Muscle endurance, strength, and active range of motion in patients with different subphenotypes in systemic sclerosis: a cross-sectional cohort study

Abstract

Objective: Proximal muscle weakness is common in patients with systemic sclerosis (SSc). Dynamic muscle endurance, muscle strength in the lower extremities, and active range of motion (AROM) in the upper extremities are less studied. We investigated functional muscle endurance, strength, and AROM, and explored differences depending on skin and/or lung involvement in SSc patients.

Method: The study divided 205 patients with limited/diffuse cutaneous systemic sclerosis (lcSSc/dcSSc) into no–mild and moderate–end-stage lung involvement, the latter based on the Medsger disease severity score. Dynamic muscle endurance in shoulder and hip flexion was assessed by the Functional Index-2, lower extremity muscle strength by the Timed-Stands Test (TST), and shoulder–arm AROM by the Functional Shoulder Assessment (FSA).

Results: Shoulder and hip flexion muscle endurance were reduced in relation to reference values median (IQR) [53% (27–100%) and 40% (23–90%), respectively, p < 0.001]. Patients with moderate–end-stage lung involvement had less endurance in shoulder [39% (21–71%) and hip flexion 35% (20–70%)] than patients with no-mild lung involvement [57% (33–99%) and 48% (28–100%), p < 0.05]. All patients, regardless of subtype/grouping, needed longer to complete the TST [21 s (17–27 s)] compared to reference values [17 s (15–18 s), p < 0.001], and patients with moderate-end-stage lung involvement had worse TST score than patients with no-mild lung involvement,  [25 s (18–30 s) vs 19 s (16–25 s), p < 0.001]. The FSA sum scores were lower compared with reference values (p < 0.01). DcSSc patients had a lower FSA-sum score [53 (48–57)] than lcSSc patients [57 (52–60), p < 0.01].

Conclusion: SSc patients have markedly reduced muscle endurance in the upper and lower extremities, reduced muscle strength in the lower extremities, and impaired AROM in the shoulders and arms. Patients with moderate–end-stage lung involvement had more impaired muscle endurance and strength but no differences were found between lcSSc and dcSSc patients. Not only muscle strength, but also dynamic muscle endurance should be measured in SSc patients.

Systemic sclerosis (SSc) is a rare systemic disorder characterized by fibrosis, vasculopathy, and autoimmunity (1, 2). In both the limited cutaneous systemic sclerosis (lcSSc) and diffuse cutaneous systemic sclerosis (dcSSc) subtypes, lung involvement, such as pulmonary hypertension and lung fibrosis, can occur (3). We previously reported reduced self-reported physical capacity, such as walking and running, in patients with SSc compared to age- and gender-matched controls from the general population (4), and similar results have been presented by others (5). In addition, more limiting factors to physical capacity, such as heart or lung disease, and reduced muscle strength and mobility, were noted in the patients with SSc (4).

Muscle weakness in SSc is mainly located in proximal muscles, such as the hip and shoulder girdle (6), but muscles in the lower extremities have also been reported as being weak (7–10). Patients with dcSSc have an almost doubled risk of developing muscle weakness in proximal limb girdle muscles (shoulder and hip girdle) compared with lcSSc (1, 3), which has been confirmed by data from the European League Against Rheumatism (EULAR) Scleroderma Trials and Research (EUSTAR) cohorts (1, 11) as well as in Japan (12). Measurement of muscle weakness is of great importance since proximal muscle weakness can predict progressive disease in SSc (13) in addition to impaired functional performance and reduced quality of life in patients with SSc (7). However, no previous study has, to our knowledge, analysed the impact of muscle endurance and muscle strength in relation to subtypes, i.e. lcSSc versus dcSSc or subgroups based on the degree of lung involvement.

Joint involvement, especially in the finger and wrist joints, is common in SSc, and a few studies have also reported impaired mobility in the elbows and shoulders (14–16). Restricted range of motion has also been reported in the lower extremities, especially in patients with dcSSc (9, 10, 16, 17). However, whether the active range of motion (AROM) differs between different degree of lung involvement has not been well explored.

