To validate manual muscle testing (MMT) for strength assessment in juvenile and adult dermatomyositis (DM) and polymyositis (PM).
To validate manual muscle testing (MMT) for strength assessment in juvenile and adult dermatomyositis (DM) and polymyositis (PM).
Patients with PM/DM (73 children and 45 adults) were assessed at baseline and reevaluated 6–9 months later. We compared Total MMT (a group of 24 proximal, distal, and axial muscles) and Proximal MMT (7 proximal muscle groups) tested bilaterally on a 0–10 scale with 144 subsets of 6 and 96 subsets of 8 muscle groups tested unilaterally. Expert consensus was used to rank the best abbreviated MMT subsets for face validity and ease of assessment.
The Total, Proximal, and best MMT subsets had excellent internal reliability (Total MMT rs = 0.91–0.98), and consistency (Cronbach's α = 0.78–0.97). Inter- and intrarater reliability were acceptable (Kendall's W 0.68–0.76, rs = 0.84–0.95). MMT subset scores correlated highly with Total and Proximal MMT scores and with the Childhood Myositis Assessment Scale, and correlated moderately with physician global activity, functional disability, magnetic resonance imaging, and axial and distal MMT scores, and, in adults, with creatine kinase level. The standardized response mean for Total MMT was 0.56 in juveniles and 0.75 in adults. Consensus was reached to use a subset of 8 muscles (neck flexors, deltoids, biceps, wrist extensors, gluteus maximus and medius, quadriceps, and ankle dorsiflexors) that performed as well as the Total and Proximal MMT, and had good face validity and ease of assessment.
These findings aid in standardizing the use of MMT for assessing strength as an outcome measure for myositis.
The idiopathic inflammatory myopathies (IIM) are systemic autoimmune diseases comprised of subgroups differentiated by clinical, histopathologic, and immunologic findings. The IIM primarily affect skeletal muscle, causing chronic inflammation, scarring, and atrophy, resulting in muscle weakness and impaired functional ability. Adult polymyositis (PM), dermatomyositis (DM), and juvenile DM are among the most frequent of the IIM (1, 2).
The IIM primarily involve proximal muscles (3, 4). A recent study verified that weakness is symmetric, that proximal extremity weakness is greater than distal, and that neck flexors are significantly weaker than extensors in PM, DM, and juvenile DM (5). In that study, the target muscles most affected were the neck flexors; the hip abductors, extensors, and flexors; and the deltoid (5).
Manual muscle testing (MMT) of proximal muscle groups, or a Total MMT score involving proximal, distal, and axial muscles, has been used as a major end point for IIM therapeutic trials and in clinical practice to follow patients longitudinally (6, 7). Traditionally, MMT has been assessed using 5-point MMT scales, including the Medical Research Council Scale. An expanded 0–10-point MMT scale (8) has also been used in recent therapeutic trials and a natural history study (6, 7), and is postulated to be more sensitive in delineating weakness. The International Myositis Assessment and Clinical Studies group and the Paediatric Rheumatology International Trials Organisation have incorporated MMT according to the Kendall 0–10-point scale as a core outcome measure for therapeutic trials (9, 10).
The MMT has been validated in some neuromuscular and musculoskeletal disorders (11, 12). The 0–10-point MMT scale has been shown to have good reliability in juvenile IIM (3). However, full validation of the 10-point MMT in PM/DM and juvenile IIM to include proximal, distal, and axial muscle groups has not previously been demonstrated.
The primary aim of the current study was to validate Total and Proximal MMT scores by demonstrating the internal reliability, convergent construct validity, and reliability and responsiveness of these scores in assessing strength in patients with adult and juvenile PM/DM. We also examined subsets of unilateral muscle groups to determine whether a subset of 6 (MMT6) or 8 (MMT8) muscle groups that contain axial, proximal, and distal muscle groups would have similar or better performance characteristics than the Total MMT score in terms of responsiveness and reliability.
