Quality of life in individuals with spinocerebellar ataxia type 10: a preliminary study

ABSTRACT Spinocerebellar ataxia type 10 (SCA10) is characterized by gait ataxia, dysarthria, nystagmus, epilepsy, reduced cognitive ability and depression, which lead to functional loss and behavioral changes. These signs gradually evolve and may interfere with the physical, emotional, and social aspects of quality of life (QoL). Objective: To assess the self-perception of quality of life and its association with disease duration, severity of ataxia, balance and functional independence. Methods: This study focused on the disease duration, ataxia severity (SARA), balance (Berg Balance Scale), functionality (FIM, Lawton IADL) and QoL (SF-36 v.2) of 15 individuals with SCA10. Results: The population sample consisted of eight females and seven males, with a mean age of 43.8 (± 8.2) years, mean age of symptom onset of 33.1 (± 8.9) years and mean disease duration of 9.8 (± 11.2) years. The mean Berg Balance Scale score was 47.2 (± 12), mean SARA score (n = 14) 11.5 (± 7.3), mean Lawton IADL score 20.4 (± 1.8) and mean FIM score 120.3 (± 5.4). Individuals with SCA10 had a greater impairment of QoL in the “role-physical” domain (p = 0.04). The longer the disease duration (p = 0.02), risk of falling (p = 0.04), severity of ataxia (p = 0.00) and functional dependence in activities of daily living (p = 0.03) and instrumental activities of daily living (p = 0.00), the worse the QoL was in the “physical functioning” domain, with a decrease of 1.62 points for each year of disease duration. Conclusion: In this sample, the greatest impairment of QoL in individuals with SCA10 was observed in “physical functioning” and “physical role”.

The characteristic symptoms are gait ataxia, dysarthria, dysphagia, nystagmus and, in some cases, epileptic seizures, reduced cognitive ability and depression 4,5 . In addition, reduced ability to perform activities of daily living (ADL) impairs quality of life (QoL) in individuals with ataxia 6 .
The World Health Organization defines QoL as the individual' s perception of their position in society and in relation to their goals, expectations and concerns 7 . A QoL assessment through generic or specific instruments is an important measure of the impact of treatments on health indicators 8 . Health is not merely the absence of disease 9 and, therefore, should be measured based on QoL 10 . This study aimed to assess the selfperception of QoL and its association with disease duration, severity of ataxia, balance and functional independence.

Participants
The study population comprised 15 individuals (eight females, seven males) with a diagnosis of SCA10 who were treated in the Ataxia Outpatient Unit. Inclusion criteria were: a) having a positive molecular genetic test for SCA10; b) in the absence of molecular results, having neurological signs characteristic of SCA10 and a family member with a positive genetic test for SCA10; c) absence of dementia; and d) having signed a voluntary informed-consent form.
The study was a prospective, descriptive, cross-sectional study. All the participants were told about the procedures and signed the voluntary informed-consent form.

Assessment instruments
The following instruments, validated in Portuguese, were used: the Medical Outcomes Study 36-Item Short-Form Health Survey Version 2 (SF-36) 11 to assess QoL; the Functional Independence Measure (FIM) 12 for ADL and the Lawton Instrumental Activities of Daily Living (IADL) scale 13 ; the Berg Balance Scale 14 for balance and fall risk; and the Scale for the Assessment and Rating of Ataxia (SARA) 15 to rate the severity of the ataxia.
SF-36: This comprises 36 questions. One measures health transition and the others cover eight domains. Scores from 0-100 are assigned to each answer, 0 corresponding to the worst health state and 100 to the best 11 . In our study, we defined a cutoff at 50 points. Scores above this value were classified as "high quality of life", and those below as "low quality of life".
FIM: This instrument provides an objective assessment of an individual's independence in terms of motor and cognitive function. It covers 18 items, and the overall score can vary from 18-126; the lower the final score, the greater the dependence 12 .
Lawton IADL Scale: This assesses an individual's ability to live independently and participate in the community. The scale comprises seven items, and the total score can vary from 7-21. The global score is interpreted as follows: ≤ 5 -totally dependent; > 5 and < 21 -partially dependent; 21 -independent 13 .
Berg Balance Scale: The Berg Balance Scale assesses static and dynamic balance during 14 tasks. A score from 0-4 is assigned to each task, and the maximum score is 56. Individuals with scores from 41-56 can walk independently (low fall risk), those with scores from 21-40 are able to walk with assistance (medium fall risk) and those with scores from 0-20 require a wheelchair (high fall risk) 14 .
SARA: This is an eight-item scale for assessing and rating ataxia. The total score ranges from 0 (no ataxia) to 40 (severe ataxia) 15 .

