Health-Related Quality of Life in Multiple System Atrophy using EQ-5D-5L: A Large Cross-Sectional Study in China

Background Multiple system atrophy (MSA) is a rare neurodegenerative disease, featuring autonomic failure plus parkinsonism and/or cerebellar ataxia. These symptoms impact the health-related quality of life (HRQoL) of MSA. Objective We aimed to evaluate the HRQoL of MSA with a preference-based instrument, the ve-level EuroQol ve-dimensions questionnaire (EQ-5D-5L), for the rst time. Methods EQ-5D-5L was used to evaluate the HRQoL. The result of HRQoL was displayed as heath utility index and visual analog scale (EQ-VAS) score. Specic scales were used to measure the disease severity, cognition, frontal lobe function, anxiety, depression, fatigue, and sleep disorders. The forward logistic model was used to explore the determinants of HRQoL in MSA. Results A total of 205 patients with cerebellar variant (MSA-C, 53.9%) and 175 patients with parkinsonian variant (MSA-P, 46.1%) patients were included in the study. The mean scores of the health utility index and EQ-VAS were 0.558 and 59.5, respectively. Mobility was reported by the largest proportion (92.1%) of MSA patients, followed by usual activities (88.7%), self-care (81.3%), anxiety/depression (72.1%), and pain/discomfort (53.9%). The determinants of the lower health utility index in MSA were female sex, greater total Unied Multiple System Atrophy Rating Scale (UMSARS) scores, fatigue, and Parkinson's disease-related sleep problems (PD-SP). Lower EQ-VAS score was associated with greater total UMSARS scores, fatigue, PD-SP, and anxiety symptom. MSA-P patients reported more frequent problems in pain/discomfort than MSA-C patients, while MSA-C patients reported more problems in mobility than MSA-P patients. Conclusion of MSA, fatigue, PD-SP and anxiety were determinants for poor HRQoL. Our research provides important information to improve the health status of patients with MSA. clinical data were collected by experienced neurologists during a face-to-face interview. The demographic characteristics including age, sex, education, disease duration, and age of onset were collected. The disease severity was measured by the Unied Multiple System Atrophy Rating Scale (UMSARS). [12] The total UMSARS score was the sum of parts I and II. The orthostatic hypotension (OH) was dened as the drop of systolic blood pressure ≥ 30mm Hg and/or diastolic blood pressure ≥ 15 mm Hg. The global cognitive function was evaluated using the Montreal Cognitive Assessment (MoCA). [13] The cognitive impairment was dened as MoCA < 19 for individuals with no more than 6 years of education, MoCA < 22 for individuals with 7 to 12 years of education, and MoCA < 24 for individuals with more than 12 years of education. [14] Frontal lobe dysfunction was dened as the frontal assessment battery (FAB) score < 16. [15] The depressive and anxiety emotion were respectively dened as the Hamilton Depression Rating Scale (HAMD, 24 items) score ≥ 8 [16] and the Hamilton Anxiety Rating Scale (HAMA) score ≥ 6. [17] Fatigue was dened as the mean score of Fatigue Severity Scale (FSS) ≥ 4. [18] Parkinson's disease-related sleep problems (PD-SP) were dened as the Parkinson's Disease Sleep Scale 2nd version (PDSS-2) score ≥ 18. [19] Excessive daytime sleepiness (EDS) dened the Epworth Sleepiness Scale (ESS) score ≥ 10. [20] Rapid eye movement sleep behavioral disorder (RBD) was dened as the Rapid Eye Movement Sleep Behavior Disorder Screening Questionnaire (RBDSQ) score ≥ 5. [21] The HRQoL was evaluated using EQ-5D-5L. This scale was developed by the EuroQol Group and was used in evaluating healthy populations and patients with different diseases. [8] The EQ-5D-5L was designed for self-completion. It contains a questionnaire with ve dimensions and a vertical visual analogue scale (EQ-VAS). The ve dimensions are Mobility, Self-Care, Usual Activities, Pain/Discomfort, and Anxiety/Depression. Each dimension has ve levels ranging from "no problems", "slight-moderate-severe" problems, to "unable to do". Descriptive levels of each dimension were dichotomized to "no problems"(level one) and "problems" (level two to ve). The descriptive scores of the ve dimensions were converted to health utility values according to the EQ-5D-5L value set for China. [22] The health utility values are anchored at 0 (death) and 1 (perfect health) and can be used to assess QALYs. The results of EQ-VAS were self-rated scores marked from 0 to 100, reecting the patient's opinion on their overall health on the day of the interview.


