|Ahead of print publication
Is polymyositis a rare or over-diagnosed entity? A descriptive follow-up study of patients initially admitted in a tertiary center with the diagnosis of polymyositis
Pablo Arturo Olivo Pallo, Matheus Santos Rodrigues Silva, Fernando Henrique Carlos de Souza, Samuel Katsuyuki Shinjo
Division of Rheumatology, Faculdade de Medicina FMUSP, Universidade de Sao Paulo, SP, Brazil
|Date of Submission||28-Apr-2020|
|Date of Acceptance||29-Jul-2020|
Samuel Katsuyuki Shinjo,
Division of Rheumatology, Faculdade de Medicina, Laboratorio de Miopatias Inflamatorias, Universidade de Sao Paulo, CEP 01246-903, Cerqueira Cesar, SP
Source of Support: None, Conflict of Interest: None
Objective: This study aimed to describe a significant sample of patients with an initial diagnosis early-stage polymyositis (PM) in who were reclassified to different diagnoses after new evaluations during follow-ups.
Subjects and Methods: In a single-center descriptive study conducted from 2000 to 2019, 108 patients with an initial presumptive diagnosis of PM were admitted to our tertiary center and were reevaluated during follow-up. Patients with promptly suspected or definite initial diagnoses of other autoimmune or hereditary myopathies were excluded.
Results: The mean age of the 108 patients when admitted initially into our center was 45.3 ± 15.4 years; they were predominantly female (70.4%) and Caucasian (81.5%). During a median follow-up period of 4.0 years and constant diagnostic reinvestigation, only 30 (27.7%) out of the 108 participants maintained the PM diagnosis, whereas the other patients were re-diagnosed with immune-mediated necrotizing myopathy (20.4%), inclusion body myositis (18.5%), muscular dystrophy 13.0% as (13.0%), antisynthetase syndrome (8.3%), metabolic myopathies (5.6%), and other muscle diseases (7.3%). The initial clinical and laboratory parameters were not distinguishable between the PM and reclassified patients. Twenty-two of 30 patients with PM continued follow-up in our service, all with clinical remission or complete clinical response, full recovery of muscle strength, and normalization of the serum levels for muscle enzymes.
Conclusions: Patients with PM should be reevaluated constantly, as PM can mimic other muscle diseases. Notably, only one-third of our large patient samples maintained diagnoses of PM until the end of the present study's data collection.
Keywords: Differential diagnoses, inflammatory myopathies, myositis, polymyositis, systemic autoimmune myopathies
|How to cite this URL:|
Olivo Pallo PA, Rodrigues Silva MS, de Souza FH, Shinjo SK. Is polymyositis a rare or over-diagnosed entity? A descriptive follow-up study of patients initially admitted in a tertiary center with the diagnosis of polymyositis. Indian J Rheumatol [Epub ahead of print] [cited 2020 Oct 27]. Available from: https://www.indianjrheumatol.com/preprintarticle.asp?id=298039
| Introduction|| |
Polymyositis (PM) is characterized by the presence of symmetrical, progressive, and predominantly proximal muscle weakness.,, The laboratory findings associated with PM are characterized by the elevated serum levels of muscle enzymes,,, a pure myopathic pattern in electroneuromyography, and the presence of endomysial lymphomononuclear infiltrate or invading nonnecrotic muscle fibers in a muscle biopsy.,
The definition of PM is based on the Bohan and Peter classification criteria, and more recently, on the classification criteria proposed by the European League Against Rheumatism/American College of Rheumatology (EULAR/ACR) 2017, which increased the specificity of the classification and definition of systemic autoimmune myopathies (SAMs), including PM. However, despite this effort, a diagnosis of PM is one of exclusion, as many entities have similar clinical characteristics, including muscular weakness and muscle enzyme elevation, leading to a misdiagnosis of PM in the early stage.,,,,,,,,,,,,,,,,,,,,,,,, Thus, there are case reports and case series of patients with initial clinical manifestations of muscle weakness diagnosed as PM in early stage; however, the patients were reevaluated and changed to other diagnoses during follow-up.,,,,,,,,,,,,,,,,,
Therefore, the aim of the present study was to describe a representative sample of patients who were admitted to our tertiary service with initial diagnoses of PM, many of whom were reclassified to different diagnoses during follow-ups. We also performed a literature review primarily focused on differential diagnoses of PM.
| Subjects and Methods|| |
In this retrospective cohort study, 108 consecutive patients with initial presumptive diagnoses of PM were admitted to our tertiary service and evaluated follow-ups were conducted between 2000 and 2019. These patients were referred to our service due to refractoriness, the need for complementary tests (e.g., a new muscle biopsy, muscle magnetic resonance imaging, or chest computed tomography) and for diagnostic reevaluations.
