Hypothyroidism

Occasionally, patients with hypothyroidism develop a mild dysequilibrium and gait ataxia in conjunction with their systemic symptoms. This is correspondent with pathological changes in midline cerebellar structure (Barnard et al., 1971). Thyroid functions should be tested in patients with progressive ataxia, since replacement treatment may dramatically improve the symptoms (Jellinek and Kelly, 1960).

Toxic Causes

Infectious and Transmissible Diseases

Ataxia can be one of the features of postinfectious encephalomyelitis but usually accompanies a more diffuse cerebral process. A more restricted cerebellar syndrome has been seen in children after a viral syndrome. Connolly et al. (1994) have proposed that this diagnosis be considered when children develop an acute ataxic disorder not associated with a more diffuse process reflected by seizures, meningismus, or obtundation. In most children, this was preceded by a nonspecific viral syndrome or varicella, with a peak incidence at the age of 5 to 6 years. A similar syndrome following Epstein-Barr virus (EBV) or vaccinations occurred in the teenage years. Cerebrospinal fluid (CSF) may show some elevation of protein and a modest mononuclear pleocytosis, and PCR may allow identification of a specific etiology such as EBV. Magnetic resonance imaging (MRI) often reveals signal density changes in the cerebellum (Fig. 72.1). Prognosis for recovery is usually excellent, with residual dysfunction in a distinct minority. However, swelling, hydrocephalus, and herniation have been described in some and may require decompressive surgery (Van Lierde et al., 2004).

Fig. 72.1 Signal density change in the cerebellum in a child with acute cerebellar ataxia of childhood.

(Courtesy Dr. V. Vedanarayanan, Department of Pediatrics, University of Mississippi Medical Center.)

Similarly, a combination of ataxia, ophthalmoplegia, and other lower cranial nerve palsies can occur as a result of brainstem encephalitis (Bickerstaff encephalitis). There appears to be significant overlap between this disease and the Miller-Fisher variant of Guillain-Barré syndrome. In a report of 62 cases diagnosed as Bickerstaff encephalitis, most had some changes in consciousness level, 40% had Babinski signs, and 60% had tetraparesis; in addition, anti-GQ1 antibody was present in 66%. MRI brain abnormalities occurred only in 30% (Odaka et al., 2003).

Human immunodeficiency virus (HIV) infection can result in many neurological syndromes including ataxia. The majority of patients with ataxia in the presence of HIV infection have discrete, well-recognized lesions such as lymphomas, chronic meningeal infection, progressive multifocal leukoencephalopathy, or toxoplasmosis. Approximately 30% of patients with HIV dementia have an ataxic syndrome at the onset of their illness before cognitive decline begins. Other authors have described isolated progressive cerebellar ataxia that does not evolve into HIV dementia in a small number of patients (Elsheikh et al., 2010; Kwakwa and Ghobrial, 2001; Tagliati et al., 1998). These patients had a rapid evolution of their ataxia leading to chair-bound status in less than a year; MRI revealed cerebellar atrophy. Pathological examination has revealed marked granular cell loss.

Creutzfeldt-Jakob disease (CJD) should be considered in patients presenting with progressive ataxia. CJD is typically a rapidly progressive dementing illness related to the accumulation of mutant prion protein (PrP), which is a posttranslational modification of the normal prion protein. Among sporadic CJD patients, over 70% have ataxia (Parchi et al., 1999). Jellinger et al. (1974) noted that the ataxic variant of CJD began with minor behavior symptoms followed by ataxia. Upper motor neuron signs were frequent; myoclonus occurred only in 25%, and dementia evolved late. Survival among these patients was slightly longer than in typical CJD, with a mean time to death of 16 months and a range of 7 weeks to 8 years. Pathologically, the cerebellum shows striking granule cell loss. A recent review noted that both the iatrogenic and variant form of CJD (human bovine spongiform encephalopathy) usually have ataxia at the onset, as does the inherited form known as Gerstmann-Sträussler-Scheinker syndrome (GSS) (Wadsworth and Collinge, 2007). Prion protein gene mutation analysis is helpful in the familial cases. Iatrogenic cases are usually homozygous for codon 129, and the same codon is all methionine in the variant cases. Tonsillar biopsy for immunocytochemistry for PrP and Western blot studies for abnormal PrP^Sc (scrapie prion protein) can be useful in the variant cases as well. CSF protein biomarker studies involving 14-3-3, S100b, and tau are often useful in diagnosing CJD (Pennington et al., 2009).

Whipple disease (WD) is a multisystemic infectious disease caused by Tropheryma whippelii. The classic symptoms include diarrhea, weight loss, and abdominal pain. The pathognomic symptoms of CNS involvement are oculomasticatory myorhythmia (OMM) and oculo-facial-skeletal myorhythmia. Many cases of CNS WD are not diagnosed until postmortem (Gerard et al., 2002), but it is postulated that all patients have CNS involvement. Cerebellar ataxia is a common feature of CNS WD. It has been reported to be present in 45% of cases, more common than OMM (Matthews et al., 2005). Since untreated WD is fatal and therapy is available, it is paramount important to consider WD in the differential diagnosis when a patient presents with ataxia along with gastrointestinal symptoms. Those with possible CNS WD should undergo multiple duodenal biopsies if needed. Polymerase chain reaction (PCR) assay of CSF, brain MRI, and a stereotactic cerebral biopsy can also be helpful. Prolonged antibiotic therapy with drugs able to cross the blood-brain barrier is recommended. A regimen with intravenous ceftriaxone sodium, 2 g twice daily for 2 weeks, followed by oral trimethoprim-sulfamethoxazole, double strength, 1 tablet twice daily, achieved good results (Matthews et al., 2005). CSF should be analyzed repeatedly to monitor efficacy. Treatment can be discontinued only when the results of CSF PCR are negative.

Autoimmune Causes

Superficial Siderosis

Superficial siderosis occurs when iron and hemosiderin are deposited in the pial and subpial regions of the brain. Clinical features include ataxia, hearing loss, and cognitive decline (Fearnley et al., 1995). Superficial siderosis results from bleeds that are often covert, such as with cryptic vascular malformations, or more obvious, such as with tumors or prior surgery. Diagnosis can be made by the characteristic hypointensity over CNS structures in T2-weighted images on MRI (Fig. 72.2). The disease can be controlled if the source of bleeding is found and obliterated.

