CONNECTIVE TISSUE DISEASE

To Be or Not To Be, Ten Years After: Evidence for Mixed Connective Tissue Disease as a Distinct Entity
Susanna Cappelli, MD,* Silvia Bellando Randone, MD,* Duanka Martinovic , MD, PhD, Maria-Magdalena Tamas, MD, Katarina Pasalic , MD, Yannick Allanore, MD, PhD, Marta Mosca, MD, PhD, Rosaria Talarico, MD, PhD, Daniela Opris, MD,** Csaba G. Kiss, MD, PhD, Anne-Kathrin Tausche, MD, Silvia Cardarelli, MD, Valeria Riccieri, MD, PhD, Olga Koneva, MD, Giovanna Cuomo, MD, PhD,*** Mike Oliver Becker, MD, Alberto Sulli, MD, Serena Guiducci, MD, PhD,* Mislav Radic , MD, Stefano Bombardieri, MD, Martin Aringer, MD, PhD, Franco Cozzi, MD, Guido Valesini, MD, Lidia Ananyeva, MD, Gabriele Valentini, MD,*** Gabriela Riemekasten, MD, PhD, Maurizio Cutolo, MD, PhD, Ruxandra Ionescu, MD, PhD,** Lszl Czirjk, MD, PhD, Nemanja Damjanov, MD, PhD, Simona Rednic, MD, PhD, and Marco Matucci Cerinic, MD, PhD*

Objectives: To determine if mixed connective tissue disease (MCTD) can be considered an independent clinical entity, to compare 3 different classication criteria for MCTD (Kasukawa, Alarcn-Segovia, and Sharp), and to dene predictors (clinical features and autoantibodies) of potential evolution toward other connective tissue diseases (CTDs). Methods: One hundred sixty-one MCTD patients were evaluated retrospectively at the diagnosis and in 2008. They were classied, at the diagnosis, according to the 3 classication criteria of MCTD (Sharp, Alarcn-Segovia, and Kasukawa) and reclassied in 2008 according to their evolution. Statistical analyses were performed to nd out predictors (clinical features and autoantibodies) of evolution into other CTDs. Results: After a mean of 7.9 years of disease, 57.9% of patients still satised MCTD classication criteria of Kasukawa; 17.3% evolved into systemic sclerosis, 9.1% into systemic lupus erythematosus,

*Department of Biomedicine, Division of Rheumatology, AOUC, DENOTHE Center, University of Florence, Florence, Italy. Department of Rheumatology, Internal Clinic, University Hospital of Split, Split, Croatia. Department of Rheumatology, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania. Institute of Rheumatology, Belgrade University, School of Medicine, Belgrade, Serbia. Universit Paris Descartes, Hpital Cochin, Service de Rheumatologie A, APHP, Paris, France. Department of Internal Medicine, Division of Rheumatology, University of Pisa, Pisa, Italy. **Sfanta Maria Hospital, UMF Carol Davila, RCRD Bucharest, Bucharest, Romania. Department of Rheumatology and Immunology, Clinic Center, University of Pcs, Pcs, Hungary. Division of Rheumatology, Department of Medicine III, University Center Carl Gustav Carus, Technical University of Dresden, Dresden, Germany. Rheumatology Unit, Department of Clinical and Experimental Medicine, University of Padua, Padua, Italy. Department of Internal Medicine, Rheumatology, Sapienza University of Rome, Rome, Italy. State Institute of Rheumatology of RAMS, Moscow, Russia. ***Rheumatology Unit, Second University of Neaple, Naples, Italy. Medical Faculty of the Charit, Department of Rheumatology and Clinical Immunology, Berlin, Germany. Research Laboratory and Academic Unit of Clinical Rheumatology, Department of Internal Medicine, University of Genova, Genova, Italy. The authors have no conicts of interest to disclose. Address for reprint requests to Susanna Cappelli, MD, Department of Biomedicine, Division of Rheumatology, Villa Monna Tessa, Viale Pieraccini 18, 50139, Firenze, Italy. E-mail: susicappelli82@yahoo.it.

0049-0172/12/$-see front matter 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.semarthrit.2011.07.010

589

590

Evidence for MCTD as a distinct entity

2.5% into rheumatoid arthritis, 11.5% was reclassied as affected by undifferentiated connective tissue disease, and 1.7% as suffering from overlap syndrome. Kasukawas criteria were more sensitive (75%) in comparison to those of Alarcn-Segovia (73%) and Sharp (42%). The presence of anti-DNA antibodies (P 0.012) was associated with evolution into systemic lupus erythematosus; hypomotility or dilation of esophagus (P 0.001); and sclerodactyly (P 0.034) with evolution into systemic sclerosis. Conclusions: MCTD is a distinct clinical entity but it is evident that a subgroup of patients may evolve into another CTD during disease progression. Initial clinical features and autoantibodies can be useful to predict disease evolution. 2012 Elsevier Inc. All rights reserved. Semin Arthritis Rheum 41:589 598 Keywords: mixed connective tissue disease, systemic sclerosis, systemic lupus erythematosus, rheumatoid arthritis, overlap syndrome, undifferentiated connective tissue disease, autoantibodies

