Abstract 103: Predictors of Transfusion in Acute Myocardial Infarction: Insights from APEX-AMI (Assessment of Pexelizumab in Acute Myocardial Infarction) Trial
Introduction: We previously demonstrated that blood product transfusion strongly and independently predicts death in a population of STEMI patients treated with primary percutaneous coronary intervention (PCI). Using multivariable regression analyses, we explored which baseline and in-hospital factors predicted transfusion.
Methods: Stepwise variable selection was performed from the database of the APEX-AMI trial, a phase 3 trial that compared pexelizumab to placebo in 5,736 patients with STEMI treated with primary PCI in 17 countries between 2004 and 2006. In-hospital CABG patients were excluded from the analysis (n=141).
Results: Of the 4429 non-CABG pts with complete baseline and in-hospital covariates, 211 (4.6%) were transfused. Most transfused pts (81%) experienced moderate or severe bleeding. Baseline independent predictors of transfusion are presented in the table⇓. After 24 hours, intra-aortic balloon pump use (OR: 6.05 95% CI 3.62–10.09, χ2 = 47.4) and nadir hemoglobin (OR: 2.57 95% CI 1.62– 4.08, χ2 = 16.0, per 10 mg/dL) were identified independent predictor of transfusion. Landmark analyses for the first 5 days maintained all predictors. Interestingly, transfusion rates were not increased in patients treated with Gp IIb/IIIa inhibitors or recatheterization.
Conclusions: In this contemporary acute MI population treated with PCI, baseline characteristics (c-statistic: 0.823) coupled with additional factors emerging during the hospital course effectively discriminate the risk of receiving blood transfusion. These data will be useful in developing strategies to reduce the need for blood transfusions so as to avoid their long term adverse effects.