Fig. 72.2 In superficial siderosis, characteristic hypointensity signals are detected by T2-weighted images on magnetic resonance imaging in the pial and subpial region around the pons and cerebellar cortex.

Autosomal Recessive Ataxias

As the autosomal recessive ataxias have been elucidated at the molecular genetic level, a semblance of a pathogenetic classification may be emerging, albeit subject to revisions or additions. The genetic abnormalities discovered so far have either pointed to abnormalities in mitochondria, oxidative stress, abnormalities of deoxyribonucleic acid (DNA) repair mechanisms, possibly abnormal protein folding, or changes in cytoskeletal proteins. Table 72.2 lists various autosomal recessive ataxias based on specific gene loci. Most of these diseases begin in childhood or early adult life; however, onset even late in life is possible with autosomal recessive ataxias. Singleton patients may occur in many families. Typically, parents do not manifest symptoms, since they are heterozygous for the mutation. If the sibship size is large, affected siblings may be found, and the disorder will affect both males and females. Parental consanguinity may be found but is not essential.

Table 72.2 Autosomal Recessive Ataxias with Known Gene Loci

Friedreich Ataxia

Clinical Features

Friedreich ataxia is the most common inherited ataxia, with a prevalence of 1 in 50,000 Caucasians. Classical descriptions of the disease include an age at onset below 25 years, typically early in adolescence (Harding, 1981). Initial symptoms include gait difficulties and clumsiness. Neurological examination reveals gait ataxia, loss of proprioceptive sense in the lower limbs, and absence of deep tendon reflexes (Pandolfo, 2003). These findings are related to early pathology in dorsal root ganglion cells, and occasional patients may be mistaken for having a hereditary sensory neuropathy. However, most will exhibit signs indicating involvement of the CNS including dysarthria, some upper motor neuron findings such as extensor plantar responses, and eye movement abnormalities such as square-wave jerks. Rarely, patients may present with cardiac disease or a spinal deformity and are later found to develop neurological disease; they tend to lose ambulation in about 9 to 15 years. At this stage, patients have increasing ataxia of both upper and lower limbs, profound proprioceptive loss, areflexia, weakness of lower-limb muscles, flexor spasms, and increasing dysarthria and dysphagia. Optic atrophy and hearing loss may occur in a minority of patients.

Systemic abnormalities include abnormal electrocardiograms, hypertrophic cardiomyopathy in about 50% of the patients, and diabetes in 10%. The hypertrophic cardiomyopathy may later become a dilated cardiomyopathy and rhythm problems such as atrial fibrillation may add to the cardiac morbidity. Skeletal abnormalities such as spinal deformities and foot deformities are common (Fig. 72.3). The mean age of death among patients with FA has been reported to be late in the fourth decade, but this information comes from clinical studies done prior to the availability of FA mutation analysis. Also, the availability of better diabetic and cardiac care as well as adaptive rehabilitation techniques can allow much longer survival. Cause of death is often cardiac, but respiratory compromise related to the motor difficulties and spinal deformity may also contribute.

Fig. 72.3 High-arched foot (pes cavus) can be seen as one of skeletal abnormalities in Friedreich ataxia. The sole of the foot is distinctly hollow when bearing weight.

Nerve conduction studies show early absence or reduction of sensory nerve potentials in a diffuse fashion, reflecting the loss of large sensory axons in peripheral nerves. Central motor conduction studies show abnormalities that evolve over the course of the disease and may reflect disease progression. MRI scans of the brain reveal no abnormalities in the cerebellum; rather, the upper cervical cord shows atrophy. Signal density changes may be seen in the posterior columns by MRI (Mascalchi et al., 1994). Functional imaging has been reported to show decreased cerebellar blood flow, but there is cerebral glucose hypermetabolism in ambulatory patients with FA (Gilman et al., 1990). Pathologically, there is loss of dorsal root ganglion cells, resultant degeneration of the dorsal columns, degeneration of spinocerebellar and corticospinal tracts, and loss of cells in the cerebellar dentate nucleus.

The Friedreich Ataxia Mutation

The mutation in FA is an unstable expansion of a repeated trinucleotide (GAA) sequence within the first intron of the FRDA gene on chromosome 9q13-21.1 (Campuzano et al., 1996) (Fig. 72.4). Expanded alleles have 66 to over 1000 repeats. In nearly 95% of affected persons, the GAA expansion occurs in both alleles (homozygous expansion), although the size of the expansion can be different in each.

Fig. 72.4 Schematic representation of the GAA expansion in the first intron of the frataxin (FXN) gene on chromosome 9 in Friedreich ataxia.

(Courtesy Dr. H. Paulson, University of Iowa.)

Since the identification of the FA mutation, we have realized that many patients with a phenotype very atypical for classical FA carry the FA mutation (Durr et al., 1996; Filla et al., 1996; Montermini et al., 1997). Some 15% of patients with homozygous GAA expansion have neurological signs identical to those of classical FA but have an age at onset later than 25 years (late-onset FA [LOFA]). Other patients with the GAA expansion have the typical age at onset but retain their tendon reflexes or even have exaggerated reflexes (FA with retained reflexes [FARR]). There are patients who combine a later onset with retained reflexes as well; in addition, very late onset of mild gait ataxia, spastic paraparesis, and even chorea have been occasionally associated with the FA GAA expansion (Hou and Jankovic, 2003). The variation of expansion size may account for the clinical diversities. As with other trinucleotide repeat diseases, there is an inverse correlation between the size of the GAA repeat and the age at onset; this correlation is better with the smaller of the two expanded alleles. Cardiomyopathy and diabetes also tend to occur in patients with larger expansions (>700 GAA repeats). However, not all of the variation in age at onset and severity of disease can be correlated with GAA size alone. Other genetic and possibly environmental factors may contribute to this variation. In addition, variation in clinical picture can result from somatic mosaicism of the repeat size, with different tissues having significant differences in the size of the expansions (Bidichandani et al., 1999). Recently it has been reported that the U mitochondrial (mt)DNA haplogroup can delay the age at onset and have a lower rate of cardiomyopathy in FA (Giacchetti et al., 2004).

