Article Text

Download PDFPDF

Extended report
Long-term outcome in juvenile-onset mixed connective tissue disease: a nationwide Norwegian study
  1. Siri Opsahl Hetlevik1,
  2. Berit Flatø1,2,
  3. Marite Rygg3,4,
  4. Ellen Berit Nordal5,6,
  5. Cathrine Brunborg7,
  6. Helene Hetland8,
  7. Vibke Lilleby1
  1. 1Department of Rheumatology, Oslo University Hospital, Rikshospitalet, Oslo, Norway
  2. 2Faculty of Medicine, Institute of Clinical Medicine, University of Oslo, Oslo, Norway
  3. 3Department of Pediatrics, St Olavs Hospital, Trondheim, Norway
  4. 4Department of Laboratory Medicine, Children's and Women's Health, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway
  5. 5Department of Pediatrics, University Hospital of North Norway, Tromsø, Norway
  6. 6Institute of Clinical Medicine, UiT—the Arctic University of Norway, Tromsø, Norway
  7. 7Oslo Centre for Biostatistics and Epidemiology, Research Support Services, Oslo University Hospital, Ullevål, Oslo, Norway
  8. 8Department of Rheumatology, Hospital of Southern Norway, Kristiansand, Norway
  1. Correspondence to Dr Siri Opsahl Hetlevik, Department of Rheumatology, Oslo University Hospital, Rikshospitalet, Postboks 4950 Nydalen, Oslo 0424, Norway; siri.opsahl{at}


Objectives To describe the characteristics, outcome and predictive factors of juvenile mixed connective tissue disease (JMCTD) in a nationwide cohort of patients.

Methods We examined 55 patients with JMCTD after a mean disease duration of 16.2 years (SD 10.0). Patients were registered according to Kasukawa's criteria. Remission criteria were defined according to those for juvenile idiopathic arthritis, plus absence of cytopenia, myositis, progressive sclerodactyly, lung and oesophageal manifestations. Organ damage was assessed with the Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology (ACR) Damage Index and the Juvenile Arthritis Damage Index (JADI). Medical records were reviewed for early predictors for outcome, which were assessed by multivariate logistic regression analyses.

Results Three patients developed systemic lupus erythematosus (SLE). Fifty-two patients had continuous JMCTD; the most common manifestations were: Raynaud (100%), arthritis (94%), puffy hands (77%) and pulmonary manifestations (58%). SLE-like, systemic sclerosis (SSc)-like and polymyositis (PM)-like findings were found in 98%, 77% and 48%, respectively. Over time, SLE-like and PM-like manifestations decreased, and SSc-like findings increased. At follow-up, 35 patients (67%) had active disease and 17 (33%) were in remission. In 34 patients (65%), SLICC or JADI≥1 assessments indicated organ damage. Active disease was associated with higher anti-ribonucleoprotein antibody titres at follow-up and positive rheumatoid factor (RF) at diagnosis and follow-up.

Conclusions Most patients with JMCTD had active disease and organ damage after a mean follow-up of 16.2 years. Active disease was associated with higher anti-ribonucleoprotein antibody levels and positive RF. The presence of RF at diagnosis predicted persistent disease activity.

  • Autoimmune Diseases
  • Disease Activity
  • Outcomes research

Statistics from


Mixed connective tissue disease (MCTD) is characterised by serum autoantibodies directed against ribonucleoprotein (anti-RNP). The clinical features include Raynaud phenomenon, puffy hands, arthritis, myositis and interstitial lung disease. The disease typically evolves over time, and it has generally been considered a chronic disease.1

A juvenile presentation occurs in 7%–23% of all MCTD cases.2 ,3 Few studies have examined disease characteristics and outcome in patients with the juvenile-onset form of MCTD (JMCTD). Those studies were mainly retrospective and based on small cohorts. Data on long-term follow-up are lacking.2 ,4 ,5 It is not clear to what extent paediatric onset differs from the adult form of the disease, but it has been suggested that children might experience a milder disease course than adult patients.2 ,6 In contrast, other studies have found less favourable outcomes in JMCTD compared with adult MCTD.7–9

Remission rates after long-term disease are uncertain. The prognosis seems to vary from mild disease to a disease with severe complications, such as pulmonary arterial hypertension (PAH) or pulmonary fibrosis. However, no predictive factors have been identified. Irreversible organ damage has been described in juvenile-onset systemic lupus erythematosus (SLE)10 and juvenile-onset dermatomyositis,11 but not in JMCTD.