The aim of this study is to examine functional dynamic muscle endurance in the shoulder and hip girdle, functional dynamic muscle strength in the lower extremities, and functional AROM of the shoulder–arm region in patients with SSc. We also explored possible differences depending on SSc subphenotypes, i.e. lcSSc versus dcSSc, and the presence of lung involvement.

Method

Study population

A cohort of 205 adult patients with SSc, at the Karolinska University Hospital, fulfilling the new 2013 American College of Rheumatology (ACR)/EULAR criteria for SSc (18), were included between September 2006 and May 2017.

Disease characteristics and subphenotypes

All patients were examined by a rheumatologist, with scoring obtained using the modified Rodnan skin score (19). The patients were classified as having either lcSSc or dcSSc (20). To evaluate disease severity, the Medsger scale of disease damage was used, with the exception that the joint/tendon and muscular items were not assessed (21). As part of routine investigation, the patients underwent echocardiography, lung function tests, and high-resolution computed tomography (HRCT) of the lungs. We used the values that were collected closest in time to the inclusion in this study. In the majority of cases these tests were performed approximately 6 months before or after inclusion in the study. The most severe value, from the different tests, was used in the scoring. Based on these tests, patients were subgrouped into no–mild or moderate–end-stage lung involvement, using the Medsger disease severity scale for the lung (0–4 points). Patients with a score of 0–1 points were classified as having no–mild lung involvement and those with a score of 2–4 points as moderate–end-stage lung involvement.

Lung fibrosis was defined as signs of fibrosis on HRCT; pulmonary arterial hypertension (PAH) was defined by right heart catheterization (22); and myositis was defined as muscular weakness, elevated creatine kinase, and signs of inflammation on magnetic resonance imaging, electromyography, or muscular biopsy (23). 

Assessments

The shoulder and the hip flexion tasks of the Functional Index-2 (FI-2) were used to assess dynamic muscle endurance in the upper and lower extremities (24). Shoulder flexion was assessed in a sitting position, with a 1 kg weight cuff around the wrist. Hip flexion was assessed in a supine position with a straight leg. The maximal number of repetitions (60 per arm and per leg) was registered and the results were presented as the percentage of maximal repetitions per muscle group. A pace of 40 beats/min resulted in 20 repetitions/min. Preliminary reference values, based on age intervals of 10 years from the general population, were collected from ongoing data collection at Karolinska University Hospital, Solna, Sweden (HA), and Oslo University Hospital, Oslo, Norway (RWH). When comparing the patients’ FI-2 values, only the right side was compared owing to the majority (88%) of reference values being assessed on the right side.

The Timed-Stands Test (TST) measures functional muscle strength in the lower extremities (25). The time needed to complete 10 full stands from a sitting position on a 46 cm high chair without using the arms was recorded with a stopwatch. Our data were compared with the age- and gender-matched healthy subjects’ predicted values, in intervals of 5 years (25).

The Functional Shoulder Assessment (FSA) measures AROM in the shoulder–arm and consists of five tasks: hand-raising, hand-to-opposite-shoulder, hand-behind-back, hand-to-neck, and hand-to-seat (26, 27). Each functional task is assessed on a six-point Likert scale (1 being the worst possible function and 6 the greatest), resulting in a total maximum score of 30 points for each shoulder-arm. Comparison between patients with SSc and individuals in the general Swedish population was carried out using the results as presented by Olofsson et al, in intervals of 10 years (28). This study (28) used an earlier version of the FSA, consisting of four tasks (i.e. not hand-to-seat), and we transformed their original reference data: hand-to-neck (1–7 points) was converted into 1–6 points; and hand-raising, hand-to-opposite-shoulder, and hand-behind-back (1–6 points each) were retained, giving a total of 24 points for each shoulder.

Procedure

The patients performed the muscle function and range of motion tests under supervision by experienced physiotherapists, who made the assessments in the following order: FSA, FI-2, and TST. The assessments took an average of 30 min to complete, with approximately 5 min rest between the FI-2 and TST.

Our study was approved by the Stockholm Regional Ethical Review Board (approval numbers: 2006/259-31/3 and 2017/591-32) in accordance with the Declaration of Helsinki, and all participants gave written informed consent.