Pathogenesis

The presence of the expanded GAA sequence in the first intron of the gene results in reduced transcriptional efficiency leading to a partial deficiency of the protein, frataxin (Lodi et al., 2001; Pandolfo, 2001; Puccio and Koenig, 2000). The reduced transcriptional efficiency of the mutated gene has been attributed to an unusual “sticky DNA” configuration of the expanded repeat (Sakamoto et al., 1999). In addition, there is evidence for chromatin condensation and remodeling, which can further suppress frataxin transcription (Pandolfo and Pastore, 2009). The exact role of frataxin in normal biology is still unclear, but many studies suggest that it is a mitochondrial protein (Fig. 72.5). Knockout of the frataxin homolog in yeast leads to accumulation of iron in the mitochondria, a finding of interest in view of the presence of iron in the cardiac muscle in human disease (Babcock et al., 1997; Pandolfo and Pastore, 2009). Activity of enzymes such as aconitase and complex I, II, and III of the respiratory chain that contain iron-sulfur clusters is reduced in cardiac muscle biopsies of patients with FA (Rotig et al., 1997). The iron-sulfur cluster synthesis defect results from the role of frataxin in the process and probably is the cause of iron accumulation. It has been hypothesized that the presence of excess iron as well as excess oxidative stress related to impaired respiratory chain function can induce oxidative stress via the Fenton reaction and cause progressive nuclear and mitochondrial damage (Karthikeyan et al., 2002, 2003; Pandolfo and Pastore, 2009).

Fig. 72.5 Frataxin is a mitochondrial protein that plays an important role in cells and tissues with high metabolic rates (e.g., Purkinje cells, heart). Mutations in FXN disrupt transcription and translation of frataxin, greatly reducing its availability in cells. Heme and iron-sulfur cluster (ISC) biosynthesis is reduced, which decreases activity of proteins that contain ISC and impairs energy production in mitochondria. Iron overload renders cells sensitive to oxidative stress and hastens subsequent cell death. Cells in brain, spinal cord, and muscles are particularly vulnerable. FA, Friedreich ataxia.

Mitochondrial Recessive Ataxia Syndrome

An autosomal recessive ataxic syndrome caused by a mutation in the polymerase γ (POLG) gene has been found to be the most common cause of recessive ataxia in Finland (Hakonen et al., 2005). Polymerase γ is encoded by the nuclear gene, POLG, in humans and is responsible for replication and repair of mtDNA. In a recent study report, 27 patients were found among 15 Finnish families with a carrier frequency of 1 : 125. In addition, patients with recessive ataxia from Norway, Belgium, and England shared the same core Finnish haplotype, suggesting a common European founder mutation and making it likely that this disease will be found in places other than Finland. The age at onset varied from 5 to 41 years. The neurological picture was heterogeneous, but ataxia, hypo- or areflexia, axonal neuropathy, and epilepsy were common. There was often mild cognitive impairment and psychiatric manifestations such as depression, anxiety, and hallucinations.

Infantile-Onset Spinocerebellar Ataxia

Infantile-onset spinocerebellar ataxia has been described among an isolated population in Finland. The disorder begins soon after children acquire the ability to walk. Neurological signs include ataxia, peripheral neuropathy with areflexia, athetosis of hands and feet, Babinski signs, and clumsiness. Speech becomes impaired, and sensorineural hearing loss occurs before the age of 5. Ophthalmoplegia and optic atrophy develop, and some patients have seizures. Patients usually have learning deficits and skeletal deformities as well. The gene responsible has recently been found to be C10ORF2, a nuclear gene encoding for Twinkle, an mtDNA-specific helicase, and a splice variant known as Twinky (Nikali et al., 2005) located on 10q24. Dominant mutations in the same gene cause a syndrome of progressive external ophthalmoplegia. In contrast to the latter, the recessive mutations in infantile-onset spinocerebellar ataxia do not appear to alter mtDNA maintenance.

Ataxia with Isolated Vitamin E Deficiency

The occurrence of a childhood-onset recessive ataxia associated with isolated vitamin E deficiency is related to mutations in the gene encoding the α-tocopherol transfer protein (α-TTP) on chromosome 8 (Ouahchi et al., 1995). A hepatic protein, α-TTP is involved in the processing of vitamin E for transport in the chylomicrons. The low vitamin E levels are related to impaired hepatic processing, not to deficient intestinal absorption. Deficiency of vitamin E, which has antioxidant properties, may cause neuronal degeneration. Purkinje cells are more vulnerable owing to their high metabolic activity and high oxygen demand. The severity of the phenotype associated with α-TTP mutations depends on residual protein activity. The childhood-onset disease has considerable resemblance to FA. There is progressive ataxia and evidence for a polyneuropathy with loss of proprioception and tendon reflexes. Retinitis pigmentosa and visual loss can accompany this syndrome. Patients often have head titubation. Cardiomyopathy can occur but is less common than in FA (Cavalier et al., 1998). Vitamin E levels should be obtained in all persons with sporadic ataxia of childhood or young-adult onset. Patients typically have less than 1.8 mg/L of vitamin E. Treatment with large doses of vitamin E stabilizes the neurological features, especially when started early in the disease, but some features may show worsening despite therapy (Mariotti et al., 2004).

Abetalipoproteinemia

Abetalipoproteinemia results from mutations in the MTTP gene, which codes for a subunit of the microsomal triglyceride transfer protein; this results in impaired lipid absorption in the intestine, including lipid-soluble vitamins like vitamin E. The ataxia resembles that in isolated vitamin E deficiency but is associated with evidence for malabsorption (Fogel and Perlman, 2007). Vitamin E and cholesterol levels are low, and apolipoprotein B is absent. Acanthocytes are found in peripheral blood smears. Transaminases may be high and INR elevated due to vitamin K deficiency. Serum lipoprotein electrophoresis can establish the diagnosis. Treatment involves dietary modification and vitamin replacement, which may prevent neurological complications.

Cayman Ataxia

This recessive ataxia of early childhood onset has been identified in the Grand Cayman island. Affected patients have been shown to carry two homozygous mutations in the gene that encodes caytaxin (ATCAY) on chromosome 19. Caytaxin contains a region with homology to a CRAL-TRIO domain, also present in α-TTP. A ligand similar to vitamin E may bind to this protein, and its deficiency may cause ataxia by a mechanism similar to that in vitamin E deficiency (Bomar et al., 2003).

