Kawasaki disease: Epidemiology and etiology

INTRODUCTION — Kawasaki disease (KD, previously called mucocutaneous lymph node syndrome) is one of the most common vasculitides of childhood, particularly in East Asia. It is typically a self-limited condition, with fever and other acute inflammatory manifestations lasting for an average of 12 days if not treated. The underlying etiology is unknown.

KD can cause a variety of cardiovascular complications, including coronary artery aneurysms, cardiomyopathy with depressed myocardial contractility and heart failure, myocardial infarction, arrhythmias, and peripheral arterial occlusion. These complications may cause significant morbidity and mortality, particularly in children who are inadequately treated. The frequency of aneurysm development and mortality has dramatically decreased as a result of intravenous immune globulin therapy. Early diagnosis is critical to achieve the optimal treatment result.

The epidemiology and etiology of KD are reviewed here. The clinical features, diagnosis, treatment, and cardiac sequelae are presented separately. Incomplete KD is also discussed separately. (See "Kawasaki disease: Clinical features and diagnosis" and "Kawasaki disease: Initial treatment and prognosis" and "Cardiovascular sequelae of Kawasaki disease: Clinical features and evaluation" and "Incomplete (atypical) Kawasaki disease".)

EPIDEMIOLOGY — KD is second only to immunoglobulin A (IgA) vasculitis (Henoch-Schönlein purpura) as the most common vasculitis of childhood [1].

Geographic variation — The incidence of KD is greatest in children who live in East Asia or are of Asian ancestry living in other parts of the world [2-4]. The incidence in underdeveloped countries is largely unknown, and ascertainment is likely to be incomplete. KD is particularly difficult to diagnose in areas where measles is still prevalent since the presentation is similar [5,6]. Many nations around the world have demonstrated an increase in the number of children diagnosed with KD since the early to mid-2000s. It is not clear, however, whether this represents an actual increase in the incidence of the disease, increased awareness of the condition, or a greater tendency to classify children with incomplete clinical features as having KD.

The following studies illustrate the geographic and ethnic variation in the incidence of KD:

●Japan – KD is a reportable disease in Japan, and almost 250,000 cases of KD have been registered there since its initial description by Tomisaku Kawasaki in 1967 [7,8]. The incidence among children younger than five years of age was approximately 215 per 100,000 per year in 2007 and 2008, with the incidence in boys aged zero to four years reaching as high as 240 per 100,000 [8]. The highest incidence was among children aged 6 to 11 months. Thus, approximately 1 in 100 Japanese children develops KD by the age of five years, making KD a very common illness in Japan.

Although epidemics of KD were observed in Japan in 1979, 1982, and 1986, there has not been an epidemic since that time. However, incidence rates now exceed the rates observed during the three previous epidemics [8]. In Japan, KD is most prevalent in the winter, with a smaller peak evident in summer [8].

●Taiwan and China – Data from the Taiwan National Health Insurance Database indicate that the incidence of KD among children younger than five years was 69 per 100,000 per year between 2003 and 2006 [9]. A hospital-based survey of KD in all 45 hospitals in Beijing reported an increase in the incidence in children younger than five years of age from 41 per 100,000 in 2000 to 51 per 100,000 in 2004 [10].

●United States – Studies of hospital discharge records by the United States Centers for Disease Control (CDC) estimated an overall annual incidence of 20 per 100,000 children younger than five years in the United States [11]. Annual incidence was highest among Asians and Pacific Islanders (30 per 100,000), intermediate among non-Hispanic African Americans (17 per 100,000) and Hispanics (16 per 100,000), and lowest among Caucasians (12 per 100,000) [11]. A winter-spring predominance of cases is characteristic, and the peak incidence of illness is at less than one year of age [11]. In contrast to Japan, surveillance in the United States is passive, and many cases may be missed.

●England – In an analysis of national admissions data in England, the annual incidence of KD for children under five years of age averaged 8 per 100,000 from 1998 to 2003 [12]. The incidences were higher for children of Chinese ethnicity and those who lived in areas of greatest urban density and with the greatest degree of poverty.