Statistical analysis

Nominal variables were analysed with Pearson’s χ² or Fisher’s exact test. For ordinal and non-normally distributed continuous variables, the Mann–Whitney unpaired test was used to analyse differences between groups. The Wilcoxon signed-ranks test was used to compare the observed values for SSc patients with age- and gender-specific reference population values. The level of significance was set at p < 0.05. All data were analysed with version 23.0 of IBM SPSS Statistics for Windows (IBM, Armonk, NY, USA).

Results

There were no significant differences in age, gender, or disease duration between lcSSc and dcSSc or the subgroups based on lung involvement (Table 1). A right-hand dominance was found in 96% of the patients. Thirteen patients (6%) were diagnosed with myositis before inclusion in the study, or during the same year (Table 1).

Table 1. Demographic data of patients with systemic sclerosis (SSc), stratified by subtype as defined by skin involvement (lcSSc or dcSSc) and subgroup as defined by lung disease (no–mild or moderate–end-stage).

	All patients (n = 205)	lcSSc (n = 169)	dcSSc (n = 36)	p	No–mild lung disease (n = 116)	Moderate–end-stage lung disease (n = 82)	p
Age (years)	60 (52–69)	60 (53–70)	60 (44–64)	0.11	59 (50–69)	61 (53–69)	0.49
Female (%)	82	84	75	0.24	85	81	0.57
Disease duration (years)	8 (3–13)	8 (4–14)	5 (2–11)	0.08	6 (3–12)	9 (4–15)	0.07
Medsger general†	88/10/1/0/1	91/9/0/0/0	78/14/5/0/3	< 0.05*	91/8/0/0/1	84/14/2/0/0	0.12
Peripheral vascular†	17/61/13/8/1^2	17/61/13/8/1^2	17/55/17/8/3	0.57	22/62/9/7/0^1	10/60/19/11/0^1	< 0.05*
Skin†	9/81/9/1/0	11/87/2/0/0	3/53/39/5/0	< 0.001*	9/83/7/1/0	10/77/12/1/0	0.4
Gastrointestinal tract†	30/55/13/1/1^13	28/58/12/1/1^9	37/41/19/3/0^4	0.83	37/49/12/1/1^8	21/62/17/0/0^3	0.06
Lung†	27/31/29/12/1^6	33/28/27/11/1^4	21/35/29/15/0^2	0.29	47/53/0/0/0	0/0/70/28/2	< 0.001*
Heart†	80/14/2/2/2^6	81/14/2/2/1^5	74/14/6/3/3^1	0.31	82/13/3/2/0^4	79/14/1/2/4^1	0.52
Kidney†	95/4/9/1/1^3	96/2/0/1/1^3	86/11/0/3/0	< 0.05*	98/2/0/0/0^2	89/7/0/3/1	< 0.01*
mRSS (points)	5 (3–9)	4 (2–7)	13 (9–19)	< 0.001*	4 (2–8)	6 (3–12)	< 0.05*
Pulmonary hypertension (%)	5	5	6	0.6	0	13	< 0.001*
Lung fibrosis (%)	38	37	44	0.45	24	60	< 0.001*
Myositis (%)	6	7	6	0.59	4	8	0.18

Values are presented as median (interquartile range) or %.

lcSSc, limited cutaneous systemic sclerosis; dcSSc, diffuse cutaneous systemic sclerosis; No–mild lung disease, Medsger disease severity score 0–1; Moderate–end-stage lung disease, Medsger disease severity score 2–4; Disease duration; time from first non-Raynaud’s symptom; Medsger general, etc.: †disease severity score, 0/1/2/3/4 (%); ⁿ missing numbers. Owing to missing data, for the diffusion capacity of the lung for carbon monoxide and/or forced vital capacity, seven patients could not be classified as having no–mild or moderate–end-stage lung disease; mRSS, modified Rodnan skin score; Pulmonary hypertension, verified by right heart catheterization at the value ≥ 25 mmHg; Lung fibrosis, defined as signs of fibrosis on high-resolution computed tomography.

*Significant difference between groups (p < 0.05).

