The Cerebellum

, Volume 11, Issue 2, pp 549–556

Cerebellar Cognitive Affective Syndrome in Machado Joseph Disease: Core Clinical Features

Authors

    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
    • Instituto Israelita de Ensino e Pesquisa Albert EinsteinHospital Israelita Albert Einstein
  • José Luiz Pedroso
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
    • Instituto Israelita de Ensino e Pesquisa Albert EinsteinHospital Israelita Albert Einstein
  • Helena Alessi
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
  • Lívia Almeida Dutra
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
    • Instituto Israelita de Ensino e Pesquisa Albert EinsteinHospital Israelita Albert Einstein
  • André Carvalho Felício
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
  • Thaís Minett
    • Department of Preventive MedicineUniversidade Federal de São Paulo
    • Department of Public Health and Primary CareUniversity of Cambridge
  • Patrícia Weisman
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
  • Ruth F. Santos-Galduroz
    • Center of Mathematics, Computer and CognitionUniversidade Federal do ABC
    • Institute of Biosciences, UNESP
  • Paulo Henrique F. Bertolucci
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
  • Alberto Alain Gabbai
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
    • Instituto Israelita de Ensino e Pesquisa Albert EinsteinHospital Israelita Albert Einstein
  • Orlando Graziani Povoas Barsottini
    • Department of Neurology and NeurosurgeryUniversidade Federal de São Paulo
    • Instituto Israelita de Ensino e Pesquisa Albert EinsteinHospital Israelita Albert Einstein
Original Paper

DOI: 10.1007/s12311-011-0318-6

Cite this article as:
Braga-Neto, P., Pedroso, J.L., Alessi, H. et al. Cerebellum (2012) 11: 549. doi:10.1007/s12311-011-0318-6

Abstract

The cerebellum is no longer considered a purely motor control device, and convincing evidence has demonstrated its relationship to cognitive and emotional neural circuits. The aims of the present study were to establish the core cognitive features in our patient population and to determine the presence of Cerebellar Cognitive Affective Syndrome (CCAS) in this group. We recruited 38 patients with spinocerebellar ataxia type 3 (SCA3) or Machado–Joseph disease (MJD)-SCA3/MJD and 31 controls. Data on disease status were recorded (disease duration, age, age at onset, ataxia severity, and CAG repeat length). The severity of cerebellar symptoms was measured using the International Cooperative Ataxia Rating Scale and the Scale for the Assessment and Rating of Ataxia. The neuropsychological assessment consisted of the Mini-Mental State Examination, Clock Drawing Test, Wechsler Adult Intelligence Scale, Rey–Osterrieth Complex Figure, Wisconsin Card Sorting Test, Stroop Color–Word Test, Trail-Making Test, Verbal Paired Associates, and verbal fluency tests. All subjects were also submitted to the Hamilton Anxiety Scale and Beck Depression Inventory. After controlling for multiple comparisons, spatial span, picture completion, symbol search, Stroop Color–Word Test, phonemic verbal fluency, and Trail-Making Tests A and B were significantly more impaired in patients with SCA3/MJD than in controls. Executive and visuospatial functions are impaired in patients with SCA3/MJD, consistent with the symptoms reported in the CCAS. We speculate on a possible role in visual cortical processing degeneration and executive dysfunction in our patients as a model to explain their main cognitive deficit.

Keywords

Spinocerebellar ataxia type 3Machado–Joseph diseaseCognitive deficitsCerebellar Cognitive Affective Syndrome

Introduction

The statement that the cerebellum is purely a motor control device no longer applies. During the last few decades, there has been increasing debate over possible non-motor functions of the cerebellum [13]. Anatomical, physiological, clinical, and functional neuroimaging data of patients with cerebellar diseases suggest the existence of cognitive and emotional neural circuits [4]. These findings might provide additional knowledge on the mechanisms underlying the Cerebellar Cognitive Affective Syndrome (CCAS) [57].

