse diameter of the cerebellum was measured on coronal images and plotted against age-matched norms (Fig. 1a).

image

Figure 1. (a) Biometry shown on T1-weighted midline sagittal image. Anteroposterior measurements were performed of the mesencephalon at the junction of the mesencephalon and pons (black dotted line), at the level of the mid-pons (black dashed line), at the level of the medulla at the pontomedullary junction (black line), and at the level of the clava (thick black line). The tegmento vermian angle is the angle between the dotted white lines drawn along the tegmentum of the brainstem and the anterior aspect of the vermis. The maximum craniocaudal height of the vermis is measured between the tip of the culmen lobule and the pyramid lobule, as shown. (b) Transverse cerebellar diameter. Fluid-attenuated inversion recovery coronal image in a child with cerebellar atrophy demonstrates marked widening of the interfoliate sulci and a bright cerebellar cortex (arrow). The white line denotes the transverse cerebellar diameter.

Download figure to PowerPoint

The tegmento vermian angle, defined as the angle between the brainstem and anterior aspect of the cerebellum, was measured (Fig. 1b). The anterior and posterior lobes of the vermis were assessed for the presence of atrophy by evaluation of the size and width of the interfoliate sulci. Atrophy of the vermian and cerebellar hemispheres was graded as 0, no atrophy (visualization of a few of the interfoliate sulci of the anterior lobule); 1, mild atrophy (visualization of most of the interfoliate sulci with separation of the folia); or 2, severe atrophy (marked prominence of the interfoliate sulci; Fig. 2). Progression of cerebellar atrophy was assessed when two or more MRI studies were obtained. T2-weighted magnetic resonance imaging and FLAIR signal abnormality of the cerebellar cortex were assessed for normal (0), mild (1), or severely increased signal intensity (2). Dentate nuclei and cerebellar white matter signal abnormalities were assessed. Signal abnormalities of the supratentorial white matter, cortex, and basal ganglia were recorded. For the purpose of our analysis, participants were divided into those with isolated cerebellar atrophy and those with additional MRI abnormalities; the latter group was designated ‘cerebellar atrophy plus’ (CA+).

image

Figure 2. Vermian atrophy. Sagittal T1-weighted image of (a) a typically developing comparison child and (b) a child with mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes reveals an increased tegmento vermian angle and mild prominence of the interfoliate sulci. (c) sagittal T1-weighted image a child with late-onset GM2 gangliosidosis demonstrates an increased tegmento vermian angle and marked prominence of the interfoliate sulci. In both (b) and (c), there is mild thinning of the brainstem when compared with the typically developing comparison child (a).

Download figure to PowerPoint

Standardized outcome measures

We carried out a prospective study of functional outcome in participants with cerebellar atrophy using a previously validated ADL assessment tool.[10] Assessment of nine domains of ADL was conducted: speech, swallowing, ability to feed self, dressing, sitting, walking, frequency of falls, bladder function, and self-hygiene. Scores ranged from 0, indicating normal function, to 36, indicating very severe functional disability (Table 2). The median age at the time of assessment was 9 years (range 16mo–18y).

Table 2. Activities of daily living assessment scorea
DomainScore
  1. a

    From Subramony et al.[10]

