Bleeding and transfusion are the most frequent complications of percutaneous coronary intervention (PCI), with a major bleeding ranging from 2.2% to 14%. The need for a blood transfusion has been shown to be closely linked with worsened patient survival and increased mortality. Lack of a validated model to predict the need for transfusion serves as an impediment to benchmarking and guiding quality improvement.
Researchers at the University of Michigan have developed a highly accurate model for prediction of need for transfusion using pre-procedural variables that are routinely collected in patients undergoing PCI. This model could help guide individualized care and guide therapeutic strategies to reduce transfusion in patients who are most at risk.
Simple tool for reliable calculation of transfusion risks in PCI patients
The algorithm has been developed and validated using data collected from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) over a 3 year period of time. The model demonstrates excellent calibration and discrimination, and is the first tool of its kind made available on the market. It accurately classifies patients into categories of low, intermediate and high risk of transfusion following PCI procedures, and also provides useful information about the impact of procedural anticoagulation and the site of vascular access on transfusion risks.
- Use as online calculator, smart device application or integrate into electronic medical records for one or more of the following purposes:
- Calculate risk adjusted transfusion rates for physicians and operators and thereby guide quality improvement efforts
- Personalize the consent process and provide the patient with their personalized risk estimate rather than the standard average risk of bleeding
- Help identify the 16% of patients who are most likely to need transfusion and thus the ideal subset for use of strategies that have been proven to reduce the risk of transfusion such as bivalirudin or use of radial approach. Help identify the large subset of patients who are at extremely low risk and in whom the use of such therapies may not be that beneficial or cost effective
- Better discrimination than existing regression based risk scores
- Can provide individual patient risks, and hence integrates well with modern personalized medicine practices (vs. comparing to general population statistics)
- Model can be applied to routine clinical practice since it reflects contemporary practices across multiple institutions and operators