Norway is well suited for long-term outcome studies on paediatric rheumatic diseases, partly due to its publicly funded, easily accessible health service, which ensures a low rate of undiagnosed children. In addition, all inhabitants are registered in the National Population Registry; thus, individuals can be readily tracked when they move to other parts of the country.

The aim of the present study was to describe the clinical characteristics and long-term outcome of JMCTD, including disease activity and damage, in a nationwide juvenile cohort, and to identify possible predictors of unfavourable outcome.



We searched for patients with JMCTD with five different overlapping acquisition methods to identify patients registered with the ICD-codes, M35.1 (ICD10) or 107.8 (ICD9):

  1. We interrogated university hospital databases for each of the four health regions in Norway, and identified a total of 31 patients between January 1996 and June 2015: Oslo University Hospital (OUS) and Hospital of Southern Norway in the southeast region (20 and 4 patients, respectively), St Olavs Hospital in the mid region (4 patients) and University Hospital of North Norway in the north region (3 patients). Databases of region west were not examined as these hospitals have referred patients with this diagnosis to OUS. Diagnosing and initial treatment of paediatric connective tissue disease in Norway is normally entrusted to the hospitals mentioned above.

  2. The Norwegian Systemic Connective Tissue Disease and Vasculitis Registry was established in 1998 and counted 2400 patients in 2012/2013. Of 17 patients registered here, 14 were not identified through hospital databases.

  3. The National Pediatric Patient Register for Rheumatic Autoimmune Disease was requested; 21 patients were registered, 4 of which were not found through the sources above mentioned.

  4. From a recently published Norwegian nationwide adult MCTD cohort, established by Gunnarsson et al,12 21 patients with juvenile onset were identified, and 9 of these were not found through the previous routes.

  5. A previous cohort of 26 patients with JMCTD was assessed by Aaløkken et al13 at our department in 2004, 7 of which were not identified through other sources.

These methods identified 65 unique patients with JMCTD, after excluding duplicate cases. The inclusion criteria were as follows: fulfilment of the Kasukawa et al14 or Alarcón-Segovia and Villareal15 criteria, symptom onset before age 18 years and a clinical diagnosis confirmed by a rheumatologist or paediatrician. Through chart examinations, three patients were excluded because they did not fulfil the inclusion criteria. Thus, we identified a total of 62 patients with JMCTD.

Informed consent was obtained from all participants and parents of patients <16 years old, according to the Declaration of Helsinki. The study was approved by the Regional Ethics Committee for Medical Research.

Clinical examination

All patients underwent a clinical examination at OUS. The examinations, performed by the main investigators (SOH or VL), included a joint count, a Rodnan skin score and a physician's global assessment of disease activity, based on a 10 cm visual analogue scale. Furthermore, pulmonary function tests, echocardiography and high-resolution CT of the lungs were performed. We evaluated participants according to the Kasukawa criteria: patients were classified with SLE-like, systemic sclerosis (SSc)-like or polymyositis (PM)-like findings.14 The classifications were recorded through examination of medical records at diagnosis, during the first year of the disease, during the entire disease course and at follow-up. Blood tests were analysed at OUS by routine laboratory methods, and previous values were recorded from medical charts. Rheumatoid factor (RF) IgA and IgM were considered positive when values were ≥25×103 IU/L.


Currently, there are no defined or validated criteria for remission in MCTD. Consequently, we used the criteria for defining remission in juvenile idiopathic arthritis (JIA),16 and added the absence of cytopenia, myositis, progressive lung and oesophageal manifestations, and progressive sclerodactyly (with or without Raynaud phenomenon). Clinical remission with medication was defined as inactive disease with immunosuppressive medication for a minimum of six continuous months. Clinical remission without medication was defined as inactive disease without immunosuppressive medication for a minimum of 12 months. Active disease was defined as the absence of remission.

