CAN THE SYMBOL DIGIT MODALITIES TEST AND THE
NINE HOLE PEG TEST PREDICT THALAMUS ATROP HY
AND MOTOR CORTEX THINNING IN SUBJECTS WITH
EARLY RRMS?
Thesis | February 2020 – July 2020 | Bachelor Psychobiology
Written by: Suzanne R de Jong (Student ID: 11673222) Supervisors: Merlin M Weeda, MSc. and Hugo Vrenken, PhD Submission date: July 19th, 2020
1 Abstract
Background The most common multiple sclerosis (MS) phenotype is relapsing remitting MS (RRMS), which is characterized by inflammatory attacks and remissions. The inflammatory attacks (i.e. relapses) can be perceived with magnetic resonance images (MRI). The relapses can lead to symptoms, including difficulty with physical functioning of the extremities, and to cognitive decline. Multiple tests are existent to assess the physical deterioration and cognitive impairment. Physical deterioration is often measured with the Nine Hole Peg Test (9-HPT) and cognitive impairment with the Symbol Digit Modalities Test (SDMT). Correlations have been found between physical decline and grey matter fraction and between cognitive impairment and thalamus atrophy. Moreover, strong relationships have been found between MRI measures and SDMT score and 9-HPT score.
Objective To investigate the longitudinal relation between thalamus volume and SDMT score and between motor cortex thickness and the 9-HPT score in subjects with early RRMS.
Methods A total of 45 subjects with early RRMS underwent two MRI examinations with a 2-year interval, including T1 and FLAIR weighted imaging. At both time points, clinical and neuropsychological performance was evaluated. Changes over time for thalamus volume, motor cortex thickness, SDMT score and 9-HPT scores were calculated as annual percentage change over two years (2-APC) with regard to baseline. Forward regression was used to find predictors for these 2-APCs. Results Over the follow-up time (2.02±0.10 years), significant atrophy could be seen for total thalamus volume (p<0.001), as well as for average motor cortex thickness (p=0.011). 2-APC of total thalamus volume could be significantly predicted by disease duration (F(1, 29)=9.002 with adjusted R2=0.211).
No predictors were found for 2-APC of average motor cortex thickness.
Significant differences over time were found for SDMT score (p<0.001) and 9-HPT scores for the non-dominant hand (p=0.022), but not for 9-HPT scores for the non-dominant hand. 2-APC of SDMT score could be predicted by educational level, AIS score and CIS score (F(3, 27)= 3.738 with adjusted R2=0.215).
2-APC of 9-HPT score for the dominant hand could be predicted by gender, EDSS score, CIS score and age (F(4, 26)= 2.918 with adjusted R2=0.204) and 2-APC of 9-HPT score for the non-dominant hand
could be predicted by HADS-A score, thickness of the motor cortex, educational level, age, EDSS score and ARR (F(7, 23)= 3.534 with adjusted R2=0.372)
Conclusion No predictive value of SDMT score and 9-HPT score could be found for thalamus atrophy and motor cortex atrophy, respectively. Furthermore, thalamus volume could not predict SDMT performance over time. However, the motor cortex thickness could predict 9-HPT performance over time of the non-dominant hand.
2 1. Introduction
Multiple sclerosis (MS) is an autoimmune disease affecting the central nervous system (CNS) (Cercignani and Wheeler‐Kingshott, 2017, Stadelmann and Brück, 2008). MS is characterized by recurrent episodes of inflammation, which cause subsequent damage of the underlying axons and demyelination, leading to brain atrophy (Miller et al., 2002). It is unknown how MS is caused, although it appears that both environmental and genetic factors influence the susceptibility to develop MS (Bishop and Rumrill, 2015). Across the world, the prevalence is approximately 33 per 100,000 people, although a great variance between countries exists (Vidal-Jordana and Montalban, 2017).
Previous studies found that the atrophy rate in MS is significantly higher compared to the atrophy rate in healthy controls caused by normal ageing (Chard et al., 2002). The atrophy perceived in MS affects both white matter (WM) and grey matter (GM) (Fisher et al., 2008), which can be measured by acquiring magnetic resonance (MR) images (Miller et al., 2002). Furthermore, MRI of the CNS is used to diagnose MS, to support the clinical grounds (Polman et al., 2011). Patients can be diagnosed with several types of MS, but the majority of patients (85%) are being diagnosed with the relapsing remitting MS (RRMS) phenotype (Vidal-Jordana and Montalban, 2017). RRMS is characterized by inflammatory attacks, or relapses, after which remissions follow, which are periods of partial or complete recovery (National Multiple Sclerosis Society).
Dependent on the size and location of these relapses in the CNS, different types of symptoms can be experienced (Huang et al., 2017). For instance, lesions in the spinal cord, brain stem and cerebrum can induce problems regarding physical functioning of the arms and the legs (Bennett et al., 2002). The most common symptoms include difficulty with mobility, decreased coordination and strength, tremor and tiredness. Physical deterioration in people with MS can be measured with the Nine Hole Peg Test (9-HPT) (Mathiowetz et al., 1985). The 9-HPT is applied in 63% of the published studies, which makes it the most frequently used upper limb measure in MS studies. The purpose of the 9-HPT is to analyze the dexterity in people with MS (Feys et al., 2017), as well as motor function of the upper extremities (Liepert et al., 2007).
Besides the physical symptoms, cognition is often negatively affected by MS. It has been learned that the impact of MS on cognition is more common and more significant than previously assumed. Cognitive impairment can lead to problems with the efficiency and speed of information processing, executive functioning and episodic memory (Bishop and Rumrill, 2015). Cognitive impairment in MS can be assessed with the Symbol Digit Modalities Test (SDMT) (Smith, 1973), which is a widely used tool to assess information processing speed in MS. It is said that the SDMT is the most sensitive test for the detection of cognitive impairment over time. Moreover, the SDMT is especially suitable for MS-related cognitive deterioration over time (Amato et al., 2010). It is known that subjects SDMT scores can be influenced by age, gender, education and the version of the SDMT (Boringa et al., 2001).
3 Moreover, factors such as mental state (depression) and fatigue can influence the information processing as well (Arnett et al., 1999, Diamond et al., 2008).