Speech0 – Normal
1 – Mildly affected: no difficulty being understood
2 – Moderately affected: sometimes asked to repeat statements
3 – Severely affected: frequently asked to repeat statements
4 – Unintelligible most of the time
Swallowing0 – Normal
1 – Rare choking (<once a month)
2 – Frequent choking (<once a week, >once a month)
3 – Requires modified food or chokes multiple times a week. Or patient avoids certain foods
4 – Requires nasogastric tube or gastrostomy feedings
Cutting food and handling utensils0 – Normal
1 – Somewhat slow and clumsy, but no help needed
2 – Clumsy and slow, but can cut most foods with some help needed. Or needs assistance when in a hurry
3 – Food must be cut by someone, but can still feed self slowly
4 – Needs to be fed
Dressing0 – Normal
1 – Somewhat slow, but no help needed
2 – Occasional assistance with buttoning (getting arms in sleeves, etc.) or has to modify activity in some way (e.g. having to sit to get dressed; use velcro for shoes, stop wearing ties, etc.)
3 – Considerable help required, but can do some things alone
4 – Helpless
Personal hygiene0 – Normal
1 – Somewhat slow, but no help needed
2 – Very slow hygienic care or has need for devices such as special grab bars, tub bench, shower chair, etc.
3 – Requires personal help with washing, brushing teeth, combing hair, or using toilet
4 – Fully dependent
Falling (assistive device = score of 3)0 – Normal
1 – Rare falling (<once a month)
2 – Occasional falls (once a week to once a month)
3 – Falls multiple times a week or requires device to prevent falls
4 – Unable to stand or walk
Walking (assistive device = score of 3)0 – Normal
1 – Mild difficulty, perception of imbalance
2 – Moderate difficulty, but requires little or no assistance
3 – Severe disturbance of walking, requires assistance or walking aids
4 – Cannot walk at all even with assistance (wheelchair dependent)
Quality of sitting position0 – Normal
1 – Slight imbalance of the trunk, but needs no back support
2 – Unable to sit without back support
3 – Can sit only with extensive support (geriatric chair, posy, etc.)
4 – Unable to sit
Bladder function (if using drugs for bladder, automatic score of 3)0 – Normal
1 – Mild urinary hesitance, urgency, or retention (<once a month)
2 – Moderate hesitance, urgency, rare retention/incontinence (>once a month, but <once a week)
3 – Frequent urinary incontinence (>once a week)
4 – Loss of bladder function requiring intermittent catheterization/indwelling catheter

Statistical analysis

The characteristics of the participants were analysed with appropriate descriptive statistics. For continuous variables, means with standard deviations were used for the normally distributed continuous data. For discrete variables, frequency and cross-tabulation tables were used to describe the distributions.

For the purpose of analysis, the total ADL score was used and the nine domains of ADL were divided into the following five subcategories: speech and swallowing, cutting food and dressing, walking and frequency of falls, bladder function and hygiene, and ability to sit. Box plots of six outcome variables (total ADL score and five subcategories score) were used to visualize the differences and distribution before the analysis. Owing to sample size limitation, the Monte Carlo estimates of permutation test was used for the exact test to compare the differences of outcomes between isolated cerebellar atrophy (ICA) and CA+ groups, and participants with and without progression of cerebellar atrophy on imaging.

A linear regression model was built for the formal analysis to arrive at a set of risk factors associated with outcome variables. Variance inflation factor was used for testing the multicollinearity of the continuous variables. We initially used univariate linear regression analysis to determine the degree of association between the outcome variable and the four predefined potential predictors (vermian atrophy, cerebellar hemisphere atrophy, age at onset, and presence of seizures). Following the univariate analysis, we conducted a series of multiple regression analyses with group (ICA or CA+) as the main confounder of the analysis. Purpose selection strategy was used for the multiple regressions. The best model was chosen based on the significance of the covariates, R2, and clinical importance. sas version 9.3, PROC GLM, and PROC NPAR1WAY with the SCORE = DATA (SAS Institute, Cary, NC, USA) option were used during the analysis.

Results

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Clinical characteristics

Forty-four participants, 25 males and 19 females, participated in the study. The median age at symptom onset was 12 months (range birth–10y). Irrespective of underlying diagnosis, the group of participants with ICA (26 out of 44) presented at a later age than the CA+ group, with a mean age at onset of 2 years (SD 2y 4mo) and 1 year and 4 months (SD 1y), respectively. Sixteen participants had onset before 1 year of age, sixteen had onset between the ages of 1 year and 2 years of age, and twelve had onset after 2 years of age. The median age at the time of assessment in the neurogenetic clinic was 9 years (range 16mo–18y).

The presenting symptoms were highly variable within the study group, and in many cases were not limited to cerebellar signs (ataxia and/or nystagmus). Fourteen participants presented with developmental delay and cerebellar signs (ataxia and/or nystagmus). Eleven participants presented with isolated cerebellar signs. Eight participants presented with global developmental delay and hypotonia, with subsequent development of cerebellar signs. Six participants presented with global developmental delay, and five participants presented with gross motor delay.

Seizures were present in 10 participants. Total ADL scores were lower in participants with seizures than in those without seizures, by 7.7 points (95% confidence interval [CI] 0.5–14.9 points; p=0.037; pooled t-test t-value −2.16; degrees of freedom [df] = 42). Three participants had complex partial or generalized tonic–clonic seizures. The seizures were frequent (>3 per wk) and required treatment with two or more antiepileptic drugs. Six participants had well-controlled seizures with a frequency of one to two episodes per year; all required at least one antiepileptic drug. One patient had very brief episodes of seizures during the first year of life, with a working diagnosis of non-febrile seizure disorder, which resolved without treatment after 1 year.

