INTRODUCTION
The vasculitides are relatively uncommon conditions whose etiology is still poorly understood, and this has made detailed epidemiology studies difficult. They can affect any age. Two diseases predominate in children—Kawasaki's disease (KD) and Henoch-Schönlein Purpura (HSP), while the other more common systemic vasculitides occur with increasing age. Changing definitions and understanding of these diseases have led to epidemiology studies being relatively recent, but the development of prospective population-based registers has been an important emerging database for comparative studies in recent years. Hopefully, this trend will be extended across the less-developed world as interest in these diseases extends.
Epidemiology studies require a sound basis for classification and that has also evolved over recent years. Of particular importance have been the American College of Rheumatology (ACR) classification criteria 1990 and the two Chapel Hill Consensus Conference (CHCC) definitions for vasculitis proposing criteria for classifying vasculitides with a variety of sensitivities and specificities. The most sensitive and specific of the seven ACR criteria were found in Churg–Strauss syndrome (now called eosinophilic granulomatosis with polyangiitis [EGPA]), giant cell arteritis (GCA), and Takayasu's arteritis (TAK). That study also identified a condition called hypersensitivity vasculitis which has disappeared (no longer recognized) from current definitions, as it has been replaced mainly by HSP and microscopic polyangiitis (MPA). The next development was the first CHCC in 1994 which included MPA but did not consider the role of antineutrophil cytoplasmic antibodies (ANCAs). Because of this and further developments in understanding pathogenesis, a further consensus conference was called in 2011 and this led to the current classification shown in Table 1.1. This recognizes the importance of ANCA in separating small-vessel vasculitis into ANCA-associated vasculitis (AAV) or immune complex-associated vasculitis, and also recognizes potential probable etiologies, as shown in the Table. It also reflects the fact that some vasculitides are restricted to a single organ and others associated with systemic disease. Furthermore, it recognizes the fact that some vasculitides can affect a wide variety of blood vessels, particularly Behçet's syndrome and Cogan's syndrome. The consensus also introduced different nomenclature, particularly for HSP—now known as immunoglobulin A (IgA) vasculitis, Churg-Strauss syndrome (now known as eosinophilic granulomatosis with polyangiitis)—and confirmed the move away from Wegener's granulomatosis to the term granulomatosis with polyangiitis. Currently, the eponyms are still included after the descriptive terms to avoid confusion, but it is expected that these will disappear with time.
It is important to recognize that the definitions of the different vasculitic diseases did not change dramatically between 1994 and 2011, and so the original criteria are still used in epidemiology studies. The problem has been that the original CHCC (1994) and ACR criteria did not have any classification criteria for MPA and polyarteritis nodosa (PAN) which has made epidemiological studies in those diseases more difficult. What follows is a brief description of current epidemiological data for the commoner types of vasculitis based on the classification system shown in Table 1.1, i.e. large-, medium-, and small-vessel disease.
LARGE-VESSEL VASCULITIS
Giant Cell Arteritis
Giant cell arteritis (GCA) is defined in CHCC 2012 as “arteritis, often granulomatous, usually affecting the aorta or 4its major branches with a predilection for the branches of the carotid and vertebral arteritis.
Often involves the temporal artery. Onset usually in patients older than 50 and often associated with polymyalgia rheumatica (PMR).”1 Difficulties in reviewing epidemiology studies relate particularly to the presence or absence of biopsy. Most older studies include only biopsy-positive cases. Biopsy rates and the intensity of the histological search for evidence of arteritis will therefore have a profound effect on any reported incidence. Giant cell arteritis is also closely related to polymyalgia and there are different epidemiology studies for polymyalgia and for GCA. It is reckoned that about 20% of patients with polymyalgia have biopsy-proven GCA, and conversely more than 40% of patients with GCA have polymyalgic symptoms.
A summary of the extensive current data on this suggests that the annual incidence is particularly high in Scandinavia, with an annual rate between 15 and 35 per 100,000 individuals over 50. Similar rates have been described in the United States and in a community study in the United Kingdom. There is an increasing incidence with age which peaks in those 80 years or more. It is very rare under the age of 50 (summarized in Ref. 2).
