Reliability study of the Dyskinesia Impairment Scale in adults with cervical dystonia
ABSTRACT
BACKGROUND – The aims of this study are (1) to assess internal consistency and interrater of the Dyskinesia Impairment Scale (DIS) in people with cervical dystonia, (2) to assess the relationship between the DIS, the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) and the Burke-Fahn-Marsden Movement Scale (BFMMS).
METHOD – In this study we included 18 adults with cervical dystonia (six males; 12 females; mean age 53y 9mo; SD 8y 1mo; range 40y 5mo – 69y 2mo). Dystonia was measured by the Dyskinesia Impairment Scale – Dystonia subscale (DIS-D).
RESULTS – The DIS-D total scale score showed moderate interrater reliability, with an ICC of 0.61 and confidence intervals between 0.35 and 0.81. The standard error of the measurement and minimal detectable difference values were 2,60% and 7,20% respectively. Cronbach’s alpha internal consistency values of the DIS-D ranged between 0.74 and 0.85. Good levels of correlations were found between the total scores of DIS-D (%NECK), BFMMS (%NECK) and TWSTRS (%TOTAL).
CONCLUSION – Results of this study show that the DIS-D is reliable in assessing adults with cervical dystonia, but might be less sensitive then the specifically developed TWSTRS.
KEY-WORDS: cervical dystonia, dystonia, Dyskinesia Impairment Scale
Dystonia is a movement disorder characterized by sustained or intermittent muscle contractions causing abnormal, often repetitive movements, postures, or both. Dystonic movements are typically patterned, twisting, and may be tremulous. Dystonia is often initiated or worsened by voluntary action and associated with overflow muscle activation.(1)
With a prevalence of 0,16 per 1000, dystonia represents the third most common movement disorder worldwide.(2-8) However, this number could be an underestimation, with many cases remaining undiagnosed.(4)
Due to a certain lack of clarity and precision in definitions, a new classification has been proposed with two different axes to describe the clinical (Axis I) and the etiological (Axis II) features.(1)
The clinical presentation of dystonia is highly variable, ranging from an overemphasized specific movement to a persistent odd posture or fixed deformity.(3) Any region of the body may be affected. Dystonia may emerge at any age and can occur in isolation or in combination with other clinical problems (table 1).(3) Apart from the large variety of clinical manifestations and the affected body region(s), the age at onset, the temporal aspects and the potential association with other clinical features are important factors for definitive diagnosis and treatment modalities of dystonia.(3)
Jinnah HA (2015)
The classification of dystonia according to etiology has undergone the most changes over the years, in part because of enormous progress in understanding their many potential causes.(9) On an etiological basis, dystonia can be classified as inherited (with a genetic origin), acquired (due to a known specific cause) or idiopathic (due to an unknown cause).(1)
Idiopathic dystonia is the most common form of adult-onset dystonia, often with focal onset. Focal dystonia affects only one body region, mostly the cervical region, referred to as cervical dystonia (CD).(10) CD is marked by deviation of the head around the horizontal (torticollis), coronal (retrocollis, anterocollis), and vertical axis (laterocollis), often associated with reduced range of motion.(11, 12) CD may spread from the neck to contiguous body areas including the face and arms, previously mentioned as overflow movements or muscle activation.(2, 13, 14). Despite the fact CD is the most common form of adult-onset focal dystonia, no validated clinical diagnostic guidelines to allow easy differentiation of CD are available, often leading to difficult diagnosis.(15, 16)
The exact pathophysiology of inherited dystonia remains currently largely unknown.(17) It is however, generally accepted that it is related to a basal ganglia disorder although no overt structural abnormalities can be found.(6) Using evidence from neuro-imaging techniques, Niethammer et al.(8) concluded that inherited dystonia has to be considered as a neurodevelopmental circuit disorder involving the cortico-striato-pallido-thalamo-cortical and the cerebello-thalamo-cortical pathways.
Despite the high prevalence of dystonia, the available treatments remain often only modestly efficacious.(3) The therapeutic aims are mostly symptomatic: diminishing the involuntary movements, pain relief, prevention and correction of contractures or postures.(18) Since treatment options are currently based on diagnosis set by a clinical measurement scale, it is essential that the applied assessment tool is capable of generating a reliable evaluation of dystonia.
A critical evaluation of all existing rating scales available for dystonia was committed in 2013, a task commissioned by the Movement Disorders Society.(19) Albanese et al. classified seven scales as ‘recommended’ for primary dystonia on a clinical basis.(19) For the evaluation of dystonia, several different rating scales have been proposed, but two have become the standard to evaluate dystonia in adults: the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) and the Fahn-Marsden Dystonia Rating Scale (BFMMS).(20, 21)
The Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) is a validated video-based assessment tool for CD and specifically developed to evaluate dystonia in the neck region. The TWSTRS is composed of three subscales measuring symptom severity, disability, and pain.(22) However, as only the neck region is scored in the TWSTRS scale, possible overflow movements to adjacent body areas cannot taken into account.