Muscle endurance in shoulder flexion and hip flexion (FI-2)

Patients with SSc had reduced muscle endurance in both the right shoulder and hip flexions compared with reference values (Table 2). The median repetitions performed were, for shoulder flexion on the right side, n = 32, and left, n = 27; and for hip flexion, right side, n = 24, and left, n = 24, out of an expected 60 repetitions in the study sample. Regarding muscle endurance, percent of expected repetitions in the shoulder and hip flexion was more reduced in patients with moderate–end-stage lung involvement than in those with no–mild lung involvement (Table 3). There were no significant differences in muscle endurance between patients with lcSSc and dcSSc.

Table 2. Shoulder and hip flexion muscle endurance on the right side in systemic sclerosis (SSc) and reference group divided into different age groups.

Age (years)		Shoulder flexion (%),0–100	p	Hip flexion (%),0–100	p
All age groups	SSc (n = 205)	53 (27–100)^7	< 0.001*	40 (23–90)^9	< 0.001*
	Reference group A (n = 85)	100 (100–100)		100 (72–100)^3
≤ 49	SSc (n = 44)	70 (40–100)^1	< 0.001*	44 (26–100)	0.001*
	Reference group A (n = 27)	100 (100–100)		100 (62–100)^1
50–59	SSc (n = 58)	70 (24–100)^1	< 0.001*	53 (23–100)^4	< 0.001*
	Reference group A (n = 22)	100 (100–100)		100 (80–100)
60–69	SSc (n = 54)	52 (27–100)^4	< 0.001*	38 (23–91)^2	< 0.001*
	Reference group A (n = 25)	100 (85–100)		100 (100–100)^2
≥ 70	SSc (n = 49)	35 (13–63)^1	0.003*	31 (17–52)^3	< 0.001*
	Reference group A (n = 11)	100 (38–100)		100 (67–100)

Values are presented as median (interquartile range).

Values for shoulder and hip flexion muscle endurance are percentage of predicted according to Functional Index-2. Reference group A is collected from Alexanderson and Walle-Hansen (unpublished data). ⁿ Missing numbers.

*Significant difference between groups (p < 0.05).

Table 3. Shoulder and hip muscle endurance according to the Functional Index-2 (FI-2), Timed-Stands Test (TST), and Functional Shoulder Assessment (FSA), in patients with systemic sclerosis (SSc) and stratified by skin involvement (lcSSc or dcSSc) and by subgroup, defined by lung involvement (no–mild or moderate–end-stage).

	All patients (n = 205)	lcSSc (n = 169)	dcSSc (n = 36)	p	No–mild lung disease (n = 116)	Moderate–end-stage lung disease (n = 82)	p
Shoulder flexion endurance sum score (%)	50 (27–92)^8	51 (28–94)^7	42 (22–81)^1	0.29	57 (33–99)^2	39 (21–71)^6	0.01*
Shoulder flexion, FI-2 (%), right	53 (27–100)^7	53 (27–100)^6	45 (18–83)^2	0.3	63 (28–100)^2	42 (20–78)^3	0.02*
Shoulder flexion, FI-2 (%), left	47 (25–90)^6	48 (27–93)^4	38 (22–79)^2	0.2	53 (32–100)^1	38 (13–69)^5	0.007*
Hip flexion endurance sum score (%)	40 (25–79)^11	40 (25–91)^9	42 (24–61)^2	0.49	48 (28–100)^2	35 (20–70)^8	0.01*
Hip flexion, FI-2 (%), right	40 (23–90)^9	40 (23–100)^7	40 (22–57)^2	0.38	48 (28–100)^2	35 (20–70)^7	0.01*
Hip flexion, FI-2 (%), left	40 (25–78)^10	38 (24–89)^8	41 (25–64)^2	0.81	47 (25–100)^2	35 (22–52)^8	0.03*
TST (s)	21 (17–27)^24	21 (17–26)^18	23 (18–29)^6	0.59	19 (16–25)^13	25 (18–30)^9	0.001*
FSA sum score, 10–60	57 (51–59)^2	57 (52–60)	53 (48–57)^2	0.003*	57 (51–60)^2	56 (50–58)	0.15
FSA, right, 5–30	28 (25–30)^2	28 (26–30)	26 (24–29)^2	0.002*	28 (26–30)^2	28 (25–29)	0.25
FSA, left, 5–30	28 (28–30)^2	29 (27–30)	27 (24–29)	0.006*	29 (26–30)^2	28 (25–30)	0.05