Time Trends
Over the last six or seven decades, incidence of GCA appears to have increased, certainly in those reported from Olmstead County (Minnesota, USA). Between 1950–1954 and 1980–1984, the annual incidence increased from 6.7 to 28.5 per year per 100,000 individual over 50. Since then, the rate appears to have stabilized. Similar changes over time were noted in Sweden and Spain. Some of this increase may reflect improved recognition of cases as well as an aging population.
Giant cell arteritis appears to be more common in higher latitudes such as Scandinavia, whereas studies from southern Europe consistently reported low incidence rates than those from Scandinavia. Similar findings have been reported within North America where the lowest incidence is reported in the south.
Ethnic Differences
Giant cell arteritis appears to be most common in Caucasians and is uncommon in non-Caucasians. The incidence is highest in Scandinavians and in populations descended from them (e.g., the population from Olmstead County). Within America, there appears to be a particularly low incidence in African Americans (0.36 per 100,000 over the age of 50) compared with 2.24 in the white population in Tennessee.2 A study from California also suggested that GCA is much less common in Asians than Caucasians. There is also a particularly low incidence in Japan compared with Europe3 but this has not been studied elsewhere in Asia.5
Genetic Factors
These have been linked particularly to the human leukocyte antigen (HLA) DR4 and HLA DRB1*0401 genes. It is interesting that these genes are much less common in the normal Japanese population, which may explain why GCA is seen much less commonly in Japan.
Environmental Factors
The latitude effect on incidence is also seen in multiple sclerosis and rheumatoid arthritis (RA), and possibilities to explain this have included the effects of exposure to UV light and its potential effect on vitamin D synthesis. In the Olmstead County cohort, they describe peaks in incidence every 7 years, which suggests a possible infectious etiology, and some studies have suggested an association with mycoplasma pneumoniae but this has never been confirmed. Some, but not all studies, have suggested a season of maximum occurrence with GCA more likely to be diagnosed in warmer summer months in the United Kingdom.
Takayasu's Arteritis
Takayasu's arteritis was described in Japan in 1908, although older case reports indicate that it was not a new disease. The 2012 CHCC definition describes TAK as “arteritis, often granulomatous, predominately effecting the aorta and/or its major branches. Onset is usually in patients younger than 50.”1 It is thought to be much more common in Asia than in the United Kingdom, with an incidence somewhere between 0.5 and 3 per million. The peak age of onset is in the third decade and it is more common in women.
Takayasu's arteritis has been described in most ethnic groups, including Africans and Asians, and may be particularly common in India. Some reports suggest a high incidence in India, with as many as 500 patients with TAK attending vasculitis clinics in Mumbai or cardiology clinics in Kolkata. Takayasu's arteritis also has a higher prevalence in Japan (40 per million) compared with Europe (4–7 per million) and is also more common in Turkey (15 per million). Different subtypes of TAK are recognized depending on the site of major vessel involvement at arteriography. In India and other South Asian countries, the renal arteritis are affected the majority of patients, leading to a high number presenting with hypertension at a young age. In South African patients, lower abdominal aorta involvement is more common than in Japan. Aneurysmal disease is also more common in Africa compared with Japan.
The etiology of TAK is unknown. The granulomatous nature has suggested links to mycobacterial disease but that has never been proven. Genetic susceptibility may be important, as it has been recorded in monozygotic twins, and about 1% of Japanese patients with TAK have an affected relative.
MEDIUM-VESSEL VASCULITIS
Kawasaki's Disease
The CHCC 2012 defined Kawasaki's disease (KD) as “arteritis associated with mucocutaneous lymph node syndrome and predominantly affecting medium and small arteries. Coronary arterities are often involved. Aorta and large arteries may be involved. Usually occurs in infants and children.”1 The most obvious difference between KD and other vasculitides is its predilection for infants and young children. It was first recognized in Japan, which still has the highest incidence of over 200 per 100,000 per year in children aged under 5.
