Development of a machine learning model to predict intraoperative transfusion and guide type and screen ordering

Assessment of machine learning classifiers for predicting intraoperative blood transfusion in non-cardiac surgery

BACKGROUND

This study aimed to develop a machine learning classifier for predicting intraoperative blood transfusion in non-cardiac surgeries.

METHODS

Preoperative data from 6255 patients were extracted from the VitalDB database, an open-source registry. The primary outcome was the area under the receiver operating characteristic (AUROC) curve of ML classifiers in predicting intraoperative blood transfusion, defined as the receipt of at least one unit of packed red blood cells. Five different machine learning algorithms including logistic regression, random forest, adaptive boosting, gradient boosting, and the extremely gradient boosting classifiers were used to construct a binary classifier for intraoperative blood transfusion, and their predictive abilities were compared.

RESULTS

337 (5%) patients received intraoperative blood transfusion. In the test-set, the logistic regression classifier demonstrated the highest AUROC (0.836, 95% CI, 0.795-0.876), followed by the gradient boosting classifier (0.810, 95% CI, 0.750-0.868), AdaBoost classifier (0.776, 95% CI, 0.722-0.829), random forest classifier (0.735, 95% CI, 0.698-0.771), and XGBoost classifier (0.721, 95% CI, 0.695-0.747). The logistic regression classifier showed a higher AUROC compared to that of a multivariable logistic regression model (0.836 vs. 0.623, P < 0.001). Among various parameters used to construct the logistic regression classifier, the top three most important features were operation time (0.999), preoperative serum hemoglobin level (0.785), and open surgery (0.530).

CONCLUSION

We successfully developed various ML classifiers using readily available preoperative data to predict intraoperative transfusion in patients undergoing non-cardiac surgeries. In particular, the logistic regression classifier demonstrated the best performance in predicting intraoperative transfusion.

Challenges in reliable preoperative blood ordering: A qualitative interview study

BACKGROUND

Presurgical blood orders are important for patient safety during surgery, but excess orders can be costly to patients and the healthcare system. We aimed to assess clinician perceptions on the presurgical blood ordering process and perceived barriers to reliable decision-making.

METHODS

This descriptive qualitative study was conducted at a single large academic medical center. Semi-structured interviews were conducted with surgeons, anesthesiologists, nurse anesthetists, nurse practitioners working in preoperative assessment clinics, and transfusion medicine physicians to assess perceptions of current blood ordering processes. Interview responses were analyzed using an inductive open coding approach followed by thematic analysis.

RESULTS

Twenty-three clinicians were interviewed. Clinicians felt that the current blood ordering process was frequently inconsistent. One contributor was a lack of information on surgical transfusion risk, related to lack of experience in ordering clinicians, insufficient communication between stakeholders, high turnover in academic settings, and lack of awareness of the maximum surgical blood ordering schedule. Other contributors included differing opinions about the benefits and harms of over- and under-preparing blood products, leading to variation in transfusion risk thresholds between clinicians, and disagreement about the safety of emergency-release blood.

CONCLUSION

Several barriers to reliable decision-making for presurgical blood orders exist. Future efforts to improve ordering consistency may benefit from improved information sharing between stakeholders and education on safe transfusion practices.