Values are presented as median (interquartile range).

lcSSc, limited cutaneous systemic sclerosis; dcSSc, diffuse cutaneous systemic sclerosis; No–mild lung involvement, Medsger disease severity score 0–1; Moderate–end-stage lung involvement, Medsger disease severity score 2–4; FI-2 (%), percentage of predicted FI-2; FSA, 10–60 points for the five included subparts: hand-raising, hand-to-opposite-shoulder, hand-behind-back, hand-to-neck, and hand-to-seat; ⁿ missing numbers. Owing to missing data, for the diffusion capacity of the lung for carbon monoxide and/or forced vital capacity, seven patients could not be classified as having no–mild or moderate–end-stage lung involvement.

*Significant difference between groups (p < 0.05).

Muscle strength (TST)

Among those completing the TST (n = 181), patients with SSc as well as the subtypes/groups (dcSSc/lcSSc and no–mild/moderate–end-stage lung involvement) had lower muscle strength when measured with TST compared to reference values (Table 4). Patients with moderate–end-stage lung involvement needed longer to complete the TST than patients with no–mild lung involvement. There was no significant difference in muscle strength between patients with lcSSc and dcSSc (Table 3).

Table 4. Timed-Stands Test (TST) in patients with systemic sclerosis (SSc) stratified by skin involvement (lcSSc or lcSSc) and lung involvement (no–mild or moderate–end-stage) in comparison with predicted reference values.

	TST (s)	Predicted reference values, TST (s)	p
All patients (n = 205)	21 (17–27)^24	17 (15–18)	< 0.001*
lcSSc (n = 169)	21 (17–26)^18	17 (16–18)	< 0.001*
dcSSc (n = 36)	25 (18–29)^6	17 (14–18)	< 0.001*
No–mild (n = 116)	19 (16–25)^13	17 (15–18)	< 0.001*
Moderate–end-stage (n = 82)	25 (18–30)^9	17 (16–18)	< 0.001*

Values are presented as median (interquartile range).

lcSSc, limited cutaneous systemic sclerosis; dcSSc, diffuse cutaneous systemic sclerosis; No–mild lung involvement, Medsger disease severity score 0–1; Moderate–end-stage lung disease, Medsger disease severity score 2–4; ⁿ missing numbers. Owing to missing data, for the diffusion capacity of the lung for carbon monoxide and/or forced vital capacity, seven patients could not be classified as having no–mild or moderate–end-stage lung involvement.

*Significant difference between groups (p < 0.05).

Shoulder–arm AROM (FSA)

The whole sample in this study, as well as patients with SSc in the age groups ≥ 60 years, had a significantly lower bilateral shoulder–arm AROM compared with the respective reference values (Table 5). Patients with dcSSc had bilaterally lower shoulder–arm AROM compared with lcSSc patients, while there were no significant differences between patients with no–mild or moderate–end-stage lung involvement (Table 3).

Table 5. Functional Shoulder Assessment (FSA) in systemic sclerosis (SSc) and reference group divided into different age groups.

Age (years)		FSA, right,4–24 points	p	FSA, left,4–24 points	p
All age groups	SSc (n = 205)	22 (20–24)^2	0.003*	23 (20–24)^2	0.001*
	Reference group B (n = 246)	23 (22–24)^3		23 (22–24)^3
≤ 49	SSc (n = 44)	22 (20–24)^2	0.42	24 (22–24)^2	0.12
	Reference group B (n = 41)	23 (22–24)^2		24 (23–24)
50–59	SSc (n = 58)	23 (21–24)	0.26	23 (21–24)	0.28
	Reference group B (n = 23)	23 (23–24)		23 (23–24)
60–69	SSc (n = 54)	22 (19–24)	0.002*	23 (20–24)	0.008*
	Reference group B (n = 74)	23 (23–23)^2		23 (23–24)^3
≥ 70	SSc (n = 49)	21 (20–23)	0.001*	21 (19–23)	< 0.001*
	Reference group B (n = 108)	22 (21–23)^1		23 (21–24)

Values are presented as median (interquartile range).