Patients with probable or definite PM according to the Bohan and Peter, and EULAR/ACR classification criteria were considered. As an initial evaluation protocol, myositis-associated and myositis-specific antibody and imaging tests were also performed for neoplasia screening.
Patients were subjected to these exclusion criteria: A promptly confirmed diagnosis or initial suspicion of classic and clinically amyopathic dermatomyositis, antisynthetase syndrome (ASS), immune-mediated necrotizing myopathy (IMNM), inclusion body myositis (IBM), overlap with other autoimmune diseases (e.g., systemic sclerosis, Sjögren's syndrome, rheumatoid arthritis, and systemic lupus erythematosus), muscular dystrophies, metabolic myopathies, neurological diseases (e.g., peripheral neuropathy, and myasthenia gravis), chronic viral infections, decompensated hypothyroidism, and neoplasm-associated myositis.
The present study was approved by the Local Ethics Committee (number 0039/10).
The following information was collected through electronic medical records: (a) demographics (gender, ethnicity, and age at PM diagnosis); (b) clinical data from the time of the PM diagnosis (constitutional symptoms such as fevers of unknown causes, involuntary weight loss, pulmonary involvement-dyspnea on mild or moderate exertion, tomographic findings, joint involvement-arthralgia/arthritis), Raynaud's phenomenon, being bedridden; “mechanic's hands”, status and muscle strength grading according to the Medical Research Council); (c) laboratory data at the time of PM diagnosis regarding maximum creatine phosphokinase (normal range: 24–173 U/L), and maximum aldolase (normal range: 1.0–7.5 U/L).
In outpatient consultations (approximately monthly during thefirst 3 months of follow-up and every 2–3 months thereafter), the disease status assessment (clinical and laboratory) and medication safety (glucocorticoids and immunosuppressive drugs) was also recorded.
Patients initially diagnosed with PM and who went to follow-ups at external services (and later referred to our service, generally due to refractory disease) were evaluated upon admission to receive myositis-associated and myositis-specific antibodies, lung-computed tomography, pulmonary function tests, magnetic resonance imaging of muscles, and a new muscle biopsy on suspicion of new diagnosis.
At each follow-up, patients were also systematically investigated for new symptoms and signs and directed complementary examinations were requested. The new symptoms were distal muscle weakness, finger flexor weakness, exercise intolerance, weakness and myalgia during exercise or fasting, the second wind phenomenon, Raynaud's phenomenon, arthralgia, fever, skin lesions, “mechanic's hands”, and progressive dyspnea.
All muscle biopsies were re-assessed and when necessary, complemented with assessment of specific muscle tissue enzymes and reaction. New muscle biopsies were performed on patients with chronic disease courses (more than 1 year) who were refractory to treatment (no improvement in muscle weakness, slow progression despite high doses of corticosteroids, an initial response to high-dose glucocorticoids and immunosuppressive drugs, sustained glucocorticoid dependence, elevated serum levels of muscle enzymes and significant fat replacement and/or muscle edema in thigh muscle magnetic resonance imaging). The exams were completed via an immunohistochemistry analysis by assessing an acid myophosphorylase and alpha-glycosidase deficiency.
In addition, antibody analysis was performed with serum samples collected at the time of initial investigation and stored at − 20°C. Identification of anti-Jo-1, anti-PL-7, anti-PL-12, anti-OJ, anti-EJ, anti-signal recognition particle (SRP), anti-Mi-2, anti-Ro-52, anti-Ku, and anti-PM/Scl was performed using a commercial kit (Myositis Profile 3, Euroimmun, Germany) according to the manufacturer's protocol. The evaluation of the results was based on the established method in a previous study. Antinuclear antibody determined by indirect immunofluorescence using Hep-2 cells as substrate. We also assayed for anti-3-hydroxy-3-methyl-glutaril-coenzyme A reductase (HMGCR) according to previous study.
