To determine the clinical and immunologic features of systemic sclerosis (SSc) in a large group of children and describe the clinical evolution of the disease and compare it with the adult form.
To determine the clinical and immunologic features of systemic sclerosis (SSc) in a large group of children and describe the clinical evolution of the disease and compare it with the adult form.
Data on 153 patients with juvenile SSc collected from 55 pediatric rheumatology centers in Europe, Asia, and South and North America were analyzed. Demographic, clinical, and immunologic characteristics of children with juvenile SSc at the onset, at diagnosis, and during the disease course were evaluated.
Raynaud's phenomenon was the most frequent symptom, followed by skin induration in ∼75% of patients. Musculoskeletal symptoms were present in one-third of patients, and the most frequently involved internal organs were respiratory and gastrointestinal, while involvement of renal, cerebral, and cardiovascular systems was extremely rare. Antinuclear antibodies were present in the sera of 81% of patients. Anti–topoisomerase I (Scl-70) and anticentromere antibodies were found to be positive in 34% and 7.1% of patients, respectively. Involvement of the respiratory, gastrointestinal, and cardiovascular systems was more frequent and occurred earlier in patients who died than in those who survived. Compared with the adult form, juvenile SSc appears to be less severe, with the involvement of fewer internal organs, particularly at the time of diagnosis, and has a less characterized immunologic profile.
This study provides information on the largest collection of patients with juvenile SSc ever reported. Juvenile SSc appears to be less severe than in adults because children have less internal organ involvement, a less specific autoantibody profile, and a better long-term outcome.
Systemic sclerosis (SSc) is a rare multisystemic disease characterized by inflammation, vascular abnormalities, and fibrosis that affects the skin and various internal organs (1). Approximately 10% of all adults with SSc report the onset of the disease during childhood, with a mean age at onset of 8 years (2).
Signs and symptoms of organ involvement usually evolve over years. Therefore it is crucial to identify these patients early, before classic skin changes occur. Except for one case collection (3), there are very few published studies of children with SSc and related disorders. For this reason, the clinical and serologic features of SSc in children are still not well known, particularly at the onset of the disease.
As part of a multiphase project, data on demographic, epidemiologic, clinical, and laboratory features of patients with juvenile SSc were retrospectively collected from several pediatric rheumatology centers around the world. In this report, we describe the clinical and serologic characteristics at the onset of the disease and at the time of diagnosis and compare these features with those seen in adult SSc.
Information on demographic, epidemiologic, clinical, and laboratory characteristics and treatment of patients with SSc with disease onset before the age of 16 years was solicited from patients seen in 270 pediatric rheumatology centers (166 in Europe, 42 in North America, 28 in South America, 30 in Asia, 2 in Australia, and 2 in Africa). These centers were obtained from the mailing lists of the Pediatric Rheumatology European Society and the Pediatric Rheumatology International Trials Organization. A special form was developed for collection of data regarding the following items.
1. Demographic characteristics (sex, age at the first signs or symptoms of the disease, age at diagnosis).
2. Family history of connective tissue or other autoimmune diseases, based on the history obtained by the individual physicians.
3. Organ involvement at the time of first manifestation and/or diagnosis, and anytime during the followup. Participants were asked to use the following recommendations to define organ involvement. Skin involvement was determined by the presence of skin induration, proximal or distal to metacarpophalangeal (MCP) joints, edema, or sclerodactyly. Peripheral vascular system involvement was diagnosed by the presence of any 1 of the following: Raynaud's phenomenon (RP), digital infarcts, digital pitting, or abnormal nailfold capillaries and capillaroscopy findings (such as megacapillaries and avascular areas). Lung involvement was defined by the presence of any 1 among dyspnea, abnormal chest radiography findings or high-resolution computed tomography (HRCT) findings, reduced diffusing capacity for carbon monoxide (DLCO), or reduced forced vital capacity (FVC), according to the normal values defined at each center. Cardiac disease was diagnosed by the presence of arrhythmias or pericarditis, heart failure, or pulmonary hypertension, seen on echocardiogram. Musculoskeletal involvement was defined by the presence of muscle weakness, arthritis, arthralgia, or tendon friction rubs. Gastrointestinal tract involvement was determined by the presence of dysphagia, gastroesophageal reflux, diarrhea, or weight loss. Renal disease was identified by the presence of raised creatinine levels, proteinuria, renal crisis, or persistent arterial hypertension. Nervous system involvement was determined by the report of seizures, peripheral neuropathy, or abnormal brain magnetic resonance imaging (MRI) findings.
