The combined prevalence of classified rare rheumatic diseases is almost double that of ankylosing spondylitis

Background Rare diseases (RDs) affect less than 5/10,000 people in Europe and fewer than 200,000 individuals in the United States. In rheumatology, RDs are heterogeneous and lack systemic classification. Clinical courses involve a variety of diverse symptoms, and patients may be misdiagnosed and not receive appropriate treatment. The objective of this study was to identify and classify some of the most important RDs in rheumatology. We also attempted to determine their combined prevalence to more precisely define this area of rheumatology and increase awareness of RDs in healthcare systems. We conducted a comprehensive literature search and analyzed each disease for the specified criteria, such as clinical symptoms, treatment regimens, prognoses, and point prevalences. If no epidemiological data were available, we estimated the prevalence as 1/1,000,000. The total point prevalence for all RDs in rheumatology was estimated as the sum of the individually determined prevalences. Results A total of 76 syndromes and diseases were identified, including vasculitis/vasculopathy (n = 15), arthritis/arthropathy (n = 11), autoinflammatory syndromes (n = 11), myositis (n = 9), bone disorders (n = 11), connective tissue diseases (n = 8), overgrowth syndromes (n = 3), and others (n = 8). Out of the 76 diseases, 61 (80%) are classified as chronic, with a remitting-relapsing course in 27 cases (35%) upon adequate treatment. Another 34 (45%) diseases were predominantly progressive and difficult to control. Corticosteroids are a therapeutic option in 49 (64%) syndromes. Mortality is variable and could not be determined precisely. Epidemiological studies and prevalence data were available for 33 syndromes and diseases. For an additional eight diseases, only incidence data were accessible. The summed prevalence of all RDs was 28.8/10,000. Conclusions RDs in rheumatology are frequently chronic, progressive, and present variable symptoms. Treatment options are often restricted to corticosteroids, presumably because of the scarcity of randomized controlled trials. The estimated combined prevalence is significant and almost double that of ankylosing spondylitis (18/10,000). Thus, healthcare systems should assign RDs similar importance as any other common disease in rheumatology.