The spinocerebellar ataxias (SCAs) are a group of genetically defined autosomal dominant neurological diseases, characterized by a heterogeneous clinical presentation. On motor evaluation, hereditary ataxias present with a broad combination of cerebellar deficits including oculomotor disorders, dysarthria, dysmetria, tremor, and ataxic gait [8]. The prevalence and severity of cognitive dysfunction vary considerably in SCA populations. Numerous differences in cognitive deficits and patterns of neuropsychological deficits have been noted over the last years [9]. SCA1, 2, and 3 are associated with greater deficits in executive functions than in verbal memory [10, 11]. Additional impairments in visual memory, visuoconstruction, and visual attention have also been described in SCA3/Machado–Joseph disease (MJD) patients [1215].

Spinocerebellar ataxia type 3 (SCA3), also known as MJD–SCA3/MJD, is the most frequent SCA worldwide. SCA3/MJD is characterized by the presence of an abnormal expansion of a CAG trinucleotide repeat on chromosome 14 at region 14q24.3-q31 that leads to an ataxin-3 mutated protein [16].

SCA3/MJD has a broad range of clinical manifestations which include cerebellar ataxia, spasticity, parkinsonism, dystonia, peripheral neuropathy, pseudoexophthalmos (bulging eyes caused by lid retraction), sleep disturbances, and hyposmia [1721]. Indeed, a recent study has demonstrated the occurrence of a more diffuse neurodegenerative process in SCA3/MJD, which may explain the non-motor symptoms seen in these patients [22].

However, despite substantial research on the cognitive aspects of cerebellar disorders, scant data exist to establish the pattern of cognitive deficits in patients with clinical and molecular diagnosis of SCA3/MJD [1015].

We investigated the cognitive manifestations of a large cohort of Brazilian patients with SCA3/MJD in an attempt to replicate and further characterize the principal cognitive and emotional features of our patient population. The aim of the present study was to establish the core cognitive features in our patient population and to determine the presence of CCAS in this group.

Methods

Subjects

A total of 38 patients with clinically and molecularly confirmed SCA3/MJD seen in our outpatient General Neurology Unit at the Universidade Federal de São Paulo, Brazil, from February 2008 to November 2010, were evaluated. Thirty-one healthy volunteers with no history of neurological or psychiatric diseases were recruited as a control group. All patients and controls met the inclusion criteria, namely: age under 60 years (to avoid any age-related cognitive phenomena, Mini-Mental State Examination (MMSE) score greater than 24, and at least 8 years of education). Medications known to induce cognitive deficits such as tricyclic antidepressants, benzodiazepines, and anticholinergic drugs were all suspended at least 1 month before clinical evaluation.

The local research ethics committee approved this study. All participants provided written informed consent for the clinical investigation and subsequent analysis.

Neurological and Psychiatric Evaluation

Clinical assessment and neurological physical examinations were performed. The ataxia clinical assessment was conducted in all subjects using the International Cooperative Ataxia Rating Scale (ICARS) and the Brazilian validated scale for the assessment and rating of ataxia (SARA) [2325]. Participants were also evaluated by the Hamilton Anxiety Scale (HAMA) and Beck Depression Inventory (BDI) [26, 27].

Neuropsychological Battery

A battery of neuropsychological tests was administered according to standard published protocols as outlined below. The methodological details of these tests have been described elsewhere in the literature. The tests used in the current study were administered according to a standard protocol. Patients were free to discontinue participation at any stage of the study in accordance with ethics committee requirements.

Global Cognitive Performance

The MMSE was used as a screening tool for general intellectual abilities [28, 29]. The Clock Drawing Test was used to screen for visuospatial and constructional abilities based on a scoring system from 0 to 15 [30].

Attention and Working Memory

The Digit Span subtest of the Wechsler Adult Intelligence Scale Revised (WAIS-R) was used to evaluate verbal attention (forward) and working memory (backward) [31]. The number of correctly reproduced items was recorded.

Visuospatial Abilities

The spatial span subtest of the WAIS-R was used to measure visual attention and working memory [31]. Forward and backward reproduction was tested separately. The number of correctly reproduced items was noted. Visual perception and attention was assessed with the picture completion subtest of the WAIS-III [32, 33]. The Rey–Osterrieth Complex Figure (ROCF) scale was used to evaluate executive visuoconstructive and planning abilities (copy) and visual episodic memory (immediate and delayed recall) [34].

Verbal Memory

The Verbal Paired Associates subtest of the Wechsler memory scale revised was applied to evaluate patients' verbal memory, learning ability, and abstraction [35].

