Nomogram for Predicting Regression of Persistent Coronary Artery Aneurysms in Kawasaki Disease: A Three-year Follow-up Cohort Study in Southwest China

Objective: Coronary artery aneurysm (CAA) is a potentially life-threatening cardiovascular complications in Kawasaki disease (KD). This study aimed to construct and then validate a prognostic model to assess the likelihood of persistent CAA regression in KD. Methods: Using Cox regression analysis, we constructed a prognostic nomogram model to estimate the probability of CAA regression at one-year, two-year, and three-year intervals in KD patients with CAA, followed by temporal external validation. The model's discriminative ability, calibration, and potential clinical benefit were evaluated by the area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis. Results: A total of 658 KD with CAA were finally included. From January 2012 to July 2021, 557 participants were enrolled and subsequently allocated into a training set (n = 390) and a testing set (n = 167) using a 7:3 random split. The temporal external validation cohort consisted of 101 cases diagnosed as KD with CAA between August, 2021 and July, 2022. Four variables were retained in the final prognostic model: age at diagnosis (HR = 0.994, 95% CI 0.990-0.998); CAA enlargement category (reference: SCAA; MCAA: HR = 0.398, 95% CI 0.299-0.530; GCAA: HR = 0.058, 95% CI 0.027-0.125); platelet count (HR = 0.999, 95% CI 0.998-1.000); and fibrinogen level (HR = 1.143, 95% CI 1.042-1.255). The model demonstrated robust discriminative performance, with AUCs of 0.854, 0.919, and 0.910 at one, two, and three years in the training set; 0.844, 0.910, and 0.913 in the testing set; and 0.763, 0.886, and 0.986 in the external validation cohort. Conclusion: We established a prognostic model capable of estimating the long-term outcomes in KD, and confirmed its favorable performance in terms of discrimination, calibration, and potential value for clinical application.