A nomogram to predict ventricular thrombus in dilated cardiomyopathy patients

Area of Science:

• Cardiology
• Medical Diagnostics
• Predictive Modeling

Background:

• Ventricular thrombus (VT) is a significant complication in patients with dilated cardiomyopathy (DCM).
• Accurate prediction of VT is crucial for effective patient management and risk stratification.

Purpose of the Study:

• To develop and validate a nomogram for personalized prediction of incident VT in DCM patients.
• To assess the predictive performance of the developed nomogram against existing risk stratification scores.

Main Methods:

• Retrospective analysis of 1267 DCM patients.
• Development of a nomogram using LASSO and multivariate logistic regression.
• Validation using AUC, calibration curves, and decision curve analysis.

Main Results:

• The nomogram identified age, LVEF, uric acid, NT-proBNP, and D-dimer as independent VT predictors.
• The nomogram demonstrated strong predictive accuracy (C-index=0.92) and good calibration.
• The nomogram outperformed CHA2DS2, CHA2DS2-VASc, and ATRIA scores in VT prediction.

Conclusions:

• A novel nomogram effectively predicts VT in DCM patients.
• The developed nomogram offers improved risk stratification compared to current scores.
• This tool facilitates personalized VT risk assessment in clinical practice.