RESUMO
BACKGROUND: Accurate classification of coronary artery abnormalities (CAAs) is essential for clinical decision-making and long-term management in Kawasaki disease (KD) patients. To date, there are several echocardiographic criteria of CAA assessment. MATERIALS AND METHODS: The Japanese Ministry of Health (JMH) criteria and the Z-score criteria from 2004 American Heart Association guidelines were adopted and their detective efficacies for CAAs were compared in 251 Chinese patients with KD Z scores were calculated by 6 published methods. RESULTS: According to the JMH criteria, 19 (7.57%) KD patients were classified as CAAs during the acute KD episode. However, the detective number of CAAs was highest and had a 0.68-fold increase by the Dallaire et al method with a Z-score cut point of ≥2.5 as compared with the JMH criteria; in contrast, more than 78.95% of patients with CAAs identified by the JMH criteria had a coronary artery Z score ≥2.5. All 6 different Z-score methods had satisfactory accuracies with a range from 93.23% to 97.61% in screening CAAs. For the 19 patients with CAAs identified by the JMH criteria, their Z scores presented the widest variation calculated by the McCrindle et al method. CONCLUSIONS: The JMH criteria underestimate the prevalence of CAAs as compared with the Z-score criteria. Quantitative assessment of coronary artery luminal dimensions, normalized as Z scores adjusted for body surface, should be recommended. The larger coronary artery luminal dimensions vary, the more heterogeneous Z scores calculated by different methods have.