Laura B. Jardim, Maria L. Pereira, Isabel Silveira, Anabela Ferro, Jorge Sequeiros, Roberto Giugliani. Neurologic Findings in Machado-Joseph DiseaseRelation With Disease Duration, Subtypes, and (CAG)n. Arch Neurol. 2001;58(6):899–904. doi:10.1001/archneur.58.6.899
Machado-Joseph disease (MJD), an autosomal dominant spinocerebellar degeneration caused by an expanded CAG repeat on chromosome 14q32.1, is a heterogeneous disorder for clinical manifestations. The reasons for the wide range of neurologic findings in this disease are poorly understood.
To explain part of this heterogeneity through the association of the neurologic findings with sex, disease duration, age of onset, clinical type, and size of CAG repeat expansion.
A case-control study.
A consecutive sample of 62 patients with MJD.
Main Outcome Measure
A direct relationship was found between the disease duration and severity of gait and limb ataxia, dysarthria, dysphagia, fasciculations, pyramidal syndrome, and ophthalmoplegia (P<.02). The most severe forms of nuclear ophthalmoplegia were associated with type 1 MJD, whereas those of supranuclear ophthalmoplegia were associated with type 3 MJD (P<.001). It was also found that higher mean (CAG)n lengths were associated with worse degrees of the pyramidal syndrome and dystonia (P<.001). The presence and severity of nystagmus, eyelid retraction, rigidity and/or bradykinesia, and optic atrophy were not clearly associated with any of the predictive variables under study.
Disease duration can explain part of the heterogeneity of ataxia, dysarthria, dysphagia, fasciculations, pyramidal syndrome, and ophthalmoplegia, in MJD. Type 1 MJD was positively associated with nuclear ophthalmoplegia; type 3 MJD was positively associated with supranuclear ophthalmoplegia. Higher mean CAG lengths were found to correlate with the pyramidal syndrome and dystonia. Nystagmus, eyelid retraction, rigidity and/or bradykinesia, and optic atrophy were hardly attributable to any known reason or variable.
MACHADO-JOSEPH disease (MJD) is a multisystem degeneration of the central nervous system, inherited as an autosomal dominant disease, and first reported in North American families of Portuguese-Azorean ancestry.1- 4 It is most often a highly incapacitating disease—typically, patients will become confined to a wheelchair and will later be beddriden. Disease manifestations usually start during adulthood, with a mean ± SD age of onset of 32 ± 12 years among Brazilian patients (L.B.J., I.S., M.L.P., A.F., J.S., R.G. and I. Alonso, BSc, M. C. Moreira, MSc, P. Mendonça, BSc, and F. Ferreirinha, BSc, unpublished data, 2000), or 37 ± 14 years among the Portuguese and those of Portuguese-Azorean descent.5 The distribution of age of onset is wide, ranging from 5 to 73 years in Portuguese patients.5 Penetrance is high but incomplete since there were rare obligate carriers free of symptoms until as late as the age of 90 years.6 The median survival time after onset varies from 14 to 25 years.2,7
The gene associated with MJD is located on chromosome 14q32.18 and contains a CAG repeat motif in the 5′ region of the coding sequence, which is expanded in patients with MJD.5 Normal alleles vary from 12 to 41 repeats, whereas expanded alleles vary between 66 and 84 repeats.9
The wide range of clinical manifestations include cerebellar ataxia, affecting gait, limb movements, speech articulation and deglutition; a pyramidal syndrome, with brisk deep tendon reflexes, Babinski sign, and spasticity; a supranuclear, progressive external ophthalmoplegia, with early limitation of upward gaze and convergence, and, less frequently, a nuclear ophthalmoplegia; extrapyramidal signs, including dystonia, rigidity and/or bradykynesia; a lower motor neuron disease, with fasciculations and amyotrophy; sensation loss; eyelid retraction; contraction fasciculations; loss of weight; and a sleep disorder.5
Three classic types were proposed4 and are widely accepted: patients with type 1 MJD have an earlier onset and a more severe course, showing severe dystonic and pyramidal signs, progressive external ophthalmoplegia, and limb and gait ataxia. Patients with type 2 MJD become sick around the general mean age of onset and have mostly cerebellar and pyramidal deficits and progressive external ophthalmoplegia. Patients with type 3 MJD have a less severe, later-onset disease, with peripheral signs and gait and limb ataxia (with or without progressive external ophthalmoplegia) with or without pyramidal syndrome. In addition to these well-known phenotypes, other variants have been proposed.10- 14 The wide variability in age of onset, the complex and heterogeneous neurologic findings of MJD, as well as the existence of 3 major phenotypes, are evidence that the disease, caused by a major single gene, is modulated by modifier factors.
