MAGNETIC RESONANCE IMAGING PHENOTYPING OF BECKER MUSCULAR DYSTROPHY NEDA FARIDIAN-ARAGH, MD,1 KATHRYN R. WAGNER, MD, PhD,2,3,4 DORIS G. LEUNG, MD,2 and JOHN A. CARRINO, MD, MPH1 1
Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins School of Medicine, 601 North Caroline Street, JHOC 5165, Baltimore, Maryland 21287, USA 2 Center for Genetic Muscle Disorders, Kennedy Krieger Institute, Baltimore, Maryland, USA 3 Department of Neurology and Neuroscience, The Johns Hopkins School of Medicine, Baltimore, Maryland, USA 4 Department of Neuroscience, The Johns Hopkins School of Medicine, Baltimore, Maryland, USA Accepted 19 March 2014
ABSTRACT: Introduction: There is little information on magnetic resonance imaging (MRI) phenotypes of Becker muscular dystrophy (BMD). This study presents the MRI phenotyping of the upper and lower extremities of a large cohort of BMD patients. Methods: In this retrospective study, MRI images of 33 BMD subjects were evaluated for severity, distribution, and symmetry of involvement. Results: Teres major, triceps long head, biceps brachii long head, gluteus maximus, gluteus medius, vasti, adductor longus, adductor magnus, semitendinosus, semimembranosus, and biceps femoris muscles showed the highest severity and frequency of involvement. All analyzed muscles had a high frequency of symmetric involvement. There was significant variability of involvement between muscles within some muscle groups, most notably the arm abductors, posterior arm muscles, medial thigh muscles, and lateral hip rotators. Conclusions: This study showed a distinctive pattern of involvement of extremity muscles in BMD subjects. Muscle Nerve 50: 962–967, 2014
Becker muscular dystrophy (BMD) is a distinct genetic muscular disorder. However, it overlaps phenotypically with other muscular dystrophies and chronic myopathies in which magnetic resonance imaging (MRI) could potentially differentiate these disorders if an MRI phenotype is well characterized. Further, MRI assessment of the degree of muscle involvement can be of value in measuring the treatment response of BMD patients in clinical trials. Although MRI has been shown to be a useful tool in the diagnosis and description of several subtypes of muscular dystrophy,1–3 there is little information on the MRI phenotype of BMD.4 In this study we evaluated the involvement of lower and upper extremities in BMD patients. Additional Supporting Information may be found in the online version of this article. Abbreviations: BMD, Becker muscular dystrophy; MRI, magnetic resonance imaging; STIR, short T1 inversion recovery; TE, echo time; TR, repetition time Key words: Becker muscular dystrophy; upper extremity; lower extremity; fatty infiltration; muscular dystrophy; MRI; phenotype Disclosure: J.A.C. serves as a research consultant for General Electric Corporation. This study was supported by research grants from Siemens AG and Carestream Health, Inc. (to J.A.C.). Correspondence to: J.A. Carrino; e-mail: [email protected] C 2014 Wiley Periodicals, Inc. V
Published online 22 March 2014 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24246
962
MRI Phenotyping of BMD
METHODS
This retrospective study was performed on a cohort of subjects in the MYO-029 trial, a double-blind, placebo-controlled clinical trial of a neutralizing antibody to myostatin in 116 adults with muscular dystrophy (including BMD, facioscapulohumeral muscular dystrophy, and limb-girdle muscular dystrophy). The inclusion and exclusion criteria for this trial have been described elsewhere.5 The retrospective analysis of the MR images collected during this study was approved by the institutional review board of The Johns Hopkins School of Medicine. The baseline MR images of all 116 subjects prior to treatment randomization were included in the study and were evaluated by a radiologist who was blinded to the diagnoses. Subsequently, the images of the 33 subjects with a confirmed molecular diagnosis of BMD were selected for further analysis. Subjects.
Muscle Imaging and Evaluation. Axial proton density [echo time (TE) 13.77 ms, repetition time (TR) 6000 ms, number of excitations 5 1, matrix 256 3 256, and acquisition time 1.2 minutes], and dual echo T2-weighted non–fat-saturated (TE 13.144 ms, TR 4900 ms, number of excitations 5 1, matrix 256 3 256, and acquisition time 2.42 minutes) MR images of bilateral shoulders, arms, pelvis, and thighs were obtained by scanners (GE Medical Systems) with magnetic field strengths of 1.5 T, slice thicknesses of 5 mm, acquisition matrices of 256 3 256, fields of view 90 cm2, and body receiver coils, with the patients lying supine. (There was variation in the fields of view between the 10 participating centers, and the average number is provided here.) The acquired images were evaluated by a radiologist. Muscles that were visualized in their entirety were scored for the presence or absence of fatty infiltration and degree of involvement (Tables 1 and 2). Muscles that were only visualized partially in a subset of subjects due to smaller field of view were evaluated for presence or absence of involvement but were not scored or evaluated for symmetry (refer to Supplementary Tables S1 and S2, available online). The scoring system used to rank degree of involvement was a MUSCLE & NERVE
December 2014
Table 1. Upper extremity muscle involvement by MRI in 33 subjects with Becker muscular dystrophy. Muscle group
Muscle
Arm abductor
Deltoid,anterior Deltoid, medial Deltoid, posterior Teres major Supraspinatus Infraspinatus Subscapularis Teres minor Triceps, long head Triceps, lateral head Triceps, medial head Coracobrachialis Biceps brachii, long head Biceps brachii, short head Brachialis
Rotator cuff
Posterior arm
Anterior arm
0 0 1 31 19 18 19 14 32 19 12 6 26 12 20
NI
NS
(0%) (0%) (3%) (93%) (57%) (54%) (57%) (42%) (96%) (57%) (36%) (18%) (78%) (36%) (60%)
(100%) (80%) (88%) (88%) (94%) (85%) (90%) (84%) (83%) (83%) (80%) (83%) (75%)
1 25 15 16 18 12 29 16 10 5 21 10 15
Av 0 0 0.04 2.5 1 0.54 0.97 0.47 2.3 0.77 0.38 0.24 1.48 0.45 1.04
NI, number of subjects exhibiting involvement of muscle; NS, number of patients showing symmetric involvement of muscle; Av, average MRI score of all subjects including both extremities.