Data collection procedures
The collection period was from August 2016 to June 2017, during the individual's multidisciplinary routine check-up at the Ataxia Outpatient Unit. The physiotherapy assessment in this outpatient unit included demographic, clinical and laboratory data (age, disease duration, duration of symptoms, number of ATTCT pentanucleotide repeats) and functional data (SARA, Berg, FIM and Lawton IADL scores). After this evaluation, the individuals were invited to participate in the study and were then assessed by one of the researchers using the SF-36 v.2.

Data analysis
The data were analyzed using Microsoft Office Excel 2013 ® .
To identify the SF-36 domains with the greatest impairment, hypothesis tests were run, in which the mean score in each SF-36 domain was compared with the cutoff (50 points). A 5% significance level was used. T-tests for equal means with unequal variances were used to confirm the hypotheses.
To confirm the hypothesis that the variables disease duration, Berg, SARA, FIM and Lawton IADL influenced the magnitude of any increase or decrease in QoL in each SF-36 domain, the correlation between each pair of variables was analyzed with ANOVA at a 5% significance level. When a statistically significant result was obtained, the linear regression equation was determined and the regression line drawn on the scatter plot. The line was fitted using the least squares method, and the linear relationship was identified as either positive or negative. Pearson's correlation coefficient was also calculated for all the SF-36 domains so that the correlation between the variables could be classified as weak, moderate or strong. Descriptive statistics including mean, standard deviation and variance were also calculated.
To analyze the SARA results (which were not obtained for one individual), the technique of replacing the median was used to preserve part of the sample variability and interfere as little as possible with the variable analyzed.