Patients with probable or definite PM/DM or juvenile IIM confirmed by the criteria by Bohan and Peter (13, 14) were seen at the National Institutes of Health (NIH) Clinical Center and 5 participating centers. Patients and/or their parents provided written informed consent and were enrolled in NIH Institutional Review Board–approved studies of myositis. Seventy-three patients with juvenile IIM participated in a multicenter observational study over 7–9 months (15), and 45 adult patients with PM/DM were screened for therapeutic trials of fludarabine or methimazole, 24 of whom entered into a trial and had a 6-month reevaluation assessment (16, 17) (Table 1). Juvenile patients with IIM were ≥4 years of age.
|Characteristic||Juvenile IIM||Adult IIM|
|Value||No. patients||Value||No. patients|
|Age, years||10.2 (5.6)||73||42 (17.2)||45|
|Sex, % female||69.9||73||77.8||45|
|Disease duration, months||33.2 (52.4)||63||60 (40.8)|
|Prednisone dose†||0.3 (0.85)||72||12.5 (10.0)||45|
|Physician global VAS, cm (0–10)|
|Disease activity||2.4 (4.0)||73||5.7 (4.0)||44|
|Disease damage||1.1 (1.9)||63||3.8 (2.8)||44|
|CMAS score (0–52)||45 (11)||69||NA|
|C-HAQ score (0–3)||0.12 (1.12)||58|
|Convery ADL score (0–91)||53 (19.5)||45|
|CK level, units/liter (0–252)||80 (73)||73||959 (2,097)||45|
|LDH level, units/liter (0–226)||238 (111)||41||364 (154)||30|
|Aldolase level, units/liter (0–7)||7 (5.2)||49||18 (19.4)||43|
|AST level, units/liter (0–34)||27 (12)||72||52 (49)||45|
|ALT level, units/liter (0–41)||25 (25)||68||53 (47)||45|
|STIR + T1-weighted MRI score (0–4)||0.8 (1.7)||31||2.0 (1.0)||45|
Patients were examined by an adult or pediatric physical therapist and rheumatologist experienced in evaluating myositis patients. Patients underwent a standardized MMT evaluation by a physical therapist using Kendall's 0–10-point scale as previously described (5, 8). All therapists were provided a standardized set of MMT commands, including a standardized order of testing and instructions to perform the examination and score the strength of each muscle group. Therapists at the NIH either attended a seminar on the 0–10-point MMT scale presented by Dr. Florence Kendall or viewed videotapes of her instruction in performing the examination. Therapists from other centers were guided on the instructional materials through telephone conversations with a physiatrist/rheumatologist (JEH). The Total MMT score included 2 axial, 7 proximal, and 4 distal muscle groups tested bilaterally, with a maximum potential value of 240 (5). The Proximal MMT score included 7 proximal muscle groups tested bilaterally, with a maximum potential score of 140. Physical therapists recorded whether a particular muscle group could not be tested accurately due to lack of cooperation, limited range of motion, calcinosis, or pain. These muscle groups were then excluded, and when possible, a contralateral muscle was substituted. Only patients who completed testing for at least 20 of the 24 muscle groups were included.
On the basis of established symmetry and patterns of weakness (5), we examined whether abbreviated subsets of muscles tested unilaterally on the right side could be validated in order to identify a more efficient strength evaluation than the Total MMT score. We generated the following types of subsets: 1) 144 subsets of 6 muscle groups (MMT6 scores) with a potential range of 0–60, which included 1 axial, 3 proximal (1 upper, 2 lower extremity), and 2 distal muscles (1 upper, 1 lower extremity); and 2) 96 subsets of 8 muscle groups (MMT8 scores) with a potential range of 0–80, which included 1 axial, 5 proximal (2 upper extremity, 3 lower extremity), and 2 distal muscles (1 upper, 1 lower extremity).