Ataxia-Telangiectasia

Clinical Features

The occurrence of ataxia-telangiectasia (AT) has been estimated to be 1 in 20,000 to 100,000 live births (Swift et al., 1993). This disorder has its onset early in the first decade with truncal instability and impaired gait. This is followed by progressive ataxia associated with polyneuropathy, hypotonia, and loss of reflexes (Fogel and Perlman, 2007). Choreiform movements are frequent. Characteristic eye movement abnormalities are seen, with the need for a rapid head thrust for fixating on targets to one side, with the eyes following later (eye-head lag). This is referred to as oculomotor apraxia. Telangiectasia tends to appear at about 5 years of age and is most frequent on the conjunctivae (Fig. 72.6) but can also occur over the earlobes, popliteal fossa, and other locations. There is evidence for an immunodeficiency, with frequent bronchopulmonary infections. These children are at risk for malignancies, especially hematological tumors like lymphomas and leukemias; other types of cancers occur if the patient reaches adulthood. Infections related to immunodeficiency are also common. The thymus gland is typically small or absent in patients with AT. Many have decreased immunoglobulin (Ig)A levels; decreased IgE and IgM levels, lymphocytopenia, and skin anergy may also occur. In addition, elevation of α-fetoprotein (AFP) in the serum is a consistent finding in AT.

Fig. 72.6 Conjunctival telangiectasia in a patient with ataxia-telangiectasia.

Ataxia–Telangiectasia-like Disorder

Ataxia–telangiectasia-like disorder (ATLD) is related to mutations in the gene coding for MRE11, another protein involved in double-stranded DNA break repair. It has a clinical picture resembling AT but has a milder course, no telangiectasia, and no elevation of serum AFP (Fogel and Perlman, 2007).

Ataxia with Oculomotor Apraxia Type 1

Ataxia with oculomotor apraxia type 1 (AOA1) is an AR ataxia of childhood onset (usually <10 years), and many patients have oculomotor apraxia akin to that in AT (Le Ber et al., 2003). It is the commonest recessive ataxia in Japan and second only to FA in Portugal. In the report from Le Ber et al., AOA1 made up 5% of non-Friedreich recessive ataxias. It has other similarities to AT including peripheral neuropathy (which may predominate the clinical picture in later stages), choreoathetosis, and dystonia. Cognitive decline is frequent. However, radiosensitivity and malignancy risk are not elevated. Serum creatine kinase (CK) and cholesterol are often high and serum albumin decreased. Mutations in the aprataxin (APTX) gene have been described in AOA1 (Date et al., 2001; Moreira et al., 2001). APTX is a nuclear protein that has an N-terminal forkhead-associated (FHA) domain resembling PNKP, a histidine triad (HIT) domain, and a C-terminal zinc finger domain. Most of the mutations affect the HIT domain. The protein may function in base excision repair, a form of single-strand DNA repair mechanism. The C terminal is also known to interact with XRCC1, which plays a role in single-strand break repair. It is unclear how this functional role is related to neurodegeneration in the cerebellum.

Ataxia with Oculomotor Apraxia Type 2

Ataxia with oculomotor apraxia type 2 (AOA2) is a progressive AR ataxia associated with cerebellar atrophy and an axonal polyneuropathy and elevation of serum AFP (Anheim et al., 2009). AOA2 is the most frequent AR ataxia next to FA in European series (Anheim et al., 2010; Le Ber et al., 2004). The age at onset is around 15 years, and the disease progresses to a stage requiring significant ambulation devices in about 15 years. Oculomotor apraxia is not universal, being seen in about 50% cases (Anheim et al., 2009; Le Ber et al., 2004). Other clinical signs noted are upper motor neuron signs in about 20%, dystonia, chorea, head tremor, and strabismus. Elevation of serum AFP to over 7 µg/L may be predictive of AOA2. Other features that have been noted infrequently include elevation of serum CK and hypogonadotropic hypogonadism. No other systemic features such as high risk for malignancy, telangiectasia, or radiation sensitivity are seen.

AOA2 has been linked to mutations in the senataxin gene (SETX) that codes for senataxin (Moreira et al., 2004). The normal function of senataxin is unclear; its C-terminal region has homology to the superfamily of helicases. DNA/RNA helicase function may contribute to the pathogenesis of neurological dysfunction (Chen et al., 2004).

Other DNA Repair Defects Causing Ataxia

Mutations in tyrosyl-DNA phosphodiesterase 1 (TDP1), a DNA repair enzyme, causes spinocerebellar ataxia with axonal neuropathy. These patients have none of the systemic features that characterize AT or ATLD (Takashima et al., 2002). Cockayne syndrome and xeroderma pigmentosum are rare diseases caused by DNA repair defects in which systemic disease predominates the phenotype, but CNS features including ataxia are often present. Mutations in genes encoding nucleotide excision repair pathways occur in these diseases (Chu and Mayne, 1996).

Autosomal Recessive Ataxia of Charlevoix-Saguenay

Autosomal recessive spastic ataxia of Charlevoix-Saguenay (ARSACS) was initially identified in children with spastic ataxia among French Canadians in the Charlevoix-Saguenay province of Quebec, Canada, an inbred population with a high carrier frequency for the gene. ARSACS has since then been documented in other geographic areas such as Turkey, Japan, Spain, and Italy. Clinically it is characterized by very early onset, prominent spasticity followed by ataxia, and amyotrophy of distal muscles (Takiyama, 2007). Progression is slow, with a mean age at death in the late 50s. The mutation causing this disease has been shown to involve a gene on chromosome 13. The protein product of this gene has similarities to chaperone proteins and has been named sacsin (Engert et al., 2000). The C-terminal region of this protein has a DnaJ motif that can interact with Hsp 70, and the N terminus has homology to HSp90. Thus SACSIN may have a chaperone function involved in protein folding.

Marinesco-Sjögren Syndrome

This rare disease characterized by ataxia, early cataracts, mental subnormality, and myopathy has been linked to mutations in the SIL1 gene. SIL1 encodes a nucleotide exchange factor for the chaperone protein, Hsp70 (Anttonen et al., 2005; Senderek et al., 2005).

Autosomal Recessive Cerebellar Ataxia Type 1

The term autosomal recessive cerebellar ataxia type 1 (ARCA1) has been used to describe an autosomal recessive ataxia in a population isolate of French Canadians from Quebec. The disorder begins in young adult life (17-46 years) and causes progressive ataxia, dysarthria, and mild eye movement problems. Reflexes were normal or slightly brisk and there was cerebellar atrophy on imaging studies. Molecular studies have revealed that this is related to mutations in the SYNE1 gene. It is thought that SYNE1 is a member of the spectrin family of cytoskeletal proteins that may be involved in anchoring actin to the plasma membrane and appears enriched in Purkinje cells (Dupre et al., 2007).

