Closed loop administration of propofol based on a Smith predictor: a randomized controlled trial

Area of Science:

• Anesthesiology
• Biomedical Engineering