‌Coronary inflammatory imaging and clinical factors: associations with coronary artery aneurysm regression in pediatric Kawasaki disease

The factors influencing the regression of coronary artery aneurysms (CAAs) in Kawasaki disease (KD) remain incompletely understood. Recent studies suggest a potential role for coronary inflammation and fat attenuation index (FAI) of perivascular adipose tissue attenuation (PCAT) derived from coronary computed tomography angiography (CCTA). This study aimed to investigate the differences in clinical, laboratory, and imaging characteristics, with a focus on pericoronary adipose tissue FAI, between patients with CAAs regression and non-regression. In this retrospective observational study, children diagnosed with Kawasaki disease accompanied by coronary artery involvement were divided into a Regression group and a Non-Regression group based on the outcome of coronary aneurysm progression. Baseline demographic, clinical, and laboratory data were collected and were comparable between groups. Key imaging parameters assessed included the maximum CAA diameter, Z-score, CAA size category (small, medium, giant), and the pericoronary fat attenuation index (FAI) measured by coronary computed tomography angiography (CCTA). Multivariable logistic regression was used to identify independent predictors of CAA regression. This retrospective observational study analyzed 40 KD patients with documented CAAs, divided into a Regression group (R group, n = 20) and a Non-Regression group (NR group, n = 20). Proportion of white blood cell count (WBC) ≥ 30 × 10⁹/L in the regression group (10.0% vs. 60.0%, p < 0.001), pericoronary CT fat attenuation index (-71.9 ± 4.5 HU vs.-66.4 ± 3.7 HU, p < 0.001), aneurysm maximum diameter (4.28 ± 2.15 mm vs. 6.86 ± 2.20 mm, p < 0.001), and Z-score (4.6 ± 2.7 vs. 8.6 ± 4.0, p < 0.001) were significantly lower than those in the non-regressive group. In addition, the distribution of coronary aneurysm size grades was significantly different between the two groups (p < 0.001), with medium (55.0%) and giant aneurysms (30.0%) predominating in the non-regression group and small aneurysms (80.0%) predominating in the regression group. Multivariate logistic regression analysis identified white blood cell (WBC) count ≥ 30 × 10⁹/L (OR 0.02, 95% CI 0.00–0.47, p = 0.016), higher pericoronary computed tomography fat attenuation index (CT FAI) (OR 0.62, 95% CI 0.43–0.89, p = 0.009), and higher Z‑values (OR 0.76, 95% CI 0.58–1.00, p = 0.049) as adverse factors for coronary artery aneurysm (CAA) regression. Based on the Youden index derived from the receiver operating characteristic curve, the optimal cutoff values for predicting CAA regression were − 68.7 HU for the mean FAI and 4.6 for the Z‑score. In KD patients with CAAs, smaller initial aneurysm Z-score, the absence of extreme leukocytosis (WBC ≥ 30 × 10⁹/L), and lower fat attenuation index (FAI) are associated with a higher likelihood of CAA regression. These findings suggest that both the initial morphological severity of the aneurysm and the state of pericoronary inflammation, as potentially reflected by FAI, are critical determinants of vascular remodeling outcomes.