Previous research has shown that physical decline, as measured with the 9-HPT, correlates with grey matter fraction (Fisher et al., 2008). Moreover, a significant decrease in primary motor cortex thickness has been observed in RRMS patients (Bergsland et al., 2015) and an association has been found between changes in motor cortex and performance on the 9-HPT (Liepert et al., 2001). Therefore, this study aims to investigate the longitudinal relationship between motor cortex thickness and 9-HPT scores in early RRMS. This will be investigated by determining the predictive value of the 9-HPT score on the motor cortex thinning, and the predictive value of the motor cortex thickness on the 9-HPT performance over time. It is hypothesized that the 9-HPT score can predict the motor cortex thinning, but it is not expected that the motor cortex thickness can predict the 9-HPT performance over time.
Furthermore, cognitive impairment, which can be measured using the SDMT, has been shown to correlate with deep grey matter (DGM) atrophy, where the thalamus is the structure most often affected (Schoonheim et al., 2012). The thalamus is a brain area which coordinates a lot of cognitive functions (Houtchens et al., 2007) and is believed to be very important for efficient information processing (Batista et al., 2012). Moreover, a very strong correlation has been found between performance on the SDMT and the thalamus volume (Batista et al., 2012, Bisecco et al., 2018). Therefore, the second aim of this study is to investigate the longitudinal relationship between thalamus volume and SDMT scores in early RRMS. This will be investigated by determining the predictive value of the SDMT score on thalamus atrophy, and the predictive value of the thalamus volume on the SDMT performance over time. It is hypothesized that the SDMT score can predict thalamus atrophy. However, no predictive value of the thalamus volume on the SDMT performance over time is expected.
4 2. Methods
2.1 Subjects
Forty early RRMS patients and fifteen age and sex matched healthy controls were included into this longitudinal study. Exact details of the inclusion and exclusion criteria can be found elsewhere (Weeda et al., 2019). In summary, patients were included when they were above 18 years of age, when they had clinical definite RRMS (Polman et al., 2011) for a maximum of five years, with minimal disability (i.e. Expanded Disability Status Scale (EDSS) score below 5.0). Patients were allowed to use first-line disease modifying treatment (DMT), but steroid use within 3 months before a visit was not allowed. Over three years, subjects visited the research center three times, with 1 year ± 3 months intervals. Each visit consisted of an interview, physical evaluation, neuropsychological evaluation, extensive MRI examination and questionnaires. An overview of the study design is depicted in Figure 1. This study was approved by the local institutional medical ethics committee and written informed consent was obtained from all the subjects, according to the Declaration of Helsinki.
Figure 1. An overview of the study for baseline, first follow-up and second follow-up including the MRI
5 2.2 Physical and neuropsychological evaluation
The demographic information (age, gender and educational level (Verhage, 1964)) and the history of the MS disease (the date and sort of the first symptoms, the date of diagnosis, relapse and treatment information and the medical history) were reported at baseline. In order to evaluate the cognitive functions, the SDMT (Smith, 1973), from the Brief Repeatable Battery of Neuropsychological Tests (BRB-N) (Buschke and Fuld, 1974), was administered to measure visual precision search, executive functions and attention. When practicing the SDMT, subjects are presented with a table which combines the digits 1-9 with nine different symbols. The test itself contains rows with only the symbols. Subjects have to report, verbally, the corresponding number. The test starts with a practice trial, which contains 10 symbols. After completing the practice trial, it is determined how many correct responses a subject can make within 90 seconds (Benedict et al., 2017). In order to discard any effect caused by different versions of the SDMT (Boringa et al., 2001), all results of the first follow-up are excluded from further calculations and statistics in this study. All of the subjects filled in different questionnaires, including the Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983), which consists out of 7 multiple-choice questions for both anxiety and depression (Giordano et al., 2011), the Athens Insomnia Scale (AIS) (Soldatos et al., 2000), and the Checklist Individual Strength (CIS-20) (Vercoulen et al., 1994). Furthermore, patients disability was evaluated with the EDSS (Kurtzke, 1983) and the 9-HPT (Mathiowetz et al., 1985). The latter is part of the Multiple Sclerosis Functional Composite Measure (MSFC) (Fischer et al., 1999). When practicing the 9-HPT, subjects first have to place nine pegs into nine holes and subsequently remove all the nine pegs. The pegs have to be placed and removed one at a time and as quickly as possible (Feys et al., 2017).
Since the 9-HPT was only performed by patients, HCs were not included in further statistics. Moreover, the SDMT scores are highly influenced by age, thus z-scores were calculated (Burggraaff et al., 2017) to correct for age and to normalize for HCs.
2.3 MRI acquisition
At each time point, MRI examination was performed on a 3T whole-body MR scanner (GE Discovery MR750) with an 8-channel phased-array head coil, with no software or hardware updates over the total study time. The protocol included a sagittal 3D T1-weighted fast spoiled gradient echo sequence (FSPGR with TR/TE/TI = 8.2/3.2/450 ms and voxel size 1.0x1.0x1.0 mm) and a sagittal 3D T2-weighted fluid attenuated inversion recovery sequence (FLAIR with TR/TE/TI = 8000/130/2338 ms with voxel size 1.0x1.0x1.2 mm).
2.4 MRI analysis
2.4.1 Lesion segmentation
Lesions were segmented using the deep-learning algorithm nicMSlesions (Valverde et al., 2017, Valverde et al., 2019), which was optimized for the current MRI protocol in a previous study (Weeda et
6 al., 2019). In summary, a full re-training of the 11 layers nicMSlesions neural network was done with the use of 14 manually segmented subjects with MS at baseline. All parameters were set at default, and probability threshold was set at 0.4. Lesion filling was performed using Lesion Automated Preprocessing (LEAP) (Chard et al., 2010).
2.4.2 Brain segmentation
For accurate brain segmentation, the longitudinal pipeline of FreeSurfer 6.0 was used (Dale et al., 1999) (Fischl et al., 1999) (Reuter et al., 2012) on lesion filled T1 weighted images. Volumetric data for whole brain, total WM, total GM, deep GM and left, right and total thalamus were obtained, as well as cortical thickness of total cortical GM, and left, right and mean motor cortex.
2.5 Statistics
Statistical analyses were performed in R studio and results were considered statistically significant upon
p<0.05. To control for baseline differences in brain volume and thickness, as well as SDMT and 9-HPT
scores, changes over time were calculated as annual percentage change over two years (2-APC). All parameters were tested for normality using a Shapiro-Wilk test. Differences between patients and HCs were calculated using an independent t-test (normal distributed data), a Mann Whitney U-test (not-normal distributed data) or a Chi Square test (categorical data). Differences over time and between 2-APCs were checked using a paired t-test (normal distributed data), a Wilcoxon Signed Ranks test (not-normal distributed data) or a Chi Square test (categorical data).