- Definitions of several myopathies:
- IBM is based on the criteria of Griggs et al.
- ASS is based on the modified criteria of Connors et al.: the presence of the antibody serum against aminoacyl-transfer ribonucleic acid synthetase and at least two of three disorders: Involving muscles, joints, and/or lungs. The criteria also include the presence of an unexplained and persistent fever, Raynaud phenomenon, and/or “mechanic's hands.”
- Muscular dystrophies are based on the reviews of muscle biopsies and are complemented by an immunohistochemical analysis for dystrophin, sarcoglycans, and dysferlin.
- Metabolic myopathy focuses on glycogenesis, including on McArdle disease (tissue experiences of myophosphorylase) and on Pompe disease (through an acid alpha-glucosidase enzymatic activity test and later confirmation through genetic analysis).
- IMNM involves reviewing muscle biopsies for muscle fiber necrosis with secondary macrophage and for antibody profiles (anti-SRP or anti-HMGCR positivity).
In each follow-up consultation, and in the presence of new clinical, laboratory, and imaging findings, patients were re-diagnosed.
Clinical remission occurred when patients had continuous 6-month periods with no evidence of disease activity while receiving no myositis therapy. Complete clinical response was comprised a continuous 6-month period with no evidence of disease activity while the patient was still receiving myositis therapy.
Data: The Kolmogorov–Smirnov test was used to evaluate the distribution of each of the continuous variables. The results were expressed as mean ± standard deviation for continuous variables and numbers (%) for categorical variables. Median values (25th–75th interquartile range) were calculated for continuous variables that were non-normally distributed.
| Results|| |
We evaluated 108 adult patients with initial diagnoses of probable or definitive PM. The mean age at the diagnosis was 45.3 ± 15.4 years, and patients were predominantly females (70.4%) and Caucasian (81.5%), as shown in [Table 1]. These patients' clinical manifestations and complementary examinations are also described in [Table 1].
|Table 1: Initial demographic, clinical and laboratory data, and complementary exams of 108 patients when admitted at our tertiary center|
Click here to view
Electroneuromyography and muscle biopsies were available in 72 and in 79 of the 108 patients, respectively.
During thefirst 6 months of follow-up, all patients received glucocorticoids and immunosuppressive drugs (e.g., azathioprine, methotrexate, cyclosporine, alone or in combination) in various therapeutic regimens.
During outpatient follow-up, 72.2% of patients showed a change in their final diagnosis [Table 2] and [Figure 1]. Of these patients, 20.4% were diagnosed with IMNM. The majority initially included external services, and antibodies were requested and muscle biopsies were re-evaluated during follow-ups. In addition, 12 patients tested positive for anti-SRP, 7 patients tested for anti-HMGCR, and three patients were negative for both antibodies. Muscle biopsies, however, showed macrophage patterns and necrosis. Moreover, 18.5% of patients had IBM (chronic clinical evolution, asymmetric muscular distribution during its evolution, and characteristic significant muscular compromise in muscular magnetic resonance imaging), 13.0% had muscular dystrophies, 8.3% had ASS (based mainly on clinical manifestation and the presence of the antibody serum against aminoacyl-transfer ribonucleic acid synthetase), 5.6% had metabolic myopathies, and 6.5% had other conditions. The remaining 30 (27.7%) patients maintained the PM diagnosis, according to the EULAR/ACR 2017 criteria.
|Figure 1: Main diseases that mimic polymyositis in the present study. ASS: Antisynthetase syndrome; HMGCR: Hydroxy-3-methyl-glutaril-coenzyme A reductase; IBM: Inclusion body myositis; IMNM: Immune-mediated necrotizing myopathies; PM: Polymyositis; SRP: Signal recognition particle|
Click here to view
These 30 PM patients had a mean age of 43.7 ± 17.4 years when they were admitted in our service [Table 1]. They were predominantly female and Caucasian. These demographic data were indistinguishable from the initial data of those 78 patients re-diagnosed with other myopathies during their follow-up. All the clinical manifestations, initial serum muscle enzymes, antibody distribution, complementary examinations (electroneuromyography, muscle biopsy, and chest-computed tomography) of PM and reclassified patients are also shown in [Table 1].