Introduction
Multiple system atrophy (MSA) is a rare, rapidly progressive neurodegenerative disease featured with autonomic failure, parkinsonism, and/or cerebellar ataxia. To date, the etiology of MSA is still uncertain and no cure for MSA is available. MSA can be subclassi ed into a cerebellar variant (MSA-C) and a parkinsonian variant (MSA-P) according to the predominant motor symptom.
[1] Except for experiencing severe disability in a short duration, many patients undergo various non-motor symptoms including urinary incontinence, dysarthria, depression, anxiety, sleep disorders, and so on. In addition, non-motor symptoms such as depression, anxiety, and autonomic de cit have a negative impact on the patients' lives and lead to a heavy disease burden. [2][3][4][5] Patients with MSA were reported to have a poorer health-related quality of life (HRQoL) than healthy controls and patients with Parkinson's disease (PD). [5,6] Compared to the common scale such as Parkinson's disease Questionnaire-39 item version (PDQ-39) which was used to measure the quality of life (QoL) of PD patients, the common symptoms of autonomic and cerebellar in MSA were not adequately re ected by PDQ-39. [7] The ve-level EuroQol ve-dimensions questionnaire (EQ-5D-5L), a preferencebased instrument developed by EuroQol groups, is one of the most frequently used tools to evaluate HRQoL worldwide.
[8] It can be used to estimate quality-adjusted life-years (QALYs), helping public health specialists and epidemiologists to plan interventions to improve patient's health status. The EQ-5D-5L was developed from the 3-level version of EQ-5D (EQ-5D-3L) and was proved to have some advantages over EQ-5D-3L. [9] However, there were only a few studies focused on the HRQoL of MSA by using EQ-5D-3L. [3][4][5]10] These studies found that the most frequently affected dimensions in MSA were mobility, self-care, and usual activities. [3][4][5]10] Besides the severity of the disease [4][5][6]11], some studies found that depression was the determinant of HRQoL of MSA. [3,5] No study on QoL of MSA evaluated by using EQ-5D-5L was available. Moreover, there was no study focused on the impact of nonmotor symptoms measuring by comprehensive assessment scales on the QoL evaluated by using EQ-5D-5L in MSA.
Therefore, in the current study, we aimed to ll the gap to evaluate the HRQoL of MSA using EQ-5D-5L. Additionally, we used speci c scales to measure the severity of non-motor symptoms including cognition, frontal lobe dysfunction, depression, anxiety, sleep disorders, and fatigue. We aimed to explore the determinants of HRQoL in MSA and the differences between MSA-C and MSA-P.

Patients and Clinical features evaluation
Consecutive patients were recruited from the Department of Neurology, West China Hospital, Sichuan University from March 2018 to July 2021. All the patients reached the diagnostic criteria of probable MSA through comprehensive medical history reviewing and physical examination. [1] The exclusion criteria were: 1) magnetic resonance imaging scan, spinal cerebellar ataxia genetic tests (SCA1, 2, 3, 6, 7), or blood test indicating a diagnosis of other neurological diseases; 2) did not nish the whole interview because of dysarthria, weakness, or other reasons. This study was approved by the Ethics Committee of West China Hospital of Sichuan University and informed consent forms were signed by all subjects.
The clinical data were collected by experienced neurologists during a face-to-face interview. The demographic characteristics including age, sex, education, disease duration, and age of onset were collected. The disease severity was measured by the Uni ed Multiple System Atrophy Rating Scale (UMSARS). [12] The total UMSARS score was the sum of parts I and II. The orthostatic hypotension (OH) was de ned as the drop of systolic blood pressure ≥ 30mm Hg and/or diastolic blood pressure ≥ 15 mm Hg. The global cognitive function was evaluated using the Montreal Cognitive Assessment (MoCA). [13] The cognitive impairment was de ned as MoCA < 19 for individuals with no more than 6 years of education, MoCA < 22 for individuals with 7 to 12 years of education, and MoCA < 24 for individuals with more than 12 years of education. [14] Frontal lobe dysfunction was de ned as the frontal assessment battery (FAB) score < 16. [15] The depressive and anxiety emotion were respectively de ned as the Hamilton

Assessment Of Hrqol
The HRQoL was evaluated using EQ-5D-5L. This scale was developed by the EuroQol Group and was used in evaluating healthy populations and patients with different diseases.
[8] The EQ-5D-5L was designed for self-completion. It contains a questionnaire with ve dimensions and a vertical visual analogue scale (EQ-VAS). The ve dimensions are Mobility, Self-Care, Usual Activities, Pain/Discomfort, and Anxiety/Depression. Each dimension has ve levels ranging from "no problems", "slight-moderate-severe" problems, to "unable to do". Descriptive levels of each dimension were dichotomized to "no problems"(level one) and "problems" (level two to ve). The descriptive scores of the ve dimensions were converted to health utility values according to the EQ-5D-5L value set for China. [22] The health utility values are anchored at 0 (death) and 1 (perfect health) and can be used to assess QALYs.
The results of EQ-VAS were self-rated scores marked from 0 to 100, re ecting the patient's opinion on their overall health on the day of the interview.

