After more than sixty years of research, the cause of Kawasaki disease remains officially unknown. The illness was first described in 1961 by a Japanese pediatrician named Tomisaku Kawasaki, and despite thousands of studies since then, no single pathogen or mechanism has been conclusively identified as the trigger. What researchers do know is that the disease behaves like an exaggerated immune response to some kind of infectious exposure in children who are genetically predisposed to it, and that exposure likely involves something in the environment that varies by season and geography.
A Disease Defined by What It Does, Not What Starts It
Kawasaki disease is classified as a systemic vasculitis, meaning it inflames blood vessels throughout the body, with a particular affinity for the coronary arteries that supply the heart. It overwhelmingly strikes children under five years old and is the leading cause of acquired heart disease in children in developed countries.1PubMed Central. Kawasaki disease as a systemic vasculitis in childhood The hallmark symptoms are familiar to pediatricians: a high fever lasting at least five days, a rash, red eyes, swollen lymph nodes, cracked lips, and a distinctive “strawberry tongue.” But the real danger is what happens inside the coronary arteries. Without treatment, roughly a quarter of children develop coronary artery aneurysms, which are balloon-like bulges in the artery walls that can persist for life.2PubMed Central. Coronary Artery Aneurysm in Kawasaki Disease: Coronary CT Angiography through the Lens of Pathophysiology and Differential Diagnosis
Since the disease lacks a known cause, it is diagnosed entirely by its clinical features. There is no blood test that confirms Kawasaki disease. This makes the search for its cause more than an academic exercise: identifying the trigger would not only explain the disease but could lead to a diagnostic test and possibly prevention.
Why Geography and Ancestry Offer Clues
One of the strongest patterns in Kawasaki disease is who gets it. The incidence varies enormously across the globe. In Japan, nearly one in every hundred children develops the disease by age five, making it extraordinarily common there. By contrast, reported rates are lowest in sub-Saharan Africa.3PubMed Central. Kawasaki Disease: Global Burden and Genetic Background Children of Japanese and Korean ancestry living in Hawaii or the continental United States also develop Kawasaki disease at higher rates than their white or Black peers, which strongly suggests a genetic component beyond simply living in East Asia.
This geographic and ethnic pattern has pushed researchers toward two complementary explanations that are not mutually exclusive: a genetic predisposition that makes certain children vulnerable, and an environmental trigger that is more common in some parts of the world than others.
The Case for an Infectious Trigger
Kawasaki disease behaves, in many ways, like an infection. It causes fever, it clusters in seasonal outbreaks (peaking in winter and early spring in Japan), and it rarely strikes the same child twice, suggesting that the immune system develops lasting protection after a first episode. Japan experienced three massive nationwide epidemics in 1979, 1982, and 1986, with case counts spiking and falling in patterns that looked like waves of a contagious illness spreading across the country.4PubMed. Kawasaki disease: A brief history
Yet decades of effort have failed to pin the disease on any single virus or bacterium. Researchers have looked at virtually every common childhood pathogen and come up empty in the sense that no one agent is consistently found in every child with the illness. That does not mean infections play no role; it may mean the trigger is something unusual or hard to detect with standard methods.
Viral Clues
Some of the most intriguing evidence for a viral cause comes from studies of lung tissue in children who died of acute Kawasaki disease. Researchers using specialized synthetic antibodies found inclusion bodies, which are dense clumps of material inside cells, in the ciliated cells lining the airways of these children. These inclusion bodies were not present in control tissue or in other cell types, and under electron microscopy they resembled the kind of protein-and-nucleic-acid aggregates that form during infection with RNA viruses.5PubMed Central. Cytoplasmic Inclusion Bodies Are Detected by Synthetic Antibody in Ciliated Bronchial Epithelium during Acute Kawasaki Disease The fact that these structures appear specifically in airway cells is consistent with a pathogen that enters through the respiratory tract, which would fit the seasonal and outbreak patterns of the disease.
Still, no specific virus has been isolated from these inclusion bodies and definitively linked to Kawasaki disease. The structures are suggestive but not proof. Some researchers have proposed that the trigger might be a novel or poorly characterized virus that standard laboratory tests simply do not detect.
