Simple risk models to predict cardiovascular death in patients with stable coronary artery disease

Area of Science:

• Cardiovascular Medicine
• Clinical Risk Stratification
• Health Economics

Background:

• Accurate risk estimation is crucial for motivating patient adherence to treatment.
• Identifying high-risk individuals with stable coronary artery disease (CAD) can guide the use of potentially expensive, warranted treatments cost-effectively.
• A simple, accessible risk model for stable CAD patients is needed.

Purpose of the Study:

• To develop a straightforward risk prediction model for patients with stable coronary artery disease (CAD).
• The model utilizes readily available clinical risk factors.
• To improve risk stratification for guiding treatment and resource allocation.

Main Methods:

• Development of risk models using the CLARIFY registry of stable CAD patients.
• Initial models incorporated simple clinical variables; subsequent models included left ventricular function, estimated glomerular filtration rate, and hemoglobin.
• Cox proportional hazards models analyzed cardiovascular death over ~5 years; calibration was assessed in the external CORONOR registry.

Main Results:

• Models based solely on simple clinical variables demonstrated good discriminatory power (c-statistics of 0.74 in CLARIFY and ≥0.80 in CORONOR).
• These models showed no lack of calibration in the external dataset.
• Formulae and integer points-based scores with risk tables were provided.

Conclusions:

• A preferred model utilizing 10 readily available variables (age, diabetes, smoking, HF symptoms, history of AFib, MI, PAD, stroke, PCI, HF hospitalization) was developed.
• This model exhibited good discriminatory power and validated well in an external dataset.
• The developed risk score offers a practical tool for managing stable CAD patients.