Background: Coronary artery (CA) z-scores are commonly used for clinical decisions in Kawasaki disease (KD). We evaluated reliability in CA measurement, reproducibility of z-score calculation, and frequency with which different z-score formulas lead to divergent management strategies.
Methods: We randomly selected 21 KD patients (pts) with ≥1 CA z-score 1.5-3 and all KD pts with ≥ 1 CA z-score 7-14 (n=20). Two echocardiographers measured LMCA, LAD and RCA. Inter- and intraobserver reliability were calculated. T-tests were used to compare CA z-score using 3 commonly used formulas (Boston, DC and Montreal).
Results: Median age at KD echo was 1.2 y (0.2-11.5 y). Interobserver reliability was high for LAD (intraclass correlation [ICC] 0.970) and RCA (ICC 0.943) and lower for LMCA (ICC 0.725). Intraobserver reliability was also high for LAD and RCA (ICC 0.991 and 0.999) and lower for LMCA (ICC 0.946). Z-scores for the 3 formulas were similar at smaller CA size, i.e., z < 3, but varied markedly at larger CA dimensions (Figure). Z-scores for the same CA dimension calculated by each of the 3 formulas resulted in disparate classification of normal vs. mild dilation in 7/21 (22%) pts, and different guidance for anticoagulation based on CA z ≥10 in 10/20 (50%) pts.
Conclusion: Although CA measurements have high inter- and intraobserver agreement, CA z-scores vary dramatically based on the z-score formula, particularly at larger CA dimensions. Discrepancies in CA z-score between calculators impacts not only the distinction between normal and mild dilatation, but most importantly, the recommendation of anticoagulation for pts with larger CA dimensions.