Disease activity and disease damage

Currently, there are no validated criteria for disease activity or damage in MCTD. Consequently, we used the following disease activity and damage scores, which were developed for assessing SLE, SSc and JIA: the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI),17 the Rodnan skin score, the Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology (ACR) Damage Index18 and the Juvenile Arthritis Damage Index (JADI).19

Statistical analyses

Differences between patient subgroups were tested with the independent samples t test for continuous, normally distributed variables, and with the Mann-Whitney U-test for non-normally distributed variables. The χ2 test or Fisher’s exact test was used to test differences in categorical variables between two groups. To identify potential risk factors for observing active disease at follow-up, we performed univariable logistic regression analyses on the patient characteristics measured at diagnosis. Subsequently, we performed multivariable logistic regression analyses with a manual backward elimination procedure to identify risk factors. Any variable with a p value <0.20 from the univariable analysis was considered a candidate for the multivariate analysis. Highly intercorrelated variables (>0.7) were avoided; therefore, age was included, but not disease duration. Age and gender were forced into the model as adjustment variables. The multivariate analysis was preceded by estimations of correlations between risk factors. For these analyses, <5% of data were missing, except for RF measured at diagnosis, for which 17% of data were missing.

The standard mortality ratio (SMR) was used to compare mortality between patients and an age-matched and gender-matched Norwegian general population in 2014 (numbers from Statistics Norway). p Values ≤0.05 were considered significant. All statistical analyses were performed with IBM SPSS V.22.0 (IBM SPSS, Chicago, Illinois, USA).


Patient characteristics

Of the 62 patients identified with a confirmed diagnosis of JMCTD, 3 were deceased, and thus, 59 patients were invited to participate. Of these, three (5%) chose not to participate and one (2%) did not respond.

Of the 55 patients with JMCTD that participated in the study, 3 patients (5%) had developed clinical SLE, and their diagnosis had been changed during the disease course. These were excluded from the analyses. Thus, 52 patients with continuous JMCTD were included in the statistical analyses.

Patient characteristics are reported in table 1. The mean disease duration, from symptom onset to the follow-up examination, was 16.2 (SD 10.3) years. Disease durations varied from 0.6 to 41.2 years. One patient was of Asian descent, all others were of European descent.

Table 1

Patient characteristics in 52 patients with JMCTD

Disease characteristics

Over the entire disease course, SLE-like disease dominated, with arthritis, leucopenia or thrombocytopenia, facial erythema, lymphadenopathy and/or serositis (98%). SSc-like disease with pulmonary manifestations, oesophageal dysmotility and/or sclerodactyly was present in 77%, and 48% had PM-like disease, with muscle weakness and/or increased creatine kinase (table 2).

Table 2

Disease characteristics in 52 patients with JMCTD

In addition to Raynaud phenomenon and positive anti-RNP, the most common manifestations were arthritis (94%), puffy hands (77%) and pulmonary manifestations (fibrosis or reduced diffusion/vital capacity) (58%). Thirty-four patients (65%) had polyarticular arthritis and 15 (29%) had oligoarticular arthritis. RF IgM or IgA was positive in 33 patients (64%) in the entire cohort, in 26 (77%) of those with polyarticular arthritis and in 6 (40%) of those with oligoarthritis. Fever was present in 15 patients (29%) at diagnosis. None of the patients had nephritis during the disease course.

From diagnosis to follow-up, the proportion of patients with SLE-like findings decreased from 94% to 52%, and the proportion with PM-like findings decreased from 35% to 4%. In contrast, SSc-like findings were found in 25% of patients at diagnosis and 69% at follow-up (figure 1).

Figure 1

Disease characteristics of juvenile mixed connective tissue disease (JMCTD), according to the Kasukawa criteria over time. The proportions of patients with systemic lupus erythematosus (SLE)-like (dotted), systemic sclerosis (SSc)-like (black) and polymyositis (PM)-like disease (diagonal stripes) were measured at diagnosis, during the first year of the disease, and at a mean follow-up of 16.2 years, in 52 patients with JMCTD.

At diagnosis, one patient had PAH, diagnosed after a right-sided heart catheterisation. At follow-up, this patient had normal echocardiography results. Another patient that had PAH died from this complication. Thus, the occurrence of PAH in our cohort of JMCTD was 3.4% (2 of 59 patients initially registered with JMCTD, excluding those that developed SLE).

Forty-nine patients (94%) received immunosuppressive medications at some time during the disease course (table 2). The most common disease modifying antirheumatic drugs (DMARD) were antimalarials (83%), corticosteroids (83%) and methotrexate (69%). Anti-tumour necrosis factor had been prescribed for three patients (6%). At the 16-year follow-up, 36 patients (69%) continued to take corticosteroids and/or DMARDs.

