Predicting In-Hospital Mortality in Patients Undergoing Percutaneous Coronary Intervention

Area of Science:

• Cardiovascular Medicine
• Health Services Research

Background:

• Standardizing risk assessment is crucial for benchmarking and quality improvement in percutaneous coronary interventions (PCI).
• The CathPCI Registry was updated in 2018 to incorporate additional variables for better classification of high-risk patients.

Purpose of the Study:

• To develop and validate a predictive model for in-hospital mortality risk following PCI, utilizing newly available variables.

Main Methods:

• A predictive model was developed and validated using data from 706,263 PCIs.
• A logistic regression model with stepwise selection was applied to a development cohort (70%) and validated on a separate cohort (30%).
• Variables selected in at least 70% of bootstrapped samples or deemed clinically relevant were included.

Main Results:

• In-hospital mortality following PCI was significantly influenced by clinical presentation, procedural urgency, cardiovascular instability, and consciousness level post-cardiac arrest.
• The developed model demonstrated excellent discrimination (C-index: 0.943) and good calibration in the validation cohort.
• The median hospital risk-standardized mortality rate was 1.9%, with an interquartile range of 1.7% to 2.1%.

Conclusions:

• The risk of mortality after PCI can be accurately predicted using a model that incorporates clinical acuity variables not previously captured.
• This validated model serves as a valuable tool for risk stratification and enhancing quality improvement initiatives in PCI procedures.