Autosomal Recessive Cerebellar Ataxia Type 2

The term autosomal recessive cerebellar ataxia type 2 (ARCA2) has been used to denote cerebellar ataxia associated with coenzyme Q10 (CoQ10) deficiency (Anheim et al., 2010). CoQ10 is a lipid-soluble component of cell membranes and is involved in transport of electrons between complex I and II of the electron transport chain to complex III. Ataxia associated with deficiency of muscle CoQ10 is heterogeneous. This combination can be seen in some patients with AOA1 as well as in some with defined abnormalities in CoQ10 metabolism. Ataxia is associated with pyramidal signs and mild cognitive decline. Some patients may have hypogonadism. CoQ10 deficiency can be associated with other phenotypes such as an infantile multisystem disease, encephalomyopathy with ragged red fibers, and Leigh syndrome. Some gene mutations have been identified in the infantile cases. The syndrome may respond to CoQ10 replacement therapy.

Childhood and Young Adult–Onset Metabolic Disorders in Which Ataxia Can Be Prominent

Many childhood-onset and young adult–onset ataxias are related to metabolic errors that can be diagnosed by specific laboratory tests rather than by gene-based tests. In many of these diseases, ataxia forms only a part of the phenotype. Table 72.3 lists some of these diseases. Some of them are amenable to various forms of therapy including dietary manipulations.

Table 72.3 Ataxias in Which Specific Abnormalities May Confirm or Point to the Diagnosis*

AOA, Ataxia with oculomotor apraxia; AT, ataxia-telangiectasia; CSF, cerebrospinal fluid; MRI, magnetic resonance imaging; MRS, magnetic resonance spectroscopy.

* Some also appear in Table 72.2, and many have additional CNS features.

Clinically Defined but (Likely) Genetically Heterogeneous Recessive Syndromes

Many children and young adults with progressive ataxia still have undefined genotypes (Table 72.4). The term early-onset ataxia with retained reflexes was originally used to describe a syndrome of ataxia with onset in childhood and some resemblance to FA but with preservation of deep tendon reflexes. We now know that many of these patients carry the FA mutation; in others, however, the genotype remains unknown. Similarly, Gordon Holmes originally described the syndrome of recessive ataxia associated with hypogonadism in 1907. This may also be genetically heterogeneous, though in some patients with such a disorder, mitochondrial mutations have been described. The term Ramsay Hunt syndrome is used for the association of ataxia and myoclonus or myoclonic epilepsy. One form of this, Unverricht-Lundborg syndrome, is related to mutations in the cystatin B gene. Others have mtDNA mutations or other disorders of intermediary metabolism such as ceroid lipofuscinosis or sialidosis.

Table 72.4 Childhood or Young Adult–Onset Ataxias with Distinctive Clinical Features*

ARSACS, autosomal recessive ataxia of Charlevoix-Saguenay; FA, Friedrich ataxia; mtDNA, mitochondrial deoxyribonucleic acid.

* But as yet poorly understood at the genetic level.

Clinical Features of Spinocerebellar Ataxias

Overall, SCAs related to different gene mutations have overlapping clinical features (Subramony, 2005). There is more experience with some of the SCAs and EAs (e.g., SCA1, 2, 3, 6, 7) than others, and the knowledge about phenotypical features in some of the SCAs is based on a limited number of families. In general, onset is in early or middle adult life, but the range of age at onset is vast, and onset in early childhood or at older ages is possible. In the better-known SCAs which are related to unstable nucleotide repeat expansions, anticipation in age at onset is seen and is usually more prominent with paternal transmission of disease; in these, age at onset is inversely correlated with the expansion size. Some of the more recently described SCAs related to more conventional mutations have childhood onset and very slow progression. Patients with SCA exhibit progressive ataxia of gait associated with clinical signs of limb ataxia such as dysmetria and dysdiadochokinesia. Speech is dysarthric. Eye movement abnormalities related to cerebellar dysfunction are frequent and include abnormal pursuit and inaccurate saccades in addition to nystagmus.

In many SCAs, ataxia is the only clinical feature. In others, clinical signs referable to pathology in structures other than cerebellum can be seen. Table 72.5 summarizes some clinical features that may help in the clinical differentiation of various SCAs. However, even in those SCAs in which non-cerebellar signs have been described, patients with a later onset may exhibit a pure cerebellar phenotype. Oculomotor abnormalities unrelated to cerebellar dysfunction include nuclear and supranuclear gaze palsy, slow saccades, ptosis, blepharospasm, and “ocular stare.” Patients may exhibit bulbar deficits such as facial atrophy, facial fasciculations, tongue atrophy and fasciculations, and poor ability to cough. Upper motor neuron signs such as brisk reflexes, with or without spasticity, and Babinski signs occur early in many SCAs. Extrapyramidal signs including akinetic-rigid syndromes, hypomimetic facies, chorea, and dystonia tend to occur variably in SCAs in different stages of the illness (Fig. 72.7). Evidence of peripheral nerve disease occurs frequently and includes distal sensory loss and loss of deep tendon reflexes as well as amyotrophy, cramps, and fasciculations. In selected SCAs, there can be associated cerebral signs such as cognitive decline or seizures; in others, there is evidence of retinal disease with visual loss (Fig. 72.8). Restless legs and sleep disturbances can occur in many. Other clinical signs that have been noted in certain genotypes include focal dystonia, postural and rest tremor, head tremor, palatal tremor, and psychiatric features such as aggressive outbursts, schizophrenic symptoms, severe depression, and apathy.

Table 72.5 Phenotypic Clues to Specific Gene Mutations in the Dominant Ataxias

Spinocerebellar Ataxia Type 1

Age at onset of SCA1 is variable from childhood to over 70 years of age, but the mean is in the 30s (Sasaki et al., 1996). The early clinical picture is one of a cerebellar syndrome associated with upper motor neuron signs. Later, ophthalmoparesis, slow saccades, a sensory predominant polyneuropathy, amyotrophy, and some extrapyramidal features such as chorea and dystonia can develop. Dysarthria is early and significant dysphagia a later symptom. There can be ineffective cough and facial twitches. Cognitive changes such as executive dysfunction occur, and the usual course before death is about 15 to 20 years. The gene mutation is an unstable CAG expansion in the ataxin-1 gene on chromosome 6, and there is an overlap between the upper end of the normal and the lower end of the pathogenic range. However, the large normal alleles can be differentiated from the small pathogenic alleles by the presence of CAT interruptions in the former, detected by the presence of a restriction site for the enzyme, SfaNI (Chung et al., 1993).

