Elsevier

Autoimmunity Reviews

Volume 13, Issue 10, October 2014, Pages 1026-1034
Autoimmunity Reviews

Review
Systemic sclerosis evolution of disease pathomorphosis and survival. Our experience on Italian patients' population and review of the literature

https://doi.org/10.1016/j.autrev.2014.08.029Get rights and content

Abstract

The clinical spectrum and prognosis of systemic sclerosis (SSc) seem to vary among patients' populations recruited during different time periods. In order to verify this possible evolution we investigated the clinico-serological and survival rate in a large Italian SSc series (821 patients; 746 females, 75 males; mean age 53.7 ± 13.9 SD years) recruited between 2000 and 2011. The observed findings were compared with previous studies of the world literature.Compared to older Italian SSc series, the present patients' population showed a significantly increased prevalence of limited cutaneous SSc (from 72 to 87.5%; p ≤ .0001) and serum anti-centromere antibodies (from 39 to 47,4%; p ≤ .001), with a significant reduction of lung (from 81 to 63.7%; p ≤ .0001), heart (from 35 to 20.5%; p ≤ .0001), and renal involvement (from 10 to 3.8%; p ≤ .0001), and skin ulcers (from 54 to 16.5%; p ≤ .0001). Cumulative 10th-year survival showed a clear-cut increase (80.7%) compared to our previous series (69.2%). These findings were mirrored by the results of survival studies published during the last five decades, grouped according to the time periods of patients’' recruitment at the referral centers. A clear progression of 10th-year survival rates was detectable, from the 54% median survival of the oldest studies (1935–1974) to 74% and 83.5% of the more recent SSc series, 1976–1999 and after 1999, respectively. In conclusion, the favorable evolution of SSc pathomorphosis and prognosis during the last decades might be related to more diffuse physician/patient awareness of this harmful disease and availability of diagnostic tools, the consequent wider recruitment of patients in the early stages of the disease, as well as to the improved therapeutic strategies.

Introduction

Systemic sclerosis (SSc) is a connective tissue disease characterized by endothelial dysfunction, dysregulation of fibroblasts with collagen overproduction, and complex immune system abnormalities [1], [2], [3]. A frequent consequence of this chronic disease is the multiple organ damage/failure responsible for marked impairment of the patient's quality of life and increased mortality [4], [5], [6], [7], [8], [9]. Since the sixties an increasing number of clinical studies evaluated the impact of the main SSc manifestations on the disease prognosis as well the cumulative SSc survival [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45]. In our previous study we underlined the progressive improvement of the mean cumulative SSc 10th-year survival as emerging from the studies published during the nineties if compared to previous reports present in the literature [5]. This observation was hypothetically related to both different compositions of scleroderma patients' series recruited during the last years and improved SSc patients' management. In order to verify this hypothesis we further investigated the composition of SSc clinical features and the cumulative 10th-year survival in a large Italian series recruited between 2000 and 2011 and compared the observed findings to that reported in our previous study [5]. In addition, we systematically reviewed the world literature on this topic; the updated findings were carefully analyzed with particular attention to the possible evolution of the SSc pathomorphosis and prognosis during the last decades.

The present study involved 3 Italian university-based divisions of rheumatology (University of Modena, University of Ferrara, and 2nd University of Napoli) with comparable, long-term experience in SSc patient management.

The study population was recruited between 2000 and 2011 at the 3 participating centers. Eight hundred twenty-one patients (746 females, 75 males; mean age at presentation, 53.7 ± 13.9 SD years) were included in the study (Table 1).

At the beginning of the follow-up, corresponding to the time of diagnosis, all patients fulfilled the 1980 American College of Rheumatology (formerly ARA) criteria for SSc classification. In all cases the diagnosis of SSc was done by an expert rheumatologist on the basis of a wider panel of clinical, serological, and capillaroscopic parameters, as previously proposed [5], [46], [47], [48], [49], [50].

At the same time, patients were also classified based on the extent of skin sclerosis as limited cutaneous scleroderma (lcSSc: sclerosis of distal extremities, not above the elbow and knees, with or without sclerosis of neck and face) and diffuse cutaneous scleroderma (dcSSc: sclerosis of both distal and proximal extremities, with or without truncal involvement) [5].