n 1972, Dr Sharp and colleagues described a new connective tissue disease, characterized by overlapping features of systemic lupus erythematosus (SLE), systemic sclerosis (SSc), and polymyositis/dermatomyositis (PM/DM) and by the presence of antibodies against the U1 small nuclear ribonucleoprotein autoantigen (U1 snRNP). This condition was termed mixed connective tissue disease (MCTD) and proposed as a distinct disease (1,2). In the original publication, describing MCTD, Sharp made the 4 following claims: (1) that the syndrome was clinically identiable by a particular group of features; (2) that the presence of high titers of antibodies to U1 snRNP was a unique (and hence diagnostic) serological feature; (3) that cerebral, pulmonary, renal involvement, and vasculitis did not occur; and (4) that the condition had a benign prognosis and responded to small doses of corticosteroids (1). Later, after observing the clinical evolution of MCTD patients, Sharp himself agreed that the original concept of MCTD had to be modied and that (1) internal organs were at risk for serious complications; (2) patients were not always steroid responsive; (3) prognosis was not always benign (3-5). However, many clinical features, originally described, are still considered typical for MCTD, namely, frequent Raynauds phenomenon (RP), swollen ngers, prominent arthritis, sometimes erosive, frequent peritendinous or subcutaneous nodules, increased risk of pulmonary involvement, and only rare signicant renal or central nervous system disease (4). The initial clinical presentation usually includes RP, swollen puffy hands, and polyarthritis or polyarthralgias. As the disease evolves, manifestations of SSc, SLE, PM, or rheumatoid arthritis (RA) can be observed (3,5-7). Over the past 4 decades the distinct entity of this disease has been criticized for the following reasons: the prognosis is worse than originally described, MCTD can evolve into other connective tissue diseases (CTDs), and anti-U1 snRNP is lacking in specicity (8). Nevertheless, many authors consider MCTD a distinct disease (6,8-11). Indeed, high-titer anti-U1 snRNP antibodies are a serologic marker for this disease and an asso-

ciation of HLA-DR4 haplotype has been observed both with anti-U1 snRNP and with MCTD per se (12-15). Three sets of classication criteria for MCTD are currently used: Kasukawa, Alarcn-Segovia, and Sharp (2,16,17) (Table 1). Alarcn-Segovias criteria are simpler compared with the others, comprising 5 clinical manifestations in addition to the serological status, while Kasukawas criteria are better suited to analyze each sign and symptom of MCTD. The original Sharps criteria are seen as rather complicated since they include a high number of clinical features and the titer of antiextractable nuclear antigen (8,18). The question of whether MCTD is a distinct disease entity still remains controversial. Many authors have tried to solve this question following MCTD patients, but these attempts have still not led to a denitive answer (19-24). For this reason, we have investigated, in a retrospective study, 161 patients with the clinical and serological features of MCTD to verify the evolution into different diseases and try to answer the question whether MCTD may remain a distinct disease. METHODS Patients In this multicentric study, clinical and serological features of 161 patients with the initial diagnosis of MCTD were retrospectively analyzed. Inclusion criterion was the diagnosis of MCTD, at disease onset, according to the expert opinion. All patients were adults and recruited by tertiary medical centers. At the diagnosis of MCTD, patients also satisfying classication criteria for a denite CTD (SSc, SLE, PM/DM) or for RA were excluded. At the beginning of the study, the 15 centers involved collected retrospectively the data of the patients at disease onset and at the time of the last evaluation (in 2008) by reviewing medical records. In 2008, mean disease duration was 7.9 5.9 years (range, 1-31 years). The female: male ratio was 11:1. After having collected the data, every center was asked to ll the charts with classication criteria for MCTD of

S. Cappelli et al.
Table 1 Classication Criteria for MCTD Sharp (1987) (2) A. Major criteria 1) Myositis, severe 2) Pulmonary involvement a. Diffusion capacity 70% of normal values b. Pulmonary hypertension c. Proliferative vascular lesions on lung biopsy 3) Raynauds phenomenon or esophageal hypomotility 4) Swollen hands or sclerodactyly 5) Anti ENA 1:10,000 and anti-U1 snRNP positive and anti-Sm negative B. Minor criteria 1) Alopecia 2) Leukopenia 3) Anemia 4) Pleuritis 5) Pericarditis 6) Arthritis 7) Trigeminal neuropathy 8) Malar rash 9) Thrombocytopenia 10) Mild myositis 11) History of swollen hands At least 4 major criteria plus anti-U1 snRNP titer of at least 1:4000 (exclusion criterion: positivity for anti-Sm); or 2 major criteria from among 1, 2, and 3 plus 2 minor criteria plus anti-U1 snRNP titer of at least 1:1000 Kasukawa et al (1987) (16) A. Common symptoms 1) Raynauds phenomenon 2) Swollen ngers or hands B. Anti-U1 snRNP antibody positive C. Mixed symptoms 1. SLE-like ndings 1) Polyarthritis 2) Lymphadenopathy 3) Facial erythema 4) Pericarditis or pleuritis 5) Leukopenia (4000/mm3) or thrombocytopenia (100,000/mm3) 2. SSc-like ndings 1) Sclerodactyly 2) Pulmonary brosis, restrictive changes of lung (VC 80%) or reduced diffusion capacity (DLCO 70%) 3) Hypomotility or dilatation of esophagus 3. PM-like ndings 1) Muscle weakness 2) Elevated serum levels of muscle enzymes (CPK) 3) Myogenic pattern on EMG At least 1 of the 2 common symptoms plus positive for anti-U1 snRNP plus 1 or more of the mixed symptoms in at least 2 of the 3 disease categories

591

Alarcn-Segovia et al (1987) (17) A. Serologic 1) Anti-U1 snRNP at a hemagglutination titer of 1:1600 B. Clinical 1) Edema in the hands 2) Synovitis 3) Myositis 4) Raynauds phenomenon 5) Acrosclerosis Serologic criterion plus at least 3 clinical criteria, including either synovitis or myositis

