Introduction
Kawasaki Disease is the leading cause of acquired heart disease in children in developed countries. Reference McCrindle, Rowley and Newburger1 While the aetiology of Kawasaki Disease remains unknown, a seasonal pattern has been well described in the northern hemisphere. Reference Burns, Herzog and Fabri2,Reference Low, McCrindle and Mueller3 The incidence of Kawasaki Disease in North America has been shown to peak in the winter months (December through April) with a trough in late summer/early fall. Reference Burns, Herzog and Fabri2,Reference Low, McCrindle and Mueller3 A seasonal environmental or infectious trigger is suspected as the cause for this pattern. Reference Rowley and Shulman4 Seasonality has been reported to be greatest for the age group with the highest incidence, those between 0–4 years of age Reference Low, McCrindle and Mueller3 , possibly reflecting a developmental susceptibility to an immunological response modulated or triggered by seasonal factors. Reference Rowley and Shulman4,Reference Manlhiot, Mueller and O’Shea5 In North America where this study was performed factors associated with an increased risk of Kawasaki Disease have included East Asian ethnicity, age < 4 years, higher pollution (nitrogen dioxide and sulphur dioxide), as well as higher rates for various types of infections. Reference Low, McCrindle and Mueller3,Reference Manlhiot, Mueller and O’Shea5 While a decrease in the incidence of Kawasaki Disease during the COVID-19 pandemic has been reported in North America Reference Sakai-Bizmark, Estevez and Wu6 , Taiwan Reference Yang and Kuo7 , Japan Reference Ae, Makino and Kuwabara8 , and Korea Reference Kang, Kim and Huh9,Reference Kim, Choi and Kim10 the impact of the pandemic on seasonality of Kawasaki Disease has not been investigated. Multisystem Inflammatory Syndrome-Children is a rare condition associated with SAR-CoV-2 and is a delayed hyperinflammatory condition occurring 2–6 weeks after an infection. Multisystem Inflammatory Syndrome-Children is characterised by multisystem involvement, resulting in organ dysfunction and shock. For Kawasaki Disease and Multisystem Inflammatory Syndrome-Children there is considerable overlap in presentation, clinical features, management and early outcomes, with the primary differentiating factor being evidence of preceding SARS-CoV-2 infection or exposure. Reference Harahsheh, Shah and Dallaire11 We sought to determine the seasonal patterns of both Kawasaki Disease and Multisystem Inflammatory Syndrome-Children during and immediately following the COVID-19 pandemic.
Materials and methods
This study was performed through the International Kawasaki Disease Registry. Enrolled patients from each centre included all patients with a site diagnosis of either Kawasaki Disease or Multisystem Inflammatory Syndrome-Children. Each site obtained local ethics board approval, and enrollment and consent procedures were as locally approved, with many sites obtaining a waiver of consent. Data from patients were submitted into a centralised REDCap database, maintained by the data coordinating centre which operated under local ethics board approval and with data sharing agreements in place with each submitting site. The data coordinating centre confirmed the site diagnosis based on submitted information. American Heart Association guidelines were utilised for Kawasaki Disease Reference McCrindle, Rowley and Newburger1 and for Multisystem Inflammatory Syndrome-Children, the 2020 Center for Disease Control guidelines with positive COVID-19 association were used for case verification. 12 As all sites were paediatric centres, adult patients more than 20 years of age were not included. Each diagnosis group was divided into either a confirmed or unconfirmed category based on the above criteria. Unconfirmed Kawasaki Disease was any patient for whom the site diagnosis could not be confirmed by the data coordinating centre (either because they had insufficient criteria or the submitted data were incomplete to confirm the diagnosis), or the patient had evidence of COVID-19 infection or exposure (hence, would overlap with Multisystem Inflammatory Syndrome-Children). Unconfirmed Multisystem Inflammatory Syndrome-Children was anyone for whom the site diagnosis could not be confirmed by the data coordinating centre, or they lacked evidence of a prior COVID-19 infection. Complete and uninterrupted data from prior to January 1, 2020 were not available. The number of patients for each diagnosis group per month was plotted over time. The number of patients/month for confirmed and unconfirmed Kawasaki Disease was tabulated for the anticipated peak Kawasaki Disease season (December–April) and the anticipated non-peak season (May–November). We have taken this to reflect the prevalence and not incidence, given that the denominator was not known and assumed to be the catchment area for each site.