Outcome measures (including physician global activity, serum levels of muscle enzymes, and physical function assessed by the Childhood Myositis Assessment Scale [CMAS] and the Childhood Health Assessment Questionnaire [C-HAQ] in children and the modified Convery activities of daily living [ADL] scale for adult subjects) were included as previously described (9, 18, 19). Physician global activity and the CMAS were assessed by the rheumatologists. A videotape demonstration and detailed instructions for the CMAS were provided to each, as well as training at an investigators' meeting (18). Axial STIR and T1-weighted magnetic resonance images (MRIs) were obtained in 31 juvenile and all adult patients with PM/DM, and were scored as described by Summers et al (20).
Data were analyzed using StatView (SAS Institute, Cary, NC), except for Cronbach's alpha and Kendall's W, which were computed using a program written with Fourth Dimension software (Fourth Dimension, San Jose, CA). MMT scores were ordinal and summarized by medians and interquartile ranges for each muscle group. Summed MMT data were expressed as a percentage of the maximum possible score to allow for comparisons among MMT6, MMT8, and Proximal MMT subscores with the Total MMT score. P values less than 0.05 were considered significant, and adjustment for multiple comparisons was not performed in this hypothesis-generating study.
Internal reliability of muscle subsets was examined by correlating the 144 MMT6 subsets and the 96 MMT8 subsets with the Total MMT score. Cronbach's alpha, an estimate of the internal consistency of constituent muscles in a subset, was calculated for the MMT6 and MMT8 subgroups and compared with the Total and Proximal MMT scores. A value >0.7 indicates acceptable internal consistency (21, 22). Intrarater reliability for MMT was assessed by using an average of the pairwise Spearman's rank correlations (rs) obtained from 2 separate assessments of 10 patients with juvenile IIM by a single physical therapist (MRS). Interrater reliability for MMT performed by physical therapists on 9 patients with juvenile IIM was evaluated using Kendall's W among each of the 4 pairs of raters, as previously described (3). A Kendall W value ≥0.7 represents very good to excellent interrater reliability (21). Convergent construct validity was determined by examining Spearman's rank correlations (rs) of the Total MMT, Proximal MMT, and MMT6 and MMT8 subsets with measures of strength, function, and disease activity. Spearman's correlations >0.70 were defined as strong, between 0.40 and 0.70 as moderate, and <0.40 as poor (22). The standardized response means (SRMs) over time were calculated by dividing the observed change by the SD of the change. Responsiveness of the MMT subsets relative to Total MMT was examined using relative efficiency, defined as the square of the ratio of the Student's t statistics of each measure to be compared (23, 24). A value >1.0 was considered to indicate greater responsiveness than the Total MMT score.
A group of 12 adult and pediatric rheumatologists and physical therapists gathered for a nominal group technique consensus formation exercise to discuss and rank-order the 8 subsets that performed at least as well as the Total MMT score for potential use in therapeutic trials and other outcome studies. These clinicians had practiced rheumatology for a mean ± SD of 14.7 ± 9.5 years and saw an average of 37 patients with IIM annually. After reviewing the MMT performance data before the session, participants silently rank-ordered their top 5 choices among the 8 subsets. Participants were asked to consider the following factors in determining their rank order for these subsets based on the data presented and clinical experience: 1) face validity: are the muscle groups in the subset often affected in myositis?; 2) impact on function: do the muscle groups in the subset have important effects on physical function when strength is impaired?; 3) performance characteristics: does the MMT subset perform well in both juvenile and adult PM/DM?; 4) ease of testing: are these muscle groups easy to test, including frequency of test position changes in the subset, fatigability, and, in younger children, cooperation issues?; 5) rater reliability: is there good inter- and intrarater reliability of the subsets and of individual muscles contained in the subset?; and 6) limitations: would joint contractures limit strength testing ability of individual subset muscle groups? After an open round-robin discussion of each person's top 3 choices and their reasons for excluding other subsets, participants silently reranked their top 3 choices based on their preference for using these MMT subsets in therapeutic trials and other outcome studies. A videotape demonstrating the performance of the top-ranked MMT8 subset is available on the International Myositis Assessment and Clinical Studies Group Web site (http://www.niehs.nih.gov/research/resources/collab/imacs/diseaseactivity.cfm). This Web site also contains the MMT commands, 0–10-point scale, and standardized positions for testing patients.
