To develop a nomogram model to predict long-term risk of coronary artery lesions (CALs) one year after diagnosis in patients with Kawasaki disease (KD).
Key Findings:
Significant predictors of long-term CAL risk included male sex, prolonged hospitalization, prolonged fever, decreased hemoglobin (Hb), decreased hematocrit (HCT), and hyponatremia.
The nomogram achieved an AUC of 0.801 in the training dataset and 0.796 in the validation dataset.
Sensitivity and specificity were 82.5% and 65.5% in the training dataset, and 66.7% and 83.5% in the validation dataset, respectively.
The calibration curve was aligned with the predicted curve, indicating good model calibration.
Decision curve analysis showed a high net benefit of the model.
Interpretation:
The nomogram prediction model demonstrated high accuracy in identifying KD patients at risk for long-term CALs.
Limitations:
Conclusion:
The nomogram can assist physicians in identifying KD patients who may develop long-term CALs.