All participants with unknown diagnosis had undergone extensive metabolic and genetic investigations including chromosomal microarray (Agilent Oligo Array – 44K; Agilent Technologies, Santa Clara, CA, USA). In the majority of participants (34 out of 44) the underlying diagnosis remained unknown. Genetic diagnosis was established in 10 participants. The genetically confirmed diagnoses in the eight participants with ICA were spastic paraplegia type 39 (PNPLA6 mutations), ataxia with oculomotor apraxia types 1 and 2 (APTX and SETX mutations), episodic ataxia 2 (CACNA1A mutation), X-linked Charcot–Marie–Tooth disease (GJB1 mutation), spinocerebellar ataxia type 7 in one participant each, and spinocerebellar ataxia type 5 in two participants. Diagnoses in the cerebellar atrophy plus (CA+) group included one participant with infantile neuronal axonal dystrophy (PLA2G6 mutations) and one with Marinesco-Sjögren syndrome (SIL1 mutations). The clinical characteristics of the participants are summarized in Table SI (online supporting information).

In the ICA group, MRI studies were performed at a median age of 8 years and 8 months (range 12mo–18y 11mo). Vermian atrophy was scored as severe in 17 participants and hemispheric atrophy scored as mild to moderate in 14 participants and severe in five participants. Both vermian and cerebellar hemispheric atrophy were present in 22 participants with ICA. In the CA+ group, MRI studies were performed at a median age of 6 years (range birth–17y 6mo). Vermian atrophy was severe in eight participants and hemispheric atrophy was mild to moderate in 13, with 17 participants having both vermian and cerebellar hemispheric atrophy. Additional MRI findings in this group included white matter signal abnormalities in 13 participants, basal ganglia or thalamus signal changes in three participants, and supratentorial grey matter atrophy in six participants. Cerebellar cortex signal abnormalities on T1-weighted images were noted in 25 participants.

Activities of daily living outcomes

ADL outcomes were better in the ICA group compared to the CA+ group. The total score for the group with ICA was 16.6 compared with 24.6 for the group with CA+, a difference of 8.0 points (95% CI 1.8–14.2 points; p=0.01, pooled t-test t-value=2.59; df=42). A comparison of the two groups showed that outcomes for speech and swallowing were worse in the CA+ group than in the ICA group, with a mean higher score of 4.9; (95% CI 3.8–6.1; p=0.001; pooled t-test t-value=−3.39; df=42), as were outcomes for bladder function and hygiene, with a mean higher score of 5.61; (95% CI 4.2–6.9; p=0.02, pooled t-test t-value=−2.50; df=42). Sitting was the function least affected in both groups and ability to walk and frequency of falls was the outcome most affected; the scores were not significantly different between the two groups (Fig. 3).

image

Figure 3. Box plot comparison of functional outcome scores (0 is normal; 36 is most severely impaired) between the isolated cerebellar atrophy (solid lines) and cerebellar atrophy plus (dashed lines) groups. p-values for statistically significant results are provided. The rectangles span the first quartile to the third quartile (the interquartile range). The horizontal line inside the rectangle shows the median and ‘whiskers’ above and below the box represent the minimum and maximum values of the activities of daily living (ADL) scores.

Download figure to PowerPoint

Follow-up MRI was carried out in 28 participants and progression of cerebellar atrophy was assessed by comparing vermian and hemispheric atrophy. On univariate analysis, the total score for the group who experienced significant radiological progression of cerebellar atrophy was 21.2, compared with 12.7 for the group with no progression, a difference of 8.5 points (95% CI 0.2–16.8 points; p=0.04), regardless of whether additional radiological abnormalities other than cerebellar atrophy were present. A large tegmento vermian angle was associated with worse outcome on univariate analysis. The total outcome score worsened by 0.7 for every 1-unit increase in the tegmento vermian angle (p=0.03). The degree of the tegmento vermian angle in the cohort ranged from 0 to 17.8°. The tegmento vermian angle was not found to be a statistically significant predictor of outcome on multiple regression analysis, probably because of the small sample size of this study.