4. Abnormal laboratory parameters at the time of diagnosis and/or anytime after diagnosis. They included white blood cell (WBC) and eosinophil count, hemoglobin count, platelet level, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, creatine kinase (CK) level, lactate dehydrogenase (LDH) level, and levels of IgG, IgA, IgM, C3 and C4. Abnormal values were determined using the normal range of laboratory standards at each participating center as a reference.
5. Serum levels of autoantibodies, including antinuclear antibodies (ANAs), anti–double-stranded DNA (anti-dsDNA), Scl-70, anticentromere antibodies (ACAs), extractable nuclear antigens (ENAs), anticardiolipin antibodies, lupus anticoagulant, and rheumatoid factor (RF). Abnormal values were determined using the normal range of laboratory standards at each participating center as a reference.
6. Type and duration of treatment.
Since patient-identifying information was not included when clinical information was collected from the charts, institutional review board approval was required only from a minority of centers, mainly in North America. Data were stored electronically on a secure computer network according to locally applicable guidelines at the participating centers.
Since there are no accepted criteria for the classification of juvenile SSc, participants were asked to report the patients diagnosed as having SSc according to the 1980 American College of Rheumatology (formerly, the American Rheumatism Association) preliminary criteria (4). They include the presence of 1 major criterion, namely symmetric thickening of the skin proximal to the MCP or metatarsophalangeal joint, or ≥2 minor criteria, including sclerodactyly, digital pitting scars, or loss of substance from the finger pad and bibasilar pulmonary fibrosis (1).
Patients with overlap syndromes and mixed connective tissue disease (MCTD) were excluded from the study. Before the analysis was completed, we carefully reviewed the characteristics of the patients and, if there was any question regarding the diagnosis of juvenile SSc, we asked the principal investigators at the participating centers to verify the accuracy of the classification of their patients in order to minimize the risk of including patients who did not have juvenile SSc.
A comparison of clinical and laboratory features between patients with juvenile SSc from the present study and adult patients was performed using published data from a large multicenter study (5). In order to make our data comparable with data from previous studies, we asked the participants to classify patients as having the diffuse or the limited form of the disease on the basis of skin involvement (6), with skin thickening above the elbows, on the trunk, or above the knees identifying diffuse disease, and skin thickening of the distal extremities identifying limited disease.
Descriptive statistics were used to report demographic, clinical, and laboratory characteristics of the patients. Data were evaluated using the chi-square test, Student's t-test, or Fisher's exact test, as appropriate. Statistical analyses were performed using SAS System Release 8 software (SAS Institute, Cary, NC). P values less than 0.05 were considered significant.
This study was conducted between January 2002 and June 2003. Of the 270 centers that had been contacted, 138 (51%) responded. Eighty-three of these centers did not have patients with SSc to include in the database. Fifty-five centers (32 European, 8 North American, 11 South American, and 4 Asian) reported 153 patients with juvenile SSc. The characteristics of the patients are summarized in Table 1. One hundred twenty patients (78.4%) were female and 33 patients (21.5%) were male, with a female:male ratio of 3.6:1. The onset of the disease occurred at a mean age of 8.1 years (range 0.4–15.6 years), and the mean time between the first sign of the disease and the diagnosis was 1.9 years (median 1 year, range 0–12.2 years).
|No. female/no. male||120/33|
|Clinical subtype, no. (%)|
|Age at onset, years|
|Disease duration at diagnosis, years|
|Duration of followup, years|
|Family history of autoimmune diseases, no. (%)||17 (11.1)|
The average followup time was 3.9 years (median 2.5 years, range 0.2–18.8 years). Among the patients studied, the earliest year of diagnosis was 1971, and the last followup visit occurred in 2005.
In 17 families at least 1 first- or second-degree relative of patients (11.1%) had an autoimmune disease. Five patients had first-degree relatives with an autoimmune disease, 9 had second-degree relatives, and 3 patients had both first- and second-degree relatives with an autoimmune disease. Three patients had a relative with SSc; in 1, it was the mother, and in 2 patients it was the grandfather. The other reported diseases were rheumatoid arthritis (RA) in 10, systemic lupus erythematosus (SLE) in 2, psoriasis in 3, and Sjögren's syndrome (SS), Crohn's disease, and autoimmune thrombocytopenia in 1 relative each.
