A prediction model for left ventricular thrombus persistence/recurrence: based on a prospective study and a...

Area of Science:

• Cardiology
• Medical Prediction Modeling

Background:

• Optimal anticoagulation duration for persistent left ventricular (LV) thrombus remains unclear.
• Identifying patients with unresolved thrombus is crucial for treatment decisions.

Purpose of the Study:

• To develop and validate a prediction model for forecasting thrombus persistence or recurrence in patients with LV thrombus.
• To identify clinical characteristics associated with unresolved LV thrombus.

Main Methods:

• Prospective and retrospective cohorts (2013-2022) combined for model derivation.
• Lasso and logistic regression used to select independent predictors.
• A nomogram risk prediction model was developed and applied for risk stratification.

Main Results:

• Six predictors were identified for the logistic regression model.
• Protuberant thrombus and higher D-dimer levels were associated with reduced persistence/recurrence risk.
• Thicker thrombus was linked to an increased rate of persistent thrombus.

Conclusions:

• The developed prediction model can forecast persistent/recurrent LV thrombus.
• The model identifies characteristics associated with unresolved thrombus.
• Prospective randomized trials are needed to validate the model and guide anticoagulation duration.