Predictors of outcome

Multiple linear regression analysis (R2=0.66) revealed that the group with CA+ had a mean higher total outcome score (corresponding to worse overall functional outcome) of 5.88 (95% CI 0.04–11.7; p=0.04) compared with the group with ICA, after adjusting for maximum transverse cerebellar diameter, age at symptom onset, age at MRI, and progression of cerebellar atrophy.

Larger size of the transverse cerebellar diameter was correlated with better functional outcome, with a decrease in total outcome score of 0.72 for every millimetre increase in the size of the transverse cerebellar diameter (p=0.004), controlling for age at MRI (Fig. 4). In our cohort, the mean maximum transverse cerebellar diameter was 91mm (SD 6.4mm). Control data for transverse cerebellar diameter were established using MRI findings in 296 healthy brains. In controls, the mean transverse cerebellar diameter at the age of 0 to 6 months was 55.8mm (SD 7.9mm), at the age of 7 to 24 months was 90.9mm (SD 4.8mm), and for children aged between 2 years and 16 years was 101.6mm (SD 4.7mm).

image

Figure 4. Regression line showing larger size of the transverse cerebellar diameter was correlated with better functional outcome, with a decrease in total outcome score by 0.72 for every millimetre increase in the size of the transverse cerebellar diameter (p=0.004). The equation for the regression line is total score = −5.88*group – 4.74*progression + 10.04*agegroup – 0.72 trans, where group = 1(cerebellar atrophy plus), progression = 1(yes), and agegroup = 2(ageonset ≥ 2) are references, and the R2-value for the model is 0.66. ADL, activities of daily living.

Download figure to PowerPoint

Age at symptom onset less than 2 years was associated with a mean higher total outcome score of 10 and worse functional outcome compared with participants with age at symptom onset ≥2 years (95% CI 3.5–16.5, p=0.004).

Discussion

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Childhood-onset cerebellar atrophy is clinically and genetically heterogeneous. The accurate clinical and genetic diagnosis of ataxia associated with cerebellar atrophy remains difficult, and diagnostic rates among previously published studies of children with cerebellar atrophy range between 33% and 47%.[2, 6] These disorders are associated with a wide range of neurodevelopmental and functional outcomes; however, there is little information for caregivers and parents regarding prognosis. We prospectively studied ADL outcomes in 44 participants with confirmed cerebellar atrophy on MRI using standardized clinical and neuroradiological protocols. Independent predictors of poorer prognosis included age at onset below 2 years, additional MRI abnormalities compared with isolated cerebellar atrophy, progression of cerebellar atrophy on repeat MRI, presence of seizures, and reduced transverse cerebellar hemispheric diameter.

There are no studies which specifically address cerebellar atrophy and functional outcome. Tavano et al.[11] studied 27 children, 17 of whom had bilateral cerebellar hypoplasia associated with volume reduction. However, they excluded participants with progressive cerebellar pathology, metabolic disorders, and other associated supratentorial abnormalities. In general, the clinical outcome was variable among this group. Some presented with severe developmental impairment, others with partially preserved motor skills. Muscle hypotonia, tremor, and ataxia were found in the most severely affected patients. Neuropsychological assessment confirmed the presence of significant impairments in attention and visuoperceptual, visuospatial, and graphic tasks. Neuropsychiatric concerns were noted in half of these participants, in particular autism spectrum disorder.

Ventura et al.[12] reviewed 14 children with cerebellar hypoplasia; they excluded those with progressive disorders. Their results suggest that cerebellar hypoplasia predisposes individuals to psychomotor delay (10 participants) and cognitive impairment (12 participants), with at least five of the latter reported to have moderate to severe impairment. Epilepsy was noted in four participants, however electroencephalography was abnormal in 11 participants. Most participants had abnormal motor examination, ataxia, and nystagmus and some had ocular motor apraxia. None of these participants had abnormal metabolic investigations or abnormal chromosomal studies.