Kasukawa, Alarcn-Segovia, and Sharp according to clinical and serological features of each patient both at disease onset and in 2008. For 2008, every center assessed also classication criteria for SLE (ARA/ACR, 1982), SSc (ARA/ACR, 1980), PM/DM (Bohan and Peter, 1975), and RA (ARA, 1987). Chart Review Charts were reviewed and patients grouped, according to their features at diagnosis, into patients who satised Kasukawa, Alarcn-Segovia, or Sharps criteria for MCTD, to dene the sensitivity of the 3 classication criteria. In 2008, patients were then reclassied according to their clinical and serological features as affected by MCTD, SSc, SLE, RA, PM/DM, overlap syndrome, or undifferentiated connective tissue disease (UCTD). We have analyzed separately the evolution of those patients who satised Kasukawa, Alarcn, or Sharps criteria at the diagnosis. To dene a patient, of 1 of the 3 categories, as still satisfying classication criteria for MCTD, he had to

fulll in 2008 the same criteria set as at the diagnosis. In 2008, patients were considered to (1) be affected by MCTD if still satisfying classication criteria for MCTD (whether or not also satisfying criteria for other CTDs); (2) suffer from SSc, SLE, RA, or PM/DM, if satisfying classication criteria for none but 1 denite connective tissue disease (CTD), without further satisfying MCTD criteria; (3) be aficted by an overlap syndrome, if satisfying classication criteria for more than 1 CTD (but not for MCTD); or (4) be affected by UCTD if having symptoms or signs of CTDs, not satisfying any criteria. We also searched for associations between clinical features and antibody prole at the rst visit and the further evolution until 2008. Antinuclear (ANA) and anticentromere (ACA) antibodies were determined by indirect immunouorescence on HEp-2 cells; anti-double-strand DNA antibodies (anti-dsDNA) by indirect immunouorescence on Crithidia luciliae; anti-U1 snRNP, antiSmith (anti-Sm), anti-SSA/ro, and antitopoisomerase I (anti-Scl70) antibodies by immunoblot and/or immuno-

592
Table 2 Clinical and Serological Features at the First Visit and in 2008 Clinical and Serologic Features Raynauds phenomenon Arthritis/arthralgia Edema of the hands Sclerodactyly Hypomotility or dilatation of esophagusb Pulmonary involvementc Pleuritis/pericarditis Facial erythema Lymphadenopathy Neurological involvement Renal involvement (nephritis) Leukopenia/thrombocytopenia Elevated CPK ANA Anti-DNAd Anti-RNPd Anti-Smd Anti-Scl70d ACAd Anti-SSA/Rod Anti-CCPd
aFor bPatients

Evidence for MCTD as a distinct entity

First Visit, No. of Patients/% 150 (93.2%) 119 (73.9%) 117 (72.7%) 47 (29.2%) 56 (34.8%) 46 (28.6%) 35 (21.7%) 32 (19.9%) 29 (18%) 9 (5.6%) 11 (6.8%) 39 (24.2%) 45 (27.9%) 156 (96.9%) 27/135 (20%) 139/139 (100%) 22/134 (16.4%) 15/116 (12.9%) 4/113 (3.5%) 31/116 (26.7%) Not done

2008,a No. of Patients/% 137 (85.1%) 80 (49.7%) 74 (46%) 69 (43%) 73 (45.3%) 71 (44.1%) 30 (18.6%) 27 (16.8%) 22 (13.7%) 18 (11.2%) 16 (9.9%) 43 (26.7%) 31 (19.2%) 151 (93.8%) 25/135 (18.5%) 126/139 (90.6%) 23/134 (17.2%) 19/116 (16.4%) 5/113 (4.4%) 31/116 (26.7%) 19/114 (16.7%)

patients who died before 2008, clinical and serological features were collected at the last visit. with hypomotility or dilatation of esophagus showed by manometry and/or esophageal barium radiograph. cPatients with evidence of interstitial lung disease (ILD) on chest radiograph (CXR) or tomography (CT) scan and/or restrictive pattern on spirometry and/or reduced diffusion capacity. dThese data were not provided for all the 161 patients included in the study.

diffusion. Enzyme-linked immunosorbent assay was used to detect anticyclic citrullinated peptide antibodies. Statistical Analysis The data for categorical variables are expressed as counts and percentages. Between-group comparisons were performed using Fishers 2-tailed exact test. Multiple variable regression analysis was applied to nd out predictors of SSc, SLE, and organ involvement. In the multiple variable model, used to nd out predictors of evolution into SLE (dependent variable), we included as independent variables SLE-like ndings: arthritis/ arthralgias, pleuritis/pericarditis, renal involvement, neurological involvement, and autoantibodies (anti-dsDNA, anti-Sm, anti-SSA/ro). In the multiple variable model, used to nd out predictors of evolution into SSc (dependent variable), we included as independent variables SSc-like ndings: sclerodactyly, pulmonary involvement, hypomotility or dilation of esophagus, and autoantibodies (anti-Scl70, anticentromere, anti-SSA/ro). To determine if a specic organ involvement (dependent variable) could be predicted by a certain autoantibody, we entered in the multiple variable model, as independent variables, the following autoantibodies: anti-dsDNA, anti-Sm, anti-SSA/ro, anti-Scl70, anticentromere. The accuracy of our results was ascertained by the repeated checks of the data included in the multiple variable models. The STATA statistical software, version 10.0 (Stata-

Corp LP, College Station, TX) was used for the analyses. P values 0.05 were considered signicant. RESULTS Clinical and Serological Features Clinical and serological features, at the rst visit and in 2008, are presented in Table 2. The most frequent signs and symptoms at disease onset were RP (93%), swollen hands (73%), and arthritis or arthralgia (74%). With disease progression, clinical features, typical for SSc, SLE, PM/DM, and RA (eg, sclerodactyly, hypomotility or dilation of esophagus, renal involvement, etc), became more frequent and could be found together in the same patient. A tendency to evolve from an inammatory phase to a sclerotic phase was observed. Indeed, the percentage of patients with swollen hands diminished (edema of the hands 73% at the rst visit versus 46% in 2008), while sclerodactyly became more frequent (29% at the rst visit versus 43% in 2008). As for antibody prole, most patients anti-U1 snRNP-positive at the onset were still positive in 2008 (100% versus 91%). Likewise, patients positive for anti-Sm, anti-dsDNA, anti-Scl70, or ACA were nearly the same at the onset and in 2008 (anti-Sm: 16% at the onset versus 17% in 2008; anti-dsDNA: 20% versus 18.5%; anti-Scl70: 13% versus 16%; ACA: 3.5% versus 4%). Five patients died: 3 of pulmonary hypertension, 1 of normotensive renal crisis, and 1 of hepatocellular car-

