Article Text

Concise report
Predictive factors of hand radiographic lesions in systemic sclerosis: a prospective study
  1. J Avouac1,
  2. G Mogavero2,
  3. H Guerini3,
  4. J L Drapé3,
  5. A Mathieu2,
  6. A Kahan1,
  7. Y Allanore1
  1. 1Department of Rheumatology A, Paris Descartes University, Cochin Hospital, APHP, Paris, France
  2. 22nd Chair of Rheumatology, University of Cagliari, Cagliari, Italy
  3. 3Department of Radiology B, Paris Descartes University, Cochin Hospital, APHP, Paris, France
  1. Correspondence to Pr Allanore Yannick, Hôpital Cochin, Service de rhumatologie A, 27 rue du faubourg Saint Jacques, 75014 Paris, France; yannick.allanore{at}


Objective To examine the outcomes of hand radiographic x-rays in patients with systemic sclerosis (SSc) and to identify risk factors for the progression of hand radiographic lesions in a prospective cohort.

Methods Dual time-point x-rays were systematically performed after a median interval of 5 years (range 4–7 years) in 103 consecutively recruited patients with SSc. Univariate and multivariate Cox proportional hazards models evaluated predictors of progression of hand radiographic lesions.

Results Radiographic progression of erosive arthritis, acro-osteolysis, calcinosis and flexion contracture occurred in 24, 22, 27 and 18 patients, respectively. Multivariate Cox regression analysis did not identify any predictor of the progression of erosive arthritis. Digital ulcers were shown independently to predict the progression of acro-osteolysis and calcinosis (HR 12.43, 95% CI 1.97 to 88.40 and 3.16, 95% CI 1.22% to 9.43%, respectively). The diffuse cutaneous subset was shown to be an independent predictor of the progression of flexion contracture (HR 7.52, 95% CI 1.21 to 43.93).

Conclusion The results highlight the striking level of hand radiographic lesions in SSc and suggest close monitoring of patients with the diffuse cutaneous subset for the occurrence or worsening of this complication. The results also show that severe peripheral vascular involvement predicts both acro-osteolysis and calcinosis, highlighting their vascular background.

Statistics from

Systemic sclerosis (SSc) is a connective tissue disease characterised by vascular, immune and fibrotic changes in the skin and some internal organs.1 Besides being an incurable disease, SSc can influence all aspects of an individual's life, including the performance of everyday occupations.2 Impaired hand function has been identified in SSc as a major source of difficulties in activities of daily living.3 Joint, bone and soft tissue involvement has been shown to contribute significantly to impaired hand function.4,,8 This is important since the striking level of clinical and radiographic hand osteoarticular involvement has recently been highlighted in SSc (see table 1 in online supplement).4 9,,13 However, none of the previous studies had longitudinal follow-up, precluding the identification of predictors of hand radiographic progression. A study was therefore undertaken to examine the evolution of hand radiographic x-rays during the course of the disease and to identify the most informative risk factors for the progression of hand radiographic lesions in patients with SSc. To this end, we performed a longitudinal study with systematic analysis of dual time-point x-rays in a cohort of patients with SSc.


Full details of the methodology are shown in the online supplement.

Study population

The study population consisted of a cohort of 120 patients consecutively referred to the Rheumatology A department over a 2-year period.

Clinical, laboratory and systemic assessments

The information collected and examinations performed at baseline in all participating patients have been published elsewhere and are detailed in the online supplement.9

Analysis of x-rays

Standard posteroanterior views of the hands and wrists were obtained for all patients at baseline and at the second evaluation time-point. Three radiographic patterns (joint, bone and soft tissue abnormalities) were predefined according to our initial study. Criteria for the evolution of hand radiographic lesions are provided in table 2 the online supplement. We considered for the analysis total radiographic progression for each pattern of abnormalities, defined by the occurrence of incident or worsened lesions compared with baseline.

Statistical analysis

Predictors of hand radiographic progression were evaluated by univariate and multivariate Cox proportional hazards models and summarised as HR and 95% CI.


Study population

We initially included 120 consecutive patients between 2002 and 2004 who had a first x-ray. From these 120 patients, 103 patients underwent a second x-ray between 2 May 2008 and 31 April 2009 and these form the basis of the final analysis. The mean time between the two x-rays was 5.3±1.4 years (median 5 years, range 4–7). The baseline characteristics of the 103 patients are shown in table 1.

Table 1

Characteristics of the 103 study patients

Evolution of hand radiographic lesions

The interobserver variation coefficient was 7.1% (95% CI 4.9% to 8.7%) (see figure 1 in online supplement). Radiographic progression of erosive arthritis, acro-osteolysis, calcinosis and flexion contracture occurred in 24, 22, 27 and 18 patients, respectively. The characteristics of patients with incident and worsened lesions are shown in table 2. Considering the three radiographic patterns, 34 patients developed at least one new radiographic lesion, in 27 patients the lesions were worse than at baseline and 52 patients had progression of hand radiographic lesions.

Table 2

Incidence of radiographic progression

Association between radiographic abnormalities

The likelihood of acro-osteolysis was higher in patients with calcinosis (46% vs 14%, p=0.001 at baseline, 42% vs 18%, p=0.009 at the time of the second x-ray) and flexion contracture (46% vs 10%, p<0.0001 at baseline and 47% vs 20%, p=0.005 at the time of the second x-ray).

Patients with flexion contracture were more likely to have joint space narrowing of the wrist and/or metacarpophalangeal joints at baseline (46% vs 18%, p=0.005) and at the time of the second x-ray (56% vs 25%, p=0.002).