semiquantitative 5-point scale, a modification of the scale initially described by Lamminen,6 which has been used in prior imaging studies in muscular dystrophy (Table 3).1,7 Muscles that were only visualized partially due to limitations in the field of view were evaluated only for the presence or absence of fatty infiltration and classified as normal or abnormal.
Distribution of Muscle Involvement. An individual muscle was considered to be affected in a subject if changes consistent with fatty infiltration were observed in at least 1 side. Involvement was considered to be frequent if it was observed in 75% of subjects and infrequent if observed in 50% of the muscle volume Areas of high signal intensity that occupy >50%, but not the entire muscle volume Muscle volume is entirely replaced by areas of high signal intensity
3 4
considered symmetric if severity scores on the left and right sides were within 1 grade of each other. For example, if the score of a muscle on a side was 1, its involvement would be considered symmetric if the score of the contralateral muscle was 0, 1, or 2. The frequency of symmetry for each muscle is expressed as the ratio of the number of subjects who showed symmetric involvement versus the number of subjects who showed any involvement of that muscle. If a specific muscle was involved symmetrically in 75% of subjects, it was classified as highly symmetric.
(such as the coracobrachialis) were found to be involved less frequently in the cohort as a whole, whereas muscles with high average severity scores, such as the long head of biceps femoris, were affected more frequently (Fig. 1). The average severity scores also showed that there can be significant variability of involvement among muscles that are contained within the same anatomic groups and perform similar functions. This is best exemplified by the findings in the arm abductors (Fig. 2), where anterior and medial deltoid were uninvolved, whereas the teres major showed 93% frequency of involvement with a score of 2.5. Also, in the posterior arm, the average score of the long head of the triceps was much higher than scores of the medial and lateral heads. In the anterior arm, the average scores of the long head of biceps brachii and brachialis were higher than the average scores of the short head of the biceps brachii and coracobrachialis. In the lower extremity, the severity of involvement within muscle groups was more consistent. An exception was the medial
Severity of Muscle Involvement. The severity of involvement in individual muscles was determined by calculating the mean score for each muscle (both right and left) for all subjects in the cohort. RESULTS
The muscles of the upper and lower extremities of our 33 subjects (100% men, mean age 39.2 6 12.9 years, age range 18–66 years) were divided by anatomic groupings and are listed in Tables 1 and 2, respectively. The absolute numbers and percentages of subjects who had visualized fatty infiltration on MRI are listed. Except for anterior and medial head of the deltoid muscles, all muscles included in this analysis were affected, each in at least 1 patient. The frequency of involvement varied widely in different muscle groups, ranging between 0% and 100% of subjects (Tables 1 and 2). The following muscles were affected frequently (abnormal in 75% of subjects): teres major; long head of triceps; long head of biceps brachii; gluteus maximus; gluteus medius; all components of the quadriceps femoris; adductor longus; adductor magnus; semitendinosus; semimembranosus; and both heads of the biceps femoris. Tables 1 and 2 also list the numbers and percentages of subjects with symmetric involvement of each muscle. All muscles that were analyzed in this cohort showed a high frequency of symmetric involvement. The fifth columns in Tables 1 and 2 list the average score for each muscle across the cohort. Muscles that had lower average severity scores 964
MRI Phenotyping of BMD
FIGURE 1. Axial proton density images of 3 Becker muscular dystrophy patients at the level of mid-thigh (supine, 1.5 T, TE 13.768 ms, TR 6000 ms, slice thickness 5 mm) demonstrating mild, moderate, and severe (top, middle, and bottom images, respectively) fatty replacement of muscles. MUSCLE & NERVE
December 2014
FIGURE 2. Axial proton density imaging of the left shoulder (supine, 1.5 T, TE 13.768 ms, TR 6000 ms, slice thickness 5 mm) demonstrates extensive fatty infiltration and replacement of the teres major (thin arrow) with relative sparing of the deltoid (thick arrow).
thigh, where there was marked sparing of the sartorius and the lateral hip rotators (Fig. 3), with relative sparing of the obturator externus. Various muscles of the upper and lower extremities were only visualized partially and therefore were not used in the analysis of severity of involvement. These muscles are listed in Tables S1 and S2 (Supplementary Material). Within this subset of muscles, frequent involvement was seen in the latissimus dorsi and the brachioradialis.
