Ancestry and Kawasaki Disease: Clinical Pathway Considerations

Abstract

Kawasaki disease (KD) is the leading cause of pediatric acquired heart disease in developed countries.1 Despite decades of research, no established biologic marker has been identified, and KD remains a clinical diagnosis.1,2 However, KD research has shown that different populations are at higher risk of complications, and early intervention and treatment of these patients may improve outcomes. The American Heart Association (AHA) and American College of Rheumatology (ACR) KD guidelines call for early identification of high-risk patients to consider intensified primary treatment.1,2 In a multiethnic population, identifying high-risk patients based on ancestry, geographic, and/or genetic lineage, is challenging. Given the absence of a biologic marker and reliance on clinical features to guide diagnostic decision-making, clinicians must carefully consider the KD literature to guide risk assessment and management. In this perspective piece, we address the complexities of incorporating Asian ancestry, which has been associated with increased risk for coronary artery aneurysms (CAA), into a local clinical KD pathway and reflect on the broader implications of considering ancestry in clinical practice guidelines.Epidemiologic studies demonstrate that KD incidence and coronary artery involvement differ across populations. Globally, incidence is highest in northeast Asian countries.1–3 Although a complete understanding of genetic contributions to KD remains elusive, there have been meaningful signals showing specific genetic loci that may influence KD susceptibility and CAA formation.4,5 These loci differ by ancestry and may help explain higher KD incidence in Asian populations.4,5 In North America, patients categorized as Asian or Pacific Islander are at greater risk for developing CAA compared to white patients, even when diagnosed early and treated promptly, suggesting these patients may need treatment intensification, which may include intravenous immunoglobulin combined with medications such as corticosteroids, tumor necrosis factor alpha inhibitors, interleukin inhibitors, and cyclosporine.1,6,7 One study found that children of Asian and Pacific Islander descent with KD had an adjusted odds ratio of 2.4 for developing CAA compared with other populations even after accounting for age, sex, and illness day at diagnosis.7 Studies examining KD incidence and CAA risk by specific Asian country of origin in US populations suggest that global incidence patterns do not cleanly map onto US populations. Specifically, although KD incidence is highest in northeast Asian countries worldwide, US data reveal a more complex pattern with some southeast Asian populations in the US exhibiting higher KD incidence and CAA risk despite lower incidence reported in their countries of origin.3,8–11 Because Asian ancestry is associated with differential risk for coronary complications, incorporating it into risk assessment can influence treatment decisions even for patients who meet diagnostic criteria for KD.Risk scores developed in Japan perform poorly in North American populations, prompting Son et al to develop and validate a risk stratification score to predict which North American patients with KD develop CAA.12 They evaluated demographic, laboratory, and echocardiographic variables and, through multivariable modeling, identified 4 risk factors that together best differentiated patients at high risk for CAA, 1 of which was reported “Asian race.” In their work, “Asian race” was defined as “one or both parents report[ing] Asian descent.”12 This definition aligns more with ancestry, defined as originating from a similar geographic region, from a similar geographic region or ancestral pedigree, or from whom one is biologically descended, than race.13 Both the AHA and ACR guidelines cite this model to identify patients who may benefit from intensification of initial KD treatment to prevent the devastating consequences associated with CAA.1,2Within our institution, the Son et al risk stratification criteria performed well in identifying patients at highest risk for developing CAA. In an unpublished retrospective analysis conducted during KD clinical pathway development, we applied the Son et al criteria to a KD cohort at our institution. The distribution of low-, moderate-, and high-risk patients mirrored that of the original development and validation cohorts in their study, suggesting that the scoring system functioned similarly in our patient population. As the primary goal of our pathway was to identify patients at highest risk for CAA at the time of diagnosis, we next examined how the Son et al criteria classified patients who ultimately developed coronary artery abnormalities. Patients at our institution who developed CAA according to AHA guidelines (n = 12) were classified as moderate or high risk by the Son et al criteria at diagnosis. Importantly, all patients with medium, large, or giant aneurysms (n = 5) were classified as high risk, and all identified as Asian race in the medical record. Taken together, this retrospective analysis demonstrated that the Son et al criteria effectively captured our highest-risk patients and identified a patient population disproportionately affected by CAA, consistent with US epidemiologic studies.However, translating these findings into the development of our local clinical pathway raised an important dilemma: would distinguishing patients with an Asian background during risk stratification perpetuate race-based medicine, or would excluding such information risk missing patients at highest risk of developing CAA? This tension highlights a broader challenge in pediatric medicine.Race is a social construct with no biological basis, whereas ancestry reflects geographic or genetic lineage and can influence disease susceptibility.13 In practice, medicine has often blurred distinctions between race and biological risk, using racial categories in research and clinical decision-making in ways that imply biologic differences.13–16 For example, within pediatrics, the 2004 hyperbilirubinemia guidelines used