Standardized criteria were followed for the evaluation of clinico-epidemiological parameters. In particular, the patient's age was evaluated at the following conventional times: a) at the appearance of isolated Raynaud's phenomenon; b) at the disease onset, considered to be the age at which the first non-Raynaud's sign(s) and/or symptom(s) compatible with the disease appeared, i.e. digital ischemic lesions, puffy hands, sclerodactyly with or without proximal scleroderma, dyspnea, and/or dysphagia; c) at the SSc diagnosis at the referral centers [5].

Organ involvement was evaluated according to the criteria previously described [2], [4], [5], [49], [50]. In particular, each type of involvement was defined as follows: 1) peripheral vascular: Raynaud's phenomenon with digital pitting and/or ulcerations or gangrene; 2) joint: presence of polyarthralgia or arthritis when inflammatory changes were observed in more than 2 joints; 3) muscle: isolated muscle weakness or weakness associated with elevated serum creatine kinase with or without electromyographic or histologic changes of inflammatory myopathy; 4) esophageal: dysphagia and/or esophageal radio-graphic dysmotility; 5) pulmonary: bibasilar fibrosis at high resolution computerized tomography and/or restrictive lung disease on pulmonary function tests, including diffusion capacity for carbon monoxide (DLCO); 6) cardiac: at least 1 of the following symptoms: pericarditis, congestive heart failure, severe arrhythmias, and/or atrioventricular conduction abnormalities; 7) systolic pulmonary arterial pressure (sPAP) calculated by means of B-mode Doppler echocardiography in order to detect patients with possible pulmonary arterial hypertension (sPAP > 40 mm Hg); and 8) renal: scleroderma renal crisis with increased diastolic blood pressure and/or progressive renal failure.

The presence of serum autoantibodies was investigated by means of standard techniques [4], [5], [39], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53]: antinuclear (ANA), anti-centromere (ACA) and anti-nucleolar antibodies (ANoA) by indirect immunofluorescence on Hep-2 cell lines (dilution 1:40); anti-extractable nuclear antigen (ENA) antibodies, including anti-Scl70, -Sm, -RNP, -SSA/SSB.

At the end of the study the vital status of all patients' lost to follow-up was established by contacting the patients themselves or relatives, their primary care physician, or the municipal death registry. For patients known to have died, hospital records, autopsy reports, or death certificates were examined, when possible, to establish the date and cause of death. On the whole, 92.7% of patient accountability was determined at the end of the study.

Causes of death were classified as definitely SSc-related (organ insufficiency, such as renal crisis, or by a manifestation attributable to no other causes or predisposing factor than SSc, for example severe lung fibrosis), possibly SSc-related (manifestation either aggravated by SSc-related organ involvement, such as pneumonia complicating severe lung fibrosis, or occurring with increased frequency in SSc, for example lung cancer), and unrelated to SSc organ injury or treatment.

Cumulative survival rates were computed by the Kaplan–Meier method [54], [55], and the difference between survival curves by the Mantel–Cox (log-rank) statistics using SPSS (IBM software, New York, version 17.0) [56].

To identify potentially important prognostic variables, the individual effect of some variables on survival was evaluated with the use of the Cox proportional hazard regression model. According to a stepwise selection process, variables were entered, or removed, from the regression equation on the basis of a computed significance probability (maximized partial likelihood ratio). Univariate analysis by Cox model and multivariate analysis were also performed separately for each of the independent variables. We examined the independent relationship to mortality risk of 8 main clinico-epidemiologic variables (patients' age, sex, skin subsets, lung, heart, and kidney involvement, sPAP, and skin ulcers) and 3 serologic parameters (erythrocyte sedimentation rate — ESR-, hemoglobin, ACA, and anti-Scl70). The hazard ratio for each independent variable was expressed as the exponential of the coefficient of the variable in the hazard equation.

Moreover, comparisons between variables were made using the chi-square test, with the Yates correction where appropriate. The unpaired 2-tailed Student t-test with the Bonferroni correction was used for multiple comparisons. A p value < 0.05 was considered significant [54].

