Risk stratification for long-term mortality after percutaneous coronary intervention

Area of Science:

• Cardiology
• Medical Informatics
• Biostatistics

Background:

• A validated, simple risk score for predicting long-term mortality after percutaneous coronary intervention (PCI) using preprocedural factors is lacking.
• Existing risk prediction models are often complex and not easily applicable in clinical practice.

Purpose of the Study:

• To develop and validate a simplified risk score for predicting long-term mortality after PCI.
• To identify key preprocedural risk factors associated with mortality post-PCI.

Main Methods:

• A Cox proportional hazards model was developed using a derivation sample of 11,897 patients who underwent PCI.
• The model was simplified to create a point-based risk score.
• The risk score was validated on a separate sample, and its predictive accuracy for 1, 3, and 5-year mortality was assessed using C statistics.

Main Results:

• The Cox model identified 12 significant risk factors for mortality, including age, BMI, multivessel disease, ejection fraction, hemodynamic state, and comorbidities.
• The simplified risk score demonstrated good agreement between observed and predicted risks of death.
• The C statistics for the model in the validation sample were 0.787 (1-year), 0.785 (3-year), and 0.773 (5-year), indicating good predictive accuracy.

Conclusions:

• A simple, point-based risk score can effectively predict long-term mortality after PCI.
• This score, derived from a Cox model, offers a practical tool for risk stratification in patients undergoing PCI.