Correlation analysis (Pearson and/or Spearman) was performed to assess possible confounders (age, education, gender, HADS score, HADS anxiety score, HADS depression score, CIS score, AIS score, disease duration, ARR and EDSS score) of SDMT and 9-HPT scores.
Predictors for 2-APC in thalamus volume, motor cortex thickness, SDMT score and 9-HPT score were assessed with forward regression, with age, education, gender, HADS score, HADS anxiety score, HADS depression score, CIS score, AIS score, disease duration, ARR, EDSS score and current treatment (baseline) as independent variables. For 2-APC in thalamus volume, SDMT and SDMT z-score were also added as independent variables, and for 2-APC in motor cortex thickness, 9-HPT z-scores were added. The other way around, thalamus volumes were added for 2-APC in SDMT score, and for 2-APC in 9-HPT scores, motor cortex thicknesses were added as independent variables.
7 3. Results
3.1 Demographics
At baseline the study sample consisted of 40 RRMS patients and 15 HCs. 9 patients and 2 HCs were excluded from the dataset, because of absence at the follow-up. The demographics of patients and HCs are shown in Table 1. No significant differences were found between patients and HCs for gender, age or education.
Disease-related variables in patients, including ARR, follow-up time, current treatment and EDSS score, but also SDMT score, z-scores for the SDMT and 9-HPT score for the dominant hand and the non-dominant hand, are given in Table 2. No significant differences were found over time for ARR (p=0.392), or EDSS score (p=0.501). However, a significant increase in SDMT score was observed over time (p<0.001), with a median 2-APC of 6.5% (interquartile range (IQR) : 0.7-10.3%), whereas a significant decline in SDMT z-score (i.e. normalized for HCs and age) was seen (p<0.001). Moreover, 9-HPT score of the non-dominant hand significantly improved over time (p=0.022), with a median 2-APC of -4.7% (IQR: -8.9–1.4%), although, the improvement for the dominant hand score, with a median 2-APC of -4.5% (IQR: -6.6–4.1%), was not significant (p=0.170).
Table 1. Baseline demographics for patients and healthy controls (HCs).
Patients (n=31) HCs (n=13) Statistics Gender (M/F) (%F) 7/24 (77%) 4/9 (69%) p=0.848
Age (mean ± SD; years) 36.2 ± 7.6 36.5 ± 12.9 p=0.954
Education (median with IQR; Verhage scale) 6 (5-7) 6 (5-7) p=0.554
8
Table 2. Table shows demographics, disease related variables and the performance on the 9-HPT and SDMT in
patients (n= 31). Moreover the percentage change over two years (2-APC) from baseline to follow-up, relative to baseline is shown as median with interquartile range (IQR).
Baseline Follow-up Statistics 2-APC ARR (median with IQR) 0 (0-1) 0 (0-0.5) p=0.392
Follow up time (mean ± SD; years) 2.02 ± 0.10
Current treatment (n, %) p=0.586
None 7 (23%) 9 (29%)
Dimethyl fumarate (Tecfidera) 13 (42%) 10 (32%) Glatiramer acetate (Copaxone) 6 (19%) 7 (23%)
Interferon-B1a (Avonex) 1 (3%) 1 (3%)
Interferon-B1a (Rebif) 3 (10%) 2 (6%)
Teriflunomide (Aubagio) 1 (3%) 2 (6%)
EDSS (median with IQR) 3.0 (2.5-3.5) 3.0 (2.5-3.5) p=0.501
SDMT (mean ± SD) 60.23 ± 12.25 65.61 ± 14.69 p<0.001 6.5% (0.7-10.3) SDMT z-scores (mean ± SD) 0.50 ± 1.16 -0.09 ± 1.23 p<0.001 9-HPT (mean ± SD) Dominant hand 19.07 ± 2.23 18.89 ± 3.14 p=0.170 -4.5% (-6.6–4.1) Non-dominant hand 20.55 ± 2.88 19.99 ± 3.56 p=0.022 -4.7% (-8.9–1.4) 3.2 Brain volumes
Patients showed a significant decrease in volume over the follow-up time for whole brain, GM and deep GM (all p<0.001). This was also seen for WM volume (p=0.003) and for total, left and right thalamus volume (p<0.001, p=0.004 and p<0.001, respectively). Furthermore, a significant increase over time was found for lesion volume (p=0.038).
Cortical thickness, as well as average motor cortex and left motor cortex thickness also significantly decreased over the follow-up time (p=0.003, p=0.011 and p=0.012, respectively). No significant decrease was found in right motor cortex thickness (p=0.068) (Table 3).
Expressing the differences as a percentage change over two years (2-APC), no significant differences were found between neither the right and left thalamus (p=0.861) nor the right and left motor cortex (p=0.455).
9
Table 3. Patients brain volumes, lesion volume and cortical thickness at baseline and the follow-up, shown as
mean with standard deviation. Moreover the percentage change over two years (2-APC) from baseline to follow-up, relative to baseline is shown as mean with standard deviation .
Baseline Follow-up Statistics 2-APC Whole brain volume; mL 1126 ± 97 1116 ± 98 p<0.001 -0.8 ± 0.9 %
White matter volume; mL 432 ± 51 430 ± 52 p=0.003 -0.6 ± 0.9 %
Grey matter volume; mL 668 ± 49 660 ± 49 p<0.001 -1.1 ± 1.1 % Deep GM volume; mL 56.99 ± 5.67 56.31 ± 5.76 p<0.001 -1.2 ± 1.2 % Thalamus volume; mL Total 13.81 ± 1.65 13.60 ± 1.72 p<0.001 -1.6 ± 2.2 % Right 6.72 ± 0.83 6.63 ± 0.84 p<0.001 -1.3 ± 1.9 % Left 7.09 ± 0.86 6.97 ± 0.92 p=0.004 -1.8 ± 3.0 % Lesion volume a; mL 4.25 ± 3.39 4.79 ± 4.04 p=0.038 6.3 ± 32.7 % Cortical thickness; mm 2.55 ± 0.10 2.54 ± 0.10 p=0.003 -0.5 ± 0.9 %
Motor cortex thickness; mm
Average 2.63 ± 0.12 2.61 ± 0.12 p=0.011 -0.7 ± 1.5 %
Right 2.60 ± 0.13 2.59 ± 0.13 p=0.068 -0.6 ± 1.6 %
Left 2.66 ± 0.12 2.64 ± 0.13 p=0.012 -0.9 ± 2.0 %
a Annual percentage change calculated over 29 patients, since 2 patients had no lesion volume at baseline
3.3 Demographic correlates of SDMT and 9HPT
In order to define the possible confounding factors for SDMT and 9-HPT score, several demographics (gender, age, education, disease duration, ARR, EDSS, HADS total score, HADS anxiety score, HADS depression score, CIS score and AIS score) were tested for their correlations with SDMT and 9-HPT (dominant and non-dominant hand) at either baseline or follow-up.
