Validation and Comparison of PROMISE and CONFIRM Model to Predict High-Risk Coronary Artery Disease in Symptomatic...

Area of Science:

• Cardiology
• Medical Imaging
• Diabetes Mellitus Research

Background:

• Accurate identification of high-risk coronary artery disease (HRCAD) is crucial for patients with diabetes mellitus (DM).
• Existing models for predicting HRCAD in DM patients require further validation and comparison.
• This study addresses the need to evaluate and compare the reliability of the CONFIRM high-risk model (CHM) and PROMISE high-risk model (PHM) in DM patients.

Purpose of the Study:

• To validate and compare the predictive performance of the CHM and PHM in symptomatic DM patients.
• To assess the accuracy of both models in identifying HRCAD using coronary computed tomographic angiography (CCTA).
• To determine which model, CHM or PHM, provides a more reliable prediction of HRCAD in this population.

Main Methods:

• A cohort of 5936 symptomatic DM patients who underwent CCTA was analyzed.
• HRCAD probability was estimated for each patient using both CHM and PHM.
• Model performance was evaluated using Area Under the receiver operating characteristic Curve (AUC), Integrated Discrimination Improvement (IDI), Net Reclassification Improvement (NRI), and the Hosmer-Lemeshow (H-L) test.

Main Results:

• The overall prevalence of HRCAD was 8% (470 patients).
• No significant difference was found in AUC between CHM and PHM (p=0.0873).
• PHM showed significantly better performance than CHM, indicated by positive IDI (3.08%, p<0.0001) and NRI (12.50%, p<0.0001), and a lower H-L chi-squared value (23.75 vs 35.81).

Conclusions:

• The PROMISE high-risk model (PHM) demonstrates superior predictive accuracy for HRCAD compared to the CONFIRM high-risk model (CHM) in symptomatic DM patients.
• PHM may offer advantages in optimizing downstream management strategies for these patients.
• The findings support the use of PHM for better risk stratification in DM patients undergoing CCTA.