小佐見 光樹

IMPORTANCE: Global studies have reported that the incidence of Kawasaki disease (KD) declined during the COVID-19 pandemic. These studies suggest that the global pandemic and its accompanying mitigation measures may provide an important opportunity to explore the hypothesis of a KD pathogenesis. OBJECTIVE: To compare changes in KD incidence in Japan before and after the start of the COVID-19 pandemic. DESIGN, SETTING, AND PARTICIPANTS: This cohort study was conducted using the data set from Japan's 26th nationwide KD survey that obtained information on patients who were diagnosed with KD in Japan from January 1, 2019, through December 31, 2020. MAIN OUTCOMES AND MEASURES: Kawasaki disease incidence rates were calculated by referring to the national population data in the vital statistics data for Japan. RESULTS: A total of 28 520 patients were identified (16 236 male individuals [56.9%]; median [IQR] age, 26 [14-44] months). A total of 17 347 patients were diagnosed with KD in 2019 and 11 173 were diagnosed in 2020, representing a 35.6% reduction in the number of patients diagnosed in 2020 compared with the previous year. Patient distributions for days of illness at the first hospital visit were almost identical in 2019 and 2020, suggesting that the decrease in KD incidence likely was not associated with pandemic-related delays in seeking treatment. The proportion of patients diagnosed with KD who were younger than 12 months was significantly larger in 2020 than in 2019 (21.6% vs 19.4%; P < .001). Compared with KD incidence among younger patients, the incidence among those 24 months and older declined rapidly after initiation of COVID-19 special mitigation measures, with a greater percentage reduction (58.3% reduction in July), but rebounded faster after the end of the special mitigation period. By contrast, the incidence among patients younger than 12 months declined moderately after the initiation of the special mitigation period, with a lower percentage reduction (40.3% reduction in October), and rebounded at a later phase. CONCLUSIONS AND RELEVANCE: In this cohort study, the number of patients diagnosed with KD decreased by approximately one-third across Japan in 2020, with no indication that parents avoided a hospital visit. Differences in KD incidence reduction patterns before and after the initiation of COVID-19 pandemic mitigation measures were found in patients with KD aged younger than 12 months compared with those 24 months or older, suggesting a potential KD pathogenesis involving transmission among children.

No studies have assessed differences between the Japanese and Z score criteria in the echocardiographic detection sensitivity of coronary artery (CA) abnormalities using large-scale data containing samples from multiple facilities engaged in daily clinical practices of Kawasaki disease (KD). We analyzed data from the 25th Japanese nationwide KD survey, which identified 30,415 patients from 1357 hospitals throughout Japan during 2017-2018. Hospitals were classified according to their use of Z score criteria. We assessed differences in hospital and patient background factors and compared the prevalence of CA abnormalities among groups using the Z score criteria. Multivariable logistic regression analyses were performed to evaluate differences in the detection sensitivity for CA abnormalities. The Z score criteria were more likely to be utilized in larger hospitals with more pediatricians and cardiologists. Even after controlling for potential confounders, detection sensitivities by the Z score criteria were significantly higher than by the Japanese criteria in patients with CA dilatations (adjusted odds ratio (95% confidence interval) 1.77 (1.56-2.01)) and aneurysms (1.62 (1.17-2.24)). No significant difference was found in patients with giant CA aneurysms. Compared with the Japanese criteria, the Z score criteria were significantly more sensitive for detecting patients with CA dilatations regardless of age, and for those with CA aneurysms only in patients aged ≤ 1 year. Our results indicate that differences in the detection sensitivity for CA abnormalities between the Z score and the Japanese criteria were dependent on the CA size and patient age.