Geographical variations and ethnic factors are important. Kawasaki's disease is for example rare in the United Kingdom but much more common in individuals from southeast Asia. A study from Birmingham showed an incidence of 14.6 per 100,000 per year in children aged under 5 in Asians from the Indian subcontinent as compared with 5.9 per 100,000 per year in black Afro-Caribbeans and 4.6 per 100,000 per year in Caucasians.4 A large case series has been recently described from Chandigarh.5 Similarly in the United States, the incidence is highest amongst southeast Asians and Pacific islanders (2.4 times more common than in the white population). Interestingly, in Hawaii, the state with the highest proportion of southeast Asians or Pacific islanders, the incidence amongst Japanese Americans is very similar to that in native Japanese and twice that in native Hawaiians or Chinese Americans. Kawasaki's disease is also slightly more common in boys and the age-specific incidence is greatest in children aged 6–11 months, with nearly 90% of cases aged <5 years.
Other Factors
A lot of research has gone into possible infectious etiology because of the predominance of the disease in young children. It is rare before 6 months, suggesting that passively acquired maternal immunity may prevent the disease and the low frequency after the age of 5 suggests a possible early exposure to a common pathogen against which children mount an appropriate protective immune response. There have also 6been “epidemics” described in Japan and a seasonal variation with a peak incidence in winter and summer. However, there has been no convincing organism detected as a possible cause.
Polyarteritis Nodosa
This was the disease originally described by Kussmaul and Maier in 1866. The CHCC 2010 definition of PAN is “necrotising arteritis with medium or small arteries without glomerulonephritis or vasculitis in arterials, capillaries or venules and not associated with ANCA.”1 One of the clinical features that differentiate PAN from the smaller vessel vasculitides is its ability to cause major organ infarction and/or hemorrhage (the latter due to ruptured aneurysms). It is important to recognize that before the 1980s there was a significant overlap between cases labelled as PAN and MPA, and it was really only the development of the CHCC definitions in 1994 that helped to distinguish it followed by the introduction of ANCA. Microscopic polyangiitis is typically associated with myeloperoxidase (MPO) ANCA, and although patients with this disease can occasionally have larger artery involvement similar to that seen in PAN, the converse is not true, i.e., PAN is defined as having an absence of involvement of smaller vessels, as defined above, and is virtually always ANCA negative.
The estimated annual incidence is therefore difficult to interpret, making an exact figure unrealistic. The impression from our local studies is that PAN is much rarer than the smaller vessel vasculitides, probably with an incidence of less than 1 per million per year. It is also apparent from studies in France that the incidence is falling as the public health impact on controlling hepatitis B virus infection by vaccination takes effect. Personal experience suggests that it is still seen regularly in India but cases are predominantly hepatitis B negative.
Ethnic Differences
A particularly high incidence of PAN was recorded in Alaskan Indians in the 1980s, but this was in a very small population and all were positive for hepatitis B infection (PAN incidence 77 per million per year). The only other studies with high incidence rates are also in populations with a high hepatitis B infection including one French series (30 per million per year) and a Swedish study (31 per million per year).
SMALL-VESSEL VASCULITIS ANCA-ASSOCIATED VASCULITIS
The CHCC 2012 split small-vessel vasculitides into those associated with ANCA and those associated with immune complexes. Some older series include patients who fulfil more than one set of criteria, particularly in the ANCA-associated group (i.e., MPA and GPA). Clearly, accurate epidemiology requires a diagnosis without any overlap and an algorithm developed to sort this has been applied to many of the populations discussed.6,7
Granulomatosis with Polyangiitis (Wegener's Granulomatosis)
Granulomatosis with polyangiitis was first described in the 1930s (by Klinger as well as Wegener) and is defined by the CHCC 2012 as “necrotizing granulomatous inflammation usually involving the upper and lower respiratory tract and necrotising vasculitis involving predominantly small vessels (e.g. capillaries, venules, arterioles). Necrotising glomerulonephritis is common.”1
Time Trends
The incidence of GPA appeared to increase in the 1980s following the introduction of ANCA. Other studies since then have suggested a relatively stable incidence and we have not seen any significant change during the 22-year period of the Norwich Vasculitis Cohort.8 However, we did see a periodicity with peaks every 7.7 years (not seen in MPA), suggesting a possible infectious etiology.