Statistical analysis
First, the Kruskal-Wallis test was performed as the normality test. Mann-Whitney test was conducted to compare the health utility values and EQ-VAS scores of patients in different subgroups regarding sex, subtype, frontal lobe dysfunction, cognitive impairment, depressive emotion, anxiety emotion, EDS, RBD, PD-SP, and fatigue. T-test or Chi-square tests were conducted to compare demographic characteristics between MSA-C and MSA-P for continuous and categorical variables. Spearman's correlation analysis was conducted to explore relationships between EQ-5D-5L scores and clinical symptoms. Forward stepwise multiple logistic regression analysis was used to explore the potential determinants of HRQoL in MSA and two subtypes. Patients were divided into two groups according to the median of the health utility value and EQ-VAS score. The health utility value or EQ-VAS score was used as the dependent variable. The demographic and clinical variables were included in the model as independent variables. A level of P < 0.05 was set for a variable entered into the model and P > 0.10 for removal from the model. Statistically signi cant was considered as P < 0.05. All analyses were performed using SPSS 22.0 (SPSS, Inc., Chicago, IL, USA).

Results
Clinical characteristics of MSA and two subtypes Finally, 205 MSA-C (53.9%) and 175 MSA-P (46.1%) were included in the study. The clinical characteristics of patients are displayed in Table 1. The average age and disease duration were 60.3 years old and 2.6 years respectively. There were 212 male patients (55.8%) and 168 females (44.2%). Patients with MSA-P had older age and age of onset and higher scores of PDSS-2 and ESS than patients with MSA-C. Patients with MSA-C had a higher proportion of OH and higher scores of RBDSQ than patients with MSA-P.
There were no signi cant differences in other clinical features between the two subtypes. values and EQ-VAS were observed between patients with MSA-C and MSA-P. The frequency of ve levels in ve dimensions of EQ-5D-5L in total MSA patients and two subtypes of patients are shown in Figure 1. For each dimension, mobility was reported by 92.1% of MSA patients, followed by usual activities (88.7%), self-care (81.3%), anxiety/depression (72.1%), and pain/discomfort (53.9%). The frequencies of reported problems of each dimension in MSA-C and MSA-P are displayed in Figure 2. The most common problem was mobility in both MSA-C and MSA-P patients, followed by usual activities, self-care, anxiety/depression, and pain/discomfort. The frequency of reported problems of pain/discomfort in patients with MSA-P was signi cantly higher than that in patients with MSA-C while the frequency of reported problems of mobility in patients with MSA-P was signi cantly lower than that in patients with MSA-C.  Related factors of EQ-5D-5L scores in total MSA patients and patients with two subtypes Spearman's analysis showed a moderate correlation between the EQ-VAS score and scores of total UMSARS, FSS, PDSS-2, HAMD, and HAMA in total MSA patients. The total UMSARS score was strongly correlated with health utility value. Scores for FAB, MoCA, FSS, PDSS-2, HAMD, and HAMA had weak to moderate correlations with health utility value in total MSA patients. (Table 3) Multivariate logistic models showed the determinants of health utility value and EQ-VAS score in total MSA and two subtypes of patients. (Table 4) In the total MSA patients, female sex, fatigue, PD-SP, and greater total UMSARS score were the determinants for the lower health utility values. Fatigue, PD-SP, anxiety emotion, and greater total UMSARS score were the determinants for the lower EQ-VAS scores. There were some differences in the determinants for health utility value and EQ-VAS score between patients with MSA-C and MSA-P. For example, cognitive impairment was the determinant for lower health utility value in patients with MSA-P but not in patients with MSA-C. And frontal lobe dysfunction was the determinant for lower EQ-VAS score in patients with MSA-P but not in those with MSA-C. More details are displayed in Table 4. Abbreviations: MSA: multiple system atrophy. EQ-VAS: visual analog scale. MSA-P: MSA with predominately parkinsonism. MSA-C: MSA with predominately cerebellar ataxia. OR: odds ratio. 95%CI: 95% con dence interval. UMSARS: the Uni ed Multiple System Atrophy Rating Scale. PD-SP: Parkinson's disease-related sleep problems. FLD: frontal lobe dysfunction. CI: cognitive impairment.