By the end of the present study, 12 of the 30 PM patients had achieved clinical remission and full recovery of muscle strength, and 10 had undergone complete clinical response, also with full recovery of muscle strength and normalization of muscle enzyme serum levels. These 22 patients all kept following up at our service. Of the remaining eight, five were lost to outpatient follow-up, and three died (two from sepsis complications and another from pancreatic neoplasia).
All the diagnostically reclassified patients with muscular dystrophies, metabolic myopathies, and other myopathies were referred to other specialties for follow-ups. Because of disease remission or stability, nine of 22 patients with MNIM, six of nine patients with ASS, and 12 of 20 patients with IBM were referred to the basic health unit for a follow-up. The remaining patients have been coming to follow-ups at our service.
The mean duration of patients' follow-ups was 4.0 (2.0–9.0) years.
| Discussion|| |
In the present study, we described a large sample of patients initially referred for having a PM diagnosis. During follow-up and reinvestigation, diagnoses were redefined in 72.3% of the patients.
Reports and case series in the literature describe various muscle diseases that mimic PM.,,,,,,,,,,,,,,,,,,,,,,,, Our study showed that several patients were misdiagnosed with PM due to diagnostic difficulty stemming from other muscular diseases sharing similar symptoms and signs,,,, laboratory findings, histological findings, antibodies, and muscle magnetic resonance imaging.,,,,, The low specificity of the Bohan and Peter criteria,, which are still widely used, should also be considered.
Until the present study, 72.3% of our patients had a change in the diagnosis after a detailed reinvestigation. Series and cases reports,,,,,,,,,,,,,,,,,, corroborate our findings, showing the diagnostic difficulty and the need to redefine diagnoses in patients initially diagnosed with PM.
Muscle weakness and increased muscle enzymes (present in 98.2% of our cohort patients) are found in other subgroups of SAMs, muscular dystrophies, metabolic myopathies, and neuromuscular diseases.,,, The myopathic electroneuromyography pattern determines the muscle origin but not the type of disease. All our patients who underwent electroneuromyography-regardless of the final diagnosis-presented an isolated myopathic pattern at the disease's diagnosis. Inflammatory infiltrate from mononuclear cells and cytotoxic CD8+ lymphocytes combined with PM patients's response to major histocompatibility complex I in histology and immunohistochemistry can also be found in IBM and in muscular dystrophies., In our group, all biopsies showed some alteration compatible with inflammatory myopathy. Antibody assessments could not distinguish patients with PM from those with neuromuscular diseases. The present study shows that these examinations were important to define patients with, for example, ASS or MNIM.
Although not made in the present study, follow-up and chronic refractories treatment may distinguish patients with PM from those with neuromuscular diseases. Indeed, our PM patients achieved clinical remission or complete clinical response. However, this parameter may not distinguish patients with PM from patients with other SAMs such as MNIM and ASS, because patients with other SAMs also achieved disease stability.
In our study, patients initially considered to have PM were re-diagnosed after further investigation, especially for IMNM (20.4%) and IBM (18.5%). Studies have observed IMNM mimicking PM in patients, and Vau de Meulen et al. observed that five of nine patients initially diagnosed with PM were actually suffering from IBM. Of our patients, 13.0% were redefined as having muscular dystrophies, and the most prevalent girdle-like dystrophy (11 out of 14 patients) coincided with the most commonly diagnosed dystrophy in adulthood. Several case reports have similarly described patients with waist muscular dystrophy,, and dysferlinopathies,, mimicking PM. Metabolic myopathies may also mimic PM., In our study, five cases were reclassified as McArdle disease and one case as Pompe disease based on new muscle biopsies complemented with specific tissue enzyme analyses. In a retrospective review of 12 patients with McArdle disease, Olivo Pallo et al. found that seven were initially diagnosed as PM carriers and they took glucocorticoids and immunosuppressive drugs.
Other diseases that can mimic PM include neuromuscular disease such as Kennedy's disease,, peripheral neuropathy, motor neuron disease, and sarcoidosis., In addition to these diseases, cases of myofibrillar myopathy and tubular aggregate myopathy appeared in the present study.