RESULTS
The mean age of the 15 individuals with SCA10 in the study was 43.8 (± 8.3) years, mean age of symptom onset 33.1 (± 9.0) years and mean disease duration 9.8 (± 11.2) years. The mean score on the Berg Balance Scale was 47.2 (± 12), and four individuals were considered dependent and needed a mobility aid when walking. Of these, three used a walking aid, and one refused to use an aid. The mean SARA score (n = 14) was 11.5 (± 7.3). The mean Lawton IADL score was 20.4 (± 1.8); six individuals were independent, three dependent for one item, and six dependent in two to four items. The items for which the most individuals were dependent were "housekeeping", "shopping" and "mode of transportation".
The mean FIM score was 120.3 (± 5.4); seven individuals were independent, five were dependent for one area and three were dependent for two to three areas. The areas with the most dependent individuals were "expression", "social interaction" and "stairs". Considering the means of the scores, most of the study population was independent in ADL and partially dependent in IADL, had a low fall risk (independent balance) and mild ataxia. Demographic, clinical, laboratory and functional data for each individual are shown in Table 1. Table 2 shows the final scores for each of the eight SF-36 domains for each individual with SCA-10. The scores were calculated according to the formulae in the SF-36 questionnaire. The mean, standard deviation and variance for each domain are also given. Mean scores for  the domains "physical functioning", "role-physical", "roleemotional and "vitality" were below 50, corresponding to poor quality of life. However, to test the significance of these means, the t-test for equal means with unequal variances was used. This revealed that the only domain for which the difference was statistically significant was "role-physical", highlighting the reduced quality of life in this domain in individuals with SCA-10.
When ANOVA was used, and the linear regression model was fitted using least squares, statistically significant relationships at the 5% level were found between the variables of disease duration, Berg, SARA, FIM, Lawton IADL and the SF-36 domain "physical functioning". The coefficients of the straight lines identified in the linear regression in the Figure  indicate whether the relationship between the variables is one of direct proportionality (rising line) or inverse proportionality ( falling line).
Linear regression between "physical functioning" and "disease duration" (A) and between "physical functioning" and the SARA score (C) at the 5% significance level resulted in fitted regression functions y = -1.6266x + 64.683 and y = -2.867x + 80.208 (Figure). These indicate an inversely proportional correlation, i.e., an increase in the SARA score and an increase in disease duration are associated with a lower score in the "physical functioning" domain. A one-point increase in the SARA score and a one-year increase in disease duration can be expected to correspond, respectively, to 2.86-point and 1.62-point decreases in the score in the "physical functioning" domain (Table 3).
Reduced quality of life in the "physical functioning" domain can be expected as disease duration (A) and severity of ataxia (C) increase. The p values used are shown in the Figure. Linear regression between "physical functioning" and the "Berg" (B), "FIM" (E) and "Lawton" (D) scores at the 5% significance level resulted in regression functions y = 1.3767x -17.738, y = 8.8807x -125.54 and y = 4.4935x -488.89 (Figure). These indicate a positive correlation and that the variables are directly proportional, i.e., a decrease in the Berg, FIM and Lawton IADL score will lead to a lower score in the "physical functioning" domain. Decreases of 1.37, 8.88 and 4.49 in the "physical functioning" score can be expected for one-point decreases in the Berg, Lawton IADL and FIM scores, respectively (Table 3).
A reduction in QoL in the domain "physical functioning" is expected as a fall risk (B), functional dependence for ADL (E) and dependence for IADL (D) increase. The p values used are shown in the Figure. The correlation coefficient (r) varies between 1 and -1. The closer r is to these values, the stronger the correlation between the variables: r = 0.10 to 0.30 corresponds to a weak correlation, r = 0.40 to 0.60 to a moderate correlation and r = 0.70 to 1 to a strong correlation. The same classification is used for negative values 16 . Pearson's correlation revealed statistically significant results at the 5% level. Moderate correlations were identified between "physical functioning" and the disease duration, SARA score, Berg score and Lawton IADL score and a strong correlation between "physical functioning" and the FIM. The r values are shown in the Figure.

DISCUSSION
This study assessed the QoL of individuals with SCA10 and identified impairment in the "role-physical" and "physical functioning" domains. Impairment in the latter was associated with longer disease duration, functional dependence, ataxia severity and fall risk. Although an understanding of how disease affects QoL is important for clinical management, few authors have investigated this relationship in individuals with spinocerebellar ataxias 17, 18,19 . The concept of QoL covers physical and psychological health, level of independence, social relationships, environment and beliefs 20 . According to Azevedo et al. 21 , individuals with chronic nervous system diseases have a lower self-perception of QoL than those with diseases in other systems.
Castilhos et al. 5 showed that Brazil has a considerable number of families with spinocerebellar ataxias and included 359 in a study, most of them in southern Brazil. The Table 3. SCA10 -Interpretation of the results of the linear regression analysis with least squares method.