For the SDMT score, age was a significant correlate at follow-up (r=-0.476, p=0.021), but not at baseline.
The 9-HPT scores of both the dominant and non-dominant hand correlated significantly with age at both baseline (r=0.544, p=0.002 and r=0.448, p=0.011, respectively) and at follow-up (rs=0.523, p=0.003 and rs=0.415, p=0.020, respectively). Moreover, EDSS score was a significant correlate for 9-HPT score. For the non-dominant hand this correlation was seen at both baseline and at follow-up (rs=0.724,
p<0.001 and rs=0.441, p=0.013, respectively), but for the dominant hand this was only seen at baseline (rs=0.418, p=0.019).
10 3.4 Regression analysis to predict changes in brain volume and thickness over time
3.4.1 Annual percentage change in thalamus volume
A significant regression equation was found where annual percentage change in total thalamus volume could be significantly predicted by disease duration (β=0.487, p=0.005) (Figure 2), where a shorter disease duration was predictive of more negative annual percentage change (i.e. atrophy) of the thalamus: F(1, 29)= 9.002, p=0.006 with adjusted R2=0.211 (Supplementary table 1). This could also
be seen for the left thalamus (β=0.410, p=0.022): F(1, 29)= 5.876, p=0.022 with adjusted R2=0.140
(Supplementary table 2). The annual percentage change in right thalamus volume could not only be predicted by disease duration (β=0.457, p=0.008), but also by baseline ARR (β=-0.344, p=0.063), CIS score (β=0.401, p=0.053) and EDSS score (β=-0.208, p=0.299): F(4, 26)=3.784, p=0.015 with adjusted R2=0.271 (Supplementary table 3). In addition to a shorter disease duration, more negative annual
percentage change of the right thalamus volume was found upon higher ARR, lower CIS and higher EDSS scores at baseline.
Figure 2. Graph demonstrates the correlation between the annual percentage change over two years (2-APC) in
total thalamus volume and disease duration in patients (β=0.487, p=0.005).
3.4.2 Annual percentage change in motor cortex thickness
A significant regression equation was found where the annual percentage change in thickness of the right motor cortex could significantly be predicted by baseline EDSS score (β=0.438, p=0.018), educational level (β=-0.379, p=0.043) and ARR (β=-0.227, p=0.221): F(3, 27)= 4.843, p=0.008 with adjusted R2=0.278 (Supplementary table 4). This equation showed that a more negative annual
percentage change in right motor cortex thickness was found upon lower EDSS score, higher
educational level and higher ARR at baseline. No significant regression equation could be calculated
-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 2 -A P C in tot al t ha la m us v ol um e ( % )
11 neither for the annual percentage change in left motor cortex thickness nor for the annual percentage change in average motor cortex thickness.
3.5 Regression analyses to predict changes in SDMT and 9-HPT scores over time
3.5.1 Annual percentage change in SDMT score
The annual percentage change in SDMT score could be significantly predicted by baseline educational level (β=0.494, p=0.009), AIS score (β=0.549, p=0.024) and CIS score (β=-0.328, p=0.150), where lower educational level, lower AIS score and higher CIS scorewere predictive of less positive annual percentage change in SDMT score: F(3, 27)= 3.738, p=0.023 with adjusted R2=0.215
(Supplementary table 5).
3.5.2 Annual percentage change in 9-HPT score
A significant regression model was found in which the annual percentage change in the 9-HPT score for the dominant hand could be significantly predicted by gender (β=-0.320, p=0.065), EDSS score (β=0.594, p=0.012), CIS score (β=-0.413, p=0.055) and age (β=-0.244, p=0.194) at baseline. The model showed that a lower EDSS score, a higher CIS score, higher age and being a female could be predictive of a more negative annual percentage change in 9-HPT score for the dominant hand: F(4, 26)= 2.918, p=0.041 with adjusted R2=0.204 (Supplementary table 6).
The annual percentage change in 9-HPT score for the non-dominant hand could be significantly predicted by baseline HADS anxiety score (β=-0.512, p=0.008), thickness of the left motor cortex (β=-1.268, p=0.060), educational level (β=-0.405, p=0.029), age (β=-0.432, p=0.027), average thickness of the motor cortex (β=1.007, p=0.130) (Figure 3), EDSS score (β=0.306, p=0.114) and ARR (β=-0.224, p=0.209). A more negative annual percentage change in 9-HPT score for the non-dominant hand was predicted by a higher HADS anxiety score, a thicker left motor cortex, a higher educational level, a higher age, a thinner average motor cortex, a lower EDSS score and a higher ARR: F(7, 23)= 3.534, p=0.010 with adjusted R2=0.372 (Supplementary table 7).
12
Figure 3. Graph demonstrates the correlation between the annual percentage change over two years (2-APC) in
nine-hole peg (9-HPT) score and the average thickness of the motor cortex (β=1.007, p=0.130).
-20 -15 -10 -5 0 5 10 15 20 25 2,2 2,3 2,4 2,5 2,6 2,7 2,8 2,9 2 -A P C in 9 -H P T sc or e ( non -dom ina nt ) (% )
13 4. Discussion
The aim of this study was to examine the longitudinal relationship between thalamus volume and SDMT score, and motor cortex thickness and 9-HPT score. It was hypothesized that change in thalamus volume could be predicted by SDMT score, and that change in motor cortex thickness could be predicted by 9-HPT score. However, neither a predictive value of thalamus volume on change in SDMT nor a predictive value of motor cortex thickness on change in 9-HPT score was expected.
