A novel model forecasting perioperative red blood cell transfusion

Area of Science:

• Medical Informatics
• Clinical Prediction Models
• Surgical Outcomes

Background:

• Perioperative blood transfusion is common but associated with risks.
• Accurate prediction of transfusion need is crucial for patient management.
• Existing models may lack comprehensive validation or clinical utility.

Purpose of the Study:

• To develop and validate a predictive nomogram for perioperative blood transfusion risk.
• To identify key clinical factors associated with transfusion requirements.
• To provide a tool for risk stratification in surgical patients.

Main Methods:

• Retrospective analysis of 97,443 surgical patients from DATADRYAD.
• Development of a predictive model using multivariate logistic regression.
• Validation of the nomogram using ROC curves, calibration plots, and decision curve analysis.

Main Results:

• A total of 6.04% of patients received more than one unit of red blood cells.
• Eight variables including age, ASA-PS, and surgery risk were identified as predictors.
• The nomogram demonstrated good predictive performance with concordance indices of 0.870 and 0.865.

Conclusions:

• A validated nomogram effectively predicts perioperative blood transfusion risk.
• The model aids in identifying patients at higher risk for transfusion.
• This tool can support clinical decision-making for transfusion management in surgery.