Modeling and Hemofiltration Treatment of Acute Inflammation

Area of Science:

• Biomedical Engineering
• Mathematical Modeling
• Immunology

Background:

• Sepsis, a life-threatening condition, arises from uncontrolled inflammatory responses to endotoxins.
• Mathematical models offer potential for optimizing treatment strategies in critical care.

Purpose of the Study:

• Develop and validate a mathematical model of the acute inflammatory response to endotoxins.
• Synthesize a model predictive control algorithm for hemoadsorption therapy in sepsis.

Main Methods:

• An 8-state differential equation model was created and calibrated using rat experimental data.
• Model predictive control and particle filter algorithms were implemented for treatment simulation and state estimation.
• Hemoadsorption device efficacy was evaluated through simulations focusing on white blood cell and cytokine capture.

Main Results:

• The model accurately predicted inflammatory responses to varying endotoxin levels in rats.
• Hemoadsorption's primary therapeutic benefit stems from white blood cell capture, not cytokine removal.
• Altering hemoadsorption filters for differential cytokine and white blood cell capture showed no significant outcome improvement.

Conclusions:

• The developed mathematical model provides a framework for understanding and potentially managing acute inflammatory responses.
• Hemoadsorption therapy is most effective through white blood cell clearance, suggesting targeted device design.
• Current hemoadsorption strategies appear robust, with limited gains from differential filtering approaches.