Skip to main content
IKDS 2024 Main banner
Icon Legend
Back

Clinical Management

Young Investigators Oral Abstract Competition

82 - PREDICTING CORONARY ARTERY ABNORMALITIES DEVELOPMENT USING CORONARY Z SCORE AND CLINICAL PARAMETERS BEFORE THE TREATMENT FOR KAWASAKI DISEASE ¦ A MULTICENTER RETROSPECTIVE STUDY IN WAKAYAMA, JAPAN

Tuesday, August 27, 2024
9:55 AM – 10:05 AM East Coast USA Time
Location: Montreal
Disclosure(s):
Takayuki Suzuki, M.D., Ph.D.: No financial relationships to disclose

Background: Predictive scores for IVIG resistance are used in the selection of initial intensive therapy for patients with KD. However, there are currently no indicators to predict the development of coronary artery aneurysms (CAA) before IVIG treatment. We reported in a single-center study that a pre-treatment coronary artery Z-score ≧ 1.43 is associated with the risk of developing CAA. We aimed to construct a predictive score for the development of CAA using indicators including the pre-treatment coronary Z-score.

Methods: We retrospectively reviewed 1,230 patients with KD who developed KD between April 2009 and March 2020 and were treated at 8 hospitals in the Clinical Research Network of the Wakayama Kawasaki Disease Study Group using the same protocol as our hospital (1st line: IVIG, 2nd line: IVIG and 3rd line: ciclosporin A).
The medical records of all enrolled patients were retrospectively reviewed, including demographics, echocardiographic findings before treatment, and laboratory tests on admission. Echocardiography was performed before treatment (Pre) and around 4 weeks after onset (Post), and the maximum Z-score of the three coronary branches excluding the LMCA was defined as Zmax, respectively. CAA was defined as Post-Zmax≧2.5, and risk factors for developing CAA were statistically examined. A multivariate analysis was performed to construct a predictive model for CAA development. The sensitivity and specificity of the model were confirmed in the validation cohort (VC1: Chiba, VC2: Wakayama).

Results: CAA developed in 57 patients (5.8%). We identified independent three risk factors associated with the development of CAA and the cutoff value for them determined from the ROC curve: Pre-Zmax≧1.6 (2 points), albumin≦3.1g/dL (1 point), and months of age≦12 (1 point). Patients with CAA risk score ≧2 are classified into a high-risk group for developing CAA (sensitivity 84.2%, specificity 60.8%). In VC1, sensitivity: 83.3%, specificity: 85.9%. In VC2, sensitivity: 72.7%, specificity: 89.6%.

Conclusion: Models including Pre-Zmax may be useful in predicting the development of CAA due to KD. Our results suggest that initial intensive treatment might be considered based on not only the risk scores of IVIG resistance but also CAA risk scores.