The clinical features at disease onset and at the time of diagnosis are summarized in Figure 1. RP was the first sign of the disease in ∼70% of the patients, and in 10% it was complicated by digital infarcts. Proximal skin induration was the second most frequent symptom, being present in 41% of the patients at onset and in 63% at diagnosis. As expected, the association of RP and skin changes, eventually with some signs of internal organ involvement as well, was the key diagnostic feature. Interestingly, nailfold capillary changes were reported in 10% of patients at the onset of the disease, in 25% at diagnosis, and in 51% during the overall disease course.
At diagnosis, the most common combination of signs and symptoms of SSc was the association of RP with skin induration, in 80 of 153 patients (52.3%), with 20 patients having concomitant respiratory involvement and 6 patients having gastrointestinal involvement.
During the overall course of the disease, RP was the most frequently reported symptom (84%), followed by skin induration in 76% of patients (Table 2). Other signs of skin involvement included sclerodactyly in 66% of patients, edema in 46%, and calcinosis in 19%.
|At diagnosis||Overall course|
|Juvenile SSc (n = 153), no. (%)||Adult SSc (n = 290), %†||P||Juvenile SSc (n = 153), no. (%)||Adult dcSSc (n = 117), %†||P|
|Edema||51/145 (35)||54||<0.001||67/145 (46)||51||NS|
|Sclerodactyly||70/153 (46)||NA||–||101/153 (66)||NA||NA|
|Skin induration||113/153 (74)||94||<0.001||116/153 (76)||95||<0.001|
|Calcinosis||14/151 (9)||26||<0.001||28/151 (19)||22||NS|
|Peripheral vascular system|
|Raynaud's phenomenon||112/150 (75)||92||<0.001||128/152 (84)||91||NS|
|Digital infarcts||28/149 (19)||17||NS||43/149 (29)||22||NS|
|Digital pitting||42/150 (28)||62||<0.001||57/150 (38)||62||<0.001|
|Abnormal nailfold capillaries||38/148 (26)||NA||–||60/150 (40)||NA||NA|
|Abnormal capillaroscopy||36/146 (25)||NA||–||78/151 (52)||NA||NA|
|Dyspnea||15/151 (10)||62||<0.001||27/151 (18)||67||<0.001|
|Abnormal chest radiography findings||18/149 (12)||64||<0.001||43/149 (29)||72||<0.001|
|Abnormal chest HRCT findings||7/148 (5)||85||<0.001||34/148 (23)||NA||NA|
|Reduced DLCO||12/150 (8)||51||<0.001||40/150 (27)||58||<0.001|
|Reduced FVC||17/150 (11)||27||<0.001||63/150 (42)||44||NS|
|Pericarditis/arrhythmias||8/153 (5)||21||<0.001||15/153 (10)||NA||NA|
|Heart failure||3/150 (2)||3||NS||11/150 (7)||3||NS|
|Pulmonary hypertension||2/153 (1)||NA||–||11/153 (7)||NA||NA|
|Muscle weakness||18/150 (12)||24||0.003||37/152 (24)||27||NS|
|Arthritis||26/150 (17)||18||NS||41/152 (27)||17||NS|
|Arthralgia||39/148 (26)||52||<0.001||54/150 (36)||60||<0.001|
|Tendon friction rubs||8/142 (6)||13||0.007||16/142 (11)||23||<0.01|
|Dysphagia||15/151 (10)||NA||–||36/152 (24)||NA||NA|
|Gastroesophageal reflux||12/149 (8)||79||<0.001||45/150 (30)||NA||NA|
|Diarrhea||3/151 (2)||11||0.002||15/151 (10)||10||NS|
|Weight loss||27/150 (18)||NA||–||41/151 (27)||NA||NA|
|Raised creatinine/proteinuria||5/153 (3)||18||<0.001||8/153 (5)||NA||NA|
|Renal crisis||0||NA||–||1/153 (1)||0||NS|
|Hypertension||2/153 (1)||18||<0.001||4/153 (3)||17||<0.001|
|Seizures||1/153 (1)||NA||–||4/153 (3)||NA||NA|
|Peripheral neuropathy||1/153 (1)||NA||–||1/153 (1)||NA||NA|
|Abnormal brain MRI findings||3/153 (2)||NA||–||4/153 (3)||NA||NA|
Abnormalities on capillaroscopy were identified in 52% of the patients while signs suggestive of severe vasculopathy, such as digital infarcts and pitting, were reported in approximately one-third of patients. Respiratory system involvement was manifested by reduced FVC (42%) and DLCO (27%). Radiologic signs of lung fibrosis were revealed in 29% of patients by standard radiography and in 23% by HRCT. Nonspecific symptoms, such as dyspnea and cough, were reported more rarely (18% and 12%, respectively).