It is worth noting, however, that greater symmetry was noted in lower extremity (16 of 19 muscles, 84%) than in upper extremity muscles (9 of 15 muscles, 60%). The greater degree of symmetry seen in the lower extremities may be due to a longer duration of involvement in these muscles, as many have reached the end-stage and are completely infiltrated by fat, thereby obscuring any initial asymmetric involvement. It is unclear why specific muscles, such as the coracobrachialis, would be affected asymmetrically in a genetic disease that presumably affects all skeletal muscles. It has been speculated that this may have a relationship to hand dominance, but the nature of this association is unclear. Although it may seem intuitive that the dominant hand would remain stronger through increased use, it is also possible that the arm used more frequently degenerates more quickly due to impaired muscle regeneration in the dystrophinopathies. Handedness of subjects was not documented in this study, but including this feature in future studies may be useful to confirm and elucidate this finding. Due to its high-contrast resolution in soft tissues, MRI can define muscle disease distribution and severity6 and can therefore identify patterns of muscle involvement in genetically distinct muscle disorders that have significant clinical overlap.8–10
DISCUSSION
In this study we have described the imaging phenotype of a cohort of 33 subjects with BMD investigated using MRI. This semiquantitative analysis of disease distribution and severity serves to define further the clinical characteristics observed in BMD. We found that the frequency and severity of muscle involvement in the lower limb was more severe than in the upper limb. This is consistent with the observed clinical progression of Becker muscular dystrophy, where abnormalities of gait are often the first reported symptom and usually precede involvement of the upper extremities. The majority of the muscles in the upper limb showed mild involvement. However, the teres major, the long head of the triceps, and the long head of biceps brachii were involved more frequently and severely, suggesting that these muscles are among the earliest involved in the upper extremity. The weakness seen in BMD is usually highly symmetric, as substantiated by our analysis, where all muscles showed highly symmetric involvement. MRI Phenotyping of BMD
FIGURE 3. Axial proton density images of the pelvis of a single subject (supine, 5 T, TE 13.768 ms, TR 6000 ms, slice thickness 5 mm). Arrows indicate sparing of the obturator internus (above) and obturator externus (below). MUSCLE & NERVE
December 2014
965
Previous studies have suggested that MRI may have advantages over clinical assessment in detection of symmetric or asymmetric involvement of individual muscles, especially those that function as part of a larger group.2,11 The muscles that comprise these groups cannot be tested individually in a clinical setting, and a greater understanding of the selectivity for specific muscles may improve our understanding of the natural progression of muscle involvement in BMD. In this study, the severity of muscle involvement within some anatomic groups, such as the rotator cuff and the anterior thigh, was quite homogeneous, although the degree of involvement in the latter group was markedly more severe. In several muscle groups, however, including the arm abductors, the posterior and anterior compartments of the upper arm, the medial thigh, and the lateral rotators of the hip, some muscles within the group were more affected than others. A better understanding of these patterns of involvement among individual muscles may help guide the diagnostic evaluation of patients with muscular dystrophy. We compared the results of our study with previously reported imaging findings in BMD. In 1985, de Visser et al. reported the results of lower limb imaging with computed tomography in 26 BMD patients.12 In that study, the extent of involvement correlated positively with the age of the subjects. As 46% (12 of 26) of their subjects were 25 years of age showed a similar extent of involvement in both studies. A study performed by Fischer et al.13 used MRI to assess 2 subjects with BMD. Involvement of the gluteus medius was reported to be mild to moderate, and involvement of the gluteus maximus was even less severe. At the level of the thigh, involvement of the anterior compartment was marked, and involvement of the posterior compartment was mild to moderate. Our results differ from these reports in that we found relatively severe involvement of the gluteal muscles (particularly the gluteus maximus). Furthermore, an examination of the average scores of the muscles comprising the anterior and posterior compartments of the thigh suggested greater involvement of the posterior compartment, with the long head of biceps femoris having the highest average score among all muscles in both compartments. Our results confirm the results of a recent study by Tasca et al.,4 who evaluated MRI data of pelvis and lower extremity muscles of 46 BMD patients. They reported a high frequency of involvement of the gluteus maximus, adductor 966
MRI Phenotyping of BMD