Frontal Executive Function

The symbol search subtest of the WAIS-III was used to assess processing speed and visuospatial function integrity. The Similarities subtest, used to identify abstraction skills, was also applied [32, 33].

The Trail-Making Test parts A and B were used to assess speed of visual search, but they also assess mental flexibility, attentional resources, and motor abilities [36]. Scoring was based on the time required to complete the task, and errors were corrected during the test. Subtraction of trail part A from part B (Trail-Making B–A) was used to control the motor and sequencing elements of the test, providing a purer measure of executive function.

The Brazilian version of the Wisconsin Card Sorting Test (WCST) was applied to assess the ability to form abstract concepts and cognitive flexibility in response to changing parameters [3739]. The total number of responses, categories achieved (ten consecutive correct responses), perseverative and random errors, rule shifting, and difficulty maintaining set errors were registered [37].

The Stroop Color–Word Test (SCWT) was used to measure selective attention, cognitive flexibility and inhibitory control. The examiner pointed out errors as and when they took place to enable all patients to complete the test. Therefore, time alone in seconds was the base score. The interference score served as the main outcome variable and was defined according to the following formula: part 3 time − mean (part 1 time + part 2 time) [37].

Verbal Fluency

The semantic verbal fluency (animals) and phonemic (letter A) tests were used to assess association fluency. The score was computed as the total number of words given by the subjects for each category in 1 min [34].

Statistical Analysis

The chi-square test (X2) (without Yates correction) was used for comparisons of categorical data. Differences in the means of continuous measurements were tested by Student's t test (t).

Separate multiple linear regressions were carried out for those neuropsychological tests which significantly differed between the two groups on the univariate analysis as dependent variables, and presence of SCA3/MJD, age, male sex, education, BDI, and HAMA as independent variables.

All tests were two tailed, and a p value <0.05 was considered to indicate statistical significance on all tests except those involving the psychiatric and cognitive performance where multiple comparisons were performed and a subsequently lower p value set (p < 0.0016) according to the Bonferroni correction. Ninety-five percent confidence intervals (CI) were calculated for the difference between means and odds ratio. The entire analysis was calculated using the statistical software Statistical Package for the Social Science (SPSS) 19 for Windows.

Results

Among patients with SCA3/MJD, mean age at disease onset was 34.7 ± 10.1 years, and mean disease duration was 7.3 ± 4.4 years. The mean CAG repetition length was 71.6 ± 3.6. Regarding clinical features, the mean score on the ICARS was 39.4 ± 23.7, and on the SARA, was 13.2 ± 9.1.

There were no differences between controls and SCA3/MJD patients regarding age (40.5 ± 10.3 versus 42.3 ± 10.1, t(67) = −0.72, 95% CI = −6.7–3.2, p = 0.473), gender (45% versus 47% males, X2(1) = 0.03, p = 0.855), or years of education (13.1 ± 2.8 versus 12.7 ± 3.0, t(67) = 0.63, 95% CI = −1.0–1.9, p = 0.528).

Table 1 shows a comparison of all neuropsychological tests performed, as well as the psychiatric assessment. Patients with SCA3/MJD had significantly more depressive and anxiety symptoms than controls.
Table 1

Comparison of psychiatric and cognitive performance of controls and patients with SCA3/MJD

 

Control

SCA3

t

df

95% CI (mean difference)

p value

Mean

SD

Mean

SD

MMSE

28.0

1.4

27.4

1.9

1.4

65

−0.2

to

1.4

0.162

HAMA

5.7

3.7

12.2

8.9

−3.6

63

−10.1

to

−2.9

0.001*

BDI

5.8

5.1

15.4

10.1

−4.6

63

−13.7

to

−5.4

<0.001*

Clock Drawing

14.9

0.4

14.8

0.6

0.8

64

−0.2

to

0.4

0.430

Digit Span

          

 Forward

5.5

1.2

5.4

1.1

0.3

67

−0.5

to

0.7

0.773

 Backward

3.8

1.3

3.7

1.0

0.2

67

−0.5

to

0.6

0.818

Spatial Span

          

