Abstract 6019: The Impact of Blood Transfusion and Other Bleeding Measures on Subsequent Mortality Among Patients Undergoing Primary PCI for Acute Myocardial Infarction: The HORIZONS-AMI Trial
Background: Bleeding complications are strongly associated with subsequent mortality among pts with ACS. The relative importance of the individual components of the composite bleeding scales used in clinical trials is poorly understood.
Objectives and Methods: Data were drawn from the HORIZONS-AMI, in which 3602 pts with STEMI undergoing primary PCI were randomized to treatment with heparin plus a glycoprotein IIb/IIIa inhibitor vs. bivalirudin alone. Bleeding and mortality outcomes were independently adjudicated using the ACUITY major bleeding scale, which incorporates laboratory data, clinical bleeding measures including blood transfusions and access site hematomas, and other parameters. The relative impacts of various bleeding components upon 30 day mortality were examined using multivariable modeling with individual components of the ACUITY bleeding scale incorporated as time-updated covariates.
Results: By 30 days, 238 pts experienced a non CABG major bleeding event (6.6%), and 93 patients died, 26 of whom (28.0%) had a prior major bleed. Bleeding was independently associated with subsequent mortality (adjusted HR 4.29 [2.57, 7.17]). Of the individual components of the ACUITY scale (each modeled independently), retroperitoneal bleeding, reoperation for bleeding, receipt of a blood transfusion, and a reduction in hemoglobin concentration >=4 g/dL were all associated with subsequent mortality after controlling for baseline characteristics and randomized treatment assignment (Table⇓).
Conclusions: Objective assessment of the impact of specific components of bleeding scales upon subsequent mortality is possible, and suggests differing relative impacts of these components. A significant proportion of attributable deaths are related to a reduction in hemoglobin >=4 gm/dL and/or receipt of a blood transfusion.