Cardiac involvement was rarely reported. Ten percent of patients developed pericarditis, 7% developed pulmonary hypertension, 7% developed heart failure, and 10% developed arrhythmias. Musculoskeletal symptoms, such as arthralgia (36%), arthritis (27%), and reduced muscle strength (24%), were present in many children with SSc.
In 30% of patients, 24-hour pH monitoring and/or gastroscopy showed gastroesophageal reflux, while gastroesophageal dysmotility, suggested by dysphagia, was described in 24% of patients. Weight loss was reported in one-third of patients, while diarrhea or constipation was reported in only 10%.
Scleroderma renal crisis was described in only 1 patient (0.7%), while 8 (5%) had increased urinary protein excretion, and 7 (5%) had elevated creatinine levels. Very few patients experienced nervous system involvement. Seizures were reported in 4 patients (3%), MRI abnormalities of the brain in 4 (3%), and peripheral neuropathy in 1 patient (1%).
Data on clinical outcome were available for 127 patients with SSc, with the remaining patients being lost to followup. Fifteen of 127 patients (11.8%) died. The causes of death were cardiac failure in 10 patients (2 of whom had pulmonary hypertension), renal failure in 2 (1 also had pulmonary hypertension), respiratory failure in 2, and septicemia in 1.
The mean age at onset of the disease in patients who died was 10.4 years (range 5.1–15 years), which was slightly higher than that in surviving patients (8.4 years [range 0.3–16 years]), while the time interval between the onset of the disease and diagnosis was significantly shorter (mean 8.8 months [range 0–49 months] in those who died versus 23 months [range 0–149 months] in survivors; P < 0.001). This observation suggests that in patients who died, the clinical manifestations were severe, leading to an earlier diagnosis from the time of symptom onset. The course of the disease was rapidly fatal in some patients and slower in others, with deaths occurring from 4 months to 18.8 years after onset of the disease (mean 4.5 years, median 3.8 years). Four of the patients (26.7%) died in the first year after diagnosis, and 11 of 15 (73.3%) died within 5 years.
All laboratory tests were performed in >80% of the patients, and the given percentages are based on the available denominators. WBC and peripheral blood eosinophil counts were elevated in 18.6% of patients. Inflammation markers such as ESR and CRP were elevated in 34.8% and 12.6% of patients, respectively. Increased levels of muscle enzymes such as CK and LDH were reported in 20.2% and 24.7%, respectively. Serum IgG levels were elevated in one-third of patients, while less frequently, increases in IgA and IgM were reported. C4 was decreased in 17% of tested patients, and C3 in 8.6%.
As shown in Table 3, results of testing for ANA were available in almost all patients (98%). They were found to be positive in the majority (80.7%). ENAs were tested in 120 of 153 patients (78.4%) and were positive in 42.5%. Scl-70 was positive in 36 of 106 tested (34%), ACAs were positive in 6 of 84 (7.1%), and Sm in 2 of 84 (2.3%). Interestingly, 7 ENA-positive sera showed no specificity.
|Serum autoantibody||Juvenile SSc, no. (%)||Adult dcSSc, no. (%)†|
|ANA||121/150 (80.7)‡||233/249 (93.6)|
|Scl-70||36/106 (34)||116/269 (43.1)|
|ACA||6/84 (7.1)§||57/253 (22.5)|
|RF||19/112 (17)||53/231 (22.9)|
|aCL||8/54 (14.8)||22/215 (10.2)|
Anti-dsDNA antibodies were reported positive in 7 of 122 patients (5.7%) and 2 of these also had decreased complement levels. None of these patients had a positive family history of SLE or had developed signs or symptoms compatible with this disease during a mean 8.0 years (range 3.5–13 years) of followup.
