10 Years of Kawasaki Disease in the Pre-COVID-19 Era: Predictive Value Of Different Scores in A Non-Japanese Population

Background: Kawasaki disease (KD) is one of the most common vasculitis in childhood and may cause serious complications, such as coronary artery aneurisms (CAA). Prompt treatment with Intravenous Immunoglobulin (IVIG) is critical to reduce cardiovascular sequelae. More recently, corticosteroids appear to be beneficial in reducing CAA in selected patients. Currently available risk prediction models for IVIG resistance are insufficiently accurate in non-Japanese children.

Objective: To evaluate in an European population the predictive value of Kobayashi Score and of a recently proposed French score for non-Asian populations (criteria used: hepatomegaly, ALT≥30 IU/L, lymphocyte<2400/mm3 and time to treatment

Methods: Retrospective study of KD in a Portuguese tertiary hospital between 2010-2019 (pre-COVID-19 era). Statistical analysis considering a significance level of 5%.

Results: 41 children included, all European Caucasian, 4.1 cases/year: 58.5% male, 76.6% under 5 years of age. Erythematous rash present in 92.7%, oral changes: 85.4%, conjunctivitis: 70.7%, extremity changes: 70.7% and cervical lymphadenopathy: 41.5%. Diagnosis of incomplete KD in 41.5% (29.4% versus 50% before and after 2015). 19.5% presented coronary alterations. 4 cases of KD shock and 2 cases of macrophage activation syndrome. 97.1% received IVIG, resistant in 22.5%. Longer fever duration and higher C-reactive Protein associated with IVIG resistance (p<0.05). Kobayashi score presented a sensitivity of 33.3% and the recently proposed score a sensitivity of 66.7%. All patient received aspirin and 23.5% corticosteroids: 80% as first-line treatment, all after 2015. 12.5% with cardiovascular changes at discharge and less than 5% long-term, with 0% mortality in the follow-up period.

Conclusion: IVIG resistance is a decisive factor in AAC risk. The recently proposed French score presented in our population a better sensibility than Kobayashi score, although lower than described in the original cohort (66.7% versus 77%). Since 2015, there seems to be a trend to diagnosing incomplete KD and use corticosteroids, likely due to increased awareness.