S. Cappelli et al.
Table 3 Patients According to the 3 Classication Criteria for MCTD at Diagnosis and in 2008 Kasukawa Alarcnb No. (%) 95 (59%) 59 (36.6%) Kasukawa Sharpb No. (%) Alarcn Sharpb No. (%) Kasukawa Alarcn Sharpc No. (%)

593

Kasukawaa No. (%) Diagnosis 2008

aPatients bPatients

Alarcna No. (%) 118 (73.3%) 70 (43.5%)

Sharpa No. (%) 67 (41.6%) 51 (31.7%)

Noned

121 (75.2%) 86 (53.4%)

65 (40.4%) 62 (38.5%) 60 (37.3%) 16 (9.9%) 39 (24.2%) 43 (26.7%) 36 (22.4%) 53 (32.9%)

who satised only 1 classication criteria (Kasukawa, Alarcn, or Sharp). who satised 2 classication criteria (Kasukawa Alarcn, Kasukawa Sharp, or Alarcn Sharp). cPatients who satised all 3 classication criteria. dPatients who did not satisfy any of the 3 classication criteria.

cinoma (in a patient with concomitant primary biliary cirrhosis). Patients at MCTD Onset According to the Three Classication Criteria According to clinical and serological features, 161 patients were divided following 3 different classication criteria (2,16,17) (Table 3). At MCTD diagnosis, only 67/161 patients satised Sharps criteria (sensitivity 42%); 118 satised Alarcons criteria (sensitivity 73%), and 121 patients satised Kasukawas criteria (sensitivity 75%). Sharps criteria were more restrictive in comparison with those of Kasukawa and Alarcn-Segovia, whereas those of Kasukawa resulted slightly more sensitive than Alarcn-Segovias criteria. In 2008, the percentage of patients who satised 1 or more of the 3 classication criteria was lower in comparison with the diagnosis (Kasukawa: 53% versus 75%; Alarcon: 44% versus 73%; Sharp: 32% versus 42%). Nevertheless, more than half of patients still satised Kasukawas criteria in 2008. The number of patients that, although they had an initial diagnosis of MCTD, did not satisfy any of the criteria was higher in 2008 than at the onset (33% versus 10%).

The Evolution of MCTD Considering Patients Who Satised Kasukawa, Alarcn, and Sharps Classication Criteria at Diagnosis Patients who satised either Kasukawa, Alarcn, or Sharps criteria at the diagnosis were analyzed separately and reclassied in 2008 (Table 4). Independently from the criteria used, approximately half of the patients still satised MCTD criteria in 2008 (to dene a patient as still satisfying classication criteria for MCTD, he had to fulll in 2008 the same set of criteria as at the diagnosis). Fifty-eight percent of patients initially classied with Kasukawa still satised these same criteria, 50% with Alarcn, 52% with Sharp. The percentage of patients who satised classication criteria for MCTD only was 17% of those classied with Kasukawa, 7.5% with Alarcn, and 13% with Sharp. The others satised both classication criteria for MCTD and for 1 or more denite CTDs (Table 4). All patients who in 2008 still fullled the same criteria set for MCTD as at the diagnosis were considered still affected by MCTD. In 2008, patients satisfying classication criteria for SSc were 32.5% of those initially classied with Kasukawa, 43% with Alarcn, and 42% with Sharp, but only 17% of those initially classied with Kasukawa and Alarcn and 13% with Sharp satised classication criteria for SSc only. The others fullled criteria for SSc and

Table 4 Evolution of Patients with the Initial Diagnosis of MCTD (Patients who satised either Kasukawa, Alarcn, or Sharps Criteria at diagnosis were analyzed separately) Patients (%) Still Affected by MCTD and Evolved into Another CTD in 2008 Criteria Fullled at the Diagnosis Kasukawa Alarcn-Segovia Sharp
aPatients

MCTD 57.9% (17.4%MCTD onlya 40.5%MCTD other CTDsb) 49.9% (7.5%MCTD onlya 42.4%MCTD other CTDsb) 52.2% (13.4%MCTD onlya 38.8%MCTD other CTDsb)

SSca 17.3% 16.9% 13.4%

SLEa 9.1% 9% 6%

RAa 2.5% 1.6% 0%

Overlap Undifferentiated Syndromec Diseased 1.6% 4.1% 10.5% 11.6% 18.5% 17.9%

who satised classication criteria for only 1 CTD (MCTD, SSc, SLE, or RA). who satised both classication criteria for MCTD and for 1 or more other CTDs (SSc, SLE, PM/DM) or for RA. cPatients who satised classication criteria for 2 or more CTDs without satisfying those for MCTD. dPatients with clinical features typical for a CTD who did not satisfy any criteria for a denite CTD.
bPatients

594
Table 5 Evolution into Another CTD in Patients with Different Disease Duration (mean 7.9 yr) (Patients who satised either Kasukawa, Alarcn, or Sharps Criteria at Diagnosis were analyzed separately) Patients (%) Evolved into Another CTD Criteria Fullled at the Diagnosis Kasukawa Alarcn Sharp Disease, yr 0 to 5 15.2% 20.4% 20.6% 5 to 10 39.2% 34.9% 43.5% 10 41.7% 50% 30%