Spinocerebellar Ataxia Type 2

SCA2 resembles SCA1 but was originally distinguished based on genetic studies of a large inbred population in Eastern Cuba (Orozco Diaz et al., 1990; Sasaki et al., 1996; Wadia et al., 1998). The presence of slow saccades early in the disease as well as peripheral neuropathy with generalized areflexia may point to SCA2. Other features that have been described in SCA2 include dystonia, l-dopa-responsive parkinsonism, and cognitive decline (Burk et al., 1999; Cancel et al., 1997). An infantile form resulting from a hyperexpanded allele has been described. The mutation is a CAG expansion in the ataxin-2 gene on chromosome 12, with normal alleles varying from 15 to 31 repeats. While normal alleles usually have CAA interruptions, pathogenic alleles are usually contiguous, but exceptions have been reported (Costanzi-Porrini et al. 2000).

Machado-Joseph Disease

MJD, or SCA3, is probably the most prevalent of the SCAs worldwide (Riess et al., 2008). Once again, the disorder has many similarities to SCA1 and 2, with an age at onset and course not too different from that of SCA1. Like SCA2, it appears to have a wider phenotypical spectrum. Apart from cerebellar signs, patients develop many brainstem signs such as facial and tongue fasciculations or myokymia, poor cough, facial atrophy, and dysphagia. Non-cerebellar eye signs such as slow saccades, disconjugate movements, ophthalmoparesis, ptosis, eyelid retraction, and blepharospasm have all been seen in MJD. In many patients, extrapyramidal features such as parkinsonism responsive to l-dopa or dystonia can dominate the clinical picture (Subramony, 1993; Subramony et al., 2002). Later-onset cases may be characterized by ataxia, amyotrophy, and peripheral neuropathy and areflexia. Other features include a cerebellar cognitive decline, sleep disturbances including restless legs, and autonomic deficits (Friedman et al., 2003; Kawai et al., 2004; Yeh et al., 2005). Impaired temperature discrimination in a generalized fashion can be a sign of MJD (Schols et al., 1996). The mutation is an unstable CAG expansion in the ataxin-3 gene on chromosome 14, and the pathogenic range from 53 to 86 is usually easily distinguished from the upper normal limit of 47 repeats.

Spinocerebellar Ataxia Type 4

SCA4 has been localized to chromosome 16q and causes a combination of sensory motor neuropathy and ataxia (Flanigan et al., 1996).

Spinocerebellar Ataxia Type 5

This dominant ataxia described in the descendents of the paternal grandparents of American President Abraham Lincoln (so-called Lincoln ataxia) has also been reported in French and German families. Age at onset is wide (10-68 years) with apparent anticipation. The disease is slowly progressive with mostly cerebellar deficits with added minor pyramidal signs. Mutations in the β-III spectrin gene (SPTBN2) have been recently described (Ikeda et al., 2006). In two of the families, small exonic deletions were found; the third family had a missense mutation (758 T to C). It is believed that β-III spectrin may have a role in stabilizing the glutamate transporter, EAAT4, at the plasma membrane.

Spinocerebellar Ataxia Type 6

The age at onset of SCA6 is higher (late 40s-50s). SCA6 usually has relatively pure cerebellar signs and occasional minor pyramidal signs. Positional vertigo can occur and downbeat nystagmus is often seen. SCA6 is often compatible with a normal lifespan. It is caused by a CAG expansion in the gene coding for the neuronal calcium channel (P/Q type, α subunit). SCA6 shares phenotypical features with hemiplegic migraine and EA2, which are allelic to it. Because of the late onset, SCA6 often may present with no positive family history.

Spinocerebellar Ataxia Type 7

This is the only SCA in which prominent visual loss related to a maculopathy occurs. Otherwise the clinical picture resembles that of SCA1 and 2 and MJD, with a combination of cerebellar ataxia and spasticity, often of an extreme nature. Slow saccades are frequent; extrapyramidal signs and peripheral neuropathy are uncommon. In different family members, visual loss and CNS features can occur in a variable combination. Subclinical visual impairment can be detected by a tritan axis defect found on the Farnsworth 15 color vision test or by electroretinogram. SCA7 is related to a CAG expansion in the ataxin-7 gene on chromosome 3. This expansion tends to be more unstable than other CAG expansions, and childhood-onset cases, usually inherited by paternal transmission, have been documented. Onset younger than age 12 was seen with repeat expansion sizes of over 67 (Johansson et al., 1998), and early-onset cases often have visual loss preceding the ataxia, a situation different from adult-onset cases. Such cases may present before the affected parent becomes symptomatic and have a more florid clinical picture including cognitive changes and seizures with a more rapid course. Similarly, a devastating infantile form of the disease due to a very large expansion can occur, and the expansion may escape detection by routine PCR testing.

Spinocerebellar Ataxia Type 8

SCA8 has been related to an unstable expansion of a CTG tract (110-250; normal, <73) in the 3′ untranslated region of the gene on chromosome 13 (Koob et al., 1999). Patients have a slowly progressive ataxia and variable spasticity and brisk tendon reflexes. Tremor and fluctuation of symptoms were noted in Finnish patients with the SCA8 expansion (Juvonen et al., 2000). The primary role of this expansion in causing ataxia has been controversial because the same CTG expansion has also been found in patients with other disorders such as stroke and Parkinson disease and even in general population samples (Stevanin et al., 2000; Worth et al., 2000). The vertical transmission of the expansion has been extensively studied by Koob et al. (1999) in only one family in which a significant logarithm of odds (LOD) score was calculated. Even in this family, there were several nonpenetrant individuals who had expansions of the repeat into larger than normal but presumably nonpathogenic range (250-800 repeats). Thus the interpretation of an expanded CTG repeat in an individual with ataxia remains controversial, and this test is probably not useful for predictive testing of at-risk individuals. However, the SCA8 expansion may be one of the more common mutations identified among “sporadic” ataxia patients.

Spinocerebellar Ataxia Type 10

Still another form of repeat expansion occurs in SCA10, a disorder typified by progressive ataxia and epilepsy. The mutation is a large unstable expansion (800-4500 repeats; normal <29) of a pentanucleotide repeat (ATTCT) on chromosome 22 (Matsuura et al., 2000). Age at onset is 14 to 45 years. Ataxia is a uniform feature and is slowly progressive. Seizures, both generalized and complex partial, occur in 25% to 85% of affected families. Variable pyramidal, peripheral nerve, and neuropsychiatric features occur (Lin and Ashizawa, 2005).