 Forward

5.3

1.5

4.3

0.8

3.5

67

0.4

to

1.5

0.001*

 Backward

4.9

1.0

3.8

1.2

4.2

67

0.6

to

1.7

<0.001*

 Similarities

21.0

5.5

16.9

7.3

2.6

66

0.9

to

7.3

0.013

 Picture completion

18.7

3.9

13.7

4.0

5.3

67

3.1

to

6.9

<0.001*

 Symbol search

29.8

4.6

16.9

8.6

7.6

67

9.5

to

16.3

<0.001*

ROCF

          

 Copy

35.2

1.5

33.9

2.5

2.5

66

0.3

to

2.3

0.014

 Immediate recall

22.0

5.7

18.5

6.5

2.4

66

0.6

to

6.5

0.021

 Delayed recall

20.9

6.3

17.5

6.4

2.2

66

0.3

to

6.5

0.033

WCST

          

 Total responses

113.2

21.2

118.7

16.2

−1.2

67

−14.5

to

3.5

0.229

 Categories achieved

3.9

2.1

3.5

1.8

0.9

67

−0.5

to

1.4

0.390

 Perseverative errors

8.4

6.7

13.7

10.3

−2.5

67

−9.6

to

−1.1

0.015

 Random errors

32.7

17.9

35.4

17.9

−0.6

66

−11.3

to

6.1

0.551

 Rule shifting errors

2.0

1.8

2.1

1.8

−0.2

66

−1.0

to

0.8

0.808

 Failure to maintain set

1.0

1.3

1.2

1.5

−0.6

66

−0.9

to

0.5

0.520

SCWT

          

 Part 1 (time)

39.4

8.7

57.6

21.2

−4.5

67

−26.3

to

−10.1

<0.001*

 Part 2 (time)

36.4

5.0

57.4

29.8

−3.9

67

−31.9

to

−10.2

<0.001*

 Part 3 (time)

67.9

20.7

100.9

37.3

−4.4

67

−48.0

to

−18.0

<0.001*

 Interference score

30.0

18.6

43.4

23.9

−2.5

67

−23.9

to

−2.9

0.013

Trail-Making Test

          

 Trail A (time)

37.6

10.5

67.4

34.9

−4.6

66

−42.8

to

−16.8

<0.001*

 Trail B (time)

86.1

29.6

118.6

45.9

−3.4

66

−51.7

to

−13.4

0.001*

 Trail B–trail A

48.4

26.9

51.2

47.7

−0.3

66

−22.1

to

16.5

0.773

Verbal paired associates

          

 First trial

4.2

1.9

3.1

1.9

2.3

67

0.2

to

2.0

0.022

 Second trial

4.8

1.9

4.4

1.8

1.1

67

−0.4

to

1.4

0.297

 Third trial

5.6

1.9

5.1

1.9

1.2

67

−0.4

to

1.4

0.247

 Delayed trial

5.3

1.6

4.6

1.8

1.7

67

−0.1

to

1.5

0.093

Verbal fluency

          

 Semantic (animals)

17.4

4.3

14.8

4.6

2.3

62

0.3

to

4.7

0.027

 Phonemic (letter A)

11.5

3.7

7.8

3.6

4.0

62

1.9

to

5.5

<0.001*

SCA3/MJD Spinocerebellar ataxia type 3/Machado–Joseph disease, df degrees of freedom, CI confidence interval, MMSE Mini-Mental State Examination, BDI Beck Depression Inventory, HAMA Hamilton Anxiety Scale, ROCF Rey–Osterrieth Complex Figure, SCWT Stroop Color–Word Test, WCST Wisconsin Card Sorting Test

*p < 0.0016 indicates statistical significance according to Bonferroni's correction

After controlling for multiple comparisons, spatial span; picture completion; symbol search; SCWT parts 1, 2, and 3; phonemic verbal fluency; and Trail-Making Tests A and B were significantly more impaired in patients with SCA3/MJD than in controls.