RF was found to be positive in 19 of 112 patients (17%). Six of them (31.6%) had arthritis, but no significant correlation was found between the presence of RF and arthritis or between RF and a family history of RA. Anticardiolipin antibodies were found in 8 of 54 patients (14.8%). None of these patients exhibited either thromboembolic symptoms or alterations of the coagulation panel.
Most of the patients (147 of 153, 96.1%) received some form of treatment. Oral corticosteroids were the most frequently used treatment (48.3% of patients), followed by methotrexate (MTX) in 43.5% of patients. Cyclophosphamide (CYC) was used intravenously in 12.9% of patients and orally in 9.5%. D-penicillamine (D-Pen) was used in half of the patients. A decreasing trend in the use of this drug was observed over time, such that while 79.1% of patients whose diagnosis was made before 1998 were treated with D-Pen, only 20.8% of those diagnosed after that date received D-Pen (P < 0.0001).
The large use of combinations of immunosuppressive treatments and the retrospective nature of the study did not allow us to verify whether a particular pattern of organ involvement was associated with a preferential use of selected drugs. The most common combinations of agents were corticosteroids (intravenous or oral) and D-Pen, which was used in 16.4% of treated patients, MTX and D-Pen, which was used in 15.6% (half of whom also received corticosteroids), corticosteroids and MTX (used in 10.2% of patients), and corticosteroids/D-Pen/MTX/CYC (8.6% of patients).
The most widely used vasodilators were calcium channel blockers in 54.4% of patients, followed by angiotensin-converting enzyme inhibitors (8.8%) and prostanoids (5.4%). H2 antagonists and prokinetics were used by 7.5% and 2.7% of patients, respectively. Almost one-fourth (23.1%) of patients took nonsteroidal antiinflammatory drugs.
This study represents the largest collection of data on patients with juvenile SSc reported to date. The participation of centers from all over the world and the good response rate make this study the most representative up to now. The information obtained is crucial to understanding the features of this disease, particularly at onset, to the identification of prognostic factors, and to future development of diagnostic and classification criteria.
The previous absence of large studies of pediatric patients with SSc makes it difficult to make any comparisons. There is 1 earlier study with data on 135 children with SSc (3). However, that study focused mainly on the clinical outcome, and organ involvement was based on the judgment of the referring physician rather than on objective data. Therefore, all available data for comparison were obtained from adult data collections.
In our study population, as in previous studies of both adults and children (7–10), juvenile SSc affected mainly females, with a female:male ratio of 3.6:1. Consistent with the findings of previously published studies (3, 7, 8), the mean age at onset was 8.1 years.
Family history of rheumatic or autoimmune disease was reported in 11% of patients with juvenile SSc, consistent with recent findings in children with localized scleroderma (11). In particular, 3 patients with juvenile SSc (2.0%) had a relative with the same disease. In adult series, cases of SSc occurring in >1 family member have been reported (12). In 2 population studies, in the US and Australia, multiple cases of SSc in a family were uncommon (1.6% and 1.7%, respectively), but familial aggregation was shown more frequently than expected in such a rare disease (13, 14).
Within the limits of a retrospective study, including self-reported information that might be biased, the significance of a positive family history of autoimmune conditions in >1 of 10 patients with juvenile SSc should be considered. This might support the hypothesis that inherited genetic factors play a role in susceptibility to autoimmune diseases, although the low concordance in twins suggests a lesser role for genetic factors or a complex trait involving many genes (15).
RP is the first symptom of the disease in ∼70% of patients and, in some instances, it precedes other manifestations by years (16). As also found in our juvenile SSc population, this symptom remained the most frequent later in the disease course, affecting almost 84% of the patients in the present study. Given the rarity of this condition in childhood and the high prevalence of ANAs in juvenile SSc, patients presenting with RP and positive ANAs must be very carefully followed up even if they do not show any other clinical features of juvenile SSc. Duffy et al (16) found that in 27 children with RP, the concomitant presence of ANA positivity and abnormal findings on nailfold capillaroscopy were highly predictive of the development of a connective tissue disease. Moreover, LeRoy and Medsger have suggested that combinations of RP, SSc-type nailfold capillary abnormalities, and SSc-selective autoantibodies, with no sign of skin sclerosis, characterize a subset of SSc in adults (17).
