The retrospective study of SSc-M patients seen at the Rheumatology Clinic, Maharaj Nakorn Chiang Mai Hospital, Thailand, between January 2010 and December 2019 was conducted. All adult SSc patients (age ≥ 18 years) fulfilled the 1980 classification criteria of SSc [14] and/or the ACR/EULAR criteria 2013 for the classification of SSc [15]. The diagnosis of SSc-M was defined as the presence of symmetrical proximal muscle weakness, elevated creatine kinase (CK) level without other explainable causes, and muscle biopsy reports compatible with myopathic change either inflammatory or fibrosing pattern. Twenty-seven patients with suspected SSc-M underwent muscle biopsy. Twenty-three of 27 SSc-M with available frozen muscle biopsy specimens for repeating immunohistochemically staining were included. For patients who had multiple muscle biopsies, the first biopsy specimen was used for analysis. Four excluded patients were DcSSc subtype with anti-topoisomerase-I antibody-positive. Clinical and laboratory data at the biopsy date were reviewed. Exclusion criteria were SSc overlapping systemic lupus erythematosus or rheumatoid arthritis or SjÖgren’s syndrome or mixed connective tissue disease.

The following data were collected from the medical records as described by the physicians as of the date of initial muscle biopsy, including demographic data, disease subtype, modified Rodnan skin score (mRSS); muscle involvement comprised myalgia and muscle strength; other organ involvement was defined as “presence” if we found its recorded data at the biopsy date including hypo-hyperpigmentation, digital ulcer, arthritis, tendon friction rub, dysphagia, gastroesophageal reflux disease (GERD), interstitial lung disease (ILD), suspected pulmonary hypertension (PH), scleroderma renal crisis (SRC) and congestive heart failure (CHF); laboratory testings comprised, creatine kinase (CK), and erythrocyte sedimentation rate (ESR), hemoglobin, and creatinine (Cr).

The SSc subtype was classified as diffuse cutaneous SSc (DcSSc) or limited cutaneous SSc (LcSSc) according to LeRoy and Medsger’s classification criteria [16]. SSc duration was defined as the interval from the first non-Raynaud’s phenomenon (NRP) contributable to SSc manifestation to undergo muscle biopsy. Myopathy duration was defined as the interval from the first muscle weakness to undergoing muscle biopsy. Skin thickening was recorded using a modified Rodnan skin score (mRSS) [17]. Muscle power was recorded on the 6 grade according to the Medical Research Council scale, which ranges from 0 (muscle weakness with no visible muscle contraction) up to 5 (normal power) [18]. We then divided the muscle power into two subgroups, including (i) severe weakness as proximal muscle power of grade ≤ 2 was classified; and (ii) mild to moderate weakness as muscle power of grade ≥ 3-4. The digital ulcer was defined as the presence of active or healed ulceration which is present at the volar aspect of the digital pulp. Arthritis was defined as the presence of joint tenderness and swelling examined by attending rheumatologists. Gastrointestinal involvements were defined as the presence of GERD or dysphagia symptoms. Interstitial lung disease (ILD) was determined by chest X-ray or high-resolution computed tomography. Suspected pulmonary hypertension (PH) was defined according to the presence of echocardiographic signs suggesting PH according to the 2015 ESC/ERS guideline [19]. Scleroderma renal crisis (SRC) was defined as present if the criteria of the International Scleroderma Crisis Study Group were fulfilled [20].

Frozen muscle biopsy specimens received from 2010 to 2019 were retrieved from a -80 °C tissue bank. Five-micron-thickness cryostat sections were done. Each specimen underwent histochemical stain with hematoxylin-eosin (H&E), immunohistochemically stained with major histocompatibility complex class I (MHC-I) and MHC-II, complement membrane attack complex (MAC), and picrosirius red stained [21]. Regarding the diagnosis of fibrosing myopathy, picrosirius red polarized images were digitally captured by an optical microscope with a polarizing filter at a magnification of 200 X and an image size of 1600x1200 pixels into red green blue (RGB) 24 bits TIFF (Tagged Image File Format) files. The collagen fibers in the fibrous tissue appeared as a bright red color. Muscle biopsies were reviewed and then the muscle pathology was classified by an experience muscle pathologist (S.S.).