The first and better studied modifying factor was the (CAG)n length itself. It was shown that the repeat length has a strong negative correlation with the disease onset and some relationship to clinical subtype.15,16
However, the variable neurologic picture could be, in some extent, related to disease duration. It is plausible that some findings are present in one patient and not in another, depending only on time of evolution.
The main purpose of this study was to help recognize which neurologic findings are attributable to disease duration, sex, subphenotype, age of onset, and CAG repeat length, and which are not attributable to any of the former, keeping in mind that the latter could be seen as independent neurologic markers of some, yet unknown, modifier factor.
Sixty-two patients with MJD were recruited among a large case-series of patients with spinocerebellar ataxia, originating from Rio Grande do Sul, the southernmost state of Brazil. All were confirmed by molecular analysis. These patients belonged to 34 families: 31 of Portuguese ancestry, 2 of African Brazilian ancestry, and 1 of German ancestry. Data, such as age at onset, sex distribution, and disease duration, are summarized in Table 1 and have been extensively commented on elsewhere (L.B.J., I.S., M.L.P., A.F., J.S., R.G. and I. Alonso, BSc, M. C. Moreira, MSc, P. Mendonça, BSc, and F. Ferreirinha, BSc, unpublished data, 2000).
All patients were interviewed and examined by the same physician (L.B.J.). A standardized procedure included a thorough neurologic examination and determining the age at onset, disease duration, first sign at onset. Onset was considered as the time when the patient or a close relative was able to date the appearance of the first symptoms.
Patients were classified among the 3 clinical subtypes, according to neurologic findings, as previously described. The age of onset, however, was not used as a criterion for their classification. Expansion of the MJD1 gene was analyzed according to conditions previously described.17
Study variables were analyzed according to the presence and degree of several neurologic findings. To evaluate which neurologic findings would be attributable to time of evolution, differences in disease duration were assessed by analysis of variance (ANOVA) with test for linear trend. Since disease duration could be debated for its subjective character, patients were also divided according to their degree of incapacitation, using the degree of gait ataxia: (1) subtle gait ataxia, observed only when walking on toes, heels, or in tandem; (2) moderate ataxia, on spontaneous gait, with gait autonomy still preserved; (3) walking with help, of another person or using a cane; and (4) wheelchair bound, the total inability to walk. Differences between the level of incapacitation and the degree of several neurologic findings were assessed by linear-by-linear association. To evaluate which neurologic findings (other than those used for their definition) could be attributable to one of the classic phenotypes, differences among the 3 clinical subtypes were assessed using the Fisher exact test. The association between the degree of each clinical variable and age at onset was tested by ANOVA; the association between the presence of a neurologic sign and sex was tested using the Fisher exact test. Finally, the association between the degree of neurologic findings and the CAG repeat length was measured by Mann-Whitney test (when comparing 2 groups) or by Kruskal-Wallis test (for multiple comparisons) since CAG length distribution was not normal.
Gait ataxia was the initial sign in all patients. Table 2 gives further details about the neurologic signs found in the entire sample. We have found 8 patients (13%) with type 1 MJD, 26 (42%) with type 2 MJD, and 21 (34%) with type 3 MJD. Their ages at onset (Table 1) were compatible with data from the literature.4,5,15,16 We were unable to classify 7 patients (11%) because they presented concurrently with severe extrapyramidal and peripheral findings, and for this reason they were labeled "unclassified." Since these patients had the longest disease duration, the difficulty in classifying them may simply reflect their advanced stage in the disease (Table 1).