The values for FSA are presented as points (4–24 points for the four included tasks: hand-raising, hand-to-opposite-shoulder, hand-behind-back, and hand-to-neck); ⁿ missing numbers. Reference group B is collected from Olofsson et al (28).

*Significant difference between groups (p < 0.05).

Discussion

This is the first study to evaluate both functional dynamic muscle endurance and strength in SSc subphenotypes, i.e. lcSSc versus dcSSc, and the presence of lung involvement. We demonstrate that patients with SSc had a markedly reduced muscle endurance and strength and also slightly lower functional AROM in the shoulder–arm region compared to reference values. In particular, patients with moderate–end-stage lung involvement had lower muscle endurance compared to patients with no–mild lung involvement, while patients with dcSSc, as expected, had more impaired shoulder–arm AROM than patients with lcSSc. This study highlights the importance of identifying SSc patients with an enhanced risk of developing reduced muscle function, for whom adapted physical exercise to improve muscle endurance and strength capacity ought to be introduced.

Surprisingly, muscle function did not differ between patients with lcSSc and dcSSc, in contrast to what was previously reported (11, 12, 29). One reason for this discrepancy may be the relatively low prevalence of patients with dcSSc in our cohort, which is in line with another Swedish study (30), whereas this was not the case in other publications (11, 31). Other reasons could be that various tools have been used to assess muscle function in previous studies in SSc, or it was not reported (11, 12, 29). The manual muscle test on eight muscle groups (MMT8), included in the Medsger disease severity scale (21) and commonly used by physicians in the clinic, measures proximal isometric muscle strength, but not muscle endurance. It has been found to be valid in patients with inflammatory myopathies, but displays limited sensitivity to detect mild muscle weakness (32, 33). How best to measure muscle strength and endurance in patients with SSc in a clinical setting should be further investigated. Isometric muscle endurance has been measured in SSc (7, 34) but measurements with expensive equipment are not feasible in a clinical setting, and the method is time consuming and requires a highly motivated patient. The same also goes for isokinetic dynamometrics, as used in a recent study (8). We chose to use the FI-2 because it is feasible to use in the clinical setting. The FI-2 probably also measures muscle strength in patients who are only able to perform up to 10 repetitions, while measuring muscle endurance in patients who can perform > 15 repetitions (35). In our study, 84% and 88% of the patients could perform ≥ 15 repetitions in the right shoulder and hip flexion, respectively, supporting that the FI-2 measured muscle endurance in the majority of cases in the study sample. The results of our study clearly demonstrate reduced dynamic muscle endurance in patients with SSc, similar to findings in patients with polymyositis and dermatomyositis (24).

To our knowledge, the TST has not previously been used in studies including patients with SSc. All patients with SSc, regardless of subtyping or lung subgrouping, needed longer time to complete the TST compared with predicted values based on healthy subjects. This impaired muscle function is similar to findings in another study with patients with SSc (9), where the participants completed fewer sit-to-stands than the normative data test when using the 30-second Sit-to-Stand test (36), although this test probably also measures muscle endurance.

SSc has a vascular component, characterized by both microvascular and macrovascular involvement (37, 38), and we found that patients with moderate–end-stage lung disease had higher Medsger scores in both the peripheral vascular and kidney organ systems than those with no–mild lung disease. The impaired circulation of the heart and the lungs also affects muscle function (37, 39). As 40% of all patients in the present study had moderate–end-stage lung involvement it is therefore not surprising that they also had worse muscle function than those with less lung involvement. These results are in line with our earlier study, where patients with more lung involvement reported worse muscle strength than those with less lung involvement (4). Similar results are presented by Lima et al (7), where the patients had lower percentages of forced vital capacity and diffusing capacity of the lungs for carbon monoxide of predicted values, lower isometric maximal voluntary contractions, and worse muscle endurance in the quadriceps muscles in comparison with healthy controls.