Evidence for MCTD as a distinct entity

those with disease duration between 5 and 10 and over 10 years (Table 5). In MCTD patients, satisfying Kasukawa and Alarcns criteria, we found, using Fishers 2-tailed exact test, a statistically signicant association between disease duration of over 5 years and evolution into other CTDs (Kasukawa, P* 0.0009; Alarcn, P* 0.0276). Clinical and Serological Features and Patient Evolution Patients with MCTD diagnosis according to the 3 criteria were further classied into those evolved versus those not evolved into SLE or SSc. The clinical and serological features at the rst visit were compared between the 2 groups. Using multiple variable regression analysis, a statistically signicant association was found between antidsDNA positivity at the rst visit and evolution into SLE in patients with a rst diagnosis of MCTD according to Kasukawa (OR 1.3; 95% CI 1.1-1.6; P 0.012) and Alarcn-Segovia (OR 1.4; CI 95% 1.1-1.7; P 0.001). We could not nd a signicant association between the other features tested in the multiple variable model (arthritis/arthralgias, pleuritis/pericarditis, renal involvement, neurological involvement, anti-Sm, and anti-SSA/ ro) and evolution into SLE. Only in patients with a rst diagnosis of MCTD according to Kasukawa, the following clinical features turned out to be independently associated with evolution into SSc: sclerodactyly (OR 1.2; 95% CI 1.0-1.4; P 0.034) and esophageal hypomotility or dilation (OR 1.4; 95% CI 1.2-1.7; P 0.001). The other features tested in the multiple variable model (pulmonary involvement, anti-Scl70, anticentromere, and anti-SSA/ro) were not signicantly associated with SSc. Autoantibodies as Predictors of Organ Involvement Since organ involvement is common in MCTD, we searched for an association between a specic autoantibody positivity at the onset and organ involvement, sclerodactyly, or serositis, in 2008 (Table 6). By multiple variable regression analysis we found signicant associations between (1) anti-Sm positivity and renal involvement (OR 1.3; 95% CI 1.1-1.5; P 0.004); (2) antiSSA/Ro positivity and neurological (OR 1.2; 95% CI

for 1 or more other CTDs (including MCTD). Twenty-four percent of patients initially classied with Kasukawa, 34.5% with Alarcn, and 34% with Sharp satised classication criteria for SLE, but between 6% and 9% only did not satisfy any other CTD criteria in addition to SLE criteria. RA criteria were fullled in 16% of patients initially classied with Kasukawa, 23.5% with Alarcn, and 16.5% with Sharp, but only a minority satised RA criteria only (Table 4). None fullled PM/DM criteria without satisfying those for another CTD (including MCTD). Six percent of patients initially classied with Kasukawa, 3% with Alarcn, and 7.5% with Sharp satised classication criteria for PM/DM and for 1 or more other CTDs (including MCTD). A MCTD patient was considered evolved into SSc, SLE, RA, or PM/DM if fullled the criteria for none but 1 denite CTD (without satisfying those for MCTD). A minority of patients (Table 4) fullled classication criteria for more CTDs without satisfying that for MCTD and was classied as affected by overlap syndrome. Some patients did not satisfy any criteria for definite CTDs and were classied as UCTD (Table 4). Evolution of Patients with Different Disease Duration The evolution into another CTD was also evaluated in patients with different disease duration (between 0 and 5, between 5 and 10, and over 10 years). The percentage of patients evolving into other CTDs was lower in patients with a disease duration between 0 and 5 years than in

Table 6 Autoantibodies (First Visit) as Predictors of a Specic Organ Involvement, Sclerodactyly, or Serositis (2008) Renal Involvement OR (95%CI) Anti-DNA Anti-Sm Anti-SSA/ro Anti-Scl70 ACA 1.1 (0.9 to 1.3) 1.3 (1.1 to 1.5) 1.0 (0.9 to 1.1) 1.0 (0.8 to 1.2) 0.9 (0.6 to 1.2) P NS 0.004 NS NS NS Neurological Involvement OR (95%CI) 1.0 (0.9 to 1.2) 1.0 (0.8 to 1.2) 1.2 (1.0 to 1.4) 0.9 (0.7 to 1.1) 1.2 (0.8 to 1.7) NS NS 0.014 NS NS Pulmonary Involvement OR (95%CI) 0.9 (0.7 to 1.2) 1.1 (0.8 to 1.4) 1.4 (1.1 to 1.7) 1.3 (0.9 to 1.7) 0.9 (0.5 to 1.5) P NS NS 0.010 NS NS

OR, odds ratio; CI, condence interval; NS: not signicant; P, multivariate analysis.

S. Cappelli et al.

595

1.0-1.4; P 0.014) and pulmonary involvement (OR 1.4; 95% CI 1.1-1.7; P 0.010); (3) anti-Scl70 positivity and esophageal involvement (OR 1.4; 95% CI 1.0-1.8; P 0.048); (4) anticentromere positivity and sclerodactyly (OR 1.9; 95% CI 1.1-3.1; P 0.019); and (5) antidsDNA (OR 1.3; 95% CI 1.1-1.5; P 0.009) and antiSSA/Ro (OR 1.3; 95% CI 1.1-1.6; P 0.001) positivity and serositis. Treatment MCTD patients often (58%) required an aggressive treatment with immunosuppressants like cyclophosphamide, cyclosporine, mycophenolate mofetil, azathioprine, methotrexate, or leunomide. In a minority of patients (3%), the disease could be controlled using symptomatic therapies only. Glucocorticoids were used in most patients (82%), and antimalarial drugs in nearly half of them (45%). By dividing patients according to their evolution, we found no signicant difference between the kind of drugs employed in patients who were still affected by MCTD in 2008 and those used in patients evolved into SSc or SLE. Considering patients with a rst diagnosis of MCTD according to Kasukawas criteria, immunosuppressants were employed in 59% of patients still classied as MCTD in 2008, in 74% of patients evolved into SSc, and in 54.5% evolved into SLE. Similar data were obtained in patients with a rst diagnosis of MCTD according to Alarcn and Sharps criteria. DISCUSSION In the last decades, the existence of MCTD as a distinct clinical entity has been often debated (4,6,13-15,20-22). Our data provide evidence that MCTD has to be considered a distinct CTD. In fact, in MCTD patients that kept their characteristics in time, we found clinical features (RP, arthralgias or arthritis, swollen hands) and a distinct autoantibody (anti-U1 snRNP) persisting throughout the course of the disease. To understand MCTD evolution, previous studies have been designed with different results (Table 7). Our data show that patients with over 5 years of disease may more frequently evolve into other CTDs. However, more than half of our patients maintained their initial clinical and serological features and, therefore, remained classied as MCTD after a mean disease duration of 7.9 years. This evidence strongly supports the hypothTable 6 Continued Hypomotility or Dilatation of Esophagus OR (95%CI) 0.9 (0.7 to 1.2) 1.0 (0.7 to 1.3) 1.2 (0.9 to 1.5) 1.4 (1.0 to 1.8) 1.3 (0.7 to 2) P NS NS NS 0.048 NS