Disease activity and disease damage at the 16-year follow-up

Remission with or without medication was observed in 17 patients (33%), of whom 8 patients (15%) were in remission without immunosuppressive medication. Active disease continued in 35 patients (67%) (table 3 and figure 2). Of the 35 patients with active disease, 12 (34%) were taking corticosteroids, 26 (74%) were taking DMARDs and 7 (20%) were not taking any antirheumatic medication.

Table 3

Disease characteristics in 52 patients with JMCTD after mean 16.2 years disease duration, according to remission and active disease

Figure 2

Active disease and remission in 52 patients with juvenile mixed connective tissue disease (JMCTD) after 16.2 years of follow-up.

Ten patients (19%) had an SLEDAI score ≥1, and 11 (21%) had at least one joint with active arthritis (table 3). A total of 34 patients (65%) had disease damage at the 16-year follow-up, defined as SLICC ≥1 (58%) and/or JADI≥1 (25%). Twenty-six patients (50%) had one or more joint with limited range of motion. The main causes of an SLICC≥1 were pulmonary manifestations (n=24 patients, 46%); of these, 22 (42%) had reduced diffusion or vital capacity and 14 (28%) had CT-verified pulmonary fibrosis. Thirteen patients (25%) had musculoskeletal damage (13 had deforming arthritis and 2 had muscle atrophy). One patient had neuropsychiatric damage (cerebrovascular event), and two had peripheral vascular damage (venous thrombosis).

Compared with patients in remission, those with active disease had higher levels of anti-RNP (p=0.04), they more frequently had anti-RNP levels >240×103 U/L (p=0.03), and they reported more severe Raynaud phenomenon (p=0.03) (table 3). A positive RF was found in more patients with active disease than in those with remission (43% vs 12%, p=0.03).

Predictors for active disease

In a multivariable model, a positive RF at diagnosis was identified as a predictor for ongoing active disease at the 16-year follow-up (OR: 6.0, 95% CI 1.3 to 27.5), after adjusting for age and gender (table 4).

Table 4

Early predictors of active disease after mean 16.2 years disease duration in 43 patients with JMCTD*


Of the 59 patients with JMCTD identified (excluding those that developed SLE), 3 died. Thus, the mortality was 5.1% over 16.8 years (mean follow-up for these 59 patients). The SMR was 5.0 (95% CI −4.8 to 14.9; p=0.99) for men, and 3.3 (95% CI −1.3 to 7.8; p=1.0) for women. The causes of death were pneumococcal meningitis, carcinomatosis of the lungs and pulmonary embolism combined with PAH.


In this Norwegian JMCTD study, the most frequent findings were Raynaud phenomenon, puffy hands, arthritis and pulmonary manifestations. SLE-like findings were most common at diagnosis, but these manifestations decreased over the disease course. In contrast, SSc-like findings became more common over time. Two thirds of patients continued to show active disease after mean disease duration of 16 years; only 15% were in remission without medication. A positive RF detected at the time of diagnosis was associated with active disease at follow-up. To our knowledge, this study is only to report predictive factors in JMCTD.

Raynaud phenomenon and a positive anti-RNP are hallmarks of MCTD. All of our patients had Raynaud phenomenon early in the disease course, and 94% had this symptom at follow-up. Both Raynaud phenomenon and a positive anti-RNP appeared to persist in the vast majority of patients with JMCTD, even in patients that achieved remission. We found significantly higher anti-RNP levels in patients with active disease than in those with inactive disease, which suggested an association between disease activity and anti-RNP levels. These findings, together with the very low drift in diagnosis (only three patients developed SLE), and the fact that none of our patients had renal involvement, support the notion that MCTD may be a distinct disease entity.

SLE-like disease was the most common manifestation at JMCTD presentation, and during the disease course (98%). The most frequent SLE-like finding was arthritis, followed by facial erythema, and leucopenia or thrombocytopenia. Fever was also frequently found at diagnosis. These findings were consistent with previous study results,2 ,4 ,20 and they are known to be typical symptoms at disease onset in juvenile-onset SLE.21 ,22 However, after a mean follow-up of 16 years, the incidence of SLE-like findings decreased, and SSc-like findings became the most common disease manifestation.