●Israel – A retrospective study using the Israel National Hospital Discharge Register indicated an increase in KD incidence in Israel from 5 per 100,000 children under five years of age from 1996 to 1999 to 7 per 100,000 in 2000 to 2004 [13]. For male infants under one year of age, the KD incidence rate doubled between 1996 to 1998 and 1999 to 2009.

Other risk factors — Boys are affected as much as 50 percent more commonly than girls [3,9,10,14,15]. Eighty to 90 percent of cases occur in children younger than five years [9,16], although KD is relatively uncommon among children younger than six months (approximately 10 percent of KD hospitalizations in the United States) [11]. Occurrence beyond late childhood is rare [16,17], although older children can develop KD and may experience delays in diagnosis and higher rates of coronary artery disease [18,19]. Fewer than 100 cases of classic KD in adults had been reported in the literature at the time of a 2010 case series and review [20].

The epidemiology of the illness is illustrated by an analysis of data from the Pediatric Health Information System (PHIS) that identified 4811 patients who were treated in 27 hospitals in the United States between 2001 and 2006 [3]. The median age at first admission was 3.4 years (range 1 month to 21.3 years), and 60 percent of the patients were between one and four years of age. Sixty percent of patients were male. Patients of Asian ancestry were overrepresented in the KD group compared with the overall patient group contained within the PHIS data set (6.9 versus 1.6 percent).

In Japan, there is a reported 10-fold increased risk of KD for children with an affected sibling and a twofold increased risk for those with a previously affected parent [21,22]. In North America, there are case reports of families with multiple affected members, but data are insufficient to determine whether there is an increased familial risk for developing KD [23].

POSSIBLE ETIOLOGIC FACTORS — The etiology of KD remains unknown. A variety of theories have been proposed based upon pathologic, epidemiologic, and demographic data [24]. Infection by one or more agents that usually cause an asymptomatic or nonvasculitic condition in most children, but which results in KD in genetically predisposed individuals, fits the epidemiologic data well [25]. Various case series over many years have reported localized outbreaks of KD, each associated with a different bacterial or viral pathogen such as parvovirus B19 [26], Propionibacterium [27], human bocavirus [28], and numerous others.

The major debate concerns whether KD is caused by a single, heretofore unidentified agent or an immunologic response to a variety of triggers. This point remains unresolved, although the possibility of a unique KD infection seems less likely with each negative serologic, genetic, and immunologic study or case series [25].

Immunologic response — KD is a systemic, inflammatory illness that particularly affects medium-sized arteries, especially the coronary arteries. Pathologic studies indicate that multiple organs and tissues are involved [29], but long-term sequelae appear to occur only in the arteries. Blood vessel damage appears to result from inflammatory cell infiltration into vascular tissues. The stimulus for this infiltration is unknown, but it is most profound in the coronary arteries and can involve destruction of luminal endothelial cells, elastic lamina, and medial smooth muscle cells in severe cases. The destruction of elastin and collagen fibers and loss of structural integrity of the arterial wall lead to dilatation and aneurysm formation. Inflammatory cells infiltrating the coronary arteries may include neutrophils, T cells (particularly CD8 T cells), eosinophils, plasma cells (particularly IgA producing), and macrophages. Macrophages are not prominent participants in any other type of vasculitis [30].

A neutrophilic infiltrate is observed in the arterial wall in KD fatalities that occur within the first two weeks after fever onset and may represent an innate immune response [31]. A study of gene expression patterns in acute KD peripheral blood using DNA microarrays reflects the predominance of neutrophils early in the course of KD [32]. Expression of genes associated with neutrophils and inflammatory processes, including adrenomedullin, grancalcin, and granulin, was high during the acute phase of illness. Levels of these transcripts tended to decrease over time, while gene expression in CD8 T cells and natural killer (NK) cells increased as the illness evolved. CD8 T cells were prevalent in the arterial wall in fatalities that occurred after two weeks in another study, consistent with the development of an acquired immune response [33]. Gene expression in the peripheral blood from individual patients tended to be more consistent than were transcript levels in different subjects on the same day of illness. This suggests the possibility that DNA polymorphisms affected individuals' gene expression.

