[Development of a new closed-loop system for controlling mivacurium-induced neuromuscular blockade]

Area of Science:

• Anesthesiology
• Biomedical Engineering
• Pharmacology

Background:

• Existing closed-loop control systems struggle with newer, short-acting muscle relaxants like mivacurium.
• Mivacurium necessitates a rapid-response closed-loop system for precise infusion control.

Purpose of the Study:

• To develop and validate a closed-loop control system for mivacurium infusion.
• To assess the system's reliability in maintaining neuromuscular blockade during surgery.

Main Methods:

• A closed-loop system incorporating a neural network predictor was designed.
• 75 patients (ASA I-II) participated in training, prediction, prediction/feedback, and validation phases.
• The final validation phase involved 20 patients to analyze prediction and system errors.

Main Results:

• A closed-loop system utilizing a neural network predictor was successfully established.
• The mean square prediction error in the validation phase was 0.1% +/- 0.2%.
• The overall system's mean square error was 0.55% +/- 0.59%.

Conclusions:

• A reliable closed-loop system for mivacurium infusion was developed.
• The system effectively maintains a 95% neuromuscular blockade during routine surgery.
• The controller's performance is comparable to current systems and warrants further validation for other muscle relaxants.