The main clinico-epidemiological features of the 821 SSc patients at the beginning and the end of the follow-up are summarized in Table 1. At the last clinical evaluation 9.1% of patients' (n 75) were deceased, while 7.3% (n 60) were lost to follow-up. At the end of the follow-up, significant variations of prevalence of some clinico-epidemiological features were observed; in particular, the percentage of skin ulcers showed a significant reduction, along with an increased female/male ratio. Table 2 reports the correlations of various scleroderma clinical–epidemiological features with cutaneous subsets and serum autoantibodies. In particular, the female/male ratio and patients' mean age at diagnosis were significantly lower in the diffuse SSc subset; hypermelanosis, skin ulcers, arthritis, myositis, and lung involvement were significantly more prevalent in the diffuse cutaneous subset, while only sicca syndrome prevailed in the limited SSc. Moreover, the percentage of deceased patients was significantly lower in ACA + compared to anti-Scl70 + or ANoA + patients; in addition the prevalence of ACA was significantly higher in female patients, older individuals, and in those with limited cutaneous SSc, conversely to that observed for anti-Scl70 and ANoA. These latter autoantibodies were more frequently associated with both lung and heart involvement. Finally, serum ACA was significantly more frequent in patients with sicca syndrome.

Causes of death were definitely ascertained in 73.3% of deceased patients (Table 3). A higher percentage of deaths were observed in males than in females (24% vs 7.6%; p  .0001). Cardiovascular and pulmonary involvement, alone or in association, were the most frequent complications (54.5%) affecting the overall disease outcome.

Other important causes of death were the neoplastic complications, responsible for 26% of deaths. These percentages were comparable with that found in our previous patients' series [18], with the exception of the lower percentage of renal involvement (from 12 to 1.7%; p = .02). Of interest, the present series showed a higher percentage of deaths not related to SSc compared to the old patients' population, namely 25.5% and 12%, respectively.

The results of survival study (Table 4) revealed a cumulative 10th-year survival of 80.7% from the time of SSc diagnosis; this percentage raised to 91.7% and 95.4% of survival calculated from SSc onset and from the appearance of Raynaud's phenomenon, respectively (Table 4; Fig. 1). Interestingly, the comparison between two subgroups of patients with different disease duration at the time of diagnosis (≤ 2 or > 2 years) showed a significantly higher survival rate calculated from diagnosis in patients with shorter disease duration (Table 4). Better survival rates were also observed for females, younger vs. older subjects (> 50 years), and limited cutaneous SSc. Moreover, teleangectasias, skin ulcers, lung, heart, and renal involvement, sPAP > 40 mm Hg, ESR > 25 mm h, and presence of serum anti-Scl70 were worse prognostic factors (Table 4; Fig. 2).

The univariate and multivariate survival analysis by Cox proportional hazards model (Table 5) evidenced some risk factors according to the above prognostic findings. In particular, univariate analysis showed 4% increase in mortality risk for every increase in year of age; 236% increase mortality for male gender; while skin ulcers, lung, heart, and kidney involvement were associated with 146%, 199%, 109%, and 234% increased risk, respectively. Diffuse cutaneous subset showed an 83% increased mortality risk compared to limited variant; while the highest risk of mortality was correlated with the presence of sPAP > 40 mm Hg. Among serologic parameters, increased ESR at diagnosis (≥ 25 mm/h) was associated with 91% excess mortality, and serum anti-Scl70 with 61% increased mortality risk at univariate analysis.

The comparison of the clinico-prognostic findings observed in the present SSc series with our previously published Italian large SSc series, recruited over a longer time interval, i.e. from 1955 to 1999 [5], is summarized in Table 3. Compared to older patients' population, the present series showed a significantly lower percentage of worse prognostic features at the end of follow-up; namely, diffuse SSc subset, serum anti-Scl70 and/or ANoA, lung fibrosis, kidney involvement, and skin ulcers (Fig. 3; Table 3). Similarly, significantly lower percentage (9.1%) of deceased patients was recorded compared to our old series (30.4%). Single organ involvement responsible for death was comparable with that found in our previous patients' series [18], with the exception of the lower rate of renal involvement in the recent series (from 12 to 1.7%; p = .02). Another significant difference was the increased percentage of deaths not correlated to SSc (from 12 to 25.5%) observed in the present patients' population (Table 3).