Plasma cells producing oligoclonal IgA antibodies are found in the arteries and respiratory tract of children with KD [34]. A synthetic monoclonal antibody derived from prevalent IgA sequences in the KD arterial wall identified intracytoplasmic inclusion bodies consistent with aggregates of proteins and nucleic acids in 85 percent of KD cases but not in infant control tissues [35,36]. Immune complexes are sometimes detected in the peripheral blood in KD, but they are not observed to form deposits in affected tissues and do not appear to correlate with the development of coronary artery disease [37,38].

Infectious etiology — Many epidemiologic data suggest that KD is caused or triggered by a transmissible agent or agents. Support for this theory is derived from the following similarities between KD and other pediatric infectious conditions [1]:

●KD is characterized by a febrile exanthem with lymphadenitis and mucositis. These are features similar to those of contagious diseases, such as adenovirus infection, measles, and scarlet fever.

●There is a seasonal increase in the winter and summer [39].

●The disease often occurs in epidemics [39,40], and a geographic wave-like spread of illness during epidemics has been noted [40].

●Boys are more susceptible than girls. In general, the "set-point" of the immune system varies between sexes, leading to an overall higher incidence and morbidity due to infections in males (eg, meningitis [41], Campylobacter enteritis [42], etc) and a higher incidence of autoimmune diseases in females [43].

●Siblings of children with KD in Japan are at increased risk for developing the disease, which usually occurs within one week of onset of the index case [40].

●The disease is common among children younger than five years but rare in those younger than six months. The rarity in infants may be explained by transfer of passive immunity to the relatively common infectious trigger(s) by transplacentally acquired maternal antibodies.

●There is spatial and temporal clustering of cases [44].

The cause(s) of KD remain(s) unknown despite clinical and epidemiologic data suggesting a relationship to infections. In fact, a retrospective review of 129 consecutive patients diagnosed during a 24-month period from 1997 to 1998 at a single, tertiary-care hospital found that 33 percent of children with typical KD had at least one confirmed infection at KD diagnosis [45]. However, a wide variety of different bacterial and viral infections were identified. Similarly, screening of patients with "universal" prokaryotic and eukaryotic primers has not identified a single infectious agent as the cause. Thus, the significance of findings, such as the presence of aggregates of particles that share morphologic features with several RNA virus families in a small number of patients with KD, remains unclear [46].

Multisystem inflammatory syndrome in children — Coronavirus disease 2019 (COVID-19) is associated with hyperinflammatory syndromes [47]. Children may develop organ failure involving the gastrointestinal tract, heart, central nervous system, kidneys, and other systems. The severe inflammation, cytopenias, coagulopathy, and hyperferritinemia are similar to macrophage activation syndrome or toxic shock in some children. Others develop mucocutaneous symptoms like those in KD [48]. In a percentage of such cases, coronary artery dilatation, and even formation of giant coronary aneurysms, may develop. (See "COVID-19: Multisystem inflammatory syndrome in children (MIS-C) clinical features, evaluation, and diagnosis".)

Genetic factors — Genetic factors appear to contribute to the pathogenesis of this disorder, as suggested by the increased frequency of the disease in Asian and Asian-American populations and among family members of an index case [1,21,49-52].

Variants or polymorphisms of the following genes are associated with an increased susceptibility to KD:

●Inositol 1,4,5-trisphosphate 3-kinase C (ITPKC) gene on chromosome 19q13.2 [53,54]. ITPKC acts as a negative regulator of T cell activation, which includes transcription of interleukin (IL) 2. The single nucleotide polymorphism (SNP) associated with KD susceptibility results in a weaker inhibitory effect upon T cell activation. Patients with this SNP may have a more vigorous T cell response during an inflammatory disease, such as KD, compared with those without this allelic change. However, this polymorphism is not common enough even in the Japanese population to explain the vast majority of cases of KD.