Adult patients with IIM were weaker than patients with juvenile IIM based on their Total and Proximal MMT scores (Table 2), as reported previously (5). The Proximal MMT score indicated a slightly greater degree of weakness than the Total MMT score in both juvenile and adult patients, based on the percentage of the maximum potential score. Among the MMT subsets that best approximated the Total MMT score, all but one indicated a greater degree of weakness based on the percentage of the maximum potential score in patients with juvenile IIM (Table 2). In adult PM/DM, 4 abbreviated subsets (MMT8 subset numbers 55, 58, 133, and 136) exhibited a greater relative degree of weakness than the Total MMT score.
|MMT score||Muscle groups included in score||Juvenile IIM (n = 73)||Adult IIM (n = 45)|
|Total MMT†||Neck flexors, neck extensors, trapezius, deltoid, biceps, iliopsoas, gluteus maximus, gluteus medius, quadriceps, wrist flexors, wrist extensors, ankle dorsiflexors, ankle plantar flexor||88.3 (79.0, 94.1)||78.2 (73.1, 84.0)|
|Proximal MMT‡||Trapezius, deltoid, biceps, iliopsoas, gluteus maximus, gluteus medius, quadriceps||87.6 (78.7, 94.2)||72.4 (66.3, 81.2)|
|MMT6 subset #118§||Neck extensors, trapezius, gluteus maximus, iliopsoas, wrist extensors, ankle dorsiflexors||86.7 (78.0, 94.3)||80.0 (74.3, 85.9)|
|Subset #13||Neck flexors, trapezius, deltoid, gluteus maximus, iliopsoas, quadriceps, wrist flexors, ankle dorsiflexors||86.8 (77.1, 92.7)||78.5 (71.7, 81.6)|
|Subset #55||Neck flexors, deltoid, biceps, gluteus maximus, gluteus medius, quadriceps, wrist flexors, ankle dorsiflexors||87.5 (76.6, 92.3)||76.8 (70.1, 81.3)|
|Subset #58||Neck flexors, deltoid, biceps, gluteus maximus, gluteus medius, quadriceps, wrist extensors, ankle dorsiflexors||86.8 (76.4, 92.5)||76.5 (69.7, 80.9)|
|Subset #88||Neck extensors, trapezius, deltoid, gluteus maximus, iliopsoas, quadriceps, wrist extensors, ankle dorsiflexors||87.5 (79.6, 94.4)||79.0 (72.8, 84.1)|
|Subset #133||Neck extensors, deltoid, biceps, gluteus maximus, iliopsoas, quadriceps, wrist flexors, ankle dorsiflexors||87.3 (77.1, 94.8)||77.3 (71.1, 81.8)|
|Subset #136||Neck extensors, deltoid, biceps, gluteus maximus, iliopsoas, quadriceps, wrist extensors, ankle dorsiflexors||87.3 (76.7, 94.4)||77.3 (70.1, 81.8)|
|Subset #142||Neck extensors, deltoid, biceps, gluteus medius, iliopsoas, quadriceps, wrist extensors, ankle dorsiflexors||89.0 (78.4, 94.3)||78.5 (70.3, 83.4)|
All 144 MMT6 subsets and 96 MMT8 abbreviated subsets significantly correlated with the Total MMT score in juvenile and adult patients with IIM (rs > 0.70, P < 0.0001). Of the top one-third subsets with the strongest correlation, 18 MMT6 and 19 MMT8 subsets were common to both juvenile and adult patients with IIM and selected for further analyses. Among the top 8 abbreviated subsets in the nominal group technique exercise, rs ranged from 0.95 to 0.99. In juvenile IIM rs = 0.96 and in adult IIM rs = 0.91 between Proximal and Total MMT scores (for all, P < 0.0001) (Table 3).