These neuropsychological tests remained significantly more impaired in patients with SCA3/MJD than in controls, independently of demographic factors, and anxiety and depressive symptoms (Table 2). This was verified by means of ten separate multiple linear regressions for the neuropsychological tests, which differed significantly between the two groups on the univariate analysis as dependent variables, and for the presence of SCA3/MJD, age, male, education, BDI, and HAMA as independent variables (Table 2). Only the coefficients for the presence of SCA3/MJD are reported.
Table 2

Multiple logistic regressions to verify association among neuropsychological tests, demographics, and disease status

Regression analysis

Neuropsychological tests

β

SE

95% CI (β)

p value

1

Spatial Span forward

−0.76

0.32

−1.4/−0.13

0.019*

2

Spatial Span backward

−0.92

0.33

−1.58/−0.26

0.007*

3

Picture completion

−3.85

1.12

−6.09/−1.62

0.001*

4

Symbol search

10.89

1.9

14.68/−7.09

<0.001*

5

SWCT part 1

12.57

4.08

4.4/20.75

0.003*

6

SWCT part 2

13.89

5.56

2.77/25.02

0.015*

7

SWCT part 3

23.02

7.64

7.72/38.32

0.004*

8

Trail-Making Test A

16.22

7.25

1.69/30.75

0.029*

9

Trail-Making Test B

26.38

10.25

5.85/46.9

0.013*

10

Fluency–phonemic

−2.92

1.01

−4.93/−0.9

0.005*

SCWT Stroop Color–Word Test

Discussion

Our results indicated that patients with clinically and molecularly proven SCA3/MJD have cognitive deficits. We described visuospatial, attention and processing speed impairment and executive dysfunction in SCA3/MJD patients.

Impaired visuospatial function in SCA3/MJD patients has been described in other studies, even when assessed using different neuropsychological tests. Deficits reported include color simultanagnosia, visual attention, visual memory, and impairments in visuoconstructive and planning abilities [12, 14, 15]. Our results failed to confirm visual memory impairment, as none of the items on the ROCF differed significantly between groups. Symbol search is a test that requires visual attention and processing speed, while picture completion measures visual perception. Our results also demonstrated significant deficits on the Spatial Span test, which also gives rise to visual attention impairments in SCA3/MJD patients. Given the fact that these three tests require fast visuomotor search, cerebellar oculomotor problems may interfere with their performance. However, the results of our study were in accordance with previous data showing that visual motion processing in cerebral cortex critically depends on an intact cerebellum and its cerebrocerebellar circuit integrity [40]. Consistent with our results, the regional metabolism of occipital cortex was found to be significantly reduced in a PET study involving SCA3/MJD patients [41].

The division of visual cortical processing was recently reviewed. The ventral stream is regarded as the “what” pathway, while the dorsal is regarded as a “how” pathway. The latter gives rise to three distinct, major pathways: a parieto-prefrontal pathway, a parieto-premotor pathway, and a parieto-medial temporal pathway, which primarily support spatial working memory, visually guided action, and spatial navigation, respectively [42]. The neurodegenerative process in patients with SCA3/MJD involves visual, auditory, vestibular, somatosensory, dopaminergic, and cholinergic systems [23]. There is evidence that nuclear ataxin-3 is widely accumulated with axonal aggregates in the cerebral cortex, which could be detrimental to the axonal transportation mechanism, thereby leading to nerve cell degeneration [43]. More specifically, the inferior and lateral subnuclei of the pulvinar were found to be affected, which are integrated into the visual attentional networks of the human brain [44]. We speculate on a possible role in visual cortical processing degeneration in our patients as a model to explain the cognitive deficits.

Previous data indicate the occurrence of executive dysfunction in patients with SCA3/MJD as measured by the WCST and SWCT [13]. Our study did not confirm these results, corroborating two other studies that also failed to show WCST impairment [10, 15]. It is possible that differences in performance on the SCWT may be related to the greater difficulty patients with oculomotor deficits have when reading across rows and columns, since the isolated items of the SCWT were equally impaired, yet the interference score was not.

Phonemic verbal fluency might be impaired in SCA3/MJD, as seen in several studies [10, 11, 13, 15]. We also found a significant impairment in phonemic verbal fluency but not in semantic category. Semantic fluency is regarded as being largely dependent upon the integrity of semantic memory, while phonemic fluency is supposedly more sensitive to executive dysfunction [45]. Indeed, phonemic and semantic fluency tasks were associated with different involvement of the frontal and temporal lobes. A recent functional magnetic resonance imaging study found greater activation in the left precentral and inferior frontal gyrus for phonemic fluency and greater, more anterior activation in the left middle frontal gyrus as well as in the left temporal lobe (fusiform gyrus) for semantic fluency [46]. Moreover, greater activation of the left occipitotemporal sulcus/posterior fusiform gyrus was also detected during word retrieval for letters than for semantic cues [46]. This phenomenon may be related to word-form processing demands, which are greater when retrieval is guided by letters rather than semantic cues [46]. Indeed, letter fluency requires initial mapping of the letter cue to phonologic and/or orthographic information, and checking that the orthography of the retrieved word matches the initial letter cue. This activity may reflect activation of the visual word formation area. Thus, phonemic verbal fluency might also be associated with visual processing deficits in our patients.