Results
Data were included for 1975 patients from 11 large North American sites which had both verified complete data and uninterrupted site reporting (Toronto, Canada, 364 patients; Washington, DC, 261 patients; Kansas City, MO, 224 patients; Columbus, OH, 197 patients; Omaha, NB, 173 patients; Montreal, Canada, 149 patients; Seattle, WA, 143 patients; Phoenix, AZ, 134 patients; Valhalla, NY, 134 patients; Edmonton, Canada, 121 patients; Palo Alto, CA, 75 patients). For Kawasaki Disease, 531 patients met the diagnosis criteria with no evidence of COVID-19 infection or exposure, and there were an additional 389 patients for whom data were insufficient or discrepant to confirm the site diagnosis. For Multisystem Inflammatory Syndrome-Children, the diagnosis was confirmed for 907, and additionally, there were 148 patients for whom the data were insufficient or discrepant to confirm the site diagnosis. Demographic characteristics for confirmed and unconfirmed patients in both diagnosis groups are shown in Table 1.
Demographiccharacteristics by diagnosis group
Data are presented as frequencies (%)or median with values at the 25th and 75th percentiles (quartiles).
MIS-C, multisystem inflammatory syndrome-children.
Enrollment was from January 2020 to July 2023. Multisystem Inflammatory Syndrome-Children cases were first reported in North America in April. 2020 (Figure 1). There were peaks in January 2021 (96 patients) and January 2022 (87 patients) for confirmed Multisystem Inflammatory Syndrome-Children (mean monthly patients 21 for the entire time period) and much smaller associated peaks for unconfirmed Multisystem Inflammatory Syndrome-Children (peaks; 2020 = 29, 2021 = 9, 2022 = 7). For Kawasaki Disease, 2020 began (January–March) with a seasonal peak in Kawasaki Disease patients/month (peak 26, mean 21). However, there was a decrease in the number of Kawasaki Disease patients/month over the remaining time period (mean 11). Of note, however, there was a peak in unconfirmed Kawasaki Disease patients that corresponded with the first smaller peak of Multisystem Inflammatory Syndrome-Children cases in May 2020 suggesting some possible diagnostic confusion during this time period partially due to an initial lack of confirmatory testing for COVID-19. After the onset of the pandemic (April 2020) there was no clear seasonal variation for confirmed Kawasaki Disease (December–April mean 12 patients/month and May–November mean 10 patients/month), with a similar trend for unconfirmed Kawasaki Disease (December–April mean 10 patients/month and May–November mean 8 patients/month). However, long after the peaks in Multisystem Inflammatory Syndrome-Children, there appeared to be a later but smaller-than-expected peak of confirmed Kawasaki Disease in April 2023 (21 patients), followed by a subsequent decline in monthly patients.
Monthly number of incident Kawasaki Disease and Multisystem Inflammatory Syndrome-Children patients. KD, Kawasaki disease; MIS-C, multisystem inflammatory syndrome-children.
Discussion
Our results show that during the pandemic in North America, the incidence of Kawasaki Disease decreased, and the usual seasonality was abolished. Early during the pandemic when no COVID-19 testing was available, there may have been overlap of Kawasaki Disease and Multisystem Inflammatory Syndrome-Children diagnoses. However, given the application and independent verification of the American Heart Association case definition and 2020 Center for Disease Control guidelines, the results are unlikely to be due to misclassification. Speculation regarding what might have affected the incidence and seasonal variability includes those factors most likely altered by the pandemic safety measures put in place (social distancing, staying home, reduction in travel, mask use, improved hygiene practices, etc.). Of the previously listed North American factors associated with increased risk of Kawasaki Disease, the primary factor affected by the pandemic safety measures is likely infectious disease exposure. In North America, numerous reports showed a decrease in the incidence of respiratory and gastrointestinal illnesses during the pandemic. At the onset of the pandemic in the United States the Center for Disease Control reported a 61% decrease in the number of clinical laboratory influenza specimens submitted pre-pandemic vs. during the pandemic and a 98% decrease in influenza as measured by the percentage of submitted specimens testing positive. Reference Olsen, Azziz-Baumgartner and Budd13 A multicentre study in 2021 demonstrated 64% lower odds of testing positive for respiratory syncytial virus and 46% lower for influenza. Reference Haddadin, Schuster and Spieker14 New York City in 2022 documented a 70–90% reduction in the incidence of numerous respiratory viruses. Reference Yuan, Yeung and Yang15 A study in Massachusetts demonstrated an 