Spinocerebellar Ataxia Type 11

SCA11 has been described in a limited number of families and causes ataxia, upper motor neuron signs, and abnormal eye movements. Mutations have been described in the tubulin tau kinase gene (TTBK2) (Houlden et al., 2007).

Spinocerebellar Ataxia Type 12

SCA12 was originally described in an American family of German descent, but most of the other families have been reported from the Indian subcontinent (Holmes et al., 2003). The usual features include an action tremor of hands and head, bradykinesia, mild ataxia, and psychiatric features, with onset usually in the 40s. MRI of the brain shows more generalized atrophy. The CAG expansion that is related to SCA12 occurs in the promoter region of the PPP2R2B gene on chromosome 5 and is not apparently coded into a polyglutamine tract.

Spinocerebellar Ataxia Type 13

Two families, one of French and another of Filipino descent, have been found to have this mutation so far. The French family had childhood onset of very slowly progressive ataxia and mild developmental delay (Herman-Bert et al., 2000). The Filipino family had a later onset age and no mental retardation (Waters, Fee et al. 2005). Both had missense mutations in exon 2 of the voltage-gated potassium channel (KCNC3), 1154G→A and 1639C→A (Waters et al., 2006).

Spinocerebellar Ataxia Type 14

SCA14 is characterized by a complex and variable clinical picture including ataxia, axial myoclonus, chorea, head tremor, and rippling muscles in some (Klebe et al., 2005; Yabe et al., 2003). Facial myokymia, diplopia, dystonia rigidity, and proprioceptive loss have also been noted. SCA14 has been linked to several missense mutations in both the regulatory and catalytic domains of the protein kinase (PK)Cγ gene on chromosome 19.

Spinocerebellar Ataxia Types 15 and 16

SCA15 and 16 were originally reported from Australia and Japan, respectively, but more recent findings suggest that the same gene mutation underlies these entities (Iwaki et al., 2008; van de Leemput et al., 2007). The disease causes pure cerebellar ataxia, sometimes with associated postural tremor. Deletions involving two contiguous genes, SUMF1 and ITPR1, were found in the Australian kindred, but in the Japanese patients, only the ITPR1 gene was deleted, suggesting that this is the critical gene.

Spinocerebellar Ataxia Type 17

SCA17 has an extremely variable phenotype that includes ataxia, upper motor neuron signs, early and severe cognitive decline with significant behavioral changes, dyskinesias, dystonia, and parkinsonian signs. This is considered to be a polyglutamine disease and is linked to an expanded CAG repeat in the TATA-binding protein (TBP) gene (pathogenic range 45-66 repeats; normal, <42). Inheritance is autosomal dominant, but occasional sporadic cases have been reported (Hagenah et al., 2004; Rolfs et al., 2003).

Spinocerebellar Ataxia Type 18

Brkanac et al. (2002) described an autosomal dominant condition characterized by both ataxia and a sensory motor neuropathy in a five-generation American family of Irish ancestry and linked this to chromosome 7q22-q32.

Spinocerebellar Ataxia Type 19

SCA19 has been reported in a family of Dutch origin with a mild ataxia, myoclonus, irregular postural tremor, and cognitive impairment. Among 11 family members with clinical data, many also had decreased proprioception and tendon reflexes. The disorder in this family was mapped to chromosome 1p21-q21 by Verbeek et al. (2002).

Spinocerebellar Ataxia Type 20

SCA20 was described in an Australian family with a wide range of age at onset, prominent dysarthria and dysphonia, and the presence of calcification in the cerebellar dentate (Knight et al., 2004). The disease is localized to chromosome 11 and appears related to the duplication of a 260-kb segment on chromosome 11q. Its pathogenesis remains unclear.

Spinocerebellar Ataxia Type 21

In a family with autosomal dominant ataxia with onset in childhood and young-adult life, a new gene locus was established at chromosome 7p (Vuillaume et al., 2002). In addition to ataxia, the patients had tremor, rigidity, and cognitive impairment.

Spinocerebellar Ataxia Type 22

Chung et al. (2003) reported a Chinese family with dominant ataxia, hyporeflexia, onset between 10 and 46 years of age, and MRI showing isolated cerebellar atrophy; locus was established at chromosome 1p, overlapping that reported in SCA19.

Spinocerebellar Ataxia Type 23

In a single family with adult-onset dominant ataxia, the locus was mapped to chromosome 20p (Verbeek 2009).

Spinocerebellar Ataxia Type 25

In a single family from France, Stevanin et al. (2004) noted childhood onset of a dominant ataxia with variable progression and an associated peripheral neuropathy. Linkage studies localized the disorder to chromosome 2p.

Spinocerebellar Ataxia Type 26

Yu et al. (2005) have described a single family of Norwegian descent with relatively pure cerebellar ataxia with little in the way of extracerebellar signs. Age at onset was from 33 to 60 years. MRI showed cerebellar atrophy, but brainstem was spared. Genetic mapping has localized this disease to chromosome 19p13.3 adjacent to the SCA6 locus.

Spinocerebellar Ataxia Type 27

In a Dutch family with ataxia, early hand tremor, and psychiatric symptoms such as aggressive outbursts, a missense mutation in the fibroblast growth factor 14 gene has been identified (van Swieten et al., 2003).

Dentatorubral Pallidoluysian Atrophy

DRPLA is most prevalent in Japan, but molecularly proven DRPLA has been reported from Europe and the United States. The original African American family in which the typical neuropathology was described many years ago is now known to carry the same mutation as the Japanese cases (Burke et al., 1994). DRPLA has a very variable age at onset from infancy to old age, and anticipation in age at onset is common. With onset younger than 20 years of age, the clinical picture is one of seizures, myoclonus, ataxia, and intellectual decline. With older onset, the disease is characterized by ataxia, chorea, dementia, and psychiatric features. Thus the differential diagnosis varies with age and includes myoclonic epilepsy syndromes in children and ataxia and Huntington disease in older persons. Diagnosis can be established by documenting an expanded CAG repeat in the atrophin-1 gene.

Sporadic Cortical Cerebellar Atrophy

Sporadic cortical cerebellar atrophy usually has onset after age 50 and results in a progressive ataxia not associated with other neurological deficits even many years after onset. MRI scans and pathological examination usually show isolated cerebellar atrophy, but this correlation is not absolute. The disease has a slow progression with a median lifespan of over 20 years after onset.

