117 - CORRELATION BETWEEN KAWASAKI DISEASE AND COMMUNITY VIRAL SURVEILLANCE IN TAIWAN

BACKGROUND

Viral infections have been suggested as a trigger for KD, although specific etiologies remain unknown. The usual patterns of common respiratory and enteroviruses circulation were disrupted during COVID-19 pandemic because of non-pharmacological interventions. We examine the trends of KD and potentially attributable viral infections during COVID-19 pandemic, and explore their correlation.

METHODS

During January 2019–December 2021, KD patients admitted to 6 tertiary and 7 secondary hospitals located in all regions of Taiwan were included. Monthly KD admission number and the patients’ demographics were analyzed. Community viral epidemic trends from the national viral laboratory surveillance conducted by the Taiwan Centers for Disease Control tested throat swab specimens collected from patients aged 0–120 years with respiratory or enterovirus infections for adenovirus, coxsackievirus A, coxsackievirus B, cytomegalovirus, enterovirus A71, enterovirus D68, non-polio enteroviruses, human metapneumovirus, influenza A, influenza B, parainfluenza, and respiratory syncytial viruses. Specimens tested positive for SARS-CoV-2 were excluded. We calculated the Pearson correlation coefficient between the monthly number of KD admissions and the number of each viral isolates with 0-, 1-, 2-, or 3-month lag.

RESULTS

During 2019–2021, a total of 459 KD patients were identified; 283 (62%) were male. Median age at diagnosis was 1.5 years (range 0.7–13.9); 119 (26%) were aged < 1 year. The number of KD admissions decreased to 107 (61%) in 2021, compared with 174 and 178 in 2019 and 2020, respectively. No obvious geographic aggregations were noted. The annual number of specimens for viral surveillance remained around 11,000, but the isolation rate for viruses decreased from 44% in 2019 to 14% and 9% in 2020 and 2021, respectively. 43% of specimens were collected from patients aged < 6 years. The monthly numbers of adenovirus and parainfluenza virus isolates were significantly correlated with the monthly number of KD admissions during the same month (adenovirus: r = 0.44, p < 0.01; parainfluenza virus: r = 0.40, p = 0.01). Influenza A virus isolation was significantly correlated with KD admissions 2 to 3 months later (2 months: r = 0.42, p = 0.01; 3 months: r = 0.37, p = 0.03).

CONCLUSIONS
With the use of non-pharmacological intervention, both the number of KD admissions and the positive rate of isolation from specimens taken for viral surveillance decreased during COVID-19 pandemic. Trends of adenovirus, parainfluenza virus, and influenza A virus isolation were significantly correlated with the number of KD during the same month or 2 months later.