Regarding limitations, the present study is descriptive, significant patients initially went to external services, and not all patients had a fully available antibody panel. Moreover, the follow-up features of patients who kept their PM diagnosis could not be compared to features of patients whose diagnosis changed because the majority of these patients were later referred to other specializations and to basic health units for follow-ups.
| Conclusions|| |
An initial diagnosis of early-stage PM should be reevaluated constantly in these patients, especially with new clinical, laboratory, and imaging findings or with unexpected developments and therapeutic responses during follow-ups., several differential diagnoses should be considered in daily clinical practice, to improve the approach and treatment of patients, daily clinical practice should consider several diagnoses with the presence of muscle weakness, elevated muscle enzymes, biopsies consistent with muscle inflammation and electroneuromyography with myopathic patterns in addition to PM. In our study, only one-third of patients maintained their PM diagnosis until the end of data collection.
Financial support and sponsorship
This work was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) 303379/2018-9 and Faculdade de Medicina da USP to SKS.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two parts). N
Engl J Med 1975;292:344-7.
Bohan A, Peter JB. Polymyositis and dermatomyositis (second of two parts). N
Engl J Med 1975;292:403-7.
Lundberg IE, Tjärnlund A, Bottai M, Werth VP, Pilkington C, Visser M, et al
. 2017 European League Against Rheumatism/American College of Rheumatology classification criteria for adult and juvenile idiopathic inflammatory myopathies and their major subgroups. Ann Rheum Dis 2017;76:1955-64.
Amato AA, Griggs RC. Unicorns, dragons, polymyositis, and other mythological beasts. Neurology 2003;61:288-9.
Szczęsny P, Świerkocka K, Olesińska M. Differential diagnosis of idiopathic inflammatory myopathies in adults-thefirst step when approaching a patient with muscle weakness. Reumatologia 2018;56:307-15.
Vattemi G, Mirabella M, Guglielmi V, Lucchini M, Tomelleri G, Ghirardello A, et al
. Muscle biopsy features of idiopathic inflammatory myopathies and differential diagnosis. Auto Immusn Highlights 2014;5:77-85.
Hilton-Jones D. Observations on the classification of the inflammatory myopathies. Presse Med 2011;40:e199-208.
Mahmood OA, Jiang XM. Limb-girdle muscular dystrophies: Where next after six decades from thefirst proposal (Review). Mol Med Rep 2014;9:1515-32.
Benveniste O, Romero NB. Myositis or dystrophy? Traps and pitfalls. Presse Med 2011;40:e249-55.
Olivo Pallo PA, Silva AM, Zanoteli E, Shinjo SK. McArdle's disease: An underestimated or underdiagnosed myopathy in rheumatologic practice? Cases series and literature review. Med Express 2018;5:Mo.008.
van der Meulen MF, Bronner IM, Hoogendijk JE, Burger H, van Venrooij WJ, Voskuyl AE, et al
. Polymyositis: An overdiagnosed entity. Neurology 2003;61:316-21.
Kissel JT. Polymyositis: Not a unicorn or mythological beast.but maybe a duck? Neurology 2008;70:414-5.
Dixit A, Abrudescu A. A case of atorvastatin-associated necrotizing autoimmune myopathy, mimicking idiopathic polymyositis. Case Rep Rheumatol 2018;2018:5931046.
Yildiz H, Roelants V, Hainaut P. Statin-induced rhabdomyolysis mimicking polymyositis on (18)F-FDG PET imaging. Acta Clin Belg 2015;70:151-2.
Xu C, Chen J, Zhang Y, Li J. Limb-girdle muscular dystrophy type 2B misdiagnosed as polymyositis at the early stage: Case report and literature review. Medicine (Baltimore) 2018;97:e10539.
Ng KP, Smith CR, Isenberg DA. Muscular dystrophy mimicking refractory idiopathic inflammatory myositis: A trio of cases. Rheumatology (Oxford) 2007;46:1618-9.
Griger Z, Nagy-Vincze M, Bodoki L, Gherardi RK, Dankó K, Hortobágyi T. Late onset dysferlinopathy mimicking treatment resistant polymyositis. Joint Bone Spine 2016;83:355-6.
Amigoa JL, Gallardo E, Gallano P, Grau-Junyentd JM. Disferlinopatía, una causa falsa de polimiositis refractaria. Med Clin (Barc) 2015;145:414-5.