magnus, gluteus medius, biceps femoris long head, and semimembranosus muscles and a low frequency of involvement of obturator internus, obturator externus, and sartorius muscles. We also compared our results to imaging findings in Duchenne muscular dystrophy. In the study by Liu et al.,14 the sartorius, gracilis, semitendinosus, and semimembranosus showed the highest frequency of selective preservation among the muscles of the pelvis and lower limbs. Although in our study the sartorius and gracilis did not demonstrate a high frequency or severity of involvement, the semitendinosus and semimembranosus did show a high frequency and degree of abnormality. Liu et al. also reported asymmetric involvement of parts of the muscle bundles of the thigh in 55% of their subjects, in contrast to our study in which all affected muscles showed symmetric involvement. Subjects in their study were younger than those in our study, with an age range of 3–13 years. This could account at least partially for the large differences between the results. A comparison could not be made between the results of our study and the recent studies in which more quantitative methods of muscle characterization were used. To our knowledge, such an approach has not been used to evaluate BMD patients, and the muscle group evaluated by Arpan et al.15 in the evaluation of Duchenne muscular dystrophy was not characterized fully in our study. Our investigation has several limitations due to its design. Because it was a retrospective study, we were unable to obtain other potentially useful MR sequences, such as short T1 inversion recovery (STIR), which are used commonly to identify edema-like changes in muscle. Limitations in the field of view also prevented complete analysis of some muscle groups, particularly the distal extremities. All subjects in our study were >18 years of age. Because BMD is a progressive disease, imaging at an earlier disease stage could potentially identify muscle involvement that precedes clinically observable muscle weakness, demonstrate patterns of muscle involvement that are more specific to BMD, and therefore be of greater utility in early diagnostic stages. A greater understanding of detailed patterns of muscle involvement in large numbers of patients with muscular dystrophy may prove to be of diagnostic value. As more genetic causes of muscular dystrophy are identified and additional genetic tests become available commercially, the recognition of patterns of muscle involvement on noninvasive imaging can be used to narrow the differential diagnosis in individual patients and improve the yield of selective genetic testing. In addition, an understanding of MRI characteristics of this MUSCLE & NERVE
December 2014
disease will be important in future clinical trials if novel outcome measures such as imaging are used. The authors thank the investigators and subjects participating in the MYO-029 trial from which the baseline MRIs were obtained. REFERENCES 1. Jungbluth H, Davis MR, M€ uller C, Counsell S, Allsop J, Chattopadhyay A, et al. Magnetic resonance imaging of muscle in congenital myopathies associated with RYR1 mutations. Neuromuscul Disord 2004;14:785–790. 2. Olsen DB, Gideon P, Jeppesen TD, Vissing J. Leg muscle involvement in facioscapulohumeral muscular dystrophy assessed by MRI. J Neurol 2006;253:1437–1441. 3. Nagao H, Morimoto T, Sano N, Takahashi M, Nagai H, Tawa R. Magnetic resonance imaging of skeletal muscle in patients with Duchenne muscular dystrophy—serial axial and sagittal section studies. No To Hattatsu 1991;23:39–43. 4. Tasca G, Iannaccone E, Monforte M, Masciullo M, Bianco F, Laschena F, et al. Muscle MRI in Becker muscular dystrophy. Neuromuscul Disord 2012;22(suppl 2):S100–106. 5. Wagner KR, Fleckenstein JL, Amato AA, Barohn RJ, Bushby K, Escolar DM, et al. A phase I/II trial of MYO-029 in adult subjects with muscular dystrophy. Ann Neurol 2008;63:561–571. 6. Lamminen AE. Magnetic resonance imaging of primary skeletal muscle diseases: patterns of distribution and severity of involvement. Br J Radiol 1990;63:946–950.
MRI Phenotyping of BMD
7. Fischer D, Kley RA, Strach K, Meyer C, Sommer T, Eger K, et al. Distinct muscle imaging patterns in myofibrillar myopathies. Neurology 2008;71:758–765. 8. Mercuri E, Counsell S, Allsop J, Jungbluth H, Kinali M, Bonne G, et al. Selective muscle involvement on magnetic resonance imaging in autosomal dominant Emery–Dreifuss muscular dystrophy. Neuropediatrics 2002;33:10–14. 9. Mercuri E, Talim B, Moghadaszadeh B, Petit N, Brockington M, Counsell S, et al. Clinical and imaging findings in six cases of congenital muscular dystrophy with rigid spine syndrome linked to chromosome 1p (RSMD1). Neuromuscul Disord 2002;12:631–638. 10. Mercuri E, Lampe A, Allsop J, Knight R, Pane M, Kinali M, et al. Muscle MRI in Ullrich congenital muscular dystrophy and Bethlem myopathy. Neuromuscul Disord 2005;15:303–310. 11. Sookhoo S, Mackinnon I, Bushby K, Chinnery PF, Birchall D. MRI for the demonstration of subclinical muscle involvement in muscular dystrophy. Clin Radiol 2007;62:160–165. 12. de Visser M, Verbeeten B Jr. Computed tomography of the skeletal musculature in Becker-type muscular dystrophy and benign infantile spinal muscular atrophy. Muscle Nerve 1985;8:435–444. 13. Fischer D, Walter MC, Kesper K, Petersen JA, Aurino S, Nigro V, et al. Diagnostic value of muscle MRI in differentiating LGMD2I from other LGMDs. J Neurol 2005;252:538–547. 14. Liu GC, Jong YJ, Chiang CH, Jaw TS. Duchenne muscular dystrophy: MR grading system with functional correlation. Radiology 1993;186:475–480. 15. Arpan I, Forbes SC, Lott DJ, Senesac CR, Daniels MJ, et al. T2 mapping provides multiple approaches for the characterization of muscle involvement in neuromuscular diseases: a cross-sectional study of lower leg muscles in 5–15-year-old boys with Duchenne muscular dystrophy. NMR Biomed 2013;26:320–328.