The reduced AROM in the shoulder–arm of patients with SSc, and the difference between lcSSc and dcSSc, confirm the results from earlier studies (15, 16). Balint et al (16) found that patients with dcSSc had significantly more contractures than lcSSc patients, and patients with lung involvement had more contractures than those without. In our study, there was a tendency (p = 0.05) towards lower AROM in the left shoulder–arm region in patients with moderate–end-stage lung involvement compared with patients with no–mild lung involvement. That we used all patients in our sample when making comparisons with the reference values, which are based on women only (28), is a limitation in our study.

The FI-2 has good to excellent interrater and intrarater reliability and good construct validity in patients with polymyositis and dermatomyositis (24). The TST has a high degree of reliability and has been validated in rheumatoid arthritis (RA), polymyositis, dermatomyositis, and other chronic diseases (25, 40). The FSA has good test–retest reliability and construct validity in patients with RA (41). These tests have not been validated for patients with SSc, but they are all functional tests not requiring formal training or expensive equipment and are thus easy to use in clinical practice. All observers collecting the reference values and patient data in this study had experience of performing the tests. Validation of FI-2, TST, and FSA in SSc is ongoing. We chose to only use two tasks of the FI-2 test, i.e. shoulder and hip flexion. Clinical experience from our group indicates that these two tasks are the most informative of the seven tasks in the FI-2. As each muscle group is validated separately, it is possible to select and perform specific tasks (24).

Reference data for the FI-2 were collected at two clinics in an ongoing study and from one of the clinics only from the dominant side. This is a limitation; nevertheless, the overwhelming majority of the general population is right-handed (42) as well as right-footed (43). The average differences between the right and left sides in FI-2 in our sample are within the measurement error of the shoulder and hip flexiontask (ten and five repetitions, respectively) (24).

The difference in numbers of patients with lcSSc and dcSSc might hamper the power of the study based on the disease subtypes. There were also missing cases in our data, since some patients had not completed either of the pulmonary function tests or TST. The missing data found in TST are mainly explained by the fact that not all patients could complete the 10 full stands; either they could not complete the 10 stands owing to weakness or they stopped because of incorrect execution (using hands to push up or for maintaining balance) (n = 13). Other reasons included those of an orthopaedic nature (n = 3), e.g. severe osteoarthritis in the hips and/or knees causing patients to stop the test prematurely because of pain, and severe cases of lung fibrosis and/or PAH causing severe dyspnoea and/or desaturation upon exertion (n = 5).

This study highlights the importance of muscle resistance training in patients with SSc. Physical exercise has proved to be safe and beneficial for improving mainly muscle strength in SSc (34, 44, 45). Future studies concerning muscle endurance training are needed, as well as studies to explore mechanisms contributing to reduced muscle function in SSc.

Conclusion

Patients with SSc have markedly reduced upper and lower extremity muscle function and impaired AROM in the shoulder–arm compared to reference values. The most severely reduced muscle function was observed in patients with SSc with moderate–end-stage lung involvement. We suggest that assessment of physical function in SSc should include functional dynamic muscle endurance and functional AROM, and not be limited to measuring overall muscle strength, especially in patients with moderate–end-stage lung involvement.

Acknowledgements

We would like to thank all the participating patients and the staff at the Unit of Rheumatology for data collection, especially the physiotherapists who supervised the tests. We thank Susanne Pettersson, RN, PhD, Adrian Levitsky, PhD, and Karen Hambardzumyan, PhD student, for their willingness to share their knowledge and advice. We would also like to thank Monica Holmner, patient partner, for valuable comments on the manuscript. Financial funding was obtained from the Doctoral School in Health care Sciences (NFV) at the Karolinska Institutet, the Swedish Rheumatism Association, Swedish Heart–Lung Foundation, Swedish Research Council, Stockholm County Council (ALF), King Gustaf V’s 80th Birthday Foundation, the Swedish Society of Medicine, Karolinska Institutet Foundations, and Stig Thune Foundation.

Disclosure statement

No potential conflict of interest was reported by the authors.