First, it was found that the 2-APC in left thalamus volume, right thalamus volume and in total thalamus volume could be predicted by disease duration, where a shorter disease duration at baseline was predictive of more negative 2-APC (i.e. atrophy) of the thalamus. In addition, right thalamus atrophy could also be predicted by higher ARR, lower CIS score (i.e. low fatigue) at baseline, and higher EDSS score at baseline. The predictive value of a shorter disease duration is contradictive to previous research (Wylezinska et al., 2003, Azevedo et al., 2018). A possible explanation could be that atrophy rates are higher at the start of the disease. Even though previous research has shown a significant effect of disease duration on grey matter fraction (Chard et al., 2002), it has also been shown that thalamus atrophy is more distinct in RRMS compared to secondary progressive MS (Rocca et al., 2010, Azevedo et al., 2018).
Next, predictors found for the 2-APC of the thickness of the right motor cortex were EDSS score, educational level and ARR, where lower baseline EDSS, higher educational level and higher baseline ARR predicted a more negative 2-APC. The finding of a lower baseline EDSS score as a predictor is contra-intuitive, but could possibly be linked to the predictive value of a shorter disease duration found for thalamus atrophy. No predictors were found for the 2-APC in either left motor cortex thickness or average motor cortex thickness, which could be due to the limitation in this study that both right and left handed subjects were included.
Third, an increase in 2-APC of SDMT score was lower in patients with a low educational level, low AIS score and high CIS score. Previous research has already shown a positive correlation between SDMT scores and educational level (Boringa et al., 2001, Amato et al., 2006). The findings from this study could add to the previous studies, since the results indicate a smaller learning effect over time in patients with a lower educational level. The predictive value of a lower AIS score (i.e. good sleep) and a higher CIS score (i.e. high fatigue), suggests that patients with increased fatigue perform poorer on the SDMT over time than patients who are less fatigued and that this is not due to sleep problems. These results are adding to previous research showing a negative impact of both cognitive and motor fatigue on the SDMT scores (Penner et al., 2009).
Last, it was found that a more negative 2-APC in 9-HPT scores of both the dominant and non-dominant hand could be predicted by higher baseline age and lower baseline EDSS score. In addition, gender (female) and higher CIS scores could also predict more negative 2-APC in 9-HPT dominant hand scores,
14 whereas higher HADS anxiety score, higher educational level, higher ARR and motor cortex thickness predicted more negative 2-APC in 9-HPT non-dominant scores.
Previous research has already shown a positive correlation between age and 9-HPT scores in RRMS patients (Solaro et al., 2019), which could also be seen in the correlation performed in this study. These results indicate that not only a cross-sectional correlation can be found, but that a longitudinal relation can be found as well. Moreover, a positive correlation between EDSS score and 9-HPT score has been found in previous studies (Solaro et al., 2019, Yozbatıran et al., 2006, Cohen et al., 2000), which has been found in this study too. Thus, the predictive value of a lower baseline EDSS score is contra-intuitive. However, no previous findings can explain this result.
The finding that a higher CIS score (i.e. high fatigue) can predict more negative 2-APC in 9-HPT dominant hand scores, adds to previous findings showing worse performance on the 9-HPT in case of higher fatigue scores (Chahin et al., 2015). However, this was only the case for physical fatigue scores, but not for cognitive or total fatigue scores, which is a distinction that was not made in this study. The strongest predictor for the 2-APC in 9-HPT score of the non-dominant hand was the HADS-A score, where a more negative 2-APC was predicted by a higher HADS-A score. This adds to previous research showing a correlation between a higher anxiety score and a lower 9-HPT score (Gold et al., 2003). Furthermore, thickness of the left motor cortex and thickness of the motor cortex were predictive of the annual percentage change in 9-HPT score of the non-dominant hand. A limitation of this study is that both right and left handed subjects were included. Hence, nothing can be concluded from the predictive value of the left motor cortex. Moreover, future studies should not include left motor cortex thickness and motor cortex thickness in the same model, since those two variables are related. A possible explanation for the predictive value of a thinner motor cortex is that damage at the start of the disease does not necessarily reflect in physical limitations, since stable functional networks have been found at the early stage of MS (Shu et al., 2016). Moreover, previous studies found that the connectivity in primary and secondary motor regions can increase after an acute relapse (Dogonowski et al., 2016) and that network reorganization occurs in early MS patients (Gamboa et al., 2014).
This study has some limitations. First, these analyses need to be replicated in healthy controls to distinguish effects due to normal ageing from effects due to pathophysiology. Secondly, brain volumes were not normalized for intra cranial volume, which may have influenced the correlation and regression results, mainly with regard to age and gender. Last, no correction was done for multiple comparisons in this study, therefore results may differ when taking the sample size and the amount of statistical tests into account. Follow-up studies should address these limitations.
15 The present study did not find the expected predictive value of SDMT score and 9-HPT score, but did find a predictive value of motor cortex thickness. Follow-up studies, when addressing the limitations, could verify the findings and these results could contribute to giving a more detailed prognosis for MS patients.
In conclusion, this study reveals that thalamus atrophy cannot be predicted by SDMT score and that motor cortex thinning cannot be predicted by 9-HPT score. Moreover, thalamus volume cannot predict SDMT performance over time, but motor cortex thickness can predict decreasing 9-HPT performance over time of the non-dominant hand.
16 5. Supplementary tables
Supplementary table 1. Forward regression model with predictor for the annual percentage change in total
thalamus volume. Model 1 Variable β Sig. Disease duration 0.487 0.005 F (df) 9.002 (1, 29) R 0.487 Adjusted R2 0.211 Sig. model 0.006
Supplementary table 2. Forward regression model with predictor for the annual percentage change in left
thalamus volume Model 1 Variable β Sig. Disease duration 0.410 0.022 F (df) 5.876 (1, 29) R 0.410 Adjusted R2 0.140 Sig. model 0.022
17
Supplementary table 3. Forward regression models with predictors for the annual percentage change in right
thalamus volume
Supplementary table 4. Forward regression models with predictors for the annual percentage change in right
motor cortex thickness
Model 1 Model 2 Model 3 Model 4 Variable β Sig. β Sig. β Sig. β Sig.