MUSCLE & NERVE
December 2014
967
Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins School of Medicine, 601 North Caroline Street, JHOC 5165, Baltimore, Maryland 21287, USA 2 Center for Genetic Muscle Disorders, Kennedy Krieger Institute, Baltimore, Maryland, USA 3 Department of Neurology and Neuroscience, The Johns Hopkins School of Medicine, Baltimore, Maryland, USA 4 Department of Neuroscience, The Johns Hopkins School of Medicine, Baltimore, Maryland, USA Accepted 19 March 2014
ABSTRACT: Introduction: There is little information on magnetic resonance imaging (MRI) phenotypes of Becker muscular dystrophy (BMD). This study presents the MRI phenotyping of the upper and lower extremities of a large cohort of BMD patients. Methods: In this retrospective study, MRI images of 33 BMD subjects were evaluated for severity, distribution, and symmetry of involvement. Results: Teres major, triceps long head, biceps brachii long head, gluteus maximus, gluteus medius, vasti, adductor longus, adductor magnus, semitendinosus, semimembranosus, and biceps femoris muscles showed the highest severity and frequency of involvement. All analyzed muscles had a high frequency of symmetric involvement. There was significant variability of involvement between muscles within some muscle groups, most notably the arm abductors, posterior arm muscles, medial thigh muscles, and lateral hip rotators. Conclusions: This study showed a distinctive pattern of involvement of extremity muscles in BMD subjects. Muscle Nerve 50: 962–967, 2014
Becker muscular dystrophy (BMD) is a distinct genetic muscular disorder. However, it overlaps phenotypically with other muscular dystrophies and chronic myopathies in which magnetic resonance imaging (MRI) could potentially differentiate these disorders if an MRI phenotype is well characterized. Further, MRI assessment of the degree of muscle involvement can be of value in measuring the treatment response of BMD patients in clinical trials. Although MRI has been shown to be a useful tool in the diagnosis and description of several subtypes of muscular dystrophy,1–3 there is little information on the MRI phenotype of BMD.4 In this study we evaluated the involvement of lower and upper extremities in BMD patients. Additional Supporting Information may be found in the online version of this article. Abbreviations: BMD, Becker muscular dystrophy; MRI, magnetic resonance imaging; STIR, short T1 inversion recovery; TE, echo time; TR, repetition time Key words: Becker muscular dystrophy; upper extremity; lower extremity; fatty infiltration; muscular dystrophy; MRI; phenotype Disclosure: J.A.C. serves as a research consultant for General Electric Corporation. This study was supported by research grants from Siemens AG and Carestream Health, Inc. (to J.A.C.). Correspondence to: J.A. Carrino; e-mail: [email protected] C 2014 Wiley Periodicals, Inc. V
Published online 22 March 2014 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24246
962
MRI Phenotyping of BMD
METHODS
This retrospective study was performed on a cohort of subjects in the MYO-029 trial, a double-blind, placebo-controlled clinical trial of a neutralizing antibody to myostatin in 116 adults with muscular dystrophy (including BMD, facioscapulohumeral muscular dystrophy, and limb-girdle muscular dystrophy). The inclusion and exclusion criteria for this trial have been described elsewhere.5 The retrospective analysis of the MR images collected during this study was approved by the institutional review board of The Johns Hopkins School of Medicine. The baseline MR images of all 116 subjects prior to treatment randomization were included in the study and were evaluated by a radiologist who was blinded to the diagnoses. Subsequently, the images of the 33 subjects with a confirmed molecular diagnosis of BMD were selected for further analysis. Subjects.
Muscle Imaging and Evaluation. Axial proton density [echo time (TE) 13.77 ms, repetition time (TR) 6000 ms, number of excitations 5 1, matrix 256 3 256, and acquisition time 1.2 minutes], and dual echo T2-weighted non–fat-saturated (TE 13.144 ms, TR 4900 ms, number of excitations 5 1, matrix 256 3 256, and acquisition time 2.42 minutes) MR images of bilateral shoulders, arms, pelvis, and thighs were obtained by scanners (GE Medical Systems) with magnetic field strengths of 1.5 T, slice thicknesses of 5 mm, acquisition matrices of 256 3 256, fields of view 90 cm2, and body receiver coils, with the patients lying supine. (There was variation in the fields of view between the 10 participating centers, and the average number is provided here.) The acquired images were evaluated by a radiologist. Muscles that were visualized in their entirety were scored for the presence or absence of fatty infiltration and degree of involvement (Tables 1 and 2). Muscles that were only visualized partially in a subset of subjects due to smaller field of view were evaluated for presence or absence of involvement but were not scored or evaluated for symmetry (refer to Supplementary Tables S1 and S2, available online). The scoring system used to rank degree of involvement was a MUSCLE & NERVE
December 2014
Table 1. Upper extremity muscle involvement by MRI in 33 subjects with Becker muscular dystrophy. Muscle group
Muscle
Arm abductor
Deltoid,anterior Deltoid, medial Deltoid, posterior Teres major Supraspinatus Infraspinatus Subscapularis Teres minor Triceps, long head Triceps, lateral head Triceps, medial head Coracobrachialis Biceps brachii, long head Biceps brachii, short head Brachialis
Rotator cuff
Posterior arm
Anterior arm
0 0 1 31 19 18 19 14 32 19 12 6 26 12 20
NI
NS
(0%) (0%) (3%) (93%) (57%) (54%) (57%) (42%) (96%) (57%) (36%) (18%) (78%) (36%) (60%)
(100%) (80%) (88%) (88%) (94%) (85%) (90%) (84%) (83%) (83%) (80%) (83%) (75%)
1 25 15 16 18 12 29 16 10 5 21 10 15
Av 0 0 0.04 2.5 1 0.54 0.97 0.47 2.3 0.77 0.38 0.24 1.48 0.45 1.04
NI, number of subjects exhibiting involvement of muscle; NS, number of patients showing symmetric involvement of muscle; Av, average MRI score of all subjects including both extremities.