Verbal Fluency tasks are also known to be affected in patients with cerebellar damage, with phonological fluency being more compromised than semantic fluency [4, 47, 48]. Cerebrocerebellar connections can provide some insight into this apparent phonemic–semantic discrepancy [47, 48]. The cerebellum is linked with prefrontal regions, but not with inferotemporal cortices, which may be important for semantic linguistic tasks [49, 50].

In our study, patients' verbal memory, tested by the verbal paired associates test, was intact. Previous data regarding verbal memory deficits in SCA3/MJD patients have shown conflicting results. When neuropsychological tests, such as the 16-item categorized and uncategorized word list and logical memory subtests 1–2 of the Wechsler Memory Scale, were applied, significant impairments were detected [10, 15]. However, another study investigated six patients with SCA3/MJD all of whom performed well on tasks of free recall and recognition of a list of words, and on story tasks, from the Hopkins Verbal Learning Test-Revised and the Wechsler Memory Scale-III, respectively [13]. A more recent study, using the Rey Auditory-Verbal Learning Test, evaluated 15 patients with SCA3/MJD but found no difference compared to controls [11]. Our results indicated no verbal memory impairment in patients with SCA3/MJD. The discordant results might be explained by the use of different neuropsychological tests. These other tests place demands on working memory, language processing, and executive functions that may also be impaired in cerebellar disorders [13].

Depression and anxiety are common in patients with SCA3/MJD, even when measured by different scales such as the Montgomery–Asberg Depression Rating Scale, State Trait Anxiety Inventory, and Hospital Anxiety and Depression Scale [1113, 15]. Our results are in accordance with those in the literature.

The CCAS is characterized by cognitive impairments in executive and visuospatial functions, and language as well as affective symptoms [5, 6]. CCAS is reported to be related to neural circuits linking prefrontal, posterior parietal, superior temporal, and limbic cortices with the cerebellum [36]. Our data do not allow us to define the role of the cerebellum in the cognitive process as patients with SCA3/MJD have extra-cerebellar involvement. However, some cognitive aspects of CCAS are present in our patients (visuospatial and executive deficits as well as affective symptoms).

Our study should be interpreted in the context of some limitations. Cortical involvement in SCA3/MJD has recently been described [51]. Unfortunately, we were unable to perform functional imaging studies in order to correlate our findings with neuropsychological studies, an approach we believe to be more appropriate for investigating the physiopathology of cognitive deficits in SCA3/MJD. However, we believe that our data are important because they reveal that visuospatial and executive dysfunctions are the core cognitive deficits in our population, consistent with the symptoms reported in the CCAS. Furthermore, this is the largest study of cognitive assessment in patients with SCA3/MJD. It is also important to mention that previous studies have described a great number of cognitive impairments, but the role of visuospatial abilities has rarely been considered an important factor of interference in other cognitive domain evaluations.

In conclusion, our study confirms executive and visuospatial impairment in patients with SCA3/MJD. We speculate on a possible role in visual cortical processing degeneration and executive dysfunction in our patients as a model to explain their main cognitive deficits. Cognitive, affective, and motor impairments in our patients may interact with each other. The description of these findings is an important clinical phenomenon that may guide physicians in disease management. Cognitive rehabilitation focusing on visuospatial function may lead to improvement in SCA3/MJD patients.

Acknowledgments

This study was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP). We would like to thank the patients and their families for participating in this study. The authors also would like to thank Dr. Jeremy Schmahmann for his comments and suggestions on this manuscript.

Conflict of Interest

We have no conflicts of interest.

Protection of Human Research Subjects and Human Subject Informed Consent

The procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national). All subjects were provided with the approved informed consent.

Copyright information

© Springer Science+Business Media, LLC 2011