Neurologic signs were classified according to their degree of severity or just according to their presence or absence (Table 3). Amyotrophy, in particular, was rarely seen (2 of 62 patients). Indeed, well-delineated muscles were seen in 15 of the 62 patients, even though several of them were already very incapacitated.
Gait ataxia, dysarthria, dysphagia, and fasciculations progressed with time (P≤.02) (Table 4), whereas nystagmus, pyramidal signs, limb ataxia, ophthalmoplegia, dystonia, rigidity and/or bradykinesia, eyelid retraction, and optic atrophy did not (Table 4). Because fasciculations are markers of type 3 MJD and because this subgroup of patients showed long disease duration (Table 1), we tested the association of disease duration with fasciculations within patients with type 3 MJD only and found a positive result, confirming former findings (P<.046, Table 4).
We chose one neurologic variable, gait ataxia, as a marker of incapacitation because disease duration being subjective information and biological time, as aging, is not the same to everyone. Gait ataxia was the initial sign in all patients, the main report in most, and showed a strong linear association with duration of disease.
Limb ataxia, pyramidal syndrome, dysarthria, dysphagia, ophthalmoplegia, and fasciculations all worsened at the same time as gait ataxia (or incapacitation) progressed (Table 4). No linear association was found between the degree of gait ataxia and nystagmus, eyelid retraction, sensation loss, dystonia, rigidity and/or bradykinesia, or optic atrophy (Table 4).
Thirty patients were male and 32 were female. Nystagmus was present in all of the women, while 5 of 30 men did not show it, a difference that was statistically significant (P<.02, Fisher exact test) (Table 4).
When the severity of each neurologic finding (other than those used for the definition of the clinical types) was analyzed among the 3 classic types, no association was found in most cases (Table 4). The severity of ophthalmoplegia, however, showed some association with phenotype (P<.001) (Table 4). Supranuclear ophthalmoplegia (manifesting as a medial longitudinal fasciculus syndrome, or a limitation in upward gaze and convergence [Table 3]) was more frequent among patients with type 3 MJD, while nuclear ophthalmoplegia—mainly of the sixth nerve—was more common in patients with type 1 MJD and in patients who were unclassified.
If neurons would be prone to disease process in specific intervals along a time line, then their signs of dysfunction would be linked to the age of onset. In this possible cause-effect relationship, disease duration is a likely confounder because, as the disease lasts longer, the more likely is the appearance of certain signs. Only patients with disease duration up to 5 years (n = 18) were used to test this hypothesis. Their mean age of onset was 27.3 years (age range, 12-47 years); earlier than that found in the global sample. In this subgroup, pyramidal signs, dysarthria, dystonic postures, and rigidity and/or bradykinesia were associated with an early onset of the disease (P = .04, P = .01, P = .003, and P = .04, respectively).For example, the 14 patients without rigidity and/or bradykinesia had a mean age of onset of 30 years, whereas the 4 patients with these symptoms had a mean age of onset of 17 years. These results are statistically significant, but should be viewed with caution, owing to the small size of the sample.
The severity of neurologic signs was also tested for a possible association with the CAG repeat length (Table 4). A higher mean CAG length was then associated with a worse degree of the pyramidal syndrome and dystonia (Table 5).
Clinical heterogeneity is one of the most puzzling characteristics within the spinocerebellar ataxias in general and in MJD in particular. The variable expressivity in age at onset, the complex and heterogeneous neurologic findings, as well as the existence of 3 subphenotypes, suggest the existence of some modifier factors.
It is not very clear to what extent the heterogeneity of neurologic signs is due to the natural history of the disease, or is due to other genetic and environmental influences. Following the chronology of symptoms in a cohort study would be the best way to find the answer. This kind of study has been partially conducted18; however, owing to the recent advances in the molecular knowledge of MJD, a more comprehensive study is still missing. We tried to circumvent this difficulty by looking for association between the severity of several findings and time of evolution of the disease, using disease duration (chronological time) and degree of incapacitation (biological time).