The Burke-Fahn-Marsden Movement Scale (BFMMS) has become a standard outcome measure for generalized inherited dystonia.(21, 23) The scale has demonstrated good reliability and validity and is also often used in the CD population.(22, 24, 25) However, notwithstanding the use in intervention studies, the sensitivity of the scale is questioned and insensitivity may be even higher in smaller regions as the neck region.(24)
In 2012, a new scale, the Dyskinesia Impairment Scale (DIS), was developed to evaluate infancy and childhood dystonia and choreoathetosis in cerebral palsy (CP) and was not listed by Albanese et al. The development of the DIS was an answer to the sensitivity limitations of the Barry–Albright Dystonia Scale (BADS), the Burke-Fahn-Marsden Movement Scale (BFMMS), and the Unified Dystonia Rating Scale (UDRS) in this condition by increasing the number of body regions and differentiating dystonia characteristics more.(24) The DIS evaluates severity of dystonia and choreoathetosis over twelve body regions using two subscales; the dystonia subscale (DIS-D) and choreoathetosis subscale (DIS-CA).(25) The DIS showed good to excellent validity, high internal consistency and proved to be reliable between raters in the evaluation of infancy and childhood dystonia and choreoathetosis in children with dyskinetic cerebral palsy (DCP).(26) The DIS has also proven good reliability in the evaluation of overflow movements in infancy and childhood dystonias.(27) However, reliability of the DIS was only assessed in CP population.
Based on (1) the inability of the TWSTRS to measure overflow activation, (2) the sensitivity limitations of the BFMMS and (3) the ability of the DIS to measure overflow movements, it is worthwhile to assess the reliability and concurrent validity of the DIS in CD population.
The objective of this study is to investigate the reliability and validity of the DIS-D compared to the BFMMS and TWSTRS assessing patients with cervical dystonia. It is hypothesized that the DIS-D can be used as an overall reliable tool to measure different forms of dystonia.
Method
Participants
In this study inclusion criteria were: (1) diagnosed with cervical dystonia and (2) older than 21 years. The exclusion criterion was suffering from other neurological conditions besides cervical dystonia. There was no control group included. Eighteen participants were included, aged between 40 and 69 years (six males; 12 females; mean age 53y 9mo; SD 8y 1mo). All participants were recruited from the University Hospitals Leuven and this study was performed in close collaboration with the medical-paramedical staff of the institute.
Assessment
The Dyskinesia Impairment Scale, Burke-Fahn-Marsden Movement Scale and the Toronto Western Spasmodic Torticollis Rating Scale were used to evaluate dystonia severity in the participants.
The DIS (figure 1) evaluates dystonia and choreoathetosis and can therefore be divided in two subscales: one for dystonia (DIS-D) and one for choreoathetosis (DIS-CA).(25)
Both subscales evaluate duration and amplitude of involuntary movement or posture in 12 body regions including the eyes, mouth, neck, trunk, and limbs. For the limbs, a distinction is made between the proximal and distal region and between the right and left side. For each of the assessed body regions, the duration refers to the amount of time that dyskinesia is present, whereas the amplitude aspect refers to the range of motion of the dyskinetic movements. All body regions are scored during two activities (action) and one resting posture (rest). Summation of the region scores gives a total action score (range 0–192) and a total rest score (range 0–96) for both subscales. The action and rest scores add up to a total score for dystonia and choreoathetosis, each with a range from 0 to 288. The total DIS score is the sum of the dystonia and choreoathetosis subscale.
FIGURE 1: Dyskinesia Impairment Scale (DIS)
Monbaliu (2012)
The BFMMS evaluates primary dystonia in nine body regions, considering a provoking factor and a severity factor. The sum of these two factors yields an individual score for each region. The total score ranges from 0 to 120.(23)
The TWSTRS scale was developed for the assessment of cervical dystonia. The scale is composed of three subscales measuring symptom severity, disability and pain. The severity score ranges from 0 to 38.
Procedure
All participants were videotaped with a standard protocol lasting 30 minutes covering all aspects of the three included scales. The video protocol consists of five parts. First, an overall view, i.e. entering the room (walking or with wheelchair), was made. The second part consists of voluntary asked activities (as described in the DIS) in sitting position (comfort position as well as active position), lying position and standing position. Recording time of each task is approximately 30 seconds and some activities are recorded from different points of view (profile view vs. frontal view). Participants get one trial in advance. After the recordings, range of motion of upper and lower limbs was measured with a goniometer. A speech interview was also performed, where some standard questions were asked, e.g. ‘what is your name?’ (Appendix A)
During a training course three raters (I.V., H.H, L.S.) were trained by a senior rater (E.M.) and thereupon interrater reliability was assessed by comparing the DIS scores of the three inexperienced raters of previously collected recordings of children with dyskinetic CP with scoring of a golden rater (E.M.) Extreme scores were discussed.