Vinit J, Samson M Jr., Gaultier JB, Laquerriere A, Ollagnon E, Petiot P, et al
. Dysferlin deficiency treated like refractory polymyositis. Clin Rheumatol 2010;29:103-6.
Scalco RS, Lorenzoni PJ, Lynch DS, Martins WA, Jungbluth H, Quinlivan R, et al
. Polymyositis without beneficial response to steroid therapy: Should miyoshi myopathy be a differential diagnosis? Am J Case Rep 2017;18:17-21.
Horneff G, Paetzke I, Neuen-Jacob E. Glycogenosis type V (McArdle's disease) mimicking atypical myositis. Clin Rheumatol 2001;20:57-60.
Pérez-López J, Selva-O'Callaghan A, Grau-Junyent JM, Gallego-Galindo L, Coll MJ, García-Morillo S, et al
. Delayed diagnosis of late-onset Pompe disease in patients with myopathies of unknown origin and/or hyperCKemia. Mol Genet Metab 2015;114:580-3.
Shi W, Kang J, Lao B, Ye X, Zeng F, Wang J, et al
. A case of Kennedy's disease misdiagnosed as polymyositis. Scand J Rheumatol 2019;48:168-70.
Harutunian GM, Beydoun SR, Rison RA. Kennedy disease misdiagnosed as polymyositis: A case report. BMC Res Notes 2013;6:389.
Kim Y, Kim Y, Jin SY, Roh HJ, Kim HS. Peripheral Neuropathy Presenting as Myopathy That was Confused With Inflammatory Myositis. Arch Rheumatol 2018;33:488-90.
Ryan A, Nor AM, Costigan D, Foley-Nolan D, El-Rafie A, Farrell MA, et al
. Polymyositis masquerading as motor neuron disease. Arch Neurol 2003;60:1001-3.
Awano N, Fukuda K, Sakayori M, Kondoh K, Ono R, Moriya A, et al
. Sarcoid Myositis with Anti-Ku Antibody Consistent with both Sarcoidosis and Polymyositis. Intern Med 2016;55:2049-53.
Sazliyana Shaharir S, Jamil A, Kosasih S, Soo Fin L, Sridharan R, Hayati Md Pauzi S. Sarcoid myopathy mimicking polymyositis: A case report and pool analysis of the literature reviews. Acta Med Iran 2017;55:800-6.
Medical Research Council. Aids to the Examination of the Peripheral Nervous System, Memorandum no. 45. London: Her Majesty's Stationery Office; 1981.
Cruellas MG, Viana Vdos S, Levy-Neto M, Souza FH, Shinjo SK. Myositis-specific and myositis-associated autoantibody profiles and their clinical associations in a large series of patients with polymyositis and dermatomyositis. Clinics (Sao Paulo) 2013;68:909-14.
De Souza FHC, Miossi R, Shinjo SK. Necrotising myopathy associated with anti-signal recognition particle (anti-SRP) antibody. Clin Exp Rheumatol 2017;35:766-71.
Griggs RC, Askanas V, DiMauro S, Engel A, Karpati G, Mendell JR, et al
. Inclusion body myositis and myopathies. Ann Neurol 1995;38:705-13.
Connors GR, Christopher-Stine L, Oddis CV, Danoff SK. Interstitial lung disease associated with the idiopathic inflammatory myopathies: What progress has been made in the past 35 years? Chest 2010;138:1464-74.
ENMC International Workshop: Development of diagnostic criteria and management strategies for McArdle disease and related rare glycogenolytic disorders to improve standards of care 17-19 April 2015 Naarden, Netherlands. Neuromuscul Disord 2017;27:1143-51.
Bembi B, Cerini E, Danesino C, Donati MA, Gasperini S, Morandi L, et al
. Diagnosis of glycogenosis type II. Neurology 2008;71:S4-11.
Borges IBP, Silva MG, Shinjo SK. Prevalence and reactivity of anti-melanoma differentiation-associated gene 5 (anti-MDA-5) autoantibody in Brazilian patients with dermatomyositis. An Bras Dermatol 2018;93:517-23.
Iaccarino L, Ghirardello A, Bettio S, Zen M, Gatto M, Punzi L, et al
. The clinical features, diagnosis and classification of dermatomyositis. J Autoimmun 2014;48-49:122-7.
[Table 1], [Table 2]