|MMT score||Internal reliability (Spearman's rank correlation with Total MMT)†||Internal consistency (Cronbach's alpha)||Intrarater reliability (Spearman's rank correlation coefficient)‡||Interrater reliability (Kendall's W)§|
|Juvenile IIM (n = 73)||Adult IIM (n = 45)||Juvenile IIM (n = 73)||Adult IIM (n = 45)||Juvenile IIM (n = 10)||Juvenile IIM (n = 9)|
|MMT6 subset #118||0.97||0.95||0.85||0.78||0.93¶||0.75|
Of the subsets selected for further analysis, 16 of 18 MMT6 subsets and all 19 MMT8 subsets had Cronbach's α = 0.70–0.91 (Table 3), suggesting good internal consistency. Cronbach's alpha values were higher in the top 8 abbreviated subsets, with α ≥ 0.78 and even higher in Total and Proximal MMT (α = 0.93–0.97).
Spearman's rank correlation, a measure of intrarater reliability, was rs = 0.90 for Total and rs = 0.95 for Proximal MMT in patients with juvenile IIM. The top 8 MMT subsets had comparable intrarater reliability, with rs ranging from 0.84 to 0.93 (for all, P < 0.01) (Table 3). Kendall's W, a measure of interrater reliability, was 0.70 for Total MMT and 0.76 for Proximal MMT in patients with juvenile IIM. Most of the top 8 MMT subsets had a Kendall's W of 0.71–0.75. A few fell just lower at 0.68–0.69 (Table 3).
Total and Proximal MMT scores correlated highly with each other, as well as with physical function assessed by the CMAS, and moderately with physician global activity, functional disability measured by the C-HAQ/HAQ, and MRI (a score reflecting an average of activity and damage) (Table 4). Total and Proximal MMT scores correlated highly with distal and axial MMT scores (rs = 0.73–0.88, P < 0.0001 in juvenile IIM; rs = 0.40–0.74, P < 0.008 in adult PM/DM). In adult PM/DM, Total and Proximal MMT scores correlated moderately with lactate dehydrogenase (rs = −0.45 to −0.50, P = 0.007–0.016), and weakly, but significantly, with creatine kinase (CK), aldolase, and aspartate aminotransferase (rs = −0.30 to −0.38, P = 0.012–0.055). MMT scores did not correlate significantly with serum muscle enzymes in patients with juvenile IIM. Prednisone dose correlated with Total, Proximal, and best MMT subset scores in juvenile IIM (rs = −0.33 to −0.39, P ≤ 0.005), but not in adult PM/DM (rs = −0.11 to −0.22, P > 0.18).
|MMT score||Total MMT||Proximal MMT||Physician Global Activity||C-HAQ||ADL||CMAS||MRI STIR + T1-weighted average||Creatine kinase|
|Juvenile IIM (n = 73)||Adult IIM (n = 45)||Juvenile IIM (n = 73)||Adult IIM (n = 45)||Juvenile IIM (n = 73)||Adult IIM (n = 44)†||Juvenile IIM (n = 58)||Adult IIM (n = 45)||Juvenile IIM (n = 69)||Juvenile IIM (n = 31)†||Adult IIM (n = 45)†||Juvenile IIM (n = 73)‡||Adult IIM (n = 45)†|
|MMT6 subset #118||0.97||0.95||0.92||0.81||−0.54||−0.31||−0.68||0.61||0.71||−0.49||−0.43||−0.13||−0.36|
For the best MMT6 and MMT8 subsets, convergent construct validity was comparable among each other and with the Total MMT score, and was generally equal to or better than the Proximal MMT score (Table 4). In a few instances, the abbreviated subsets did not have as strong a correlation with other myositis assessment measures as the Total MMT score had. This included correlation of the subsets with physician global activity and with serum CK in patients with juvenile IIM (MMT8 subset numbers 13, 55, and 58), and with the Convery ADL in adult PM/DM (Table 4). In some cases, subsets correlated better than the Total MMT score, including with physician global activity in adult IIM, with the C-HAQ (MMT6 subset number 118 and MMT8 subset number 88), or with thigh muscle MRI in adult IIM (many MMT8 subsets and the Proximal MMT score) (Table 4).