Cervical dystonia patients were assessed with the full DIS protocol, but during scoring only the subitems showing variability were taken into account. The choreoathesis subscale was scored, but not taken into account in the data analysis as the main interest of this study was the evaluation of dystonia symptoms in the cervical dystonia population.
Statistical Analysis
To assess interrater reliability, the intraclass correlation coefficients (ICCs) and 95% confidence intervals (CIs) were used for the total scores and item scores of the DIS-D, the BFMMS and the TWSTRS. An ICC higher than 0.90 is indicated as excellent, an ICC between 0.75 and 0.90 as good, and an ICC<0.75 as moderate to poor.(28) To interpret the ICC scores <0.75, we considered ICC values between 0.60 and 0.75 as moderate and less than 0.60 as poor.(24)
The ICCs of the training study were calculated for each body region separate, as well as for the total dystonia score, the total choreoathetosis score and the total scale score.
The ICCs of the cervical dystonia patients were calculated for each of the three rating scales. ICCs of the DIS-D were calculated based on the total dystonia score, which was a summation of eyes and neck scores as they were the only subitems in which dystonia was scored. For these two body regions, interrater reliability was calculated as well. ICCs of the BFMMS were calculated for eyes and neck, for the total provoking score, the total severity score and the total scale score. Interrater reliability for total %NECK was also computed, which is a summation of the provoking and severity subscale of the neck region. ICCs of the TWSTRS were calculated for four out of six subscales, namely maximal excursion (subdivided into six movements), duration, shoulder elevation and range of motion. Subscales ‘effect of sensory tricks’ and ‘time’ were not taken into account since the DIS video protocol did not allow scoring of these subitems.
For a more in-depth analysis, the Shapiro-Wilkinson test of normality was used to check normality of the interrater reliability data from the DIS-D. If this test showed a normal distribution, a t-test for dependent samples was performed to check for significant differences between the raters. Whereas, a Wilcoxon signed rank test was used if the data was not normally distributed. The standard error of measurement (SEM) and smallest detectable difference (SDD) were calculated using the formula SEM = standard deviation x √(1-ICC) and MDD = SEM*1.96*√2. The internal consistency was evaluated by Cronbach’s alpha. Spearman’s correlation coefficient, was used to assess the correlation between the TWSTRS (%TOTAL), BFMMS (%NECK) and DIS-D (%NECK).(28) All statistics were calculated with SPSS 25.0 (IBM Corp. Armonk, NY).
Results
Interrater reliability training study
The ICC value for the total DIS was 0.80 (95%; CI= 0.54-0.94). For the DIS-D and the DIS-CA, ICC values were 0.83 (95%; CI= 0.60-0.95) and 0.62 (95%; CI= 0.22-0.88) respectively.
For the DIS-D (table 2), ICCs ranged from 0.70 to 0.82 for the total scale score. During activity a total ICC of 0.82 was calculated, whereas the total ICC value during rest was 0.72. For the proximal body regions, ICC values were moderate to poor in both rest and activity. For the arm distal regions, ICC values were high during activity, but moderate during rest. For the other extremities, ICC values were moderate to poor both during rest and activity.
For the DIS-CA (table 3), ICCs ranged from 0.48 to 0.67 for the total scale score. During activity a total ICC of 0.63 was calculated, whereas the total ICC value during rest was 0.55. For the region scores, both during rest and activity, good reliability was found for the neck region and distal arm region. Moderately high values were found for the proximal part of the right arm. For the eye region, ICC values were moderately high during activity, but not during rest whereas the ICC values in the mouth region were higher during rest than activity. The other body regions presented moderate to poor reliability values.
The interrater reliability was considered sufficient for the trained raters to continue scoring.
HIER PFD VAN TABLE 2 en 3
Interrater reliability CD patients
The DIS-D total scale score showed moderate interrater reliability, with an ICC of 0.61 and confidence intervals between 0.35 and 0.81.
The ICCs for the total scores of the DIS-D showed poor to moderate interrater reliability (table 4). ICCs of the total scores of the duration aspect, amplitude aspect, and the summation of both were 0.47, 0.59, and 0.56, respectively, during activity and 0.49, 0.58, and 0.56 respectively, during rest.
DIS-D total %NECK revealed good interrater reliability, with an ICC of 0.78 and confidence intervals between 0.59 and 0.90. Region scores for eyes ranged between 0.20 and 0.44, region scores for neck ranged between 0.59 and 0.78.
The BFMMS total score showed poor interrater reliability, with an ICC of 0.53 and confidence intervals between 0.25 and 0.76.
BFMMS %NECK revealed good interrater reliability with an ICC of 0.86 and confidence intervals between 0.73 and 0.94.
ICCs of the provoking subscale ranged between 0.72 and 0.97. ICCs of the severity subscale ranged between 0.38 and 0.74. Results are shown in table 5.