2002 defined a disease as rare when it affects less than 200,000 people. Patients frequently remain undiagnosed for many years, and treatment is often inefficacious. Data on prevalence, disease burden, treatment regimens, access to healthcare systems, and mortality are, to a great extent, unknown. Thus, RDs remain an unresolved challenge in modern medicine.
The classification, overall prevalence, and treatment options of rare rheumatic diseases are poorly defined. Disorders may affect the musculoskeletal apparatus with arthritis and myalgia but also involve other tissues in the form of myositis, vasculitis, autoimmune organ involvement, or bone diseases. Patients are often diagnosed as having a psychosomatic disorder due to missing or unrecognized somatic and/or objective findings. Patients not only have to cope with their disease burden, but are also at risk of developing additional psychiatric comorbidities. For example, patients with undiagnosed diseases in Germany have a prevalence of depressive symptoms three times higher than the average population (22% vs. 8.1%) [1,2]. On the other hand, a verified diagnosis may aid the patient in accepting their diagnosis and coping with the ensuing symptoms and challenges. However, due to the scarcity of randomized controlled trials, treatment options for RDs are often limited and remain empirical. In addition, the definitions of key symptoms often vary in clinical studies, hampering uniform analyses.
RDs in rheumatology may be analyzed systematically by prevalence, genetic background and pathogenesis, clinical involvement, treatment options, and prognosis. Prevalence data vary according to age, as well as global and ethnic background. For example, seronegative symmetrical synovitis with pitting edema (RS3PE) has a high prevalence among the elderly, with 0.09% of all individuals over the age of 50 years being affected in Japan [3], but seems to be a quite rare disorder among younger individuals. Similarly, the prevalence of adult onset Still's disease (AOSD) varies globally: 3.4-6.9/100,000 in Norway [4], 6.77/100,000 in Turkey [5], and 3.9/100,000 in Japan [6]. With respect to ethnic background, the estimated worldwide prevalence of Gaucher's disease is 1-2/100,000, but in Ashkenazi-Jews the prevalence may be as high as 1/850 [7].
The pathogenesis and genetic backgrounds of some RDs in rheumatology have been studied increasingly in recent years, and in some cases well elucidated. Blau syndrome was described in 2001 and is characterized by mutations in the CARD15/NOD2 gene [8] and overexpression of autoinflammatory cytokines [9]. Interestingly, mutations in CARD15/NOD2 are also associated with other diseases with inflammatory involvement, such as Crohn´s disease and arthritis [10]. Familial cold urticaria (FCU), Muckle-Wells syndrome (MWS), and neonatalonset multisystem inflammatory disease (NOMID or chronic infantile neurologic cutaneous articular syndrome [CINCA]) were originally thought to be three similar but distinct diseases. Further evidence has shown that all three syndromes are the result of mutations in the same gene, CIAS1. They are now referred to as different phenotypes of the same disorder, namely cryoporin-associated periodic syndrome (CAPS) [11,12]. The MEFV gene, best known for causing familial Mediterranean fever (FMF) equally demonstrates the importance of genetics in RDs [13,14]. Heterozygous mutations in MEFV are also found in many children with periodic fever, stomatitis, pharyngitis, adenitis (PFAPA) [15,16]. Findings suggest that exon variants in MEFV may also be associated with AOSD [17].
With respect to pathogenesis, infectious agents may also play a role in RDs. As some diseases follow a distinct seasonal pattern, infectious pathogenesis has been suggested for Kawasaki disease [18,19], and IgA-vasculitis (formerly Henoch-Schönlein purpura) [20,21].
Treatment options often, but not exclusively, include corticosteroids, such as in RS3PE [32] and eosinophilicmyalgia syndrome [33]. Although research on RDs is often limited to retrospective, single center trials or case reports only, randomized controlled trials (RCTs) have been increasingly available in recent years for some conditions, such as granulomatosis with polyangiitis [34], PFAPA syndrome [35], and FMF [36].
The disease course may be classified as self-limited (e.g., Kawasaki disease), chronic with a variable remitting-relapsing course during treatment (e.g., Takayasu arteritis), and chronic with a predominantly progressive and difficult to control course (e.g., systemic sclerosis).
The prognosis for RD varies and may be affected by the primary disease, complications, and treatment, especially long-term immunosuppression.
The objective of this study was to analyze the complexity of RDs in rheumatology. Based on data from the literature, we identified some of the most important sets of rheumatic diseases and calculated their combined prevalence. Our data may give better insight into this area of rheumatology, aid specialized centers for RDs, and raise overall awareness in the healthcare field.

Results
The diseases extracted from the databases are summarized in Table 2 of Appendix. The 76 syndromes and diseases were classified as follows: vasculitis/vasculopathy (n = 15), arthritis/arthropathy (n = 11), autoinflammatory syndromes (n = 11), myositis (n = 9), bone disorders (n = 11), connective tissue diseases (n = 8), which include inflammatory as well as non-inflammatory conditions, overgrowth syndromes (n = 3), and others (n = 8). A definitive genetic cause was identified in 26 diseases (34%). Out of the 76 conditions, 34 diseases (44%) were classified as chronic, primarily progressive, and difficult to control. Twenty-seven diseases (35%) were classified as chronic with a variable remitting-relapsing course. Six diseases (7%) were classified as self-limited. Acute phase reactants may be elevated in 49 diseases (64%). Corticosteroids are used as a therapeutic option in 49 diseases (64%). The mortality was variable and could not be determined precisely, but nine diseases (11%) were considered severe and potentially lethal if left untreated.
Prevalence data were available for 28 syndromes and diseases. For an additional five diseases, estimated prevalence data were already available. For another eight diseases, only incidence data were available. The prevalence of 38 diseases was estimated at 1/1,000,000, for 4 diseases at 1/100,000, and for 1 disease at 1/10,000. The combined prevalence per 10,000 is given in Table 1 (see also Figs. 1 and 2). The summed prevalence of all available and estimated RDs was 28.8/10,000.

