Kawasaki Disease and Respiratory Viruses: EcologicalSpatiotemporal Analysis

Background: Kawasaki disease is an uncommon vasculitis affecting young children. Its etiology is not completely under-stood, although infections have been frequently postulated as the triggers. Respiratory viruses, specifically, have often beenimplicated as causative agents for Kawasaki disease presentations. Objective: We aimed to conduct an ecological spatiotemporal analysis to determine whether Kawasaki disease incidence wasrelated to community respiratory virus circulation in a shared region and population, and to describe viral associations beforeand during the COVID-19 pandemic. Methods: We obtained independent statewide data sets of hospital admissions of Kawasaki disease and respiratory multiplexpolymerase chain reaction tests performed at two large hospital networks in Victoria, Australia, from July 2011 to November2021. We studied spatiotemporal relationships by negative binomial regression analysis of the monthly incidence of Kawasakidisease and the rate of positive respiratory polymerase chain reaction tests in different regions of Victoria. Peak viral seasons(95th percentile incidence) were compared to median viral circulation (50th percentile incidence) to calculate peak seasonincreased rate ratios. Results: While no seasonal trend in Kawasaki disease incidence was identified throughout the study period, we found a1.52 (99% CI 1.27-1.82) and a 1.43 (99% CI 1.17-1.73) increased rate ratio of Kawasaki disease presentations in associationwith human metapneumovirus and respiratory syncytial virus circulation, respectively, before the COVID-19 pandemic. Norespiratory viral associations with Kawasaki disease were observed during the COVID-19 pandemic. Conclusions: Our large ecological analysis demonstrates novel spatiotemporal relationships between human metapneumo-virus and respiratory syncytial virus circulation with Kawasaki disease. The disappearance of these associations in theCOVID-19 pandemic may reflect the reduced circulation of non-SARS-CoV-2 viruses during this period, supporting theprepandemic associations identified in this study. The roles of human metapneumovirus and respiratory syncytial virus inKawasaki disease etiology warrant further investigation