Discussion
Our research measured the HRQoL of MSA using EQ-5D-5L, providing important data on the QoL of Chinese MSA patients. We found that patients with MSA in a Chinese population had poor HRQoL. Mobility was the most commonly reported problem and pain/discomfort was the least common problem, which was consistent with the previous study using EQ-5D-3L. [4] Besides disease severity, sex, fatigue, and PD-SP were found to be the determinants for health utility value of MSA, and anxiety symptom, fatigue, and PD-SP were the determinants for EQ-VAS of MSA. Additionally, we found there were some differences in the determinants for HRQoL of MSA between MSA-C and MSA-P patients.
The mean health utility index (0.558) and the mean EQ-VAS score (59.5) of patients with MSA were lower than those of the Chinese healthy population (0.943, 82.9) and those of patients with coronary heart disease and stroke history (0.850, 61.0). [23,24] The mean EQ-VAS score of Chinese patients with MSA (59.5) was higher than the results of the European study (44.5 and 36.9). [4,5] This may be explained by the shorter disease duration and mild disease severity of patients enrolled in our study. Consistent with the previous study, we also found that the total UMSARS score was the determinant of health utility value and VAS scores in the MSA and two subtypes. [4] We found that mobility was the most commonly reported problem and pain/discomfort was the least common problem, which was consistent with previous studies. [4,5] Furthermore, we found that a higher frequency of problem of mobility was reported in patients with MSA-C than patients with MSA-P, which was consistent with the previous study assessed by EQ-5D-3L. [4] However, we found there was no difference in the mean scores of motor examination and global disability scale of UMSARS between the two subtypes. Our nding suggested that UMSARS has limited ability to re ect mobility in patients with MSA-C. In addition, we found that patients with MSA-P reported more pain/discomfort than patients with MSA-C, which was consistent with the previous study assessed with EQ-5D-3L. [5] This may be due to the severer damage in the basal ganglia in MSA-P than in MSA-C. [25] Depression was reported to be a determinant of the QoL of MSA. [3,4,26,27] In our study, the depressive emotion was related to the EQ-VAS score of MSA-C. We also found that anxiety played an important role in the decreasing EQ-VAS score of MSA. [3] Attempts to treat these symptoms are important to improve the HRQoL of patients with MSA. Similar to the result of the previous study, we found that female sex was a negative determinant of health utility values of MSA. [4] Additionally, females were more likely to initially manifest motor symptoms and recurrent falls within three years.
[28] More researches should be made to understand the gender difference in MSA.
Fatigue is common in several neurological diseases including PD and multiple sclerosis. [29] There were 29-82% of MSA patients who suffered from fatigue.
[30], which was signi cantly higher than that of patients with PD and healthy control.
[31] In the current study, we found that fatigue was negatively correlated with health utility value and EQ-VAS score of MSA, which was consistent with our previous study evaluating QoL of MSA patients using PDQ-39. [11] The pathological mechanisms of fatigue are still uncertain.
A study found that fatigue in MSA was related to the decrease of the 5-hydroxytryptamine receptor in the raphe nuclei and brain stem.
[31] We should pay more attention on fatigue when managing patients with MSA.
In our previous study, sleep-related disorders including PD-SP, RBD, and EDS were found to be associated with higher disease severity in MSA.
[32] In the current study, PD-SP but not RBD and EDS was found to be a determinant of the health utility value and EQ-VAS score of MSA. Since PD-SP includes motor symptoms at night, PD symptoms at night, and disturbed sleep, the etiology of PD-SP in MSA is complex and is still under research. Management of PD-SP in MSA is necessary.
Recent studies found that a varying degree of cognitive impairment was common in MSA assessed by comprehensive neuropsychological testing. Additionally, we found that frontal lobe dysfunction was the determinant of EQ-VAS score in MSA-P but not in MSA-C, which was similar to our previous nding that frontal lobe dysfunction was the determinant of HRQoL assessed by PDQ-39 in MSA. [26] Further study should pay attention to the cognitive function in patients with MSA.
This research had several limitations. First, this cross-sectional research explored the features and determinants of the EQ-5D-5L score in MSA. Further study should concentrate on the longitudinal change of the EQ-5D-5L scores of MSA. Second, the diagnosis of MSA was based on the clinical features but not the autopsy. Further study should build autopsy con rmed cohort to study the HRQoL of MSA.

Conclusion
The current research found that MSA patients had poor HRQoL and EQ-5D-5L was a validating tool to evaluate HRQoL in MSA. The determinants of lower health utility value were female sex, greater scores of total UMSARS, fatigue, and PD-SP while that of lower EQ-VAS scores were greater scores of total UMSARS, fatigue, PD-SP, and anxiety symptom. In addition, the determinants of EQ-5D-5L were different between patients with MSA-C and MSA-P. This research provided important information on the HRQoL and potential determinants of MSA, helping developing intervention strategies to improve their health status.