semiquantitative 5-point scale, a modification of the scale initially described by Lamminen,6 which has been used in prior imaging studies in muscular dystrophy (Table 3).1,7 Muscles that were only visualized partially due to limitations in the field of view were evaluated only for the presence or absence of fatty infiltration and classified as normal or abnormal.
Distribution of Muscle Involvement. An individual muscle was considered to be affected in a subject if changes consistent with fatty infiltration were observed in at least 1 side. Involvement was considered to be frequent if it was observed in 75% of subjects and infrequent if observed in 50% of the muscle volume Areas of high signal intensity that occupy >50%, but not the entire muscle volume Muscle volume is entirely replaced by areas of high signal intensity
3 4
considered symmetric if severity scores on the left and right sides were within 1 grade of each other. For example, if the score of a muscle on a side was 1, its involvement would be considered symmetric if the score of the contralateral muscle was 0, 1, or 2. The frequency of symmetry for each muscle is expressed as the ratio of the number of subjects who showed symmetric involvement versus the number of subjects who showed any involvement of that muscle. If a specific muscle was involved symmetrically in 75% of subjects, it was classified as highly symmetric.
(such as the coracobrachialis) were found to be involved less frequently in the cohort as a whole, whereas muscles with high average severity scores, such as the long head of biceps femoris, were affected more frequently (Fig. 1). The average severity scores also showed that there can be significant variability of involvement among muscles that are contained within the same anatomic groups and perform similar functions. This is best exemplified by the findings in the arm abductors (Fig. 2), where anterior and medial deltoid were uninvolved, whereas the teres major showed 93% frequency of involvement with a score of 2.5. Also, in the posterior arm, the average score of the long head of the triceps was much higher than scores of the medial and lateral heads. In the anterior arm, the average scores of the long head of biceps brachii and brachialis were higher than the average scores of the short head of the biceps brachii and coracobrachialis. In the lower extremity, the severity of involvement within muscle groups was more consistent. An exception was the medial
Severity of Muscle Involvement. The severity of involvement in individual muscles was determined by calculating the mean score for each muscle (both right and left) for all subjects in the cohort. RESULTS
The muscles of the upper and lower extremities of our 33 subjects (100% men, mean age 39.2 6 12.9 years, age range 18–66 years) were divided by anatomic groupings and are listed in Tables 1 and 2, respectively. The absolute numbers and percentages of subjects who had visualized fatty infiltration on MRI are listed. Except for anterior and medial head of the deltoid muscles, all muscles included in this analysis were affected, each in at least 1 patient. The frequency of involvement varied widely in different muscle groups, ranging between 0% and 100% of subjects (Tables 1 and 2). The following muscles were affected frequently (abnormal in 75% of subjects): teres major; long head of triceps; long head of biceps brachii; gluteus maximus; gluteus medius; all components of the quadriceps femoris; adductor longus; adductor magnus; semitendinosus; semimembranosus; and both heads of the biceps femoris. Tables 1 and 2 also list the numbers and percentages of subjects with symmetric involvement of each muscle. All muscles that were analyzed in this cohort showed a high frequency of symmetric involvement. The fifth columns in Tables 1 and 2 list the average score for each muscle across the cohort. Muscles that had lower average severity scores 964
MRI Phenotyping of BMD
FIGURE 1. Axial proton density images of 3 Becker muscular dystrophy patients at the level of mid-thigh (supine, 1.5 T, TE 13.768 ms, TR 6000 ms, slice thickness 5 mm) demonstrating mild, moderate, and severe (top, middle, and bottom images, respectively) fatty replacement of muscles. MUSCLE & NERVE
December 2014
FIGURE 2. Axial proton density imaging of the left shoulder (supine, 1.5 T, TE 13.768 ms, TR 6000 ms, slice thickness 5 mm) demonstrates extensive fatty infiltration and replacement of the teres major (thin arrow) with relative sparing of the deltoid (thick arrow).
thigh, where there was marked sparing of the sartorius and the lateral hip rotators (Fig. 3), with relative sparing of the obturator externus. Various muscles of the upper and lower extremities were only visualized partially and therefore were not used in the analysis of severity of involvement. These muscles are listed in Tables S1 and S2 (Supplementary Material). Within this subset of muscles, frequent involvement was seen in the latissimus dorsi and the brachioradialis.