esis that MCTD is a distinct entity. There is also evidence that some clinical features of MCTD may change over time, and that the disease may evolve from an inammatory phase to a sclerotic one (20,24). In fact, in some patients edema of the hands subsided during the course of the disease and changed into more sclerotic skin changes; this sometimes led to a situation where patients did not fulll MCTD criteria any more. Still, as criteria stand now, some patients with sclerodactyly could not be classied as SSc, but remained in the MCTD spectrum. This condition might obviously change in the future if more stringent classication criteria will be provided. In this study, we dened evolution into another CTD (in 2008) if patients did not satisfy criteria for MCTD anymore but satised criteria for another denite CTD. Indeed, clinical and serological features of patients who fullled both classication criteria for MCTD and for 1 or more denite CTDs remained typical of MCTD. We dened patients who did not satisfy any sets of criteria for CTDs (in 2008) affected by UCTD. This category is presented in the article to differentiate patients who still satised criteria for MCTD (in 2008) from those who did not. Nevertheless, since they had a previous diagnosis of MCTD, they might also be considered as still affected by MCTD with disease in remission or changes in disease phenotype. Clearly, our data point out the need for unied classication criteria. In fact, the same patient, at the time of diagnosis frequently fullled 1 set of MCTD criteria, but not another. At present, sensitive classication criteria are needed to reduce the number of patients who remain in limbo without a diagnosis. In our study, Kasukawas criteria were the most sensitive and with these criteria a higher number of patients were still classied as MCTD in 2008. This evidence may suggest that Kasukawas criteria can be presently the most useful in clinical practice. According to some authors, anti-Sm and anti-dsDNA are usually not found or can be detected in only low titers in patients with MCTD (1,6). Some of our patients showed features typical for more CTDs (eg, they had sclerodactyly and malar rash at the same time) and were positive for both anti-U1 snRNP and anti-Sm or/and anti-dsDNA. In our opinion, such patients should be still classied as affected by MCTD and anti-Sm and anti-dsDNA positivity should not be con-

Sclerodactyly OR (95%CI) 0.9 (0.7 to 1.1) 0.9 (0.7 to 1.2) 1.0 (0.8 to 1.3) 1.0 (0.7 to 1.3) 1.9 (1.1 to 3.1) P NS NS NS NS 0.019

Pleuritis/Pericarditis OR (95%CI) 1.3 (1.1 to 1.5) 1.2 (0.9 to 1.4) 1.3 (1.1 to 1.6) 1.1 (0.9 to 1.3) 1.2 (0.9 to 1.7) P 0.009 NS 0.001 NS NS

596
Table 7 Evolution of MCTD in Previous Works Author (yr) Nimelstein et al (1980) (24) Van den Hoogen et al (1994) (21) Gendi et al (1995) (19) Evolution 59% SSc, 5.9% SLE, 5.9% RA 46% MCTD, 21% SSc, 15% SLE, 9% RA, 9% Overlap syndrome 36% MCTD, 34% SSc, 25% SLE, 5% RA Disease Duration 8 yr (follow-up) 5 yr (follow-up)

Evidence for MCTD as a distinct entity

Number of Patients 25 46

Conclusions MCTD spontaneously evolves into SSc The majority of patients have or will develop a classied CTD within 5 years MCTD is, for most patients, an intermediate stage in a progression to an other CTD It does not appear that MCTD frequently evolves into SSc or SLE More than half of patients, maintained their initial clinical and serological features and remained classied as MCTD; this supports the existence of MCTD as a distinct entity

20 yr (mean disease duration)

46

Burdt et al (1999) (20) Cappelli et al (present 2011)

21% SSc, 13% SLE

15 yr (mean duration of follow-up) 7.8 yr (mean disease duration)

47

57.9% MCTD, 17.3% SSc, 9.1% SLE, 2.5% RA, 1.6% Overlap Syndrome, 11.6% UCTD (Kasukawa) 49.9% MCTD, 16.9% SSc, 9%SLE, 1.6% RA, 4.1% Overlap syndrome, 18.5% UCTD (Alarcn) MCTD 52.2%, 13.4% SSc, 6% SLE, 0% RA, 10.5% Overlap syndrome, 17.9% UCTD (Sharp)

161

sidered an exclusion criterion for this disease but as a warning of a possible evolution into SLE. Antibodies to anti-U1 snRNP are not conned to MCTD but can be found also in other CTDs. According to different studies, between 20% and 40% of patients with SLE (25-27), 2% and 14% with SSc (28-30), and 6% and 9% with myositis (31,32) are positive for anti-U1 snRNP. Patients with RA, instead, are generally negative (33,34). In MCTD patients, organ involvement, leading to lethal complications, and evolution into other CTDs should be prevented and predictors should be identied. In the present study we have looked for clinical and serological features that might predict evolution into another CTD. Recently it has been hypothesized that capillaroscopy may be useful in predicting evolution. In particular, in MCTD patients evolving into SSc, the presence of angiogenesis (branched capillaries) is often the dominant hallmark (5,35). Of particular interest is the role of autoantibodies as predictors of organ involvement. In the literature, associations between certain autoantibodies and clinical features of MCTD patients have already been reported (9,20,36,37). The association between anticentromere antibodies and sclerodactyly as well as between anti-Scl70 antibodies and esophageal involvement in SSc (38-42), or between anti-dsDNA and serositis in SLE, has already been described (43,44). The association between antiSSA/Ro and pulmonary involvement was unexpected. In a previous study anti-SSA/Ro antibodies correlated with