race as an independent risk factor for hyperbilirubinemia, with east Asian race listed as a risk factor whereas Black race was listed as a protective factor, contributing to delayed recognition and intervention for kernicterus in Black infants.17 These guidelines were revised in 2022 to instead reflect family history or genetic ancestry suggestive of inherited red blood cell disorders as the relevant risk factor for bilirubin neurotoxicity.18 In health research, unclear definitions of race and ancestry can contribute to inconsistent data collection and inaccurate classifications, whereas treating racial categories as biological proxies can lead to misleading results and false assumptions that observed racial differences reflect biological risk.13 Because clinical practice guidelines are built on such research, these errors can influence patient care. A study of pediatric clinical practice guidelines found that race was frequently used in ways that reinforced inequities: normalizing the majority group, conflating race with culture, and perpetuating stereotypes.19 Recognizing these potential harms, professional organizations, including the American Academy of Pediatrics, American Medical Association, and National Academy of Medicine, have advocated for a shift from race-based medicine, which treats race as an inherent biological risk factor, to race-conscious medicine, which recognizes racism rather than race as a modifier of health outcomes.14,20–23In light of these considerations, we incorporated Son et al’s scoring system into our clinical pathway to identify patients at the highest risk of developing CAA but adapted the language to replace “Asian race” with “Asian ancestry.” This decision was informed by multiple converging lines of evidence including global epidemiologic data demonstrating population differences in KD outcomes, Son et al’s finding that Asian ancestry is associated with increased coronary risk among North American patients, and a retrospective review of our institution’s KD patient population that showed a similar pattern.1–3,12 Importantly, the operational definition used by Son et al aligned more closely with Asian ancestry rather than race, supporting our decision to retain the risk stratification scoring while revising the language used in the pathway.Although ancestry is a more biologically meaningful construct in contrast to race, it is not without pitfalls and must be contextualized carefully to avoid reinforcing bias. We therefore incorporated additional safeguards in pathway implementation. Although KD incidence differs across populations, KD remains a clinical diagnosis without a definitive biological marker. Ancestry may contribute to these epidemiologic differences in disease incidence, but evidence supporting its role in diagnostic decision-making is less robust than evidence supporting its role in postdiagnosis risk stratification. Thus, we emphasized that ancestry should only be considered after patients met diagnostic criteria for KD, intentionally separating risk assessment from diagnostic evaluation. Once KD was diagnosed, we encouraged front-line physicians to ask if biological parents were of Asian descent, allowing a more accurate capture of “ancestry” compared to demographic data reported in the electronic health record.24 Although we did not provide a formal script, physicians were counseled to frame this question in its clinical context, explaining that ancestry helps identify patients who may be at higher risk for complications of KD and thus benefit from treatment intensification. Within the pathway, Asian ancestry was operationalized inclusively: physicians were not asked to specify country of origin and patients with either sole or mixed Asian ancestry received one point on the multifactor risk stratification score. We retained this broader operational definition in our clinical pathway, consistent with Son et al’s risk model and recognizing that data on subregional CAA risk remain incomplete, particularly in underresourced regions, and that available evidence is mixed.3,8–12,25 Using more granular ancestry categories could inadvertently imply clinically meaningful distinctions when none are currently established. Importantly, patients identified as high risk do not automatically receive treatment intensification. Instead, pediatric rheumatology is consulted, and treatment decisions are individualized based on the overall clinical picture.Although our experience pertains to KD, the questions we encountered are likely familiar to clinicians caring for children with other conditions that lack well-defined biologic or genetic markers. In these settings, diagnosis and management rely on clinical judgment, risk stratification tools, and population-level data. When evidence suggests population-level differences in risk across groups, often described using demographic variables without clear biological explanations, the question is not simply whether to include or exclude such variables but how they should be justified and applied in practice. Answering that question requires clinicians to understand the difference between race and ancestry, critically appraise the available evidence from both the literature and local data, deliberately justify whether and how this information should inform patient care, and carefully operationalize and evaluate its use over time. In the absence of definitive diagnostic markers, clinical pathways often need to distill heterogeneous clinical and epidemiologic information into actionable guidance that supports real-time decision-making. Simply changing terminology (ie, replacing “race” with “ancestry”) is insufficient to prevent bias because how information is applied in practice matters just as much as the language used. Formal evaluation of pathway implementation and its downstream effects are important next steps. As evidence evolves, and as more granular epidemiologic and genetic data become available, clinicians must continually reassess how demographic information is defined, justified, and operationalized to ensure epidemiologic risk informs rather than undermines equitable, high-quality patient care.