TABLE V – Intraclass correlation coefficients (ICC) with 95% Confidence Intervals (CI) for the Burke-Fahn-Marsden Movement Scale
Provoking Severity Provoking x Severity
ICC 95% CI ICC 95% CI ICC 95% CI
1 Eyes 0.71 0.50-0.87 0.38 0.10-0.66 0.71 0.48-0.87
2 Neck 0.97 0.93-0.99 0.74 0.53-0.88 0.82 0.66-0.92
3 Total score
0.87
0.74-0.94
0.54
0.27-0.77
0.53
0.25-0.76
ICC = Intraclass Correlation Coefficient; CI = Confidence Interval
The TWSTRS total score showed good interrater reliability, with an ICC of 0.80 and confidence intervals between 0.58 and 0.92 (table 6). ICCs of lateral and sagittal shift were poor, whereas the ICCs of other region scores showed moderate to good interrater reliability.
TABLE VI: Intraclass correlation coefficients (ICCs) with 95% confidence intervals (CI) for the Toronto Western Spasmodic Torticollis Rating Scale
ICC 95% CI
1 Rotation 0.71 0.48 – 0.86
2 Laterocollis 0.91 0.82 – 0.96
3 Anterocollis 0.91 0.81 – 0.96
4 Retrocollis 0.79 0.61 – 0.91
5 Lateral shift 0.34 0.05 – 0.64
6 Sagittal shift 0.20 -0.06 – 0.53
7 Shoulder elevation 1.0
8 Duration 0.79 0.61 – 0.91
9 Range of motion
0.65 0.39 – 0.83
10 Total score 0.80 0.58 – 0.92
ICC = Intraclass Correlation Coefficient; CI = Confidence Interval
An in-depth analysis of interrater reliability values of the DIS-D was performed to see if there was a difference in scoring between the three raters. First, a Shapiro-Wilkinson was performed to see if the different subitems of the DIS-D were normally distributed.
For the neck total values, results showed a normal distribution for rater 1 and 3, but not for rater 2. In order to compare all raters, non-parametric tests were used. A Wilcoxon signed rank test was executed to see if there were significant differences between the raters. For the eyes total values, results showed a normal distribution for rater 2, but a non-normal distribution for rater 1 and 3. Therefore, non-parametrical tests were used. A Wilcoxon signed rank test was performed to check for significant differences between rater 1, 2 and 3. For neck+eyes total values, the Shapiro-Wilkinson test showed that all data was normally distributed. Therefore, a t-test for dependent samples was used to analyze whether the attributed scores were significantly different between raters. Results are shown in table 7.
Table VII: In-depth analysis of interrater reliability values of Dyskinesia Impairment Scale – dystonia subscale
Z-value P-value ICC 95% CI
Neck Rater 1 – Rater 2 -0.05 0.959 0.80 0.55-0.92
Rater 1 – Rater 3 -0.06 0.954 0.83 0.60-0.93
Rater 2 – Rater 3 -0.03 0.977 0.73 0.40-0.89
Eyes Rater 1 – Rater 2 -2.56 0.010 0.51 0.05-0.79
Rater 1 – Rater 3 -1.96 0.050 0.01 -0.46-0.47
Rater 2 – Rater 3 -2.36 0.018 0.12 -0.22-0.50
T-value P-value ICC 95% CI
Neck + Eyes Rater 1 – Rater 2 -2.11 0.050 0.79 0.50-0.92
Rater 1 – Rater 3 0.64 0.531 0.56 0.14-0.81
Rater 2 – Rater 3 1.76 0.096 0.48 0.06-0.76
Z-value = measure of standard deviation; P-value = probability score; T-value = ?; ICC = Intraclass Correlation Coefficient; CI = Confidence Interval
For the neck region, scores were not significantly different, proving good agreement between the raters when scoring dystonia in the neck region. ICCs between the different raters ranged from 0.73 to 0.80. For the eyes region, results showed less agreement. The Wilcoxon signed rank test pointed out that all score means were significantly different between the raters. ICC values were also poor between the different raters.
For the eye and neck region combined, P-values ranged from 0.05 to 0.53, implying that scores were not significantly different between raters. ICC values ranged from poor to good, depending on the raters.
Standard error of measurement and smallest detectable difference CD patients
For interrater reliability, the SEM and MDD values for the DIS-D total score were 2.60% and 7.20% respectively. For the DIS-D %NECK, the SEM was 7.04% and the MDD 19.5%. For the BFMMS total score, SEM and MDD were 1.32% and 3.65% respectively. For the BFMMS %NECK, the SEM was 12.67% and the MDD was 35.13%. Calculating the SEM and MDD for the TWSTRS scale resulted in 5.97% and 16.54% respectively.
Internal consistency CD patients
Results for the internal consistency of the three raters for the DIS-D ranged from 0.74 to 0.85, from 0.34 to 0.64 for the BFMMS and from 0.38 to 0.46 for the TWSTRS (table 8).