The SRM for Total MMT was 0.56 in juvenile IIM and 0.75 in patients with adult PM/DM. Using relative efficiencies, the Proximal MMT score was slightly less responsive relative to the Total MMT score in juvenile IIM, but slightly more responsive in adults with PM/DM (Table 5). Several subsets were more responsive than the Total MMT score in both juvenile and adult patients with IIM, including MMT8 subset numbers 58, 88, 136, and 142 (Table 5). Only one abbreviated subset, MMT8 subset number 13, was slightly less responsive than the Total MMT score in both juvenile and adult IIM, but it had a relative efficiency >0.90 in both populations. MMT6 subset number 18 and MMT8 subset number 55 were responsive in the juvenile but not in the adult IIM cohort (Table 5).
|MMT score||Juvenile IIM (n = 49)||Adult IIM (n = 24)|
|MMT6 subset #118||1.25||0.94|
The top-ranked MMT subset chosen for use in therapeutic trials and interim assessments in clinical studies was MMT8 subset number 58, which included neck flexors, deltoid, biceps, gluteus maximus, gluteus medius, quadriceps, wrist extensors, and ankle dorsiflexors (Table 6), followed closely by MMT8 subset number 13, with iliopsoas substituted for gluteus medius and wrist flexors substituted for wrist extensors. MMT8 subset number 55 ranked third, with wrist flexors substituted for wrist extensors. These subsets were ranked highest because they provided the best balance of function between upper and lower extremities; they consisted of muscle groups frequently affected in myositis, including lower extremity muscles more so than upper extremity muscles. The top-ranking subset was more responsive than the Total MMT score, and these subsets performed well in both juvenile and adult IIM.
|MMT8 subsets||Votes||Rank order|
In this study, using a 0–10-point MMT scale (8) to test 24 bilateral proximal, axial, and distal muscles, we preliminarily validated the Total and Proximal MMT scores by demonstrating internal reliability and consistency, rater reliability, convergent construct validity, and responsiveness in assessing strength in patients with adult PM/DM and juvenile IIM. Most therapeutic PM/DM trials have used unvalidated 5-point MMT scales that included varying sets of proximal muscles and some distal muscle groups (6, 25). The lack of consistency in scoring or in the content of specific muscle groups makes it difficult to compare data across studies. More recent trials used the 0–10-point MMT of a total score of muscles, because this expanded scale is thought to enhance the sensitivity of strength testing, particularly in the fair-to-good muscle strength range (8). Inclusion of a uniform set of muscle groups with standardized scoring (5) that has been validated should provide clinicians and researchers with more reliable, standardized MMT data in patients with adult and juvenile PM/DM.
Total MMT testing has several disadvantages, including that testing of 24 muscle groups is time consuming, patients often experience fatigue during the testing, patients occasionally experience muscle pain that makes muscle testing unpleasant and stressful, and children frequently are not able to cooperate for the entire 24 muscle group test, resulting in incomplete results or inconsistent strength evaluation (19). Testing numerous muscles and changing test positions can add to overall and local muscle fatigue and decrease test reliability due to decreased muscle force. Documenting patterns and symmetry of weakness in our previous study in juvenile and adult PM/DM (5) allowed us to evaluate Total and Proximal MMT scores as measures of strength testing in patients with myositis. In the present study, to approximate the Total MMT score while reducing its disadvantages, we also examined unilateral muscle subsets of 6 or 8 muscle groups that included proximal, distal, and axial muscles with more emphasis on proximal and lower extremity muscle groups. We postulated that testing a subset of muscles might be more sensitive to change over time and have better reliability. A validated abbreviated MMT score that includes fewer muscles and 3 instead of 5 test positions is not only more efficient but might also improve testing reliability.