Variable Interpretation
Disease duration vs.  present study assessed individuals with SCA10 in southern Brazil, where this type of SCA is the second most prevalent after SCA3.
The mean age of symptom onset in the study sample was 33 years, corroborating the literature, in which a mean of 33-35 years has been reported 5,22 , and the individuals were young when  they began to gradually lose their autonomy. Two of them had had the disease for over 30 years. The different series of SCA10 patients in the literature show that the mean disease duration is very variable, with a mean survival of 13-20 years 5,22,23 .
Although it is a generic instrument, the SF-36 is highly reliable, providing valid data covering the patient' s physical and mental health 24 . It has also been used by other authors for individuals with ataxia 19,25 . The "role-physical" domain focuses on how difficulties in performing tasks at work (reflected in, for example, fewer hours spent working) and ADL affect the QoL 26 .
The "physical functioning" domain considers the limitations on performing physical activities of different intensities 26 .
The lower scores in the "role-physical" and "physical functioning" domains in this sample corroborate the findings of Sánchez-López et al. 19 . Unlike the present study, which assessed individuals with SCA10, the study by Sánchez-López et al. 19 assessed individuals with different types of ataxias, allowing the authors to observe the impact of the disease in the "general health", "vitality", "bodily pain" and "social functioning" domains. Tai et al. 23 observed a reduction in the QoL of individuals with Friedreich' s ataxia in all domains of the SF-36. The lower scores in the "role-physical" domain in the present study may be related to the fact that most of the participants were not working, as only two individuals kept their jobs. Work is one of the main activities carried out by adults and, in addition to providing an income, it is associated with satisfaction, social interaction, coexistence and personal fulfilment 27 .
The lower scores in the "physical functioning" domain in the present study were related to longer disease duration, corroborating the findings of Tai et al. 23 . As the symptoms of ataxia progress, the need for help from others increases. In this sample, this was evidenced by the FIM scale for stairs, expression and social interaction. No relationship was observed between the FIM cognitive aspects and the SF-36 mental component, probably because of the sample size, as both the ability to interact socially and to express oneself are considered in the "social role" domain of the SF-36 and are important skills for the individual' s well-being.
Dependence in more complex activities of daily living is associated with a lower QoL, considering the "physical functioning" domain. The individuals in this sample were independent for a single item on the Lawton IADL scale (responsibility for their own medications), while activities such as housekeeping, shopping and traveling required help from others. Moreira et al. 25 , when assessing the QoL of individuals with Parkinson's disease, associated dependence on others for ADL with a worse QoL.
In the present study, a greater severity of ataxia was associated with worse QoL in the "physical functioning " domain, corroborating the findings of Graves et al. 4 . These authors assessed individuals with episodic ataxia type 1 and observed that those who had persistent ataxia also had worse ataxia severity and a longer disease duration, as well as difficulties in performing ADL and an increase in falls. In addition, all the SF-36 domains were below normal, particularly the "role-emotional", although the QoL was generally lower in the physical component.
In our study, there was a trend toward a correlation between QoL and the "role-emotional" domain. Ten individuals had a final score below the average of 50 in the "role-emotional" domain, indicating a low QoL, which may have been a result of the mild mood disorder (depression) observed in individuals with SCA10 28 . The fact that the other five patients had scores of 80 in this domain may explain the lack of correlation with QoL in this sample.
The mean score on the Berg Balance Scale indicated a low fall risk in this sample. Nevertheless, this score was associated with a worse QoL in the "physical functioning" domain. A study by Filippin et al. 17 assessed balance in individuals with SCA3 and the relationship to all SF-36 domains. They observed that a worse QoL in the "physical functioning" domain was associated with a worse static and dynamic balance and even with a mean Berg score of 41.85, showed a correlation with fall risk. According to Aizawa et al. 18 , signs and symptoms other than ataxia (such as ataxia severity and greater functional impairment) can influence static and dynamic balance.
In a study of 16 individuals with SCA2, Amarante et al. 29 described how balance, fall risk and ataxia severity evolved over one year of disease. In our study, we found that for every year of disease there was a 1.62-point decrease in the "physical functioning" score, reflecting an increase in disease severity, functional dependence and fall risk.
The small population sample in this study prevented us separating the groups by age, gender, disease duration, age of onset of symptoms, employment status and presence of a caregiver, which may have enabled us to identify significant changes in other SF-36 domains. Further studies with larger population samples are therefore required to confirm our findings.
In conclusion, reduced QoL in individuals with SCA10 in this cohort was associated with the "role-physical" and "physical functioning" domains. Self-perception of QoL was influenced by disease severity, disease duration, balance impairment and level of dependence on others for ADL and IADL.