The Superantigen Hypothesis
A competing theory focuses on bacterial toxins called superantigens. Normal immune responses involve a small fraction of the body’s T cells recognizing a specific invader. Superantigens bypass that specificity and activate a huge number of T cells at once, causing a massive inflammatory surge. Researchers have isolated superantigen-producing bacteria, particularly certain strains of Staphylococcus aureus and Streptococcus, from children with acute Kawasaki disease.6Progress in Pediatric Cardiology. The etiology of Kawasaki disease: a superantigen-mediated process
The superantigen theory would explain the intense, widespread inflammation seen in the disease. But like the viral hypothesis, it has a problem: not all children with Kawasaki disease carry these bacteria, and plenty of children who carry superantigen-producing bacteria never develop the illness. This has led many researchers to suspect that Kawasaki disease is not caused by one specific microbe but rather by an abnormal immune reaction that can be set off by several different infectious triggers in the right host.
A Windborne Agent From Central Asia
One of the more surprising lines of research links Kawasaki disease outbreaks to large-scale wind patterns. Analyses of Japan’s three major epidemics and the year-to-year fluctuations in case numbers found a consistent connection: more cases occurred when strong tropospheric winds blew from central Asia across the north Pacific.7Scientific Reports. Association of Kawasaki disease with tropospheric wind patterns The same pattern held for Hawaii and San Diego, which are thousands of miles farther east but connected by the same wind corridor. This could explain why Kawasaki disease peaks simultaneously in Japan, Hawaii, and coastal California each winter, despite the vast distances separating them.
Further atmospheric modeling pinpointed the densely cultivated farmlands of northeastern China as a likely source region for whatever the wind is carrying.8PubMed Central. Tropospheric winds from northeastern China carry the etiologic agent of Kawasaki disease from its source to Japan One leading candidate is Candida, a common fungus. Candida species have been found in seasonal dust carried by these wind currents, and separate remote sensing studies have documented how dust cycles from agricultural regions correlate with annual spikes in Kawasaki disease cases in Japan.9PubMed Central. Remote sensing observation of annual dust cycles and possible causality of Kawasaki disease outbreaks in Japan
This hypothesis gains support from animal experiments. When mice are injected with a water-soluble extract of Candida albicans cell walls, they develop a form of vasculitis that mimics Kawasaki disease, including coronary artery inflammation.10PubMed. Crucial role of NLRP3 inflammasome in a murine model of Kawasaki disease That does not prove Candida causes the human disease, but it shows that fungal components can reproduce the core pathology in an experimental setting. The wind hypothesis remains controversial, and many researchers are skeptical that a single airborne agent explains a disease seen worldwide, including in regions not downwind of central Asia.
Genetic Susceptibility
The stark differences in incidence across ethnic groups point clearly to genetic factors. Genome-wide studies have identified several genes that influence whether a child develops the disease at all and how severe it becomes. One well-replicated finding involves a variant in the CASP3 gene, which encodes a protein involved in programmed cell death. A specific change in the regulatory region of CASP3 disrupts how immune cells bind to the surrounding DNA, altering the gene’s activity in immune cells and increasing susceptibility to Kawasaki disease. This association held in both Japanese and European American populations.11PubMed Central. Common variants in CASP3 confer susceptibility to Kawasaki disease
Beyond susceptibility, genetics also influence how a child responds to treatment. Kawasaki disease is treated with high-dose intravenous immunoglobulin (IVIG), a pooled antibody preparation given through a vein. About ten to twenty percent of children do not respond to the first round. A genome-wide study identified eleven genetic variants that, when combined into a risk score, could predict which children were more likely to be IVIG-resistant.12PubMed Central. Prediction for Intravenous Immunoglobulin Resistance by Using Weighted Genetic Risk Score Identified From Genome-Wide Association Study in Kawasaki Disease Children who do not respond to IVIG face a higher risk of developing coronary artery aneurysms, so identifying them early matters for clinical decisions.13PubMed. Risk Factors for Coronary Artery Aneurysms in Children With Kawasaki Disease
Epigenetic Changes During the Illness
