Anticoagulation therapy advisor: a decision-support system for heparin therapy during ECMO

Area of Science:

• Biomedical Engineering
• Pharmacometrics
• Clinical Informatics

Background:

• Extracorporeal membrane oxygenation (ECMO) requires precise anticoagulation management.
• Heparin anticoagulation during ECMO is challenging, with risks of thrombosis or bleeding.
• Current methods for monitoring anticoagulation may lack sufficient precision.

Purpose of the Study:

• To develop and validate a pharmacokinetic-pharmacodynamic (PK-PD) model for predicting activated clotting times (ACT) during ECMO.
• To integrate the PK-PD model into a decision-support system for improved anticoagulation management.
• To assess the accuracy and bias of the model in predicting ACT values.

Main Methods:

• Development of a PK-PD model using the NONMEM program.
• Integration of the PK-PD model into a clinical decision-support system.
• Validation of the model using an independent dataset and nonlinear regression for individualization.

Main Results:

• The population PK-PD model achieved a mean absolute prediction error of 33.5 seconds for ACT.
• Individualized model parameter estimates reduced the mean absolute error to 25.5 seconds.
• The model demonstrated a reduction in mean bias from -14.3 seconds to -3.1 seconds with individualization.

Conclusions:

• A validated PK-PD model can significantly improve the prediction and control of anticoagulation during ECMO.
• Individualized model parameter estimation enhances prediction accuracy and reduces bias.
• The developed decision-support system offers a comprehensive environment for managing ECMO anticoagulation.