Disease duration 0.448 0.011 0.424 0.014 0.438 0.010 0.457 0.008 ARR -0.260 0.120 -0.385 0.034 -0.344 0.063 CIS 0.298 0.096 0.401 0.053 EDSS -0.208 0.299 F (df) 7.295 (1, 29) 5.130 (2, 28) 4.650 (3, 27) 3.784 (4, 26) R 0.448 0.518 0.584 0.607 Adjusted R2 0.173 0.216 0.267 0.271 Sig. model 0.011 0.013 0.010 0.015
Model 1 Model 2 Model 3 Variable β Sig. β Sig. β Sig.
EDSS 0.485 0.006 0.377 0.033 0.438 0.018 Education -0.298 0.087 -0.379 0.043 ARR -0.227 0.221 F (df) 8.903 (1, 29) 6.351 (2, 28) 4.843 (3, 27) R 0.485 0.559 0.591 Adjusted R2 0.209 0.263 0.278 Sig. model 0.006 0.005 0.008
18
Supplementary table 5. Forward regression models with predictors for the annual percentage change in SDMT
score
Supplementary table 6. Forward regression models with predictors for the annual percentage change in 9HPT
score for the dominant hand
Model 1 Model 2 Model 3 Variable β Sig. β Sig. β Sig.
Education 0.379 0.036 0.507 0.009 0.494 0.009 AIS 0.330 0.076 0.549 0.024 CIS -0.328 0.150 F (df) 4.852 (1, 29) 4.326 (2, 28) 3.738 (3, 27) R 0.379 0.486 0.542 Adjusted R2 0.114 0.181 0.215 Sig. model 0.036 0.023 0.023
Model 1 Model 2 Model 3 Model 4 Variable β Sig. β Sig. β Sig. β Sig.
Gender -0.331 0.069 -0.339 0.058 -0.293 0.092 -0.320 0.065 EDSS 0.261 0.138 0.466 0.029 0.594 0.012 CIS -0.360 0.090 -0.413 0.055 Age -0.244 0.194 F (df) 3.577 (1, 29) 3.033 (2, 28) 3.207 (3, 27) 2.918 (4, 26) R 0.331 0.422 0.513 0.557 Adjusted R2 0.079 0.119 0.181 0.204 Sig. model 0.069 0.064 0.039 0.041
19
Supplementary table 7. Forward regression models with predictors for the annual percentage change in 9HPT
score for the dominant hand.
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Variable β Sig. β Sig. β Sig. β Sig. β Sig. β Sig. β Sig. HADS A -0.460 0.009 -0.434 0.013 -0.465 -0.008 -0.568 0.002 -0464 0.014 -0.518 0.008 -0.512 0.008 Thickness lef t motor cortex -0.229 0.171 -0.237 0.152 -0.364 0.036 -1.246 0.070 -1.272 0.062 -1.268 0.060 Education -0.226 0.172 -0.338 0.048 -0.380 0.028 -0.328 0.060 -0.405 0.029 Age -0.360 0.055 -0.340 0.066 -0.417 0.033 -0.432 0.027 Average thickness motor cortex 0.906 0.179 0.975 0.146 1.007 0.130 EDSS 0.229 0.214 0.306 0.114 ARR -0.224 0.209 F (df ) 7.789 (1, 29) 5.011 (2, 28) 4.112 (3, 27) 4.440 (4, 26) 4.059 (5, 25) 3.741 (6, 24) 3.534 (7, 23) R 0.460 0.513 0.560 0.637 0.669 0.695 0.720 Adjusted R2 0.185 0.211 0.237 0.314 0.338 0.354 0.372 Sig. model 0.009 0.014 0.016 0.007 0.008 0.009 0.010
20 6. References
AMATO, M., PORTACCIO, E., GORETTI, B., ZIPOLI, V., RICCHIUTI, L., DE CARO, M., PATTI, F., VECCHIO, R., SORBI, S. & TROJANO, M. 2006. The Rao’s Brief Repeatable Battery and Stroop Test: normative values with age, education and gender corrections in an Italian population.
Multiple Sclerosis Journal, 12, 787-793.
AMATO, M. P., PORTACCIO, E., GORETTI, B., ZIPOLI, V., IUDICE, A., DELLA PINA, D., MALENTACCHI, G., SABATINI, S., ANNUNZIATA, P., FALCINI, M., MAZZONI, M., MORTILLA, M., FONDA, C., DE STEFANO, N. & TU, S. S. G. 2010. Relevance of cognitive deterioration in early relapsing-remitting MS: a 3-year follow-up study. Mult Scler, 16, 1474-82.
ARNETT, P. A., HIGGINSON, C. I., VOSS, W. D., WRIGHT, B., BENDER, W. I., WURST, J. M. & TIPPIN, J. M. 1999. Depressed mood in multiple sclerosis: relationship to capacity-demanding memory and attentional functioning. Neuropsychology, 13, 434-46.
AZEVEDO, C. J., CEN, S. Y., KHADKA, S., LIU, S., KORNAK, J., SHI, Y., ZHENG, L., HAUSER, S. L. & PELLETIER, D. 2018. Thalamic atrophy in multiple sclerosis: a magnetic resonance imaging marker of neurodegeneration throughout disease. Annals of neurology, 83, 223-234. BATISTA, S., ZIVADINOV, R., HOOGS, M., BERGSLAND, N., HEININEN-BROWN, M., DWYER, M. G.,
WEINSTOCK-GUTTMAN, B. & BENEDICT, R. H. 2012. Basal ganglia, thalamus and neocortical atrophy predicting slowed cognitive processing in multiple sclerosis. Journal of neurology, 259, 139-146.
BENEDICT, R. H., DELUCA, J., PHILLIPS, G., LAROCCA, N., HUDSON, L. D., RUDICK, R. & CONSORTIUM, M. S. O. A. 2017. Validity of the Symbol Digit Modalities Test as a cognition performance outcome measure for multiple sclerosis. Multiple Sclerosis Journal, 23, 721-733.
BENNETT, F., CLEMMONS, D. C. & FRASER, R. T. 2002. Multiple sclerosis: Psychosocial and vocational
interventions, Demos Medical Publishing.
BERGSLAND, N., LAGANÀ, M. M., TAVAZZI, E., CAFFINI, M., TORTORELLA, P., BAGLIO, F., BASELLI, G. & ROVARIS, M. 2015. Corticospinal tract integrity is related to primary motor cortex thinning in relapsing–remitting multiple sclerosis. Multiple Sclerosis Journal, 21, 1771-1780.