The prevalence, however, has increased over time, but this probably reflects better treatment leading to improved mortality. The recent estimates for Norwich record a prevalence of 146 per million in 2008 similar to a study from Lund with a rate of 160 per million in 2003.8, 9
Geographical Factors
The incidence of GPA appears to be higher in northern Europe as compared with southern Europe and is much more common in northern Europe than in Japan where GPA is relatively rare. Studies from India suggest that some patients with GPA may be misdiagnosed because of similarities in pulmonary presentation.7
Ethnic Factors
The incidence rates are similar in most Caucasian populations, but the incidence is lower (approximately halved) in nonwhite Caucasian populations in Europe.10 Similarly in New Zealand, GPA is twice as common in Europeans as compared with Maoris or Asians.11 In Japan, GPA is much less common than MPA and also the incidence of the associated antibody cANCA or PR3 ANCA is much rarer in Japan than pANCA or MPO ANCA. Similar findings have been reported from China.
Genetic Factors
Various HLA associations have been studied but often in too small populations, but a recent genome-wide association study has shown significant associations. PR3 ANCA was associated with three specific genes—HLA DP, SERPINA 2, and PRTN3. SERPINA 1 encodes alpha-1 antitrypsin which may also be important.
Environmental Factors
A variety of environmental factors have been suggested. Seasonal differences have been described by some but not others; infection may be important, particularly upper respiratory infection with or nasal carriage of Staphylococcus. The presence of Staph aureus has been associated particularly with an increased risk of relapse. The predominance of GPA in northern versus southern European populations has led to a suggestion that UV radiation might be important (as was mentioned previously in association with GPA).
There have been extensive studies looking at potential occupational exposure. Of note has been an association with silica but that is perhaps more closely linked to MPA than GPA. There are also some data reporting links with occupational exposure to hydrocarbons such as paints and glues, and our own study showed an association with farming. It was not possible to distinguish what type of farming was important although the association appeared stronger in livestock than crops.12
Microscopic Polyangiitis
The first description of the disease we now call MPA, in 1948 by Davson and colleagues, used the term microscopic polyarteritis. MPA became the preferred term with the description of the association of this disease with MPO ANCA and at the time of the 1994 Chapel Hill Conference. Chapel Hill Consensus Conference 2012 defines MPA as “necrotising vasculitis with few or no immune deposits, predominantly affecting small vessels (i.e. capillaries, venules, arterioles). Necrotising arteritis involving small or medium arteries may be present. Necrotising glomerulonephritis is very common. Pulmonary capillaritis often occurs. Granulomatous inflammation is absent.”1 As previously described in the section on PAN, the literature regarding epidemiology has to be carefully interpreted because of the overlapping features of these two diseases prior to the introduction of CHCC definitions.
Time Trends
The incidence of MPA appeared to increase in the 1980s following the introduction of ANCA and our most recent studies suggest an annual incidence of 5.9 per million per year. Other studies from Stockholm also showed an increasing incidence between 1986 and 1992 although we have seen no obvious change in incidence since then. It seems likely that some of the increase in incidence was due to better recognition and changing ideas regarding diagnosis and classification.
Geographical Factors
The north/south gradient relating to the incidence of MPA is the inverse of that seen in GPA. Microscopic polyangiitis is more common in southern Europe (e.g., Lugo, Spain) as compared with Tromsø in northern Norway.13 The two geographical regions with an apparently high incidence of MPA and PAN are Japan and Kuwait, but the reasons for this are unclear.
Ethnic Differences
The Kuwait study reported only Kuwaitis with vasculitis. In Japan, studies from the Miyazaki prefecture showed an incidence of MPA much higher than European Caucasian populations and as outlined above, MPO ANCA much more frequent than PR3 ANCA-associated disease.
Environmental Factors
Microscopic polyangiitis has been associated with a number of environmental factors, particularly silica, hydrocarbons, 8drugs, and infections. Drug-associated vasculitis appears to be particularly associated with MPO ANCA and most commonly mimics MPA. The drugs most commonly associated with this are propylthiouracil and hydralazine although others have been implicated.
Case reports have described vasculitis in association with pulmonary silicosis, and as described for GPA, there has been a significant association with silica in many studies, more associated with MPA than GPA. An interesting study after a major earthquake in Kobe, Japan, in 1995 showed a trebling of the incidence of MPO ANCA-associated vasculitis, thought possibly to relate to high levels of silica dust caused by the earthquake. Studies in Norwich showed no significant association between MPA alone and occupational silica exposure, but there was an association overall with the whole range of AAVs.