Muscle pathological findings were classified as the followings:(i) Polymyositis (PM) - the presence of endomysial inflammatory cell infiltration which is surrounding or invading non-necrotic fibers [22]; (ii) Dermatomyositis (DM) - presence of perifascicular atrophy [22]; (iii) Inclusion body myositis (IBM) - presence of endomysial inflammatory cell infiltration and rimmed vacuoles [23]; (iv) Immune-mediated necrotizing myopathy (IMNM) - presence of scattered necrotic and regenerating muscle fibers, sparse inflammation and presence of sarcolemma immunohistochemically staining for MAC [24]; (v) Non-specific myositis (NsM)- presence of scattered perivascular, perimysium or endomysial inflammatory infiltration that does not fulfill to the definition of PM, DM, or IMNM [22]; and (vi) Fibrosing myopathy (FM) - presence of predominantly fibrosis without fulfilled pathological changes of the above-mentioned myopathies [9, 10].

The descriptive data are presented as frequency (percentage: %), mean ± standard deviation (SD), or median (interquartile range 1, 3: IQR 1, 3). Three subgroups of muscle pathologic findings in this study included SSc-IMNM, NsM, and PM. Therefore, a comparison of categorical variables among the three subgroups was analyzed using Fisher’s exact test. Continuous variables between the three subgroups were compared using the Kruskal-Wallis test or the one-way ANOVA test. p-values < 0.05 were considered statistically significant. Statistical analyses were performed using IBM SPSS Statistics for Windows, version 20 (IBM Corp., Armonk, N.Y., USA).

A total of 23 SSc-M patients that fulfilled the inclusion criteria (14 female [60.9%], 19 DcSSc [82.6%], 17 anti-topoisomerases I antibody-positive [73.9%]) had a mean ± SD age of 53.6 ± 7.7 years; median (IQR 1, 3) duration of SSc and myopathy was 11 (4, 14) months and 2 (0, 3) months, respectively. There were no patients with anti-centromere antibody-positive in our study population. At the biopsy date, their mean ± SD mRSS was 23 ± 15.3, CK levels were 2442.8 ± 1431.2 U/L (median: 2425.0 U/L [1567.0, 3368.0 ]) and ESR levels were 72.8 ± 34.2 mm/hr. There were 17 (73.9%) SSc-M patients with mild to moderate weakness and 6 (26.1%) with severe weakness. There were six patients (26.1%) with the digital ulcer, 15 (65.2%) with arthritis, 4 (17.4%) with tendon friction rub, 17 (73.9%) with dysphagia, 13 (56.5%) with GERD, 16 (69.6%) ILD, 6 (26.1%) suspected PH, and 4 (17.4%) CHF. Out of 15 patients with clinical arthritis, two (13.3%) have rheumatoid factor positive-low liter. Out of the 23 muscle histological findings after undergoing the staining methods, 14 (60.8%) patients had IMNM, 6 (26.1%) NsM, and 3 (13%) PM. There were no FM, IBM, and DM in our population. The patients were then divided into three subgroups: (i) IMNM, (ii) NsM, and (iii) PM.

The comparative analysis of the baseline clinical characteristics and laboratory findings among the three inflammatory patterns is shown in Table 1. Regarding the demographic data, the IMNM subgroup tended to have more prevalence of female gender and had longer disease duration of SSc than the others. Whereas, the NsM subgroup tended to have a higher proportion of DcSSc subtype, anti-topoisomerase-I antibody-positive and shorter time from muscle weakness symptoms to muscle biopsy than the other group. Regarding organ involvement, a higher prevalence of severe muscle weakness and arthritis was observed in the IMNM subgroups. There were no statistical differences in other clinical manifestations, including digital ulcer, tendon friction rub, dysphagia, GERD, ILD, suspected PH, and CHF. No scleroderma renal crisis was observed in our population. In addition, mean or median baseline laboratory values, including CK level, ESR, Hb, and Cr, as well as current medical treatment (data not shown), showed no statistically significant within the three inflammatory patterns.