The disease progression, measured either by duration or by incapacitation, seemed to be responsible for the worsening that occurred in gait and limb ataxia, dysarthria, dysphagia, fasciculations, and pyramidal signs (Table 4). No association was found between the duration of the disease process and nystagmus, eyelid retraction, dystonic postures, rigidity and/or bradykinesia, and optic atrophy. Other reasons should explain their presence. To look for these reasons, the severity of neurologic signs was assessed according to MJD type, age of onset, sex, and CAG repeat length.
The few patients with a short disease duration prevented our attempt to relate the different neurologic signs to age of onset. Also, the small numbers that link nystagmus with female sex demand some caution.
The CAG repeat length of the expanded allele also did not explain the presence of nystagmus, eyelid retraction, rigidity and/or bradykinesia, and optic atrophy. Therefore, the heterogeneity in these neurologic signs remained unexplained.
The only association found between sign and clinical type was those related to ophthalmoplegia. The supranuclear ophthalmoplegia was more severe among patients with type 3 MJD, whereas nuclear ophthalmoplegia was more common in patients with type 1 MJD and in those who were unclassified. Because type 1 MJD is the most severe form and patients who were unclassified were those who have a longer duration of illness, nuclear ophthalmoplegia could be linked with the severity of the general clinical picture. Nuclear ophthalmoplegia was also associated with large CAG tracts. Since large CAG tracts are the main determinants of type 1 MJD,15 this result seems to confirm the former. Severity of gait and limb ataxia and dysarthria seemed to be exclusively attributed to disease progression.
The pyramidal syndrome did not differ statistically among the 3 subtypes, whereas we expected to find the more severe forms among patients with type 1 MJD.4,5,16 Because we observed that its severity was associated with the CAG repeat length (Table 5), we can speculate that the former lack of statistical significance could be due to a small sample of type 1 MJD cases.
The present findings related to disease duration almost reproduce those described by Barbeau et al19 and by Coutinho,18 but the same cannot be said about the other results, mostly because of differences in the way the neurologic examination was plotted in the 3 studies. Barbeau et al found that "nystagmus, when present, does not appear to be progressive."19(p522) Unfortunately, they did not mention eyelid retraction, optic atrophy, and rigidity and/or bradykinesia.19 Coutinho followed up 67 patients for 3.5 years, on average, and found associations with disease duration that were similar to our data, the exception being the pyramidal syndrome. She also noted that eyelid retraction did not worsen with disease duration.18 Both previous studies were performed before the molecular discoveries, so that these data were lacking.18,19 A third study described the relationship of some variables of the natural history of MJD.7 They found that the CAG repeat length increased the rate of disease progression, whereas the age of onset did not. Our results agreed with the latter findings, but not with the former findings. Finally, although it had a design that was similar to ours, the study of Klockgether et al20 evaluated only the risk factors for peripheral neuropathy in MJD.
Heterogeneity of several neurologic findings seem to be partly explained by duration of the disease, clinical type, and CAG repeat length. Nystagmus, eyelid retraction, rigidity and/or bradykinesia, and optic atrophy, however, were hardly attributable to any factor. We speculate that they could be seen as independent neurologic markers of some unknown modifier factor.
Accepted for publication November 28, 2000.
This work was partly supported by Fundação de Amparo à Pesquisa do Rio Grande do Sul (FAPERGS), Porto Alegre, Brazil, Comissão de Aperfeiçoamento de Pessoal do Ensino Superior (CAPES), Brasília, Brazil, and Instituto de Cooperaçã Científica e Tecnológica Internacional (ICCTI) Lisboa, Portugal.
We gratefully acknowledge the statistical review of Bernardo Lessa Horta, MD, PhD, Escola de Medicina e de Psicologia, Universidade Católica de Pelotas, Pelotas, Brazil.
Corresponding author: Laura B. Jardim, MD, Medical Genetics Service, Hospital de Clínicas de Porto Alegre, Rua Ramiro Barcelos 2350, 90035-003 Porto Alegre, Brazil (e-mail: email@example.com).