All SSc-like findings increased from onset to follow-up. Pulmonary manifestations were the most common features among SSc-like findings; 28% of patients had CT-verified pulmonary fibrosis after 16 years. This rate was lower than that reported in a Norwegian study on adult patients with MCTD,23 but the same as that found in a previous Norwegian JMCTD study.13 Mier et al reported that 30% had fibrosis on CT, but only a few selected patients were examined.4 Lung disease might be less common in the juvenile than in the adult population, as shown by Kotajima.2 Also, none of the deaths in our cohort were caused by Interstitial Lung Disease (ILD), in contrast to findings by Gunnarsson et al23 in adult MCTD showing a significant association between severe ILD and increased mortality. Further studies are needed to elucidate the prognosis of pulmonary complications in JMCTD.

The occurrence of PAH in our cohort was comparable to that found in a Norwegian study on PAH in adult MCTD (3.4%).24 Studies on cohorts with JMCTD have reported variable PAH incidences, from 6% to 14%.2 ,4 ,7 The lower prevalence among our patients might be explained by the fact that they were relatively unselected; in contrast, previous studies assessed patients that had been referred to tertiary care units.

PM-like findings occurred less frequently than SLE and SSc-like manifestations. Also, the frequency of PM-like findings decreased during the disease course, consistent with findings in previous studies.4 ,9 Taken together, our results suggested that inflammatory presentations in JMCTD, such as myositis, arthritis and serositis, decreased over time, and the more fibrotic scleroderma-like features increased with disease progression. This observation was also described in previous studies.4 ,5 ,20

In this study, 67% had active disease after a mean of 16 years, and only 15% were in remission without medication. Tellier et al reported similar results in a recent study; 62% of their patients had continuous active disease.25 Additionally, Michels found low remission rates9; however, comparisons should be made with caution, due to the lack of validated criteria for remission in MCTD.

A positive RF was found in a large proportion of our patients (64%). This was consistent with previous JMCTD studies, where proportions varied between 37% and 81%,4 ,5 ,20 ,25 ,26 and comparable to that found in adult rheumatoid arthritis (RA) (50%–80%),27 but much higher than what is reported in JIA (approximately 5%).28 ,29 We also found that the presence of RF was an early predictor of ongoing active disease after 16 years. To our knowledge, this study is the first to describe predictive factors in JMCTD. In RA and JIA, it is well known that RF is a prognostic factor for both continued disease activity30 and increased joint damage,28 ,31 ,32 but the significance of the marker in MCTD is not known.

SLICC assessments showed that over half of our patients had irreversible organ damage, mainly caused by pulmonary manifestations. Previously, SLICC has not been assessed in JMCTD. However, in a Norwegian cohort with juvenile SLE, Lilleby et al21 found that 61% had an SLICC score ≥1, most frequently due to neuropsychiatric, renal and musculoskeletal damage. Thus, juvenile systemic lupus erythematosus (JSLE) and JMCTD may therefore cause damage in comparable proportions of patients, but in different organs.

We found a mortality rate of 5.1% over a period of 16.8 years. Mortality in previous JMCTD reports varied from 2.8%2 to 29%.7 A review from 20054 estimated an annual disease-specific mortality of 3–4 per 1000 patients, similar to our annual mortality of 3 per 1000 patients. Although we observed 5.1% mortality in our young patients, based on the SMR, JMCTD did not significantly increase the risk of death. However, this analysis may have been limited by our small sample size.

This study had some limitations. First, it included a relatively small sample size; nevertheless, our cohort was one of the largest examined in JMCTD. Second, we lack validated disease activity scores and damage scores for MCTD. This presented a challenge, because we had to use scores that were validated for other diseases, which may not have been directly applicable to MCTD.

The strengths of this study were the long follow-up time, and the fact that all patients had been thoroughly examined by the same two physicians. Previous studies were mostly based on chart reviews or a cross-sectional examination of a small group of patients. Our study population was designed to be representative of all patients with JMCTD, because we identified patients from the entire country and used multiple acquisition routes.

In conclusion, in this cohort of JMCTD, we found that most patients had ongoing active disease after mean of 16 years, and more than half of the participants had irreversible organ damage. Patients with active disease more often displayed RF positivity and had higher levels of anti-RNP antibodies than patients in remission. The presence of RF at diagnosis was a significant prognostic factor for long-term, ongoing active disease.


Helga Sørhøy for practical assistance in examining the patients, Torhild Garen for help with preparation of questionnaires, Silje Reiseter for help with patients and Øyvind Molberg for administrative support.



  • Handling editor Tore K Kvien

  • Funding The study was funded by The Norwegian Rheumatism Association.

  • Competing interests None declared.

  • Ethics approval The Regional Ethics Committee for Medical Research, REK.

  • 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.