It is worth noting, however, that greater symmetry was noted in lower extremity (16 of 19 muscles, 84%) than in upper extremity muscles (9 of 15 muscles, 60%). The greater degree of symmetry seen in the lower extremities may be due to a longer duration of involvement in these muscles, as many have reached the end-stage and are completely infiltrated by fat, thereby obscuring any initial asymmetric involvement. It is unclear why specific muscles, such as the coracobrachialis, would be affected asymmetrically in a genetic disease that presumably affects all skeletal muscles. It has been speculated that this may have a relationship to hand dominance, but the nature of this association is unclear. Although it may seem intuitive that the dominant hand would remain stronger through increased use, it is also possible that the arm used more frequently degenerates more quickly due to impaired muscle regeneration in the dystrophinopathies. Handedness of subjects was not documented in this study, but including this feature in future studies may be useful to confirm and elucidate this finding. Due to its high-contrast resolution in soft tissues, MRI can define muscle disease distribution and severity6 and can therefore identify patterns of muscle involvement in genetically distinct muscle disorders that have significant clinical overlap.8–10
DISCUSSION
In this study we have described the imaging phenotype of a cohort of 33 subjects with BMD investigated using MRI. This semiquantitative analysis of disease distribution and severity serves to define further the clinical characteristics observed in BMD. We found that the frequency and severity of muscle involvement in the lower limb was more severe than in the upper limb. This is consistent with the observed clinical progression of Becker muscular dystrophy, where abnormalities of gait are often the first reported symptom and usually precede involvement of the upper extremities. The majority of the muscles in the upper limb showed mild involvement. However, the teres major, the long head of the triceps, and the long head of biceps brachii were involved more frequently and severely, suggesting that these muscles are among the earliest involved in the upper extremity. The weakness seen in BMD is usually highly symmetric, as substantiated by our analysis, where all muscles showed highly symmetric involvement. MRI Phenotyping of BMD
FIGURE 3. Axial proton density images of the pelvis of a single subject (supine, 5 T, TE 13.768 ms, TR 6000 ms, slice thickness 5 mm). Arrows indicate sparing of the obturator internus (above) and obturator externus (below). MUSCLE & NERVE
December 2014
965
Previous studies have suggested that MRI may have advantages over clinical assessment in detection of symmetric or asymmetric involvement of individual muscles, especially those that function as part of a larger group.2,11 The muscles that comprise these groups cannot be tested individually in a clinical setting, and a greater understanding of the selectivity for specific muscles may improve our understanding of the natural progression of muscle involvement in BMD. In this study, the severity of muscle involvement within some anatomic groups, such as the rotator cuff and the anterior thigh, was quite homogeneous, although the degree of involvement in the latter group was markedly more severe. In several muscle groups, however, including the arm abductors, the posterior and anterior compartments of the upper arm, the medial thigh, and the lateral rotators of the hip, some muscles within the group were more affected than others. A better understanding of these patterns of involvement among individual muscles may help guide the diagnostic evaluation of patients with muscular dystrophy. We compared the results of our study with previously reported imaging findings in BMD. In 1985, de Visser et al. reported the results of lower limb imaging with computed tomography in 26 BMD patients.12 In that study, the extent of involvement correlated positively with the age of the subjects. As 46% (12 of 26) of their subjects were 25 years of age showed a similar extent of involvement in both studies. A study performed by Fischer et al.13 used MRI to assess 2 subjects with BMD. Involvement of the gluteus medius was reported to be mild to moderate, and involvement of the gluteus maximus was even less severe. At the level of the thigh, involvement of the anterior compartment was marked, and involvement of the posterior compartment was mild to moderate. Our results differ from these reports in that we found relatively severe involvement of the gluteal muscles (particularly the gluteus maximus). Furthermore, an examination of the average scores of the muscles comprising the anterior and posterior compartments of the thigh suggested greater involvement of the posterior compartment, with the long head of biceps femoris having the highest average score among all muscles in both compartments. Our results confirm the results of a recent study by Tasca et al.,4 who evaluated MRI data of pelvis and lower extremity muscles of 46 BMD patients. They reported a high frequency of involvement of the gluteus maximus, adductor 966
MRI Phenotyping of BMD