TABLE VIII: Internal consistency (Cronbach’s alpha) of the Dyskinesia Impairment Scale – dystonia subscale (DIS-D), Burke-Fahn-Marsden Movement Scale (BFMMS) and Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS)
DIS-D BFMMS TWSTRS
Rater 1 0.74 0.51 0.46
Rater 2 0.85 0.34 0.40
Rater 3 0.85 0.64 0.38
Concurrent validity CD patients
Moderate to high levels of correlation were found between the total scores of DIS-D (%NECK), BFMMS (%NECK) and TWSTRS using the Spearman’s correlation coefficient (table 11). The correlation between the DIS-D (%NECK) and the BFMMS (%NECK), and between the DIS-D (%NECK) and the TWSTRS for one rater are presented in a scatterplot (figure 2 and 3), showing an uphill pattern.
Table IX: Spearman correlation between the Burke–Fahn–Marsden Movement Scale (BFMMS), Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) and the Dyskinesia Impairment Scale – dystonia subscale (DIS-D)
Rater 1 Rater 2 Rater 3
TWSTRS – BFMMS 0.78** 0.51* 0.81**
TWSTRS – DIS-D 0.68** 0.63* 0.55**
BFMMS – DIS-D 0.55* 0.45 0.57*
TWSTRS – BFMMS (%NECK) 0.82** 0.51* 0.82**
TWSTRS – DIS-D (%NECK) 0.68** 0.63** 0.51*
BFMMS (%NECK) – DIS-D (%NECK) 0.87** 0.45 0.70**
* p = 0.001, ** p = 0.005
Figure 2: Scatterplot of correlation between BFMMS (%NECK) and DIS-D (%NECK)
Figure 3: Scatterplot of correlation between TWSTRS and DIS-D (%NECK)
Discussion
Given the multitude of presentations of dystonia and the variety of available treatments, reliable and valid assessments of the severity, distribution and evolution of dystonia are indispensable.
The aim of this study was to examine if the DIS-D could be used to assess adults diagnosed with inherited and/or idiopathic dystonia. So far, no research about the reliability of the DIS had been performed in an adult population. Due to recruitment difficulties, only participants with cervical dystonia participated in this study. Therefore, the initial aim to ‘assess reliability of the DIS in adults with inherited dystonia’ was reformulated to ‘assess the reliability of the DIS in participants with focal – and more specifically – cervical dystonia’.
Interrater reliability training study
DIS-D total scores showed moderately high to good interrater reliability, these results are in line with the results described in junior PT’s (Monbaliu et al., 2013) and were considered sufficient for the inexperienced raters to continue scoring.
During action overall moderate scores were found. Values less than 0.40 were calculated for leg RD and eyes. Our research team as well as other observers described observation issues scoring the eyes region using the DIS and other rating scales. Comella et al. (2003) also observed the lowest level of interrater agreement in the eyes regions (and lower face, jaw and upper face) when evaluating patients with other clinical measurement scales.(22) Our observation adds supplementary evidence to the fact that the assessment of dystonia in this anatomical region is perceived more difficult than other body regions. Additionally, as also stated by Comella et al., it might be possible that videotape examination for this body region does not provide sufficient information for correct rating.
A possible explanation for the low scores obtained in the assessment of leg RD is the low variation in the scores, which may have contributed to the generation of lower ICCs.(28)
During rest, same overall moderate scores were found. For the region scores, the values of trunk, leg RD and leg LD assessment appeared to be low. The same reflection can be made with regards to the low ICCs for legs in rest as what was stipulated earlier for the legs in action. In the assessment of the trunk, a large range of zero values was scored. The zero values in combination with a couple outliers could explain the negative value for this particular region.
DIS-CA total scores showed moderate to moderately high scores. Unlike the results of the junior PT’s in the study of 2013, the reliability of the choreoathetosis subscale was found to be lower than the dystonia subscale.
During activity, body regions showing the lowest level of interrater agreement were trunk, arm RD and LD and leg RP, LP, RD and LD with values less than 0.40. Analyzing these scores, a large range of zero values and some outliers were scored, explaining low ICCs for these regions.
Considering the interrater agreement for individual items during rest; eyes, arm LP, leg RP/LP/LD showed values less than 0.40. For the eyes, argumentation as in de DIS-D is followed. Trunk, arm LP and leg RP showed outlier scores, resulting in lower ICCs. Leg LP and leg LD showed zero variance.
Interrater reliability CD patients
The results of the reliability study in participants with cervical dystonia showed overall moderate ICC scores for DIS-D. Including ‘mouth’ in the ICC calculations lowered ICC values. An analysis of the individual score forms, pointed out that this body region had only been scored in three patients. Due to this fact, the region score for mouth was excluded in ICC calculations.
Neck values showed ICCs that were comparable with the ICC values of the BFMMS. ICC values of the total neck score of the BFMMS and TWSTRS showed good interrater reliability, confirming our skills as reliable raters in the assessment of dystonia.
The assessment for eyes showed low ICC values during rest as well as in activity, therefore an in-depth analysis was carried out. P-values for the eyes showed less agreement, pointing out that all score means were significantly different between the raters, confirming the difficulty of scoring dystonia in the eyes region, as previously mentioned in the reliability study of DCP patients.