●Angiopoietin 1 (ANGPT1) and vascular endothelial growth factor A (VEGFA) genes [55-57]. Expression of angiopoietin 1 is upregulated and VEGF is downregulated in patients with acute versus convalescent KD, suggesting disruption of vascular homeostasis.

●The genes encoding the chemokine receptor CCR5 and its major ligand CCL3L1 [58].

●The adenosine triphosphate (ATP) binding cassette, subfamily C, member 4 (ABCC4) gene. ABCC4 is a cyclic nucleotide transporter involved in migration of dendritic cells and cellular efflux of prostaglandin [59].

Genome-wide association studies have revealed other potential susceptibility loci, including a functional polymorphism in the immunoglobulin G receptor gene (FCGR2A) [60-64]. In addition to being a putative susceptibility factor, ITPKC is associated with an increased risk of coronary artery aneurysms. Other genetic factors related to the development of coronary lesions in KD are discussed separately. (See "Cardiovascular sequelae of Kawasaki disease: Clinical features and evaluation", section on 'Coronary artery abnormalities'.)

ENVIRONMENTAL FACTORS — Environmental factors also have been proposed as the triggers of KD. Overlap between the cutaneous and acral manifestations of KD to those of acrodynia, a disease caused by mercury toxicity, led to proposals that mercury might play a causative role in KD [65]. Similarly, the frequent association between KD and atopy, particularly atopic dermatitis, has inspired a search for a common immune susceptibility [66]. This has led to various hypotheses involving dust mites, rug shampoo, and pollen release in the etiology of KD, although, once again, additional data accrued over time have not provided support to these theories. Analysis of seasonal variations in epidemics of KD suggested to one group that cases are linked to largescale wind currents from Central Asia, potentially carrying an airborne antigenic trigger in the troposphere [67].

Despite the plausibility of these and numerous other theories, it is difficult to understand how a single environmental factor could account for a ubiquitous disease such as KD that occurs at all times of the year and in virtually every country on Earth. Thus, like other pediatric vasculitides such as IgA vasculitis (Henoch-Schönlein purpura) [68], KD may occur in genetically susceptible individuals following exposure to any of a variety of infectious and/or environmental triggers [69].

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Kawasaki disease (The Basics)")

SUMMARY

●Epidemiology – Kawasaki disease (KD) is one of the most common vasculitides of childhood. The incidence of KD is greatest in children who live in East Asia (eg, Japan, Korea, Taiwan) or are of Asian ancestry living in other parts of the world. Other risk factors include male sex, age between six months and five years, and family history of KD. (See 'Epidemiology' above.)

●Etiology and pathogenesis – The etiology of KD remains unknown, but several etiologic factors have been postulated:

•Immunologic response – Inflammatory cell infiltration into KD vascular tissue leads to vascular damage, but the stimulus for this inflammatory infiltration has not been identified. (See 'Immunologic response' above.)

•Infectious etiology – The similarities between KD and other pediatric infectious conditions suggest that KD is caused by a transmissible agent. However, no studies have convincingly identified a specific virus, bacterium or bacterial toxin, or other pathogen associated with KD. The etiology may be a previously unidentified infectious agent. An alternative theory to a specific inciting agent is that KD represents a final common pathway of immune-mediated vascular inflammation following a variety of inciting infections and/or environmental antigens. (See 'Infectious etiology' above.)

•Genetic factors – Genetic factors appear to contribute to the pathogenesis of this disorder, as suggested by the increased frequency of the disease in Asian and Asian-American populations and among family members of an index case. A number of gene polymorphisms are associated with an increased susceptibility to KD, and some of these variants are also associated with coronary artery lesions and aneurysm formation. (See 'Genetic factors' above.)