Survival prediction models for coronary intervention: strategic decision support

Area of Science:

• Cardiovascular Medicine
• Medical Decision Making
• Biostatistics

Background:

• Coronary artery disease (CAD) management requires selecting optimal revascularization therapy.
• Uncertainty exists regarding whether percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) offers superior long-term survival for CAD patients.

Purpose of the Study:

• To develop predictive survival models for CABG and PCI (using bare-metal stents [BMS] and drug-eluting stents [DES]).
• To create a decision-support tool based on these models to aid clinical decision-making.
• To identify patient subgroups who may benefit from specific revascularization strategies.

Main Methods:

• Survival models were developed for 23,182 patients undergoing CABG or PCI (BMS/DES) between 1995-2007.
• Models incorporated 25 risk factors identified from early and mid-term mortality data.
• A decision-support tool was programmed to predict 5-year survival differences among therapies for individual patients.

Main Results:

• Unadjusted 5-year survival rates were 86% (CABG), 83% (BMS), and 84% (DES).
• Risk factor analysis identified distinct predictors for early and mid-term mortality.
• Patients benefiting most from DES had acute infarctions requiring emergency revascularization; those benefiting most from CABG had extensive CAD and comorbidities.

Conclusions:

• Detailed prognostic models for PCI and CABG are valuable.
• These models support a clinically relevant decision-support tool.
• The tool can guide clinicians and patients in selecting revascularization strategies for potential long-term survival benefits.