The importance of nailfold capillary abnormalities is confirmed by our data, in which half the patients with SSc had a pathologic pattern during capillaroscopy and almost 40% had visible capillary abnormalities suggestive of scleroderma. These results are possibly biased by the limited availability of nailfold capillary microscopy in all centers collaborating in the study, but confirm results of previous studies indicating its utility both in diagnosis and in followup of connective tissue diseases such as SSc (16, 18).
In our study, the course of the disease was rapidly fatal in some patients and slower in others. In more than one-fourth of the deceased patients, death occurred in the first year after diagnosis, and in 73.3% of patients, it occurred within 5 years of diagnosis. In a smaller study (19), a few children had very rapid development of internal organ failure, particularly cardiac, leading to severe disability and eventually to death, while other patients experienced a slow insidious course with lower mortality. An in-depth analysis of the mortality data and prospective studies are needed to further evaluate these observations. The only available data on a large number of pediatric patients are consistent with ours, regarding both the prevalence and the causes of death (3).
Comparison with adult studies is difficult because in children the limited cutaneous form of SSc is rare, whereas it is far more frequent in adults. Furthermore, our data, within the limits of a retrospective study, show that none of the clinical and immunologic features that characterize the limited and the diffuse forms in adults are expressed at significantly different rates in children (data not shown). It has been found that in a substantial number of patients with childhood-onset SSc, the diagnosis is confirmed either during adolescence or as young adults (20), and it is possible that the limited subset might be underdiagnosed in younger children because of the lack of a full clinical picture.
Unfortunately, since there is no validated skin score for children, the usual parameters for adult skin scores cannot necessarily be extrapolated to a pediatric population. Indeed, it has recently been observed that healthy children have increased values on the modified Rodnan skin score (21), which is validated in adults with SSc and used as a primary outcome measure in clinical trials (22).
We compared the clinical and laboratory characteristics, at the time of diagnosis and during the overall disease course, of children in our study with those reported in adults in a large multicenter European study of 290 patients with SSc (5). Although the different methodologies of the studies as well as the diversity in diagnostic tests and treatments among the sites may limit the results, this analysis provides interesting thoughts for discussion. The comparison of the clinical features at the time of diagnosis is most reliable because it reduces the confounding signs and symptoms (e.g., concomitant diseases or unhealthy lifestyle, such as smoking, obesity, etc.) that are particularly present in adults and during long-term followup.
As shown in Table 2, at the time of diagnosis children showed significantly less frequent involvement of all organs, except for the prevalence of arthritis and the severity of RP, as shown by digital infarcts. The differences between adult and juvenile SSc became less evident during followup. During the overall course of the disease, children showed signs of interstitial lung involvement less frequently than adults. This frequency is lower than in a previous study of 13 patients, in which pulmonary function test results were abnormal in 12 of 13 patients (23). For other internal organs, a similar pattern of involvement has been seen in adult and pediatric patients, with the exception being gastroesophageal dysmotility (which has been reported in ∼25% of our patients and in 50–80% of adults with diffuse cutaneous SSc), arterial hypertension, and musculoskeletal symptoms, which are significantly much more common in adults (5, 24). In contrast, in a recent study (20), muscle inflammation was observed in up to 38% of children with SSc and was a distinguishing feature from adult SSc. These results, however, could be influenced by the fact that the population examined in that study included patients with overlap syndromes and MCTD, which have more frequent muscle involvement.
Other differences with SSc in adults (5) can be seen in the prevalence of arthritis, which is slightly more common in children, RP, and skin sclerosis, which are somewhat less frequent in children. It is interesting to note that in our population the prevalence of renal crisis was very low despite the relatively frequent use of corticosteroids, which are known to be associated with increased risk of this complication. With regard to immunologic features, adults with SSc seem to have a significantly higher prevalence of ANAs and ACA positivity than children. No significant difference between the adult and juvenile forms was found in the frequency of Scl-70, RF, and ACAs (Table 3).
In conclusion, juvenile SSc appears to be less severe than the adult form. Children have less internal organ involvement, a less specific autoantibody profile, and a better long-term outcome. This study provides reliable information on the pattern of organ involvement and the outcome in patients with juvenile SSc, and represents a starting point for the development of a more comprehensive and pediatric-specific classification.
The authors are grateful to Dr. Christopher Denton for his thoughtful comments and to Francesca Loro and Fabio Vittadello for statistical support.