Eosinophilic Granulomatosis with Polyangiitis (Churg-Strauss)
Eosinophilic granulomatosis with polyangiitis was first described in 1951 by two American pathologists who described features of patients at postmortem.14 The CHCC 2012 defined EGPA as “eosinophil rich and necrotising granulomatous inflammation often involving the respiratory tract and necrotising vasculitis affecting predominantly small to medium vessels and associated with asthma and eosinophilia. ANCA is more frequent when glomerulonephritis is present.”1 The incidence of EGPA appears broadly similar in all populations and is less common than GPA or MPA, with an incidence of between 0.5 and 2 million per year. Prevalence data show that EGPA is rarer than GPA or MPA in all countries studied, including the United Kingdom, New Zealand, and France (range 10–20 per million).
Environmental Factors
The cause of EGPA is unknown but the strong association with asthma suggests that inhaled antigens may be important. Eosinophilic granulomatosis with polyangiitis has also been described in association with vaccination and desensitization and with some drugs including sulfonamides. It has also been linked to the use of leukotriene inhibitors. Initial studies suggested that this was due to a masking of disease and steroid reduction, but a study from New Zealand showed that the drug may actually act as a trigger for the development of EGPA.
SMALL-VESSEL VASCULITIS-IMMUNE COMPLEX
IgA Vasculitis (Henoch-Schönlein Purpura)
This disease was first described by Heberden in the late 18th century but is attributed to Schönlein (1837), who described the purpura and arthritis, and Henoch (1874), who described the additional features of abdominal pain and kidney involvement. The importance of IgA has been recognized over recent decades, and it is clear that the original ACR studies included patients who we would now define as HSP but who were then classified as hypersensitivity vasculitis. There has therefore been a change in the definition of IgA vasculitis/HSP such that the CHCC 2012 defined IgA vasculitis as “vasculitis with IgA 1 dominant immune deposits, affecting small vessels (predominantly capillaries, venules or arterioles). Often involves skin and gut and frequently causes arthritis. Glomerulonephritis indistinguishable from IgA nephropathy may occur.”1 IgA vasculitis occurs predominantly in children, the peak onset at 5–6 years. It is relatively rare in adults with an incidence of 3 per million per year in adults as compared to 10–20 per 100,000 per year in children <17.
IgA vasculitis is slightly more common in Asians from the Indian subcontinent as compared with white Caucasians and blacks—24 per 100,000 per year in Asians, 17 per 100,000 per year in Caucasians, and 6 per 100,000 per year in blacks, all aged less than 17 years. Drugs have often been implicated in the etiology of IgA vasculitis (these include antibiotics, beta-lactams, analgesics, and nonsteroidal anti-inflammatory drugs).
Vasculitis Associated with Specific Conditions
Vasculitis has been described for many years associated with diseases such as RA, systemic lupus erythematosus, and Sjogren's syndrome. There are few classification criteria for these and less data than in the primary vasculitides.
Systemic Rheumatoid Vasculitis
A histological description of vasculitis in a patient with RA goes back to 1898, but the clinical studies really appeared in the 1940s and 1950s when patients presented with peripheral gangrene, mononeuritis multiplex and a variety of other systemic involvement.15,16 All types of blood vessels can be 9involved, with small-vessel vasculitis being more common. This can manifest as a relatively benign nail fold or nail edge infarcts, but this may also be associated with more widespread systemic disease. Systemic rheumatoid vasculitis is most commonly seen in patients with long-standing seropositive erosive RA and more common comparatively in males than females.
There was a strong association between the onset of rheumatoid vasculitis and unrestrained glucocorticoid use in the 1950s and 1960s, suggesting that it may be an important causative factor. It is clear that rheumatoid vasculitis has become less common with the reducing use of glucocorticoids, but even in the original studies, patients were seen who had never been treated with these drugs. The incidence peaked at over 10 per million in the 1970s and 1980s but has now fallen to <4 per million per year in our studies. The relative rarity of rheumatoid vasculitis was first noted in North America and appeared to occur in the United Kingdom in the 1990s so is probably related to better controlled disease with drugs such as methotrexate but not linked to the introduction of biologic agents.
