Risk stratification algorithm for clinical outcomes in anemic patients undergoing percutaneous coronary intervention

Area of Science:

• Cardiology
• Hematology
• Medical Informatics

Background:

• Percutaneous coronary intervention (PCI) is a common procedure for coronary artery disease.
• Anemia is a prevalent comorbidity in patients undergoing PCI.
• The relationship between hemoglobin levels and adverse cardiovascular outcomes post-PCI requires further investigation.

Purpose of the Study:

• To examine the association between baseline and visit hemoglobin levels and major adverse cardiovascular and cerebral events (MACCE) in PCI patients.
• To develop and validate risk stratification models for predicting MACCE in this population.

Main Methods:

• Retrospective cohort study of 8,298 PCI patients (2013-2020).
• Multivariate Cox proportional hazards models and machine learning algorithms (Gradient Boosting) were used for analysis.
• Model performance was assessed using Brier scores and receiver-operating characteristic curves.

Main Results:

• A total of 1,636 patients (19.71%) experienced MACCE during a median 4-year follow-up.
• Visit hemoglobin and hemoglobin change were significantly associated with MACCE risk (HR < 1).
• Low estimated glomerular filtration rate and high HbA1c were key predictors; a risk tree model identified high-risk patients.

Conclusions:

• Hemoglobin levels during and after PCI are more predictive of MACCE than baseline levels.
• Optimizing hemoglobin levels may improve long-term prognosis for anemic PCI patients.
• A novel risk stratification model can enhance screening for high-risk anemic patients undergoing PCI.