Predictors of total radiographic progression

Joint pattern

Univariate analyses identified flexion contracture and increased systolic pulmonary artery pressure as predictors of radiographic arthritis progression (table 3 and also table 3 in online supplement). Antibody status for rheumatoid factor and anti-CCP2 antibodies, elevated acute reactants and functional disability did not predict the occurrence of new erosions and progression of radiographic arthritis. Multivariate Cox regression analysis did not identify any independent predictor of radiographic progression.

Table 3

Predictors of radiographic progression in univariate and multivariate Cox regression analysis

Bone pattern

Patients who developed acro-osteolysis and those in whom acro-osteolysis was worse on the second x-ray were more likely to have criteria of severe disease at baseline, especially flexion contracture, severe vascular and interstitial lung involvement. Only two disease characteristics were confirmed as independent predictors in multivariate analysis: the presence at baseline of subcutaneous calcifications (HR 6.17, 95% CI 1.16 to 47.23) and of digital ulcers (HR 12.43, 95% CI 1.97 to 88.40).

Soft tissue pattern

Men with the diffuse cutaneous subtype of SSc, digital ulcerations and decreased carbon monoxide diffusion capacity/alveolar volume were more likely to develop progression of radiographic calcinosis (table 3). Among these factors, multivariate analysis identified male gender (HR 3.99, 95% CI 1.45 to 12.13) and digital ulcerations (HR 3.16, 95% CI 1.22 to 9.43) as independent predictors.

Further development or worsening of flexion contracture was more likely in patients with the diffuse cutaneous subtype of SSc. The likelihood of radiographic progression of flexion contracture was also significantly higher in patients with acro-osteolysis, radiographic arthritis and functional disability (table 3). In multivariate Cox regression analysis, the diffuse cutaneous subtype was the only independent predictor of flexion contracture progression (HR 7.52, 95% CI 1.21 to 43.93; p=0.03).


This is the first prospective study to determine the outcomes and identify predictors of hand radiographic lesions in patients with SSc. This systematic examination of dual time-point x-rays confirms the striking level of hand radiographic lesions in SSc.

Radiographic progression of erosive arthritis occurred in 23% of patients with SSc. For comparison, in patients with rheumatoid arthritis (RA) not treated with biological agents, radiographic progression was reported to occur in 52% of patients at 2 years and in 59% of patients at 10 years.14 15 All patients who developed incident erosive arthritis and 13 of 14 patients in whom erosive arthritis worsened did not fulfil the American College of Rheumatology classification criteria for the diagnosis of RA, suggesting the existence of primary erosive arthropathy in SSc which is unexplained by overlap with RA.9 10 13 16,,19 However, only 5 of 24 patients in whom there was progression of erosive arthritis had evidence of persistent clinical synovitis. This low frequency of synovitis associated with the high frequency of distal interphalangeal joint involvement in this sample of postmenopausal women supports cartilage damage—which could be related to osteoarthritis or another unidentified mechanism—rather than erosive inflammatory joint disease.

Digital ulcers were identified as independent predictors of the radiographic progression of acro-osteolysis, which confirms our previous findings of an association between acro-osteolysis and vascular digital and systemic complications.9 These findings support the primary targeting of SSc vasculopathy for the management of acro-osteolysis, although other pathogenic hypotheses such as vitamin D deficiency must also be taken into consideration.16 17

The predictive value of digital ulcers for the radiographic progression of calcinosis is in accordance with our previous results which showed a higher likelihood of calcinosis in patients with severe vascular digital involvement.9 Thus, vascular injury may be involved in the pathogenesis of such lesions and treatment of calcinosis, like acro-osteolysis, should primarily target SSc vasculopathy. Unfortunately, there are no robust data showing that calcium channel blockers have a positive effect on calcinosis, and there is still no effective treatment for calcinosis.18 19

We identified the diffuse cutaneous subset as an independent predictor of radiographic progression of flexion contracture, which highlights the need for close monitoring of patients with this cutaneous subset for the occurrence or worsening of this complication. Moreover, future therapies aimed at the prevention of flexion contraction should focus on this subgroup of patients.

Our study has several limitations that deserve consideration. The number of patients with new lesions was too low to study accurately the relationship between the time between the first and second x-rays and the number of new lesions. Clinical and laboratory parameters were collected only at baseline, which meant we were unable to study their change over time and to correlate them with radiographic progression. Analysis of the clinical features of the hand, especially synovitis, as predictors of hand radiographic progression was not beyond the scope of this study and should be further assessed. To that end, determination of the predictive value of clinical articular involvement for the worsening of SSc is an ongoing project developed using prospective data from the EUSTAR registry. Values of MRI/ultrasound synovitis on osteo-articular outcomes will also have to be investigated in the future.

The design of our study did not allow the determination of the annual incidence of radiographic lesions and it is noteworthy that the interval between the two x-rays was not the same for all patients. All patients with SSc were treated with calcium channel blockers and some also received physical treatment for flexion contracture which may have influenced our results. Our evaluation of x-rays was only qualitative for each pattern of radiographic involvement. Thus, no correlation could be identified with the severity and extent of the hand x-ray lesions as no tool or radiographic score is yet available and their validation would need a very large sample of patients.

In conclusion, our study provides reasonably accurate information that was not previously available concerning the hand radiographic outcome of patients with SSc. In addition, it has identified simple markers of hand radiographic lesions that have predictive value in this patient sample and showed a good reproducibility of x-ray analysis.


View Abstract


  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the local ethics committee from Paris/Cochin.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.