In addition to inherited DNA variants, the disease appears to reshape how genes are switched on and off through chemical modifications to DNA. Studies comparing blood samples from children in the acute phase of Kawasaki disease against healthy controls have found widespread changes in DNA methylation, the chemical tagging system cells use to silence or activate genes. In acute Kawasaki disease, genes involved in inflammation and innate immune responses tend to become less methylated (more active), while genes governing adaptive immune functions like T cell activation become more methylated (quieter).14PubMed. Epigenome-Wide DNA Methylation Profiling of Peripheral Blood Shows Lymphocyte Dysfunction in Children with Kawasaki Disease
One study found that genes in the S100A family, which are involved in how immune cells cross blood vessel walls, were turned up during the acute phase and then turned back down during recovery, corresponding to methylation changes that tracked with disease activity.15PubMed Central. Multiomics analyses identified epigenetic modulation of the S100A gene family in Kawasaki disease and their significant involvement in neutrophil transendothelial migration Broader analyses have identified hundreds of genes with altered methylation patterns in Kawasaki disease patients, many clustered in pathways related to inflammation, blood clotting, and innate immunity.16PubMed. Comprehensive analyses of DNA methylation and gene expression profiles of Kawasaki disease Whether these epigenetic shifts are a cause, a consequence, or an amplifier of the disease process is an open question, but they help explain why the immune response in Kawasaki disease is so intense and so specifically damaging to arteries.
How the Immune System Damages the Arteries
Whatever the initial trigger, the downstream damage in Kawasaki disease is relatively well understood. The immune system mounts an aggressive response that converges on medium-sized arteries, especially the coronary arteries. Part of this involves the NLRP3 inflammasome, a molecular alarm system inside immune cells. In both mouse models and human patients, NLRP3 activation drives the production of inflammatory signaling molecules like IL-1 and IL-18, and promotes a form of inflammatory cell death called pyroptosis.17PubMed. Potential roles of NLRP3 inflammasome in the pathogenesis of Kawasaki disease
The structural damage to artery walls is driven in large part by enzymes called matrix metalloproteinases, particularly MMP-9. In animal models, MMP-9 breaks down elastin, the stretchy structural protein that gives arteries their flexibility. When researchers knocked out the MMP-9 gene in mice, the animals still developed vascular inflammation after disease induction, but elastin breakdown and aneurysm formation were virtually eliminated.18PubMed. Matrix metalloproteinase 9 activity leads to elastin breakdown in an animal model of Kawasaki disease In human patients, MMP-9 levels surge during the acute phase and the enzyme stains diffusely through coronary artery lesions.19PubMed. Matrix metalloproteinase-9 in vascular lesions and endothelial regulation in Kawasaki disease
There is also an autoimmune dimension. Children with acute Kawasaki disease produce antibodies that bind to the cells lining their own blood vessels. These anti-endothelial cell antibodies activate the endothelial cells, causing them to pump out inflammatory signals and express adhesion molecules that attract more immune cells to the vessel wall, creating a self-reinforcing cycle of inflammation.20PubMed Central. The role of anti-endothelial cell antibodies in Kawasaki disease – in vitro and in vivo studies
An Antigen-Driven Response, Not Random Chaos
A key question in understanding any immune-mediated disease is whether the immune response is specific, meaning directed at a particular target, or nonspecific, meaning a generalized alarm that attacks everything. In Kawasaki disease, the evidence tilts toward specificity. Studies of coronary artery tissue from fatal cases found that the IgA-producing immune cells infiltrating the artery walls produced a restricted, repetitive set of antibodies rather than a random assortment. This oligoclonal pattern, where a small number of antibody types dominate, is the hallmark of an immune system that has recognized a specific invader and is ramping up antibodies against it.21PubMed. Oligoclonal IgA response in the vascular wall in acute Kawasaki disease
This finding is significant because it argues against the superantigen theory as a complete explanation. Superantigens activate T cells in a broad, nonspecific way, but the antibody response in the artery walls looks focused and deliberate. It supports the idea that a conventional pathogen, one that the adaptive immune system recognizes through normal antigen-specific pathways, is involved in triggering the disease. The challenge is that researchers have not yet identified what antigen those antibodies are targeting.