magnus, gluteus medius, biceps femoris long head, and semimembranosus muscles and a low frequency of involvement of obturator internus, obturator externus, and sartorius muscles. We also compared our results to imaging findings in Duchenne muscular dystrophy. In the study by Liu et al.,14 the sartorius, gracilis, semitendinosus, and semimembranosus showed the highest frequency of selective preservation among the muscles of the pelvis and lower limbs. Although in our study the sartorius and gracilis did not demonstrate a high frequency or severity of involvement, the semitendinosus and semimembranosus did show a high frequency and degree of abnormality. Liu et al. also reported asymmetric involvement of parts of the muscle bundles of the thigh in 55% of their subjects, in contrast to our study in which all affected muscles showed symmetric involvement. Subjects in their study were younger than those in our study, with an age range of 3–13 years. This could account at least partially for the large differences between the results. A comparison could not be made between the results of our study and the recent studies in which more quantitative methods of muscle characterization were used. To our knowledge, such an approach has not been used to evaluate BMD patients, and the muscle group evaluated by Arpan et al.15 in the evaluation of Duchenne muscular dystrophy was not characterized fully in our study. Our investigation has several limitations due to its design. Because it was a retrospective study, we were unable to obtain other potentially useful MR sequences, such as short T1 inversion recovery (STIR), which are used commonly to identify edema-like changes in muscle. Limitations in the field of view also prevented complete analysis of some muscle groups, particularly the distal extremities. All subjects in our study were >18 years of age. Because BMD is a progressive disease, imaging at an earlier disease stage could potentially identify muscle involvement that precedes clinically observable muscle weakness, demonstrate patterns of muscle involvement that are more specific to BMD, and therefore be of greater utility in early diagnostic stages. A greater understanding of detailed patterns of muscle involvement in large numbers of patients with muscular dystrophy may prove to be of diagnostic value. As more genetic causes of muscular dystrophy are identified and additional genetic tests become available commercially, the recognition of patterns of muscle involvement on noninvasive imaging can be used to narrow the differential diagnosis in individual patients and improve the yield of selective genetic testing. In addition, an understanding of MRI characteristics of this MUSCLE & NERVE
December 2014
disease will be important in future clinical trials if novel outcome measures such as imaging are used. The authors thank the investigators and subjects participating in the MYO-029 trial from which the baseline MRIs were obtained. REFERENCES 1. Jungbluth H, Davis MR, M€ uller C, Counsell S, Allsop J, Chattopadhyay A, et al. Magnetic resonance imaging of muscle in congenital myopathies associated with RYR1 mutations. Neuromuscul Disord 2004;14:785–790. 2. Olsen DB, Gideon P, Jeppesen TD, Vissing J. Leg muscle involvement in facioscapulohumeral muscular dystrophy assessed by MRI. J Neurol 2006;253:1437–1441. 3. Nagao H, Morimoto T, Sano N, Takahashi M, Nagai H, Tawa R. Magnetic resonance imaging of skeletal muscle in patients with Duchenne muscular dystrophy—serial axial and sagittal section studies. No To Hattatsu 1991;23:39–43. 4. Tasca G, Iannaccone E, Monforte M, Masciullo M, Bianco F, Laschena F, et al. Muscle MRI in Becker muscular dystrophy. Neuromuscul Disord 2012;22(suppl 2):S100–106. 5. Wagner KR, Fleckenstein JL, Amato AA, Barohn RJ, Bushby K, Escolar DM, et al. A phase I/II trial of MYO-029 in adult subjects with muscular dystrophy. Ann Neurol 2008;63:561–571. 6. Lamminen AE. Magnetic resonance imaging of primary skeletal muscle diseases: patterns of distribution and severity of involvement. Br J Radiol 1990;63:946–950.
MRI Phenotyping of BMD
7. Fischer D, Kley RA, Strach K, Meyer C, Sommer T, Eger K, et al. Distinct muscle imaging patterns in myofibrillar myopathies. Neurology 2008;71:758–765. 8. Mercuri E, Counsell S, Allsop J, Jungbluth H, Kinali M, Bonne G, et al. Selective muscle involvement on magnetic resonance imaging in autosomal dominant Emery–Dreifuss muscular dystrophy. Neuropediatrics 2002;33:10–14. 9. Mercuri E, Talim B, Moghadaszadeh B, Petit N, Brockington M, Counsell S, et al. Clinical and imaging findings in six cases of congenital muscular dystrophy with rigid spine syndrome linked to chromosome 1p (RSMD1). Neuromuscul Disord 2002;12:631–638. 10. Mercuri E, Lampe A, Allsop J, Knight R, Pane M, Kinali M, et al. Muscle MRI in Ullrich congenital muscular dystrophy and Bethlem myopathy. Neuromuscul Disord 2005;15:303–310. 11. Sookhoo S, Mackinnon I, Bushby K, Chinnery PF, Birchall D. MRI for the demonstration of subclinical muscle involvement in muscular dystrophy. Clin Radiol 2007;62:160–165. 12. de Visser M, Verbeeten B Jr. Computed tomography of the skeletal musculature in Becker-type muscular dystrophy and benign infantile spinal muscular atrophy. Muscle Nerve 1985;8:435–444. 13. Fischer D, Walter MC, Kesper K, Petersen JA, Aurino S, Nigro V, et al. Diagnostic value of muscle MRI in differentiating LGMD2I from other LGMDs. J Neurol 2005;252:538–547. 14. Liu GC, Jong YJ, Chiang CH, Jaw TS. Duchenne muscular dystrophy: MR grading system with functional correlation. Radiology 1993;186:475–480. 15. Arpan I, Forbes SC, Lott DJ, Senesac CR, Daniels MJ, et al. T2 mapping provides multiple approaches for the characterization of muscle involvement in neuromuscular diseases: a cross-sectional study of lower leg muscles in 5–15-year-old boys with Duchenne muscular dystrophy. NMR Biomed 2013;26:320–328.
MUSCLE & NERVE
December 2014
967
