A random forest based risk model for reliable and accurate prediction of receipt of transfusion in patients
Hitinder S Gurm1, Judith Kooiman2, Thomas LaLonde3
1Department of Internal Medicine, Division of Cardiovascular Medicine, University of Michigan, Ann Arbor, Michigan, United States of America.
Insights
A new tool accurately predicts blood transfusion risk after Percutaneous Coronary Intervention (PCI). This helps clinicians identify high-risk patients for better decision-making and quality assessment.
Area of Science:
- Cardiology
- Medical Informatics
- Health Services Research
Background:
- Blood transfusion is a frequent complication of Percutaneous Coronary Intervention (PCI), linked to adverse outcomes.
- No existing risk model effectively identifies patients at high risk for transfusion post-PCI.
Purpose of the Study:
- To develop and validate a predictive tool for blood transfusion in patients undergoing PCI.
- To create a reliable method for assessing transfusion risk in contemporary PCI procedures.
Main Methods:
- Random forest models were built using 45 pre-procedural variables.
- An abbreviated model was created using the most influential variables.
- Model performance was assessed using AUC and NRI in an independent validation dataset.
Main Results:
- The study included 103,294 PCI procedures; models were trained on 72,328 and validated on 30,966.
- Both full and reduced models showed excellent predictive performance (AUC: 0.888 and 0.880).
- Procedural anticoagulation and radial access impacted transfusion rates in intermediate/high-risk groups, but not low-risk patients.
Conclusions:
- A novel, user-friendly computational tool can reliably calculate transfusion risk post-PCI.
- This risk prediction algorithm aids bedside clinical decision-making and quality assessment.
Background:
Transfusion is a common complication of Percutaneous Coronary Intervention (PCI) and is associated with adverse short and long term outcomes. There is no risk model for identifying patients most likely to receive transfusion after PCI. The objective of our study was to develop and validate a tool for predicting receipt of blood transfusion in patients undergoing contemporary PCI.
Methods:
Random forest models were developed utilizing 45 pre-procedural clinical and laboratory variables to estimate the receipt of transfusion in patients undergoing PCI. The most influential variables were selected for inclusion in an abbreviated model. Model performance estimating transfusion was evaluated in an independent validation dataset using area under the ROC curve (AUC), with net reclassification improvement (NRI) used to compare full and reduced model prediction after grouping in low, intermediate, and high risk categories. The impact of procedural anticoagulation on observed versus predicted transfusion rates were assessed for the different risk categories.
Results:
Our study cohort was comprised of 103,294 PCI procedures performed at 46 hospitals between July 2009 through December 2012 in Michigan of which 72,328 (70%) were randomly selected for training the models, and 30,966 (30%) for validation. The models demonstrated excellent calibration and discrimination (AUC: full model = 0.888 (95% CI 0.877-0.899), reduced model AUC = 0.880 (95% CI, 0.868-0.892), p for difference 0.003, NRI = 2.77%, p = 0.007). Procedural anticoagulation and radial access significantly influenced transfusion rates in the intermediate and high risk patients but no clinically relevant impact was noted in low risk patients, who made up 70% of the total cohort.
Conclusions:
The risk of transfusion among patients undergoing PCI can be reliably calculated using a novel easy to use computational tool (https://bmc2.org/calculators/transfusion). This risk prediction algorithm may prove useful for both bed side clinical decision making and risk adjustment for assessment of quality.
