Prediction of death after percutaneous coronary interventional procedures

Area of Science:

• Cardiology
• Interventional Cardiology
• Health Outcomes Research

Background:

• Predicting procedural death after percutaneous coronary intervention (PCI) is challenging due to variations in patient populations and practices.
• The influence of current patient selection and technological advancements on PCI outcomes remains unclear.
• Accurate risk prediction is crucial for comparing PCI outcomes and improving patient care.

Purpose of the Study:

• To evaluate the accuracy of a risk equation based on patient variables in predicting procedural death after PCI.
• To determine if an existing risk score can be reliably applied in the current era of interventional cardiology.

Main Methods:

• Analysis of the Mayo Clinic Coronary Interventional Database (1995-1997).
• Calculation of expected mortality using the New York State multivariate risk score.
• Comparison of observed versus expected mortality rates in 3387 patients undergoing 3830 PCI procedures.

Main Results:

• The New York State risk score demonstrated high predictive accuracy for procedural death (chi-square = 213.8; P <.0001).
• Observed mortality (2.38%) closely matched the expected mortality (2.32%) (P = not significant).
• High-risk lesion characteristics (calcium, thrombus, type C) showed a modest association with mortality.

Conclusions:

• The New York State multivariate risk score accurately predicted procedural death in the studied PCI patient cohort.
• This validated risk model can be utilized for comparing outcomes across different institutions and practice patterns.
• The findings support the use of established risk scores in the current landscape of interventional cardiology.