Lessons From MIS-C During the Pandemic
When COVID-19 swept the world in 2020, pediatricians began seeing a new condition in children called multisystem inflammatory syndrome in children, or MIS-C. It appeared weeks after a SARS-CoV-2 infection and looked strikingly similar to Kawasaki disease: fever, rash, red eyes, and coronary artery involvement. Early reports described MIS-C as essentially Kawasaki disease triggered by a known virus.22PubMed Central. A comparison of Kawasaki Disease and multisystem inflammatory syndrome in children
As more data accumulated, the two conditions turned out to be more different than they first appeared. MIS-C tends to affect older children, is more common in Black and Hispanic children rather than East Asian children, and has distinct immunological features. Children with MIS-C show pronounced drops in certain T cell populations and different patterns of inflammatory signaling compared with Kawasaki disease patients. The autoantibody profiles also differ, with some antibodies against blood vessel proteins elevated in both conditions but others unique to Kawasaki disease.23Nature Reviews Rheumatology. Multisystem inflammatory syndrome in children and Kawasaki disease: a critical comparison Some researchers have concluded that MIS-C shares more in common with toxic shock syndrome than with Kawasaki disease in terms of its underlying immune mechanisms.24PubMed Central. Kawasaki Disease and Multisystem Inflammatory Syndrome in Children: Common Inflammatory Pathways of Two Distinct Diseases
The MIS-C experience is instructive for Kawasaki disease research. It demonstrated that a known virus can provoke a Kawasaki-like syndrome in genetically susceptible children, lending credibility to the idea that Kawasaki disease itself is triggered by one or more common infectious agents. But it also showed that the specific features of the host immune response, not just the trigger, determine which disease develops. Two different triggers can produce similar-looking vascular inflammation through different immune pathways.
Animal Models and Why the Puzzle Persists
Much of what researchers know about Kawasaki disease mechanics comes from mouse models, and these reveal something interesting about the cause question. Two different experimental approaches can produce coronary arteritis that looks like Kawasaki disease under a microscope. One uses an extract from Lactobacillus casei, a common bacterium, whose cell wall components activate toll-like receptors on immune cells and trigger focal coronary artery inflammation.25PubMed. TLR2 and MyD88 contribute to Lactobacillus casei extract-induced focal coronary arteritis in a mouse model of Kawasaki disease The other uses the Candida albicans extract mentioned earlier. Two completely different organisms, one bacterial and one fungal, produce essentially the same arterial disease in mice.
This convergence hints at a frustrating possibility: the specific trigger might matter less than the immune pathway it activates. Both models work by stimulating pattern-recognition receptors on innate immune cells, including the dectin-1 receptor for fungal components and TLR2 for bacterial ones.26PubMed. Dectin-1/Syk signaling is involved in Lactobacillus casei cell wall extract-induced mouse model of Kawasaki disease Both depend on the involvement of T cells to produce full-blown coronary arteritis.27PubMed Central. CD8+ T Cells Contribute to the Development of Coronary Arteritis in the Lactobacillus casei Cell Wall Extract-Induced Murine Model of Kawasaki Disease If Kawasaki disease in humans works the same way, there may not be a single “cause” to find. The disease could be the end result of several different microbial triggers funneling into the same inflammatory cascade in a child whose genetics make that cascade unusually aggressive.
That would explain one of the most puzzling features of Kawasaki disease history. When Tomisaku Kawasaki first published his report in Japanese in 1967, the disease was rapidly recognized in countries around the world throughout the 1960s and 1970s. One explanation offered at the time was that it was a new disease spreading outward from Japan. But another possibility, the one that many researchers now favor, is that Kawasaki disease had been around for a long time and was only recognized as a distinct entity once declining rates of scarlet fever and other bacterial rash illnesses in the antibiotic era made it visible as a separate condition.4PubMed. Kawasaki disease: A brief history If the cause were a single novel pathogen, it would be strange for it to have appeared simultaneously on multiple continents. If the cause is a common class of triggers in the right host, the pattern makes much more sense.