The explicit purposes of developing these abbreviated muscle subsets are to enhance consistency among international myositis therapeutic trials with adult and pediatric patients and for brief clinical followup visits. By nominal group technique consensus formation, we rank-ordered the best 8 muscle group subsets containing unilateral-representative axial muscle groups, as well as upper and lower extremity proximal and distal muscle groups (Table 6). These MMT subsets had results comparable with those of Total and Proximal MMT scores for internal reliability, consistency, and construct validity. These 8 muscle group subsets were in a comparable range for rater reliability in patients with juvenile IIM. They were slightly more responsive than the Total or Proximal MMT score, which is especially important for therapeutic trials. The 8 muscle group subsets performed better than the 6 muscle group subsets, and no 6 muscle group subsets achieved consensus for replacing the Total MMT score. The top-rated 8 muscle group subsets also had acceptable face validity, included frequently involved muscle groups, and were felt to be easier to test, even in patients with joint contractures or calcinosis. However, they require prospective validation, including in therapeutic trials, to further define their performance characteristics in other populations.
MMT has been widely used to test muscle strength in the IIM (26), and is a preferred method of testing strength in patients with IIM because weakness encompasses the full range, from no to full strength, and MMT tests this full range; because it is simple, easy to use, and available for all examiners internationally; and because validated, standardized MMT scales allow comparison between studies (19).
Although validation of the Total MMT and MMT8 subsets in this study is well substantiated, the study had some limitations. Although we included the most common forms of IIM (adult and juvenile PM/DM), we did not include patients with inclusion body myositis and myositis associated with malignancy. Of the 45 adult patients, 24 were accepted into therapeutic trials for refractory disease and may have been weaker than those regularly encountered in clinical practice. Also, we did not include recently diagnosed (possibly weaker) patients. However, the remaining patients were either adults who did not meet weakness criteria for therapeutic trials or patients with juvenile IIM who were enrolled in a natural history study. These patients might be more representative of patients at other centers. Another potential limitation is that our patients enrolled from rheumatology centers that specialize in IIM where the therapists have many years of myositis testing experience, which could result in better-quality data than if they were enrolled at clinics serving fewer patients with IIM.
Although the MMT8 has several advantages, the top-rated MMT8 subsets might not yet contain the most representative muscles. The highest ranking subset contained 4 of the 5 weakest muscles identified in the IIM (neck flexor, deltoid, gluteus maximus, and gluteus medius), but it did not include hip flexors, the other weakest muscle group in adult and patients with juvenile IIM (5). It did include the proximal quadriceps and 2 distal muscles (ankle dorsiflexors and wrist extensors), which are functionally important (27, 28). Clinically, some muscles are easier to test than others; those that require position changes for testing are generally not as popular with clinicians. More research is needed to determine and validate the best MMT8 subsets.
The current study is the only one to our knowledge to validate the bilateral Total MMT score consisting of 24 muscles, a Proximal MMT score including 14 muscle groups, and unilateral 8 muscle group MMT subsets, chosen by consensus formation, for patients with adult and juvenile PM/DM. The MMT8 subsets, which take less time and involve less patient effort, performed as well as or better than Total MMT in terms of responsiveness, content validity, and construct validity. This preliminary validation of Total MMT, Proximal MMT, and an abbreviated subset of 8 key muscles can also help standardize the testing and reporting of MMT, an important measure of strength and a core outcome measure (9, 10) for myositis clinical studies.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Rider had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Rider, Giannini, Feldman, Wright, Lindsley, Pachman, Miller, Hicks.
Acquisition of data. Rider, Jain, Smith, Whitney-Mahoney, Feldman, Wright, Lindsley, Pachman, Villalba, Bowyer, Plotz, Hicks.
Analysis and interpretation of data. Rider, Koziol, Giannini, Lovell, Miller, Hicks.
The authors thank Dr. Robert Wesley for statistical support; Drs. Elizabeth Adams, Holly Cintas, and Kathleen Haines, and Ms Jacqueline Gilbert for assistance with data collection; and Drs. Holly Cintas and Jerome Danoff for their critical reading of the manuscript.