BISECCO, A., STAMENOVA, S., CAIAZZO, G., D’AMBROSIO, A., SACCO, R., DOCIMO, R., ESPOSITO, S., CIRILLO, M., ESPOSITO, F. & BONAVITA, S. 2018. Attention and processing speed
performance in multiple sclerosis is mostly related to thalamic volume. Brain imaging and
behavior, 12, 20-28.
BISHOP, M. & RUMRILL, P. D. 2015. Multiple sclerosis: Etiology, symptoms, incidence and prevalence, and implications for community living and employment. Work, 52, 725-34.
BORINGA, J., LAZERON, R. H., REULING, I. E., ADER, H., PFENNINGS, L. E., LINDEBOOM, J., DE SONNEVILLE, L. M., KALKERS, N. & POLMAN, C. 2001. The brief repeatable battery of neuropsychological tests: normative values allow application in multiple sclerosis clinical practice. Multiple Sclerosis Journal, 7, 263-267.
BURGGRAAFF, J., KNOL, D. L. & UITDEHAAG, B. M. 2017. Regression-based norms for the symbol digit modalities test in the Dutch population: Improving detection of cognitive impairment in multiple sclerosis. European neurology, 77, 246-252.
BUSCHKE, H. & FULD, P. A. 1974. Evaluating storage, retention, and retrieval in disordered memory and learning. Neurology, 24, 1019-1019.
CERCIGNANI, M. & WHEELER‐KINGSHOTT, C. 2017. From micro-to macro-structures in multiple sclerosis: What can we learn from diffusion imaging. NMR Biomed.
CHAHIN, S., MILLER, D., SAKAI, R. E., WILSON, J. A., FROHMAN, T., MARKOWITZ, C., JACOBS, D., GREEN, A., CALABRESI, P. A. & FROHMAN, E. M. 2015. Relation of quantitative visual and neurologic outcomes to fatigue in multiple sclerosis. Multiple Sclerosis and Related Disorders, 4, 304-310.
CHARD, D. T., GRIFFIN, C. M., PARKER, G. J., KAPOOR, R., THOMPSON, A. J. & MILLER, D. H. 2002. Brain atrophy in clinically early relapsing-remitting multiple sclerosis. Brain, 125, 327-37.
21 CHARD, D. T., JACKSON, J. S., MILLER, D. H. & WHEELER‐KINGSHOTT, C. A. 2010. Reducing the impact
of white matter lesions on automated measures of brain gray and white matter volumes.
Journal of magnetic resonance imaging, 32, 223-228.
COHEN, J., FISCHER, J., BOLIBRUSH, D., JAK, A., KNIKER, J., MERTZ, L., SKARAMAGAS, T. & CUTTER, G. 2000. Intrarater and interrater reliability of the MS functional composite outcome measure.
Neurology, 54, 802-806.
DALE, A. M., FISCHL, B. & SERENO, M. I. 1999. Cortical surface-based analysis: I. Segmentation and surface reconstruction. Neuroimage, 9, 179-194.
DIAMOND, B. J., JOHNSON, S. K., KAUFMAN, M. & GRAVES, L. 2008. Relationships between information processing, depression, fatigue and cognition in multiple sclerosis. Archives of
clinical neuropsychology, 23, 189-199.
DOGONOWSKI, A.-M., BLINKENBERG, M., PAULSON, O. B., SELLEBJERG, F., SØRENSEN, P. S., SIEBNER, H. R. & MADSEN, K. H. 2016. Recovery from an acute relapse is associated with changes in motor resting-state connectivity in multiple sclerosis. Journal of Neurology, Neurosurgery &
Psychiatry, 87, 912-914.
FEYS, P., LAMERS, I., FRANCIS, G., BENEDICT, R., PHILLIPS, G., LAROCCA, N., HUDSON, L. D., RUDICK, R. & CONSORTIUM, M. S. O. A. 2017. The Nine-Hole Peg Test as a manual dexterity
performance measure for multiple sclerosis. Multiple Sclerosis Journal, 23, 711-720.
FISCHER, J. S., RUDICK, R. A., CUTTER, G. R. & REINGOLD, S. C. 1999. The Multiple Sclerosis Functional Composite Measure (MSFC): an integrated approach to MS clinical outcome assessment. National MS Society Clinical Outcomes Assessment Task Force. Mult Scler, 5, 244-50.
FISCHL, B., SERENO, M. I. & DALE, A. M. 1999. Cortical surface-based analysis: II: inflation, flattening, and a surface-based coordinate system. Neuroimage, 9, 195-207.
FISHER, E., LEE, J. C., NAKAMURA, K. & RUDICK, R. A. 2008. Gray matter atrophy in multiple sclerosis: a longitudinal study. Ann Neurol, 64, 255-65.
GAMBOA, O. L., TAGLIAZUCCHI, E., VON WEGNER, F., JURCOANE, A., WAHL, M., LAUFS, H. &
ZIEMANN, U. 2014. Working memory performance of early MS patients correlates inversely with modularity increases in resting state functional connectivity networks. Neuroimage, 94, 385-395.
GIORDANO, A., GRANELLA, F., LUGARESI, A., MARTINELLI, V., TROJANO, M., CONFALONIERI, P., RADICE, D., SOLARI, A. & GROUP, S. I.-T. 2011. Anxiety and depression in multiple sclerosis patients around diagnosis. J Neurol Sci, 307, 86-91.
GOLD, S. M., SCHULZ, H., MÖNCH, A., SCHULZ, K.-H. & HEESEN, C. 2003. Cognitive impairment in multiple sclerosis does not affect reliability and validity of self-report health measures.
Multiple Sclerosis Journal, 9, 404-410.
HOUTCHENS, M., BENEDICT, R., KILLIANY, R., SHARMA, J., JAISANI, Z., SINGH, B.,
WEINSTOCK-GUTTMAN, B., GUTTMANN, C. & BAKSHI, R. 2007. Thalamic atrophy and cognition in multiple sclerosis. Neurology, 69, 1213-1223.
HUANG, W. J., CHEN, W. W. & ZHANG, X. 2017. Multiple sclerosis: pathology, diagnosis and treatments. Experimental and therapeutic medicine, 13, 3163-3166.
KURTZKE, J. F. 1983. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology, 33, 1444-52.