Sporadic Ataxia with Added Non-Cerebellar Deficits

Patients with sporadic ataxia with added non-cerebellar deficits initially have slowly progressive ataxia but soon develop upper motor neuron signs, ophthalmoplegia, parkinsonian features, and autonomic failure. Both MRI and pathological examination show cerebellar and brainstem degeneration. Autonomic structures may be affected as well, so the clinical picture of this disorder merges with that of the olivopontocerebellar atrophy form of multiple system atrophy (MSA). Gilman et al. (2000) have shown that among patients with idiopathic progressive ataxia, 25% to 35% will transition to probable MSA over 5 to 10 years, and a similar rate was observed by Abele et al. as well (2002). Clinical evidence for this in most patients will be added parkinsonian and autonomic deficits, but in a few, only autonomic failure develops. Signs of autonomic failure include orthostatic hypotension, incontinence, and erectile dysfunction. Parkinsonian signs include rigidity and facial hypomimia. Older age at onset and a more rapid progression to a disabled state confer a higher risk of such a transition; median survival after such transition was only 3.5 years. Sphincter electromyography (EMG) and cardiovascular autonomic testing may be useful to confirm the diagnosis of MSA. Signs of autonomic failure such as orthostatic hypotension and cardiac denervation are uncommon in the inherited ataxias, though bladder symptoms are often present in late stages.

Clinical Approach to Patients with Progressive Cerebellar Ataxia

A careful clinical approach to patients presenting with progressive ataxia will allow accurate diagnosis and appropriate management (Fig. 72.13). The age at onset, the tempo of progression, associated neurological and systemic signs, and the availability of family history can all be considered in making a diagnosis. Imaging studies, especially MRI of the brain, allow the diagnosis of those disorders in which characteristic morphological abnormalities underlie the ataxia (e.g., infarcts, mass lesions, demyelinating plaques). Among the degenerative ataxias, MRI can be helpful by documenting isolated cerebellar versus pontocerebellar atrophy and the presence of additional abnormalities such as cortical atrophy or hyperintensity in the middle cerebellar peduncle. Other laboratory studies of value in a patient with progressive ataxia include thyroid function tests, vitamin E and B₁₂ measurements, serology for syphilis, gliadin and GAD antibodies, paraneoplastic antibodies, and possibly CSF examination for pleocytosis, such as can be seen with meningeal carcinomatosis, and for oligoclonal bands. PCR analysis for the WD organism may be useful. In young adults and children, when appropriate clinical circumstances exist, specialized laboratory studies including those in Table 72.3 are indicated. Careful elicitation of family history and the details of the illness in other family members are often of value.

Fig. 72.13 Algorithm for a diagnostic approach to patients with progressive ataxia.

The art and science of molecular testing in inherited ataxias is an evolving one. As of now, DNA tests are available for the CAG expansions involved in SCA1, 2, 3 (MJD), 6, 7, 12, 17, and DRPLA. The CTG expansion associated with SCA8 and the ATTCT expansion in SCA10 are other existing tests. In addition, tests that detect conventional mutations in SCA5, SCA13, SCA14, and EA2 are available. Among the AR ataxias, testing can be done for the FA GAA expansion and mutations in the POLG, aprataxin, and senataxin genes. Tests for the common mutations in mtDNA that may be associated with ataxias such as MERRF, MELAS, and NARP are offered. Gene tests are also possible for AT, ARSACS, AVED, and EA syndromes and many other recently defined gene mutations can be tested for by selected labs or research facilities (see www.geneclinics.org). Patients with features compatible with an inherited ataxia as described in this chapter are possible candidates for gene testing. Such testing is very accurate and usually less expensive than many tests usually employed in neurological diagnosis. Often the choice of tests can be made based on the inheritance pattern and some of the clinical clues alluded to earlier. When family history suggests a dominant mode of inheritance, mutation analysis for the SCAs may be the only test needed. In addition, the prevalence of different genotypes can vary in different areas, probably due to genetic bias caused by strong founder effects. For instance, the FA mutation is confined to Indo-Caucasian populations. The dominant ataxias have different prevalence rates in different areas (Schols et al., 2004) (Fig. 72.14). With regard to the nucleotide repeat expansion disorders, interpretive caution may be required for many reasons. The alleles may be classified into those with normal numbers of repeats (nonpathogenic) and clearly pathogenic ( fully penetrant) alleles. In addition, alleles that have repeat numbers in the gray zone between normal and pathogenic can be found in many diseases. These have been classified as mutable normal (i.e., an allele size that is not yet known to be associated with disease signs but may expand into pathogenic range when transmitted to the next generation) and low penetrance alleles that can cause disease at a late age. The website www.geneclinics.org is an excellent resource for information regarding the evolving information about such tests, and Tables 72.9 and 72.10 summarize some of this information. The FA mutation analysis is indicated in anyone who has a phenotype compatible with classical FA, LOFA, or FARR. Since the phenotypical variability of FA has been found to be extensive, one should consider doing this test in almost all childhood or young adult–onset sporadic ataxia that cannot be readily diagnosed. In the dominant ataxias, some of the phenotypical features noted in Table 72.5 may be of use, but these serve only as very rough guides. Sequence analysis for point mutations is becoming increasingly available; a particular sequence alteration reported may be difficult to interpret, but it may be one already known to be pathogenic. If not, whether a given alteration is the real cause of the disease can only be judged based on more research—for example, the alteration occurs in a highly conserved site and can be predicted to alter protein function, or the alteration segregates with many other affected individuals in the family and cannot be found in control chromosomes from the same ethnic group. There are also no firm guidelines for the use of ataxia tests among patients with sporadic ataxia. If the disorder cannot be adequately diagnosed, at least FA, SCA6, SCA8, and fragile X analysis should be done (see also Sporadic Ataxias). If the clinical features are more typical for an inherited ataxia despite a negative family history or the family history is sketchy, the entire dominant ataxia panel should be performed.

Fig. 72.14 The dominant ataxias have different prevalence rates in different areas.

(Courtesy Dr. Ludger Schöls, University of Tübingen, Germany.)

Predictive testing for the ataxias in at-risk persons is possible if the specific mutation in the particular family is known. It is best done within the framework of a formal predictive testing program such as has been recommended for Huntington disease. Children at risk should not be subjected to predictive testing.