Other Vasculitides
The list of vasculitides shown in Table 1.1 indicates that we have not covered disease such as Cogan's, Behçet's, cryoglobulinemia, or anti-GBM disease. Behçet's disease is more difficult because the clinical features of Behçet's disease are not always due to vasculitis, and the epidemiology of that disease is outside the remit of this chapter. The other diseases have not been well studied in terms of epidemiology. Here, as indicated in the major syndromes discussed above, there is a tremendous need for prospective population-based studies such as disease registers. The increasing interest in systemic vasculitis needs to be accompanied by a strong move away from individual stamp collecting of rarities to organized collaborative systemic studies.
This chapter was adapted from Epidemiology of Vasculitis by Watts and Scott 201417 (see references therein).
REFERENCES
- Jennette JC, Falk RJ, Bacon PA, et al. 2012 revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides. Arthritis Rheum. 2013;65:1–11.
- Smith CA, Fidler WJ, Pinals RS. The epidemiology of giant cell arteritis. Report of a ten-year study in Shelby county, Tennessee. Arthritis Rheum. 1983;26:1214–9.
- Kobayashi S, Yano T, Matsumoto Y, et al. Clinical and epidemiologic analysis of giant cell (temporal) arteritis from a nationwide survey in 1998 in Japan: the first government-supported nationwide survey. Arthritis Rheum. 2003;49:594–8.
- Gardner-Medwin JM, Dolezalova P, Cummins C, et al. Incidence of Henoch–Schonlein purpura, Kawasaki disease, and rare vasculitides in children of different ethnic origins. Lancet. 2002;360:1197–202.
- Singh S, Aulakh R, Bhalla AK, et al. Is Kawasaki disease incidence rising in Chandigarh, north India? Arch Dis Child. 2011;96:137–40.
- Watts R, Lane S, Hanslik T, et al. Development and validation of a consensus methodology for the classification of the ANCA-associated vasculitides and polyarteritis nodosa for epidemiological studies. Ann Rheum Dis. 2007;66:222–7.
- Liu LJ, Chen M, Yu F, et al. Evaluation of a new algorithm in classification of systemic vasculitis. Rheumatology (Oxford). 2008;47:708–12.
- Watts RA, Scott DG. ANCA vasculitis: to lump or split? Why we should study MPA and GPA separately. Rheumatology (Oxford). 2012;51:2115–7.
- Mohammad AJ, Jacobsson LT, Mahr AD, et al. Prevalence of Wegener's granulomatosis, microscopic polyangiitis, polyarteritis nodosa and Churg–Strauss syndrome within a defined population in southern Sweden. Rheumatology (Oxford). 2007;46:1329–37.
- Mahr A, Guillevin L, Poissonnet M, et al. Prevalences of polyarteritis nodosa, microscopic polyangiitis, Wegener's granulomatosis, and Churg–Strauss syndrome in a French urban multiethnic population in 2000: a capture–recapture estimate. Arthritis Rheum. 2004;51:92–9.
- O’Donnell JL, Stevanovic VR, Frampton C, et al. Wegener's granulomatosis in New Zealand: evidence for a latitude-dependent incidence gradient. Intern Med J. 2007;37:242–6.
- Lane SE, Watts RA, Bentham G, et al. Are environmental factors important in primary systemic vasculitis? A case–control study. Arthritis Rheum. 2003;48:814–23.
- Watts RA, Gonzalez-Gay MA, Lane SE, et al. Geoepidemiology of systemic vasculitis: comparison of the incidence in two regions of Europe. Ann Rheum Dis. 2001;60:170–2.
- Churg J, Strauss L. Allergic granulomatosis, allergic angiitis, and periarteritis nodosa. Am J Pathol. 1951;27:277–301.
- Scott DG, Bacon PA, Tribe CR. Systemic rheumatoid vasculitis: a clinical and laboratory study of 50 cases. Medicine (Baltimore). 1981;60:288–97.
- Bywaters EG, Scott JT. The natural history of vascular lesions in rheumatoid arthritis. J Chronic Dis. 1963;16:905–14.
- Ball GV, Fessler BJ, Bridges SL Jr. Oxford Textbook of Vasculitis. Oxford University Press; 2014.