an early onset and a rapid development of lung involvement in patients with SSc (45). Associations between anti-SSA/Ro and serositis in SLE and between anti-Sm and renal involvement have already been reported (37,44). Our data detected a correlation between antiSSA/Ro and neurological involvement: previously an association between vasculitis and anti-SSA/Ro has been described (46) and vasculitis is a well-known cause of neurological symptoms. In many of our patients, glucocorticoids could not control the disease and immunosuppressants were needed. The evidence that the kind of drugs employed in patients who remained affected by MCTD did not differ signicantly from those used in patients evolved into SSc or SLE may suggest that MCTD prognosis is not much different from that of SSc and SLE. Due to the type of study (retrospective), some data, in particular the dosage of autoantibodies at the rst visit, were not available for all patients. Missing data may be responsible for an undervaluation of patients that at the diagnosis fullled Kasukawa, Alarcon, and Sharps classication criteria. Moreover disease evolution may have been inuenced by therapy. In 2008 some patients did not satisfy any criteria for denite CTDs and this may be the consequence of the treatment they received. As regards therapy and its inuence on evolution, it must be admitted that in a retrospective analysis the dropout from a treatment may not be well calculated. However these limitations do not diminish the validity of

S. Cappelli et al.

597
16. Kasukawa R, Tojo T, Miyawaki S. Preliminary diagnostic criteria for classication of mixed connective tissue disease. Mixed connective tissue disease and antinuclear antibodies. Kasukawa R, Sharp GC, editors. Amsterdam: Elsevier; 1987. p. 41-7. 17. Alarcn-Segovia D, Villareal M. Classication and diagnostic criteria for mixed connective tissue disease. Mixed connective tissue disease and antinuclear antibodies. Kasukawa R, Sharp GC, editors. Amsterdam: Elsevier; 1987. p. 33-40. 18. Maddison PJ. Mixed connective tissue disease, overlap syndromes, and eosinophilic fasciitis. Ann Rheum Dis 1991;50: 887-93. 19. Gendi NST, Welsh KI, van Venrooij WJ, Vancheeswaran R, Gilroy J, Black CM. HLA type as a predictor of mixed connective tissue disease differentiation: ten-year clinical and immunogenetic follow-up of 46 patients. Arthritis Rheum 1995;38:256-66. 20. Burdt MA, Hoffman RW, Deutscher SL, Wang GS, Johnson JC, Sharp GC. Long-term outcome in mixed connective tissue disease. Arthritis Rheum 1999;42:899-909. 21. Van den Hoogen FH, Spronk PE, Boerbooms AM, Bootsma H, de Rooji DJ, Kallenberg CG, et al. Long-term follow up of 46 patients with anti-(U1)snRNP antibodies. Br J Rheumatol 1994; 33:1117-20. 22. Venables PJ. Mixed connective tissue disease. Lupus 2006;15: 132-7. 23. Lzaro MA, Maldonado Cocco JA, Catoggio LJ, Babini SM, Messina OD, Garcia Morteo O. Clinical and serologic characteristics of patients with overlap syndromes: is mixed connective tissue disease a distinct clinical entity? Medicine (Baltimore) 1989;68: 58-65. 24. Nimelstein SH, Brody ST, McShane D, Holman HR. Mixed connective tissue disease: a subsequent evaluation of the original 25 patients. Medicine (Baltimore) 1980;59:239-48. 25. Faria AC, Barcellos KS, Andrade LE. Longitudinal uctuation of antibodies to extractable nuclear antigens in systemic lupus erythematosus. J Rheumatol 2005;32:1267-72. 26. Migliorini P, Baldini C, Rocchi V, Bombardieri S. Anti-Sm and anti-RNP antibodies. Autoimmunity 2005;38:47-54. 27. Wang CL, Ooi L, Wang F. Prevalence and clinical signicance of antibodies to ribonucleoproteins in systemic lupus erythematosus in Malaysia. Br J Rheumatol 1996;35:129-32. 28. Ihn H, Yamane K, Yazawa N, Kubo M, Fujimoto M, Sato S, et al. Distribution and antigen specicity of anti-U1RNP antibodies in patients with systemic sclerosis. Clin Exp Immunol 1999;117: 383-7. 29. Avouac J, Air P, Dieude P, Caramaschi P, Tiev K, Diot E, et al. Associated autoimmune diseases in systemic sclerosis dene a subset of patients with milder disease: results from 2 large cohorts of European Caucasian patients. J Rheumatol 2010;37:608-14. 30. Jacobsen S, Halberg P, Ullman S, van Venrooji WJ, HierMadsen M, Wiik A, et al. Clinical features and serum antinuclear antibodies in 230 Danish patients with systemic sclerosis. Br J Rheumatol 1998;37:39-45. 31. Brouwer R, Hengstam GJ, Vree Egberts W, Ehrfeld H, Bozic B, Ghirardello A, et al. Autoantibody proles in the sera of European patients with myositis. Ann Rheum Dis 2001;60:116-23. 32. Ghirardello A, Zampieri S, Iaccarino L, Tarricone E, Tonello M, Bendo R, et al. Myositis specic and myositis associated autoantibodies in idiopathic inammatory myopathies: a serologic study of 46 patients. Reumatismo 2005;57:22-8. 33. Hassfeld W, Steiner G, Studnicka-Benke A, Skriner K, Graninger W, Fisher I, et al. Autoimmune response to the spliceosome. An immunologic link between rheumatoid arthritis, mixed connective tissue disease, and systemic lupus erythematosus. Arthritis Rheum 1995;38:777-85. 34. Paulus HE, Wiesner J, Bulpitt KJ, Patnaik M, Law J, Park GS, et al. Autoantibodies in early seropositive rheumatoid arthritis, before and during disease modifying antirheumatic drug treatment. J Rheumatol 2002;29:2513-20.

