The diagnosis of coronary artery abnormalities (CAA), including dilation and aneurysm, in patients with Kawasaki disease is paramount to treatment planning. CAA are defined using various standards, which makes diagnosis difficult. The aims of this study were to determine the variability of CAA prevalence according to existing guidelines and Z score formulas and to examine the discrepancies in widely used Z score formulas.

Using data from a Korean national survey on Kawasaki disease, 6,889 patients were included and analyzed. The overall prevalence of CAA and the prevalence for subgroups were compared on the basis of aneurysm severity, age, and body surface area. Finally, discrepancies among five Z score formulas were evaluated by comparing two of the formulas in pairs.

According to the Japanese criteria, the prevalence of CAA was 18%. According to the American Heart Association criteria, the prevalence of dilation or aneurysm was about 21% to 42%, and that of aneurysm of the left anterior descending artery or right coronary artery was about 8% to 27%. The prevalence of CAA and that of left anterior descending or right coronary artery aneurysm was significantly different, with discrepancies between the Japanese and AHA Z score criteria, as well as among the five Z score formulas. Additionally, misclassification of aneurysm severity was observed for each criterion or Z score formula. There was significant variation among calculated Z scores. The more extreme the Z score values, the more discrepancy was observed.

Different guidelines and Z score formulas yield significantly different prevalence rates and classifications of CAA. In addition, more discrepancies were observed with higher Z score values. As CAA or aneurysm severity could be changed by guidelines or Z score formulas, they should be chosen carefully, and when a particular formula is chosen, consistency is needed.