A nomogram predicting 30-day mortality in patients undergoing percutaneous coronary intervention

Area of Science:

• Cardiology
• Medical Informatics

Background:

• Percutaneous coronary intervention (PCI) is a common procedure.
• Early detection of mortality post-PCI is critical for patient management.
• Existing risk models often use outdated data, necessitating updated prediction tools.

Purpose of the Study:

• To develop and validate a nomogram for predicting 30-day mortality after PCI.
• To provide a more accurate and current risk assessment tool for clinicians.

Main Methods:

• A cohort of 10,444 patients undergoing PCI was analyzed.
• Logistic regression with stepwise backward selection identified key predictive features.
• A nomogram was constructed using selected parameters.

Main Results:

• The nomogram incorporated age, male sex, cardiac dysfunction, STEMI, and TIMI 0-2 post-PCI.
• The nomogram demonstrated strong predictive performance with an AUC of 0.881.
• It outperformed a previous risk model (AUC = 0.7).

Conclusions:

• A novel nomogram effectively predicts 30-day mortality in unselected PCI patients.
• This tool can significantly enhance clinical risk stratification.
• The nomogram offers a valuable resource for improving patient outcomes.