this study, which includes a very large number of MCTD patients whose clinical and serological features have been analyzed during the disease course. In conclusion, our data provide evidence that MCTD may have the dignity of a distinct disease but also suggest the need to identify predictors of organ involvement and evolution into other CTDs, to start a suitable aggressive treatment in due time. Very likely, the autoantibody prole might identify those patients at risk for severe complications. ACKNOWLEDGMENTS We thank Dr Irene Miniati for her help in performing statistical analyses and Dr Patrizia Cerboni for secretarial assistance. REFERENCES
1. Sharp GC, Irvin WS, Tan EM, Gould RG, Holman HR. Mixed connective tissue disease: an apparently distinct rheumatic disease syndrome associated with a specic antibody to an extractable nuclear antigen (ENA). Am J Med 1972;52:148-59. 2. Sharp GC. Diagnostic criteria for classication of MCTD. Mixed connective tissue disease and antinuclear antibodies. Kasukawa R, Sharp GC, editors. Amsterdam: Elsevier; 1987. p. 23-32. 3. Sharp GC, Anderson PC. Current concepts in the classication of connective tissue disease. J Am Acad Deramtol 1980;2:269-74. 4. Maddison PJ. Mixed connective tissue disease: overlap syndromes. Baillires Best Pract Res Clin Rheumatol 2000;14: 111-24. 5. Bellando Randone S, Cutolo M, Czirjk L, Matucci Cerinic M. Mixed connective tissue disease, a roundabout to rheumatic diseases? Curr Rheumatol Rev 2009;5:133-40. 6. Smolen JS, Steiner G. Mixed connective tissue disease: to be or not to be? Arthritis Rheum 1998;41:768-77. 7. Farhey Y, Hess EV. Mixed connective tissue disease. Arthritis Care Res 1997;10:333-8. 8. Kasukawa R. Mixed connective tissue disease. Intern Med 1999; 38:386-93. 9. Setty YN, Pittman CB, Mahale AS, Greidinger EL, Hoffman RW. Sicca symptoms and anti-SSA/Ro antibodies are common in mixed connective tissue disease. J Rheumatol 2002;29:487-9. 10. Sharp GC, Hoffmann RW. Overlap syndromes: mixed connective tissue disease and Sjgrens syndrome. Systemic lupus erythematosus. Lahita RG, editor. New York: Academic Press, 1999. p. 551-73. 11. Sharp GC, Hoffmann RW. Mixed connective tissue disease. Connective tissue disease. Belch JJF, Zurier RB, editors. London: Chapman & Hall; 1995. p. 151-78. 12. Aringer M, Smolen JS. Mixed connective tissue disease: what is behind the curtain? Best Pract Res Clin Rheumatol 2007;21: 1037-49. 13. Hoffman RW, Rettenmaier LJ, Takeda Y, Hewett JE, Pettersson I, Nyman U, et al. Human autoantibodies against the 70-kd polypeptide of U1 small nuclear RNP are associated with HLA-DR4 among connective tissue disease patients. Arthritis Rheum 1990; 33:666-73. 14. Hameenkorpi R, Ruuska P, Forsberg S, Tiilikainen R, Makitalo R, Hakala M. More evidence of distinctive features of mixed connective tissue disease. Scand J Rheumatol 1993;22:63-8. 15. Dong RP, Kimura A, Hashimoto H, Akizuki M, Nishimura Y, Sasazuki T. Difference in HLA-linked genetic background between mixed connective tissue disease and systemic lupus erythematosus. Tissue Antigens 1993;41:20-5.

598
35. Cutolo M, Grassi W, Matucci Cerinic M. Raynauds phenomenon and the role of capillaroscopy. Arthritis Rheum 2003;48: 3023-30. 36. Okawa-Takatsuji M, Aotsuka S, Uwatoko S, Takaono M, Iwasaki K, Kinoshita M, et al. Endothelial cell-binding activity of antiU1-ribonucleoprotein antibodies in patients with connective tissue diseases. Clin Exp Immunol 2001;126:345-54. 37. Reichlin M, van Venrooij WJ. Autoantibodies to the URNP particles: relationship to clinical diagnosis and nephritis. Clin Exp Immunol 1991;83:286-90. 38. Miyawaki S, Asanuma H, Nishiyama S, Yoshinaga Y. Clinical and serological heterogeneity in patients with anticentromere antibodies. J Rheumatol 2005;32:1488-94. 39. Johanet C, Agostini MM, Vayssairat M, Abuaf N. Anti-Scl70 and anti-centromere autoantibodies. Biological markers of 2 forms of systemic scleroderma. Presse Med 1989;18:207-11. 40. Cozzi F, Zucchetta P, Durigon N, Marzola MC, Bullo A, Favaro M, et al. Changes in esophageal peristalsis in diverse clinical forms

Evidence for MCTD as a distinct entity

and antibody specicity in scleroderma: a scintigraphic study in 100 cases. Reumatismo 2003;55:86-92. Kinuya K, Nakajima K, Kinuya S, Michigishi T, Tonami N, Takehara K. Esophageal hypomotility in systemic sclerosis: close relationship with pulmonary involvement. Ann Nucl Med 2001;15:97-101. Ho KT, Reveille JD. The clinical relevance of autoantibodies in scleroderma. Arthritis Res Ther 2003;5:80-93. Vil LM, Molina MJ, Mayor AM, Peredo RA, Santaella ML, Vil S. Clinical and prognostic value of autoantibodies in puerto Ricans with systemic lupus erythematosus. Lupus 2006;15:892-8. Gmez J, Surez A, Lpez P, Mozo L, Daz JB, Gutirrez C. Systemic lupus erythematosus in Asturias. Spain: clinical and serologic features Medicine (Baltimore) 2006;85:157-68. Rebora A, Parodi A. Mixed connective tissue disease and correlated diseases. G Ital Dermatol Venereol 1990;125:357-62. Alexander EL, Provost TT, Stevens MB, Alexander GE. Neurologic complications of primary Sjgrens syndrome. Medicine (Baltimore) 1982;61:247-57.

41. 42. 43. 44. 45. 46.