The survival study revealed a better cumulative 10th-year survival in the present Italian SSc series compared to older patients' population [5], with a clear-cut progression in the survival rates of patients recruited during three different time intervals, namely 1955–85, 1986–99, and 2000–2011, with a survival of 60.6, 76.8, and 80.7%, respectively (Fig. 4).

A careful review of previous studies published since 1961 confirmed some interesting variations in the SSc survival rates during the last five decades [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45]. Table 6 reports the survival studies classified in three main groups according to different time periods of patients' recruitment, as well as survival rates calculated from either SSc diagnosis, disease onset, or appearance of Raynaud's phenomenon. The majority of studies reported the cumulative 10th-year survival calculated from SSc diagnosis. Interestingly, a progressive increase of the median 10th-year survival rate calculated from diagnosis was detectable. With the exception of two reports regarding Asian patients [25], [31], the oldest studies showed very low survival rates, with median 54% of 10th-year survival from diagnosis, raising to 72.6% and 83.5% of the second and third groups, respectively (Fig. 4). Out of 39 survival studies published since 1961 [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], only five reported the 10-year survival calculated from disease onset, and four from the appearance of Raynaud's phenomenon. These 9 studies were often characterized by small patients series (range 72–405) and a large distribution of survival rates [14], [19], [20], [21], [23], [24], [33], [36], [37].

The present study on large Italian SSc patients' population evidenced a clear-cut evolution of disease pathomorphosis and prognosis during the last years. Scleroderma patients recruited between 2000 and 2011 showed a significant number of favorable clinical and prognostic parameters compared to those observed during the past decades in the same geographical areas of Italy [4], [5]. In particular, the percentage of patients with limited cutaneous subset or serum ACA was statistically higher compared to our old series, while a significant reduction of well-known adverse prognostic factors, such as the prevalence of diffuse cutaneous subset and anti-Scl70 or ANoA, was also recorded; moreover, some important clinical features, namely lung, heart, and renal involvement, skin ulcers, and SSc-related deaths were significantly less frequent; all together these findings were mirrored by a significant increase of cumulative 10th-year survival and the results of Cox model analysis.

The lower rate of skin ulcers observed either at the beginning and at the end of follow-up, compared with our old series [5] might be related to both shorter disease duration at the first visit and usefulness of systemic and local treatment strategies during the last years [57], [58].

The increased 10th-year survival rate of the present series was largely confirmed by the systematic review of the world literature, regardless the geographical origin of patients' populations [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45]. Survival studies published during the last five decades revealed a progressive improvement of the overall SSc prognosis; the median of cumulative 10th-year survival rates calculated from SSc diagnosis was markedly lower in old SSc series compared to recent patients' populations (54% vs. 83.5%). The improved outcome of SSc during the last decades seems to be independent of different genetic and/or environmental pathogenetic cofactors [59], [60], and/or of different methodological approaches [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45]. From one side, it could be related to the early diagnosis and wider patient recruitment at specialist referral centers, which may reflect the entire scleroderma clinical spectrum. The early SSc diagnosis may be correlated to the widespread information about the disease clinical presentation among general practitioners and patient associations, as well as to systematic use of suitable diagnostic tools, such as capillaroscopic examination and serological markers [46], [48], [49], [50], [61]. From the other side, the contribution of recently available treatments also should be taken into account [57], [58], [62], [63], [64], [65], [66]. Pathogenetic–symptomatic treatments, mainly immunomodulating and vasoactive drugs, are earlier employed for the whole disease and/or single organ damage, that is, scleroderma renal crisis, lung or heart involvement, and digital ulcerations [57], [58], [62], [63], [64], [65], [66]. Pathogenetic–symptomatic treatments, mainly immunomodulating and vasoactive drugs, are earlier employed for the whole disease and/or single organ damage, that is, scleroderma renal crisis, lung or heart involvement, and digital ulcerations.