LIEPERT, J., UHDE, I., GRÄF, S., LEIDNER, O. & WEILLER, C. 2001. Motor cortex plasticity during forced-use therapy in stroke patients: a preliminary study. Journal of neurology, 248, 315-321. LIEPERT, J., ZITTEL, S. & WEILLER, C. 2007. Improvement of dexterity by single session low-frequency
repetitive transcranial magnetic stimulation over the contralesional motor cortex in acute stroke: a double-blind placebo-controlled crossover trial. Restorative neurology and
neuroscience, 25, 461-465.
MATHIOWETZ, V., WEBER, K., KASHMAN, N. & VOLLAND, G. 1985. Adult norms for the nine hole peg test of finger dexterity. The Occupational Therapy Journal of Research, 5, 24-38.
22 MILLER, D. H., BARKHOF, F., FRANK, J. A., PARKER, G. J. & THOMPSON, A. J. 2002. Measurement of
atrophy in multiple sclerosis: pathological basis, methodological aspects and clinical relevance. Brain, 125, 1676-95.
NATIONAL MULTIPLE SCLEROSIS SOCIETY. Relapsing-remitting MS (RRMS) [Online]. Available:
https://www.nationalmssociety.org/What-is-MS/Types-of-MS/Relapsing-remitting-MS
[Accessed 30 Apr 2020].
PENNER, I., RASELLI, C., STÖCKLIN, M., OPWIS, K., KAPPOS, L. & CALABRESE, P. 2009. The Fatigue Scale for Motor and Cognitive Functions (FSMC): validation of a new instrument to assess multiple sclerosis-related fatigue. Multiple Sclerosis Journal, 15, 1509-1517.
POLMAN, C. H., REINGOLD, S. C., BANWELL, B., CLANET, M., COHEN, J. A., FILIPPI, M., FUJIHARA, K., HAVRDOVA, E., HUTCHINSON, M. & KAPPOS, L. 2011. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Annals of neurology, 69, 292-302.
REUTER, M., SCHMANSKY, N. J., ROSAS, H. D. & FISCHL, B. 2012. Within-subject template estimation for unbiased longitudinal image analysis. Neuroimage, 61, 1402-1418.
ROCCA, M. A., MESAROS, S., PAGANI, E., SORMANI, M. P., COMI, G. & FILIPPI, M. 2010. Thalamic damage and long-term progression of disability in multiple sclerosis. Radiology, 257, 463-469.
SCHOONHEIM, M. M., POPESCU, V., RUEDA LOPES, F. C., WIEBENGA, O. T., VRENKEN, H., DOUW, L., POLMAN, C. H., GEURTS, J. J. & BARKHOF, F. 2012. Subcortical atrophy and cognition: sex effects in multiple sclerosis. Neurology, 79, 1754-61.
SHU, N., DUAN, Y., XIA, M., SCHOONHEIM, M. M., HUANG, J., REN, Z., SUN, Z., YE, J., DONG, H. & SHI, F.-D. 2016. Disrupted topological organization of structural and functional brain
connectomes in clinically isolated syndrome and multiple sclerosis. Scientific reports, 6, 1-11. SMITH, A. 1973. Symbol digit modalities test, Western Psychological Services Los Angeles.
SOLARO, C., CATTANEO, D., BRICHETTO, G., CASTELLI, L., TACCHINO, A., GERVASONI, E. &
PROSPERINI, L. 2019. Clinical correlates of 9-hole peg test in a large population of people with multiple sclerosis. Multiple sclerosis and related disorders, 30, 1-8.
SOLDATOS, C. R., DIKEOS, D. G. & PAPARRIGOPOULOS, T. J. 2000. Athens Insomnia Scale: validation of an instrument based on ICD-10 criteria. J Psychosom Res, 48, 555-60.
STADELMANN, C. & BRÜCK, W. 2008. Interplay between mechanisms of damage and repair in multiple sclerosis. Journal of neurology, 255, 12-18.
VALVERDE, S., CABEZAS, M., ROURA, E., GONZÁLEZ-VILLÀ, S., PARETO, D., VILANOVA, J. C., RAMIÓ-TORRENTÀ, L., ROVIRA, À., OLIVER, A. & LLADÓ, X. 2017. Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach.
NeuroImage, 155, 159-168.
VALVERDE, S., SALEM, M., CABEZAS, M., PARETO, D., VILANOVA, J. C., RAMIÓ-TORRENTÀ, L., ROVIRA, À., SALVI, J., OLIVER, A. & LLADÓ, X. 2019. One-shot domain adaptation in multiple sclerosis lesion segmentation using convolutional neural networks. NeuroImage: Clinical, 21, 101638. VERCOULEN, J. H., SWANINK, C. M., FENNIS, J. F., GALAMA, J. M., VAN DER MEER, J. W. &
BLEIJENBERG, G. 1994. Dimensional assessment of chronic fatigue syndrome. J Psychosom
Res, 38, 383-92.
VERHAGE, F. 1964. Intelligentie en leeftijd: Onderzoek bij Nederlanders van twaalf tot
zevenenzeventig jaar, Van Gorcum.
VIDAL-JORDANA, A. & MONTALBAN, X. 2017. Multiple Sclerosis: Epidemiologic, Clinical, and Therapeutic Aspects. Neuroimaging Clin N Am, 27, 195-204.
WEEDA, M. M., BROUWER, I., DE VOS, M. L., DE VRIES, M. S., BARKHOF, F., POUWELS, P. J. W. & VRENKEN, H. 2019. Comparing lesion segmentation methods in multiple sclerosis: Input from one manually delineated subject is sufficient for accurate lesion segmentation.
Neuroimage-Clinical, 24.
WYLEZINSKA, M., CIFELLI, A., JEZZARD, P., PALACE, J., ALECCI, M. & MATTHEWS, P. 2003. Thalamic neurodegeneration in relapsing-remitting multiple sclerosis. Neurology, 60, 1949-1954.
23 YOZBATıRAN, N., BASKURT, F., BASKURT, Z., OZAKBAS, S. & IDIMAN, E. 2006. Motor assessment of
upper extremity function and its relation with fatigue, cognitive function and quality of life in multiple sclerosis patients. Journal of the neurological sciences, 246, 117-122.
ZIGMOND, A. S. & SNAITH, R. P. 1983. The hospital anxiety and depression scale. Acta psychiatrica