Bolduc et al.[13] studied 49 children with cerebellar malformations, which included infants born at term with Dandy–Walker malformation, cerebellar hypoplasia, vermis hypoplasia, rhomboencephalosynapsis, and Joubert syndrome. They performed MRI studies, neurological examination, and standardized neurodevelopmental assessments, and important predictors of outcome in this group included the presence of central nervous system or chromosomal anomalies, and a diagnosis of autism spectrum disorder. The presence of a malformation affecting the vermis had the most noticeable effect on expressive language and gross motor skills. This group has subsequently characterized a group of 32 participants with cerebellar malformations using volumetric MRI.[14] Consistent with their previous findings, participants in this group with cerebellar malformation with associated central nervous system or chromosomal anomalies had the worst outcomes. In patients with isolated cerebellar malformations, decreased total cerebellar volumes were associated with global developmental delay, cognitive delays, deficits in gross and fine motor skills, and with expressive language skills. Vermian volume loss was significantly associated with global developmental delay, defects of cognition, expressive language, and gross motor and fine motor skills, and early signs of autistic features. There were no significant associations between mid-hemispheric or medial hemispheric volumes and developmental and functional scores in their cohort.

We specifically characterized participants with vermian and cerebellar hemisphere atrophy; participants with cerebellar malformations or hypoplasia were excluded. Cerebellar hypoplasia was defined by the presence of a small vermis and a cerebellum with normally sized interfoliate sulci and sulcation. Repeat brain MRI is usually helpful to differentiate between these entities, by assessing space/volume loss between vermian and hemispheric folia.

In our cohort, the presence of other supratentorial MRI abnormalities was associated with worse outcomes, as was age at symptom onset below 2 years and smaller transverse cerebellar diameter, after controlling for other parameters. The presence of high cerebellar cortex signal intensity on FLAIR sequences was not significantly associated with functional outcome. Large tegmento vermian angle was also associated with worse functional outcomes; however, the degree of the tegmento vermian angle was found to be significantly correlated with functional outcome on univariate but not multivariate analysis. Interestingly, cerebellar vermian atrophy per se was not found to be a significant predictor of total functional outcome, even after adjusting for group (ICA or CA+). These observations may be limited by the relatively small sample size of our cohort, and two-dimensional methods might not be as capable of detecting maximum vermian height changes at early stages of the disease as volumetric methods.

There is no ADL assessment tool that has been designed or validated specifically for children with cerebellar atrophy, and this is a critical need for clinicians and researchers in this field. The core elements of the ADL assessment (speech and swallowing, cutting food and dressing, walking and frequency of falls, bladder function and hygiene, and ability to sit) which were assessed in this study have been utilized in a population-based study of disability in children,[15] which was the rationale for choosing this outcome measure.

Other limitations of this study include potential selection bias, as these participants were assessed in an outpatient setting; thus, very sick new-born infants with more severe presentations would have not been included in the study. An inherent source of bias is the high proportion of participants without a diagnosis (34 of 44 participants), and the likely heterogeneous aetiologies of the cerebellar atrophy, which will have differential impact on outcome and prognosis. As the underlying diagnoses are likely to be distinct between patients with ICA and CA+, we used these groups as a surrogate means of differentiating between disorders of varying aetiology based on neuroradiological evaluation, in the absence of a confirmed genetic diagnosis. There were additional limitations with respect to the MRI data. As the imaging was performed on MRI magnets from a variety of vendors, three-dimensional data were not uniformly obtained in the majority of cases and volumetric data are not available.

The major strengths of this study include its prospective design and standardized neuroradiological and clinical assessments carried out by a single paediatric neuroradiologist and a single neurogeneticist, which reduces bias caused by interrater discrepancies.

Conclusions

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

In this study we present the results of a study of clinical and radiological predictors of functional outcome in participants with childhood-onset cerebellar atrophy. We found that the most significant imaging predictors of poor outcome included the degree of atrophy of the cerebellar hemispheres, progression of cerebellar atrophy on MRI, and the presence of supratentorial signal abnormalities or volume loss, while the most significant clinical predictors were age at onset of clinical symptoms and presence of seizures. This information may be useful in the diagnosis and future management of this complex group of disorders.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank the participants and their families for participating in this study. This study was supported by a grant from the Rare Disease Foundation and the BC Children's Hospital Foundation. The study sponsor had no role in study design, data collection, data analysis, manuscript preparation, and/or publication decisions. The authors declare no potential conflicts of interests with respect to the authorship and/or publication of this article.

References

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Conclusions
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
dmcn12289-sup-0001-TableS1.xlsxapplication/msexcel21KTable SI: Clinical characteristics of the study participants. ICA, isolated cerebellar atrophy; CA+, cerebellar atrophy plus; GDD, global developmental delay.

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.