Considering the numerous discrepancies among survival studies present in the world literature, a number of important clinical and methodological aspects deserve to be addressed. Firstly, the extent of cutaneous involvement is considered one of the most reliable prognostic parameters; we previously demonstrated the better discriminating power of the 3-subset classification, i.e. limited, intermediate, and diffuse cutaneous SSc [5], also supported by other clinical studies [4], [18]. However, during the last years, the majority of Authors adopted the 2-subset classification, i.e. limited and diffuse cutaneous SSc [11], [12], [15], [17], [21], [22], [24], [25], [28], [30], [32], [34], [35], [36], [38], [39], [40], [43], [66], [67], [68]; in order to avoid possible misinterpretations among different studies, we recently used this latter two-subsetting classification. Rather, this option follows from the basic consideration that the disease outcome should be better predicted by multiple prognostic factors, including both the extent of skin fibrosis and the presence of some clinico-serological abnormalities [1], [2], [3], [5], [11], [12], [13], [14], [43], [44], [45], [66], [67], [68], [69], [70], [71].

SSc is characterized by a higher mortality rate if compared with other connective tissue diseases. The worse prognosis may be the consequence of severe single organ injury, mainly the kidney, lung, or heart, or more often it may be the result of multiple visceral organ involvement [1], [2], [3], [5], [11], [12], [13], [14], [43], [44], [45], [66], [67], [68]. Thus, a single clinical manifestation, even the extent of cutaneous sclerosis, per se is not necessarily a sufficient prognostic indicator in a given patient. In fact, patients with very limited or absent skin sclerosis, such as SSc sine scleroderma, may develop one or more severe visceral organ involvement that may affect the overall disease outcome [5], [44], [57], [66], [67], [68]. In the single patient the thoroughly clinical assessment and the early detection of organ damage are decisive for a correct prognostic evaluation and effective therapeutical interventions [5], [8], [60], [66]. Moreover, the employed classification criteria, the modalities of patient selection, including the concentration in the past decades of more severe cases at the referral centers, and survival calculation are particularly critical; they may explain some discrepancies among different studies [4], [5], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45]. In principle, survival rates should be calculated on the basis of disease duration, i.e. from disease onset to death; however, this modality can be misleading. Generally, there is a long delay between the actual SSc clinical onset and the diagnosis, especially for slow progressive variants of the disease; in these instances it is often arbitrary to establish the date of disease onset, even for the patients themselves. Actually, patients referred to specialist centers did not represent the entire SSc population, but a surviving cohort; that is the so-called immortality bias [72]. Certainly, the survival calculated from disease onset underestimates the true mortality from SSc; but more misleading is the survival calculated from the onset of Raynaud's phenomenon, which per se cannot identify patients with true SSc. The time interval between ‘apparently isolated’ Raynaud's and SSc onset largely varies among patients; generally, it is inversely correlated with the severity of SSc [4], [5]. In the early stages of the disease, patients with ‘suspected secondary’ Raynaud's phenomenon should be classified as having true SSc only in the presence of typical clinical, serological, and/or capillaroscopic alterations [5], [47], [51], [73], [74]. Patients with early SSc are increasingly recruited at specialist referral centers during the last years. In this respect, we observed that survival rate calculated from diagnosis in patients referred within the first 2 years of non-Raynaud's first symptoms was higher compared to those with longer disease duration. Therefore, prospective studies on SSc patient populations referred early in the course of the disease may further minimize the above-mentioned difficulties still present in retrospective clinico-prognostic investigations.

In conclusion, we can affirm that the pathomorphosis of SSc and in particular its overall prognosis is changing considerably during the latest 1–2 decades; it is suggested by the comparative analysis of clinico-serological and prognostic findings emerging from present series and the review of previously published survival studies. It is very likely that improved physician/patient awareness of this harmful disease and the availability of reliable diagnostic tools permit to recruit SSc series that may better reflect the actual scleroderma spectrum, with increasing percentage of individuals with early stages of the disease and mild clinical variants. Moreover, the role of novel pathogenetic treatments, more often introduced early in the course of the disease, can be also taken into account.

Section snippets

Take-home messages

  • The study underlines the evolution of SSc pathomorphosis and prognosis during the time.

  • Worse prognostic factors are less frequent in recently recruited Italian SSc series.

  • The prevalence of diffuse cutaneous SSc and anti-Scl70 Ab is significantly reduced.

  • Skin ulcers and main visceral organ involvement are less commonly detected.

  • Review of world literature shows increased median 10-year survival, from 54% to 83.5%.

References (74)

  • I. Marie et al.