Discussion
RDs are challenging for patients, healthcare professionals, and societies. Signs and symptoms often remain unrecognized and patients are excluded from specific treatment. Patients are frequently diagnosed with a psychosomatic disorder. Conversely, some patients with true extrasomatic diagnoses insist on the presence of a RD and cause significant expenses to healthcare systems. To overcome this bias, the importance of RDs should be recognized in public healthcare.
The knowledge of RDs is mostly available from case reports or case studies. On one hand, these studies are important in order to document essential information, such as commonly reported symptoms, different treatment regimens, and outcomes. However, such studies may involve reporting bias and, thus, are difficult to compare. For example, multicentric histiocytosis is a disease primarily reported in Caucasian women. However, this may be due to increased awareness in Western countries. Furthermore, women may consult a doctor more often and simulate a female Caucasian predominance [37].
Most studies scrutinizing RDs are designed as single center, retrospective studies due to a lack of patient numbers or networks. Larger registries would provide an opportunity to conduct retrospective or prospective and multicenter studies with a greater number of participating patients. Therefore, further development of international expert centers and registries is in great demand. In recent years, advances have been made due to the establishment of international expert/reference centers. For example, the Eurofever project for autoinflammatory diseases provided classification criteria and evaluated treatment options for multiple disorders [38,39].
In addition, genetic testing has become increasingly available and, thus, more important in recent years. For example, whole exome sequencing in patients with similar symptoms without previous knowledge of candidate genes led to the identification of WISP 3 and SLCO2A1 as the pathogenic mutations in progressive pseudorheumatoid dysplasia [40] and primary hypertrophic osteoarthropathy [23].
In this study, we aimed to identify and classify RDs in rheumatology. We were able to show that the most common symptoms in rare rheumatic diseases are arthritis (31.0%, 89.3/100,000), followed by vasculitis (26.6%, 76.8/100,000), and connective tissue involvement (16.0%, 46.1/100,000), which in this study includes inflammatory as well as non-inflammatory conditions. Importantly, the total prevalence of a symptom was commonly dominated by only a few comparatively prevalent diseases. For example, systemic sclerosis (prevalence: 22.5/100,000) makes up 48.8% of all rare rheumatic diseases with connective tissue involvement.
Our study has several limitations. First, the nomenclature for diseases and syndromes is often used interchangeably, and the same disease or a variation in the same group of diseases is sometimes named differently. For example, CAPS is formerly known as Muckle-Wells, CINCA/NOMID, or familial cold autoinflammatory syndrome [12]. This may lead to incomplete search results and impede the comparability of available studies. Second, most prevalence data are almost exclusively based on retrospective analyses of hospital information or questionnaires. Prevalence data also vary according to ethnic background, geography, and age, which makes the use of overall prevalence data quite uncertain. For example, FMF or Behçet´s are more common in Mediterranean and Middle Eastern populations [41,42] and rare in other regions, demonstrating the difficulty in using a local geographic prevalence. Similarly, although currently considered a RD by the European definition, an Italian study found an unexpectedly high overall prevalence of 8.5/10,000 for cryoglobulinemic vasculitis, which would no longer be defined as a RD. Although this study had some limitations, including the implementation of a questionnaire leading to a higher participation rate in the elderly population than the younger age groups, we decided to exclude cryoglobulinemic vasculitis from our list of rare rheumatological diseases. Our reasoning for this was the methodically more accurate estimation of prevalence by this study [43]. A well-known, verified RD, systemic sclerosis is quite common in Choctaw Native Americans (prevalence 66/100,000 in Oklahoma Choctaws and 469/100,000 in full-blooded Choctaws [44]), but rare among all other studied populations [45]. This may be due to a unique HLA haplotype, but other environmental factors may also play a role. Furthermore, our estimates of prevalence data may be somewhat inaccurate. We probably underestimated the prevalence by choosing 1 in 1,000,000 rather than 1 in 100,000, and the total prevalence of rare rheumatic diseases is likely to be even higher than our estimate of 29.6/10,000.
Prevalence data may also differ depending on age. For example, giant cell arteritis is probably extremely rare in younger patients but quite frequent in patients older than 55 years of age (UK 250/100,000 [46]), with ageindependent prevalence data being difficult to obtain. Because the overall estimated prevalence may be even higher than 1/10,000, we also excluded giant cell arteritis from our list of RDs in rheumatology.
Similarly, we also excluded systemic lupus erythematodes, one of the better known "rare" diseases in 76 rheumatology from our study, as recent epidemiological studies suggest that it is probably not a rare disease by the above mentioned European definition [47]. Furthermore, we also excluded rhupus syndrome, because it is suspected to be an overlap of systemic lupus erythematodes and rheumatoid arthritis and thus its prevalence may be difficult to obtain and distinguish [48].
The classification system we used also has its limitations. Conditions can be classified by their etiology or by their clinical appearance. As the etiology of many rare diseases remains unknown, we decided to classify diseases according to their main organ system involved in the clinical appearance of the disease. Exceptions include the category of autoinflammatory and overgrowth syndromes, where multiple organ systems may be involved, and the conditions in one group have a basic (suspected) etiology in common. However, overlaps between included conditions may have occurred because of the complex nature of some RDs.
Another pitfall is that diagnostic and/or classification criteria may differ in varying definitions of the diseases (e.g., familial Mediterranean fever), are not well established, and have been suggested based on radiographic or histological findings. In most cases, no validation studies are available to confirm specificity and sensitivity.
We observed that RCTs are available only for some RDs, such as Behcet´s disease [49] or ANCA-associated vasculitis [34]. Furthermore, most follow-ups of patients with RDs are rarely published. Larger patient cohorts would be necessary to obtain reliable data on outcome, disease progression, morbidity, and mortality.  2 Flow chart of syndrome inclusion in the analysis. After conducting the literature search and analysis with regard to prevalence, four diseases were excluded due to high prevalence, and seven diseases were merged into three, resulting in a total of 76 syndromes included in further analysis disease itself or adverse events to immunosuppressive treatment. These data merit in-depth analyses, as they may shed more light on the disease and its pathophysiology or potential treatment options. Incidence and prevalence data have become available for increasingly more RDs and, in some cases, the prevalence has even increased in recent years (e.g., Kawasaki disease) [50]. This trend may be due to a true increase in incidence, increased diagnosis of previously undiagnosed patients due to an upsurge in physician or patient awareness (internet, patient support groups, RD associations, etc.), or simply due to better national reporting systems and databases. In addition, an increased overall life expectancy in the general population has led to an increased incidence and prevalence of conditions that predominantly affect the elderly (e.g., GCA). Continuously better treatment options may also lead to prolonged overall survival with an increase in prevalence rates.
In this study, we found that the cumulative prevalence of RDs in rheumatology is at least 28.8/10,000, which is almost double that of ankylosing spondylitis (18/10,000) [51], a rather common disease seen in practice. This observation suggests that symptoms should be carefully acknowledged in all patients, especially when an overt psychopathology is present, as many patients with a yet undiagnosed disorder show signs of depression or other psychosomatic disorders [1], which can obscure the differentiation between primary disease and secondary complication even more for the treating physician. Our study may aid physicians as a simple tool for diagnosing patients with an undiagnosed rheumatic disease by comparing the symptoms, prevalence, and likeliness of one disorder to another.