88% decrease in the daily diagnosis rate for gastroenteritis Reference Hatoun, Correa, Donahue and Vernacchio16 and another in Colorado described a 52% decrease in outpatient acute gastroenteritis visits. Reference Armistead, Tran, White, Wilson and Scallan Walker17 An exception to this was infection with respiratory syncytial virus, which showed a seasonal peak between July and November for 2021 and 2022 18 and an increase in hospitalisations in Canada. Reference Bourdeau, Vadlamudi and Bastien19 This decrease in population exposure to infectious diseases may have resulted in a reduction in the incidence of Kawasaki Disease as well as its seasonality. Reluctance of Kawasaki Disease patients to seek hospital care during the pandemic, rather than the seasonality of Kawasaki Disease-related pathogen epidemics may be argued as a reason for the decrease in Kawasaki Disease incidence. While this study was not able to compare the number of days prior to diagnosis/hospitalisation pre- and post-pandemic there are a number of studies in North America, Reference Burney, Roberts and Dehaan20 Denmark, Reference Ünlü, Holm, Krusenstjerna-Hafstrøm, Glerup, Bjerre and Herlin21 and Japan Reference Katsumata, Harama and Toda22 that have shown there was no difference in the number of days to presentation before and during the pandemic. An additional factor that may have been affected by the pandemic safety measures is pollution. A global study of pollution during the pandemic described a 20–30% reduction in NO2 emissions, with 30% for the United States. Reference Muhammad, Long and Salman23 While it is true that there was a decrease in pollution during the pandemic, historically, one of the most common factors speculated to be involved is preceding viral infections, and this is the most likely contributing factor. Reference Rowley and Shulman4 A further consideration may be that the decline in the number of Kawasaki Disease cases may have also been influenced by health care behaviour and health care system factors during the pandemic. Families may have been reluctant to present to sites, and accessibility to care may have been limited. This would lead to the assumption that an unknown number of patients with Kawasaki Disease would have the diagnosis missed. This is unlikely, given that Kawasaki Disease is an acute and serious condition, and also that there was likely heightened awareness of Kawasaki Disease-like illness during the pandemic given the emergence of Multisystem Inflammatory Syndrome-Children.
Just prior to the onset of the pandemic, the percentage of unconfirmed Kawasaki Disease (compared to total Kawasaki Disease cases) was lower than during the pandemic (23% vs 45% respectively). COVID-19 is speculated to result in immune system hyperactivity acting as a trigger for various hyperinflammatory diseases, potentially including Kawasaki Disease. Reference Zebardast, Hasanzadeh, Ebrahimian Shiadeh, Tourani and Yahyapour24 A portion of the unconfirmed Kawasaki Disease (who may have had exposure to COVID-19 with or without documentation) and the unconfirmed Multisystem Inflammatory Syndrome-Children patients may have been patients who experienced a hyperinflammatory reaction that did not fully manifest as a confirmed case for either condition. The number of Kawasaki Disease patients/month (both confirmed and unconfirmed) after the onset of the pandemic remained low, reflecting the reduced Kawasaki Disease incidence during the pandemic. Additionally, the trends for both the confirmed and unconfirmed Kawasaki Disease patients showed no seasonal variability. While some unconfirmed Kawasaki Disease patients may not have been Kawasaki Disease, if even a small percentage were true Kawasaki Disease seasonality still would have been seen, and this was not the case.
The primary study limitation is that data from prior to January 2020 were not available, as well as more recent data after the end of the pandemic. Therefore, comparisons were made to known historical trends and to the available months prior to the onset of the pandemic.
During both 2020 and 2021, the peak of confirmed Multisystem Inflammatory Syndrome-Children was seen during January which corresponds to the anticipated seasonal peak of Kawasaki Disease. This suggests that a COVID-19/Multisystem Inflammatory Syndrome-Children surge somehow replaced the Kawasaki Disease peak. However, debate continues regarding whether Multisystem Inflammatory Syndrome-Children is actually a specific form of Kawasaki Disease. Overall, these data show that during the pandemic in North America, the incidence of Kawasaki Disease decreased, and the usual seasonality was abolished. Further detailed epidemiologic data are needed to verify this observation, including ongoing surveillance to determine if seasonality returns.
Financial support
None.
Competing interests
The authors declare none.
Open access