    Prospective study to evaluate the association between systemic sclerosis and occupational exposure and review of the literature

    Autoimmun Rev

    (2014)
  • S.A. Jimenez et al.

    Potential role of human-specific genes, human-specific microRNAs and human-specific non-coding regulatory RNAs in the pathogenesis of systemic sclerosis and Sjögren's syndrome

    Autoimmun Rev

    (2013)
  • D. Rossi et al.

    The role of nail-videocapillaroscopy in early diagnosis of scleroderma

    Autoimmun Rev

    (2013)
  • M. Antic et al.

    Treating skin and lung fibrosis in systemic sclerosis: a future filled with promise?

    Curr Opin Pharmacol

    (2013)
  • P. Cipriani et al.

    Stem cells in autoimmune diseases: implications for pathogenesis and future trends in therapy

    Autoimmun Rev

    (2013)
  • M. Elhai et al.

    Systemic sclerosis at the crossroad of polyautoimmunity

    Autoimmun Rev

    (2013)
  • M.F. Careta et al.

    Prospective study to evaluate the clinical and radiological outcome of patients with scleroderma of the face

    Autoimmun Rev

    (2013)
  • M. Koenig et al.

    Predictive value of antinuclear autoantibodies: the lessons of the systemic sclerosis autoantibodies

    Autoimmun Rev

    (2008)
  • E.C. LeRoy et al.

    Scleroderma (systemic sclerosis): classification, subsets and pathogenesis

    J Rheumatol

    (1988)
  • T.A. Medsger

    Systemic sclerosis (scleroderma): clinical aspect

  • T.A. Medsger et al.

    Classification and prognosis

  • C. Ferri et al.

    Cutaneous and serological subset of systemic sclerosis

    J Rheumatol

    (1991)
  • C. Ferri et al.

    Systemic sclerosis: demographic, clinical, and serologic features and survival in 1,012 Italian patients

    Medicine

    (2002)
  • M. Elhai et al.

    Trends in mortality in patients with systemic sclerosis over 40 years: a systematic review and meta-analysis of cohort studies

    Rheumatology

    (2012)
  • A. Komócsi et al.

    The impact of cardiopulmonary manifestations on the mortality of SSc: a systematic review and meta-analysis of observational studies

    Rheumatology

    (2012)
  • E. Krishnan et al.

    Systemic sclerosis mortality in the United States: 1979–1998

    Eur J Epidemiol

    (2005)
  • R.D. Altman et al.

    Predictors of survival in systemic sclerosis (scleroderma)

    Arthritis Rheum

    (1991)
  • M.C. Arias-Nuñez et al.

    Systemic sclerosis in northwestern Spain: a 19-year epidemiologic study

    (2008)
  • C. Bryan et al.

    Survival following the onset of scleroderma: results from a retrospective inception cohort study of the UK patient population

    Br J Rheumatol

    (1996)
  • L. Czirják et al.

    Survival and causes of death in 366 Hungarian patients with systemic sclerosis

    Ann Rheum Dis

    (2008)
  • R.J. Eason et al.

    Progressive systemic sclerosis in Auckland: a ten year review with emphasis on prognosis feature

    Aust N Z J Med

    (1981)
  • A.C. Gelber et al.

    Race and association with disease manifestations and mortality in scleroderma. A 20-year experience at the Johns Hopkins Scleroderma Center and review of the literature

    Medicine

    (2013)
  • M. Giordano et al.

    Different antibody patterns and different prognosis in patients with scleroderma with various extent of skin sclerosis

    J Rheumatol

    (1986)
  • A. Hashimoto et al.

    Predictors of survival and causes of death in Japanese patients with systemic sclerosis

    J Rheumatol

    (2011)
  • R. Hesselstrand et al.

    Mortality and causes of death in a Swedish series of systemic sclerosis patients

    Ann Rheum Dis

    (1998)
  • P. Hissaria et al.

    Survival in scleroderma: results from the population-based South Australian Register

    Intern Med J

    (2011)
  • A.M. Hoffmann-Vold et al.

    Survival and causes of death in an unselected and complete cohort of Norwegian patients with systemic sclerosis

    J Rheumatol

    (2013)
  • Cited by (0)

    View full text