Conclusion
Our study shows that RDs in rheumatology are not as rare as previously thought, affecting at least 28.8/10,000 people. Therefore, for every patient diagnosed with ankylosing spondylitis, 1.6 patients may suffer from a rare rheumatic disease. Although research and case reports of RDs are important, international expert centers are necessary to initiate and perpetuate patient cohorts and registries, establish classification/diagnostic criteria, and conduct clinical trials. Standard questionnaires and laboratory analyses may aid in obtaining better insight into the pathophysiology of RDs.

Methods and materials
The abstract archives of the European League Against Rheumatism, the American College of Rheumatology, Orpha.net, and the PubMed database were searched for the following terms: "rare" in combination with arthritis, arteritis, connective tissue disease, rheumatic, and vasculitis. Furthermore, terms were used in various combinations including arthralgia, autoimmune, fever, inflammation, joint pain, muscle pain, myalgia, and swollen joint. Identified syndromes were then classified according to their main appearance under the following terms: arthritis/arthropathy, bone disorders, autoinflammatory syndromes, connective tissue diseases, myositis/ myopathy, overgrowth syndromes, vasculitis/vasculopathy, and others. Furthermore, we conducted a search of the literature and analyzed each disease according to the following criteria: prevalence, genetics/pathogenesis, diagnostic criteria, symptoms, laboratory findings, therapy, and prognosis. Diseases and syndromes were assessed for their overall prevalence and excluded if they did not meet the European definition of a RD (< 5/10,000) (see Fig. 3).
If more than one prevalence was available, the prevalence data were averaged accordingly.
For statistical reasons, we estimated the prevalence for diseases for which no epidemiological data were available as one of three possible values: < 1/1,000,000, 1/100,000 or 1/10,000. The overall total prevalence for all RDs in rheumatology was estimated by summing up the available or estimated individual prevalence of each disease.

Appendix
See