Closed-loop anesthetic drug concentration estimation using clinical-effect feedback

Area of Science:

• Anesthesiology
• Pharmacokinetics
• Control Theory

Background:

• Current target-controlled infusion systems use open-loop prediction for anesthetic drug administration.
• Open-loop systems are susceptible to patient variability in pharmacokinetics and pharmacodynamics.
• Accurate anesthetic drug concentration estimation is crucial for patient safety and effective anesthesia.

Purpose of the Study:

• To present a novel closed-loop approach for anesthetic drug concentration estimation.
• To improve the robustness of drug concentration estimation against patient variability.
• To reduce drug concentration errors compared to open-loop prediction methods.

Main Methods:

• Developed a closed-loop estimation system utilizing clinical-effect measurement feedback.
• Designed a robust estimator using μ-synthesis theory to process drug administration and effect measurements.
• Validated the closed-loop estimation principle through Monte Carlo simulations with diverse patient models.

Main Results:

• Closed-loop estimation demonstrated improved robustness against patient pharmacokinetic and pharmacodynamic variability.
• Statistically significant reductions in median percentage, median absolute percentage, and maximum absolute percentage drug-concentration errors were observed.
• The novel approach outperformed open-loop prediction in accuracy and reliability.

Conclusions:

• Closed-loop anesthetic drug concentration estimation offers superior accuracy and robustness.
• This method enhances patient safety by minimizing drug concentration errors.
• The μ-synthesis based robust estimator provides a promising framework for future anesthetic drug delivery systems.