High p-values for the neck were shown and resulted in a good agreement between the raters when scoring this region. A T-test for eyes and neck combined, resulted in low P-values as well as poor ICCs between the different raters. As mentioned before, the complexity of scoring dystonia in these anatomical regions lowers overall interrater reliability scores. These results are in agreement with results of Monbaliu et al. (2012, 2013).
Standard error of measurement and smallest detectable difference CD patients
In addition to the interrater reliability scores, SEM and MDD were calculated for all ICC values. The SEM and MDD values for the DIS-D were 2.60% and 7.20% respectively, these values are slightly lower than the values of the senior and junior PTs (Monbaliu et al, 2013.). A MDD of 7% implies that a score variation of approximately 7% is the minimum change detectable beyond what may be due to error, e.g. to be sure a ‘true’ improvement has occurred, rather than the difference being due to measurement errors. This value is lower than the measurement error for the BFMMS %NECK (35%) and for the TWSTRS (16%) but slightly higher than the measurement error for the BFMMS total score (4%). Low MDD values will benefit the sensitivity.
Internal consistency CD patients
Internal consistency was calculated using Cronbach’s alpha. Correlating the score for each scale item with the total score for each observation, then comparing it to the variance for all individual item scores computes Cronbach’s alpha. The values ranged from 0.743 to 0.845 and were comparable with the values for the senior PTs (Monbaliu et al., 2013) which points out that the DIS-D is a stable rating construct in measuring dystonia.
Concurrent validity CD patients
Finally, concurrent validity was assessed using Spearman’s correlation coefficient because on-normal distributed data. Moderate correlation was found between the DIS-D and TWSTRS, as between the DIS-D and BFMMS. The scatterplots showed moderate to high correlation between the DIS-D (%NECK) and the BFMMS (%NECK) as well as between the DIS-D (%NECK) and the TWSTRS.
Comparing the three rating scales, we conclude that the DIS is convenient to use as each item of the scale is clearly defined, except for the definition of the eyes region. The initial interrater reliability study shows high ICC values, confirming the statement by Monbaliu et al. (2012) that the DIS is an efficient and easy scoring system yielding consistent results, even when used by inexperienced raters.(26) However, an initial training is necessary and the video-protocol needs to be followed rigorously.
Some remarks can be made after assessing the TWSTRS. While scoring we noticed there was no definition of midline and extreme opposite position, making interpretation difficult, as already pointed out by Comella et al. (1997) and Jost et al. (2012). Another remark is the fact that items ‘maximal excursion’ and ‘range of motion’ basically describe the same aspects of abnormal posture, i.e. the ratings for these items are not independent from each other. The TWSTRS does not take overflow movements into account, which can lead to an underestimation of symptom severity.
Limitations in assessing CD patients with the BFMMS are the inclusion of weighting factors that halve the contribution of dystonia in eyes, mouth and neck to the total score. For this reason, weighting factors were excluded in the calculations of the total scores. Concerning subitems the BFMMS is focused more on the provoking factor during movement, whereas the DIS focuses on the distinction between duration and amplitude of dystonia. The provoking factor is used to evaluate the situation under which dystonia is present, a maximum score of 4 is given when dystonia occurs at rest, a score of 3 is given when dystonia occurs intermittently at rest. This means that even when the dystonia shows up briefly during rest, a relative high score should be given.
In a review study, executed by Jahanshahi et al. (1990), one-third of the included cervical dystonia patients showed overflow movements with involvement of the arms, jaws or trunk.(14) Due to this reason, the DIS-D was included to evaluate overflow muscle activation. Variability in scores for the eyes region showed that eye activity might be overflow movements. There was no variability observed in other body regions, concluding that cervical dystonia remained focal in this patient population.
Although CD is limited to dystonia in the neck region, the symptoms were variable ranging from patient to patient. Therefore the TWSTR might be a better scale to assess CD, while the DIS-D is probably not sensitive enough to respond to all expressions of cervical dystonia. Notwithstanding, the ICC value for the total neck score showed good interrater reliability with a value of 0.78, concluding the DIS-D region scores can be used to assess specific body parts in patients with rare forms of focal dystonia, where specific measurement scales may be not available.
Lastly important to mention is that the difference between clinical characteristics of CD patients and DCP patiens should be kept in mind during the interpretation of the low ICCs of the DIS-D in cervical dystonia patients. Cervical dystonia has a variable presentation of symptoms. Although the pathogenesis of CD remains obscure, multiple characteristics play a role in the development of this movement disorder, being genetic factors, trauma, the sensory system and impaired basal ganglia disfunction.(29) Unlike DCP patients, magnetic resonance imaging (MRI) is seldom used in the diagnosis of CD.
We acknowledge that our study has limitations. An overall remark that can be made is the use of a videotape protocol, which might be too complex to be applicable in routine clinical practice. The scoring time needed to finish the DIS can be a drawback when applying the scale in clinical practice. The three scales were assessed using the Dyskinesia Impairment Scale video protocol. As a result, scoring ‘time’ and ‘effect of sensory tricks’ in the TWSTRS Torticollis Severity Scale was not possible. Another remark that can be made is the homogeneity of the patient population. All participants were suffering from cervical dystonia, which is unfavorable for the generalization of results.
Conclusion
Results of this study show that the DIS-D is reliable in assessing adults with cervical dystonia, but might be less sensitive then the specifically developed TWSTRS. We conclude that further research is recommended to unravel the clinical characteristics of cervical dystonia as they are clearly distinct from cerebral palsy. Further assessment of the value of these measurement techniques will also require a larger sample of subjects. On the other hand, further research in the field of clinical measurement scales will not exclude subjective interpretation. Therefore, we urge for the development of a scoring system, including 3-dimensional movement analysis to increase objectivity.
References
1. Albanese A, Bhatia K, Bressman SB, Delong MR, Fahn S, Fung VS, et al. Phenomenology and classification of dystonia: a consensus update. Mov Disord. 2013;28(7):863-73.
2. Albanese A. The clinical expression of primary dystonia. J Neurol. 2003;250(10):1145-51.
3. Jinnah HA, Factor SA. Diagnosis and treatment of dystonia. Neurol Clin. 2015;33(1):77-100.
4. Steeves TD, Day L, Dykeman J, Jette N, Pringsheim T. The prevalence of primary dystonia: a systematic review and meta-analysis. Mov Disord. 2012;27(14):1789-96.
5. Albanese A, Asmus F, Bhatia KP, Elia AE, Elibol B, Filippini G, et al. EFNS guidelines on diagnosis and treatment of primary dystonias. Eur J Neurol. 2011;18(1):5-18.
6. Kojovic M, Pareés I, Kassavetis P, Palomar FJ, Mir P, Teo JT, et al. Secondary and primary dystonia: pathophysiological differences. Brain. 2013;136(Pt 7):2038-49.
7. Contarino MF, Berger-Plantinga E, Foncke EM, Ritz K, Mellema J, Baas F, et al. Clinical and genetic characterization of a large Dutch family with primary focal dystonia. Mov Disord. 2008;23(14):1998-2003.
8. Niethammer M, Carbon M, Argyelan M, Eidelberg D. Hereditary dystonia as a neurodevelopmental circuit disorder: Evidence from neuroimaging. Neurobiol Dis. 2011;42(2):202-9.
9. Jinnah HA, Albanese A. The New Classification System for the Dystonias: Why Was it Needed and How was it Developed? Mov Disord Clin Pract. 2014;1(4):280-4.
10. Crowner BE. Cervical dystonia: disease profile and clinical management. Phys Ther. 2007;87(11):1511-26.
11. Caputi F, Nashold BS, Spaziante R. Analysis of static and kinetic abnormalities typical of torticollis, and postoperative changes. Stereotact Funct Neurosurg. 1995;64(1):16-31.
12. T W, S B. Clinical diagnosis and management of dystonia. London: Boca Raton; 2007.
13. Greene P, Kang UJ, Fahn S. Spread of symptoms in idiopathic torsion dystonia. Mov Disord. 1995;10(2):143-52.
14. Jahanshahi M, Marion MH, Marsden CD. Natural history of adult-onset idiopathic torticollis. Arch Neurol. 1990;47(5):548-52.
15. Dauer WT, Burke RE, Greene P, Fahn S. Current concepts on the clinical features, aetiology and management of idiopathic cervical dystonia. Brain. 1998;121 ( Pt 4):547-60.
16. Defazio G, Jankovic J, Giel JL, Papapetropoulos S. Descriptive epidemiology of cervical dystonia. Tremor Other Hyperkinet Mov (N Y). 2013;3.
17. Breakefield XO, Blood AJ, Li Y, Hallett M, Hanson PI, Standaert DG. The pathophysiological basis of dystonias. Nat Rev Neurosci. 2008;9(3):222-34.
18. Robert I, E. MM. Rehabilitation in Movement Disorders. Cambridge: Cambridge University Press; 2013.
19. Albanese A, Sorbo FD, Comella C, Jinnah HA, Mink JW, Post B, et al. Dystonia rating scales: critique and recommendations. Mov Disord. 2013;28(7):874-83.
20. Comella CL, Stebbins GT, Goetz CG, Chmura TA, Bressman SB, Lang AE. Teaching tape for the motor section of the Toronto Western Spasmodic Torticollis Scale. Mov Disord. 1997;12(4):570-5.
21. Burke RE, Fahn S, Marsden CD, Bressman SB, Moskowitz C, Friedman J. Validity and reliability of a rating scale for the primary torsion dystonias. Neurology. 1985;35(1):73-7.
22. Comella CL, Leurgans S, Wuu J, Stebbins GT, Chmura T, Group DS. Rating scales for dystonia: a multicenter assessment. Mov Disord. 2003;18(3):303-12.
23. Elze MC, Gimeno H, Tustin K, Baker L, Lumsden DE, Hutton JL, et al. Burke-Fahn-Marsden dystonia severity, Gross Motor, Manual Ability, and Communication Function Classification scales in childhood hyperkinetic movement disorders including cerebral palsy: a 'Rosetta Stone' study. Dev Med Child Neurol. 2016;58(2):145-53.
24. Monbaliu E, Ortibus E, Roelens F, Desloovere K, Deklerck J, Prinzie P, et al. Rating scales for dystonia in cerebral palsy: reliability and validity. Dev Med Child Neurol. 2010;52(6):570-5.
25. Monbaliu E, Ortibus E, De Cat J, Dan B, Heyrman L, Prinzie P, et al. The Dyskinesia Impairment Scale: a new instrument to measure dystonia and choreoathetosis in dyskinetic cerebral palsy. Dev Med Child Neurol. 2012;54(3):278-83.
26. Monbaliu E, Ortibus E, Prinzie P, Dan B, De Cat J, De Cock P, et al. Can the Dyskinesia Impairment Scale be used by inexperienced raters? A reliability study. Eur J Paediatr Neurol. 2013;17(3):238-47.
27. Vanmechelen I, Bekteshi, Bossier K, Feys H, Monbaliu E, et al. Dystonia and choreoathetosis overflow movements in participants with dyskinetic cerebral palsy. Submitted DMCN
28. Portney L, Watkins M. Foundations of Clinical Research: Application to Practice, 3rd edn. New Jersey: Pearson Prentice Hall; 2009.
Appendix D: Populaire samenvatting
De Dyskinesia Impairment Scale (DIS) is een klinische meetschaal die in de eerste plaats gebruikt wordt voor het evalueren van kinderen met dyskinetische cerebrale parese (DCP). De betrouwbaarheid en validiteit werden reeds nagegaan voor deze patiëntenpopulatie. In deze studie werden de psychometrische eigenschappen van de DIS nagegaan bij patiënten met een cervicale dystonie.
De term ‘dystonie’ verwijst naar een bewegingsstoornis die wordt gekenmerkt door onvrijwillige spiercontracties en bijgevolg voor abnormale bewegingen of lichaamshoudingen zorgt. Er wordt over cervicale dystonie gesproken als voornamelijk de nekregio is aangedaan. Ondanks de grote prevalentie van deze aandoening, zijn er slechts een beperkt aantal behandelingsmethoden voorhanden. Gegeven het feit dat behandelingen worden afgestemd op diagnose, die gesteld wordt met klinische meetschalen, is er een grote nood aan betrouwbare en valide meetinstrumenten.
In het verleden werden al heel wat meetschalen ontworpen die dystonie kunnen evalueren, maar er zijn slechts enkelen methodologisch kwaliteitsvol. Verder onderzoek naar de betrouwbaarheid en validiteit van de Dyskinesia Impairment Scale bij volwassenen met cervicale dystonie is dan ook een noodzakelijke stap in de verruiming van inzichten in de ziektestatus van deze doelgroep. Een betrouwbare evaluatie zal bovendien ook bijdragen tot het verbeteren van gerichte behandelingstechnieken.
Een eerste stap die in dit onderzoek gezet moest worden, was nagaan of de beoordelaars betrouwbaar zijn in het scoren van dystonie en choreo-athetose bij kinderen met DCP. Dit werd gedaan aan de hand van tien video’s. Uit de statistische resultaten voor de intraclass correlatie coëfficiënt (ICC) scoren wij, als onervaren raters, goede waarden die tussen 0.70 en 0.81 liggen, op de dystonie subschaal (DIS-D).
Later werd de DIS-D geëvalueerd bij patiënten met een cervicale dystonie. Voor deze studie werden 18 proefpersonen (zes mannen; 12 vrouwen; gemiddelde leeftijd 53j 9ma) geïncludeerd. Tijdens deze evaluatie werd het videoprotocol van de DIS gevolgd, de patiënten werden gescoord met drie meetschalen: de Dyskinesia Impairment Scale, de Burke-Fahn-Marsden Movement Scale en de Toronto Western Spasmodic Torticollis Rating Scale.
Tijdens de statistische analyse werden de ICC, interne consistentie en concurrent validiteit nagegaan. Uit de resultaten van de ICC komen lage waarden naar voor. Om deze resultaten verder te kunnen verklaren, werd een analyse uitgevoerd die de scores van de verschillende beoordeelaars met elkaar vergelijkt. Hieruit blijkt dat het scoren van de ogen significant verschillende waarden oplevert tussen de raters, en bevestigt zo ook de complexiteit van scoring in deze regio.
Op basis van bovenbeschreven resultaten, kan een algemeen besluit gevormd worden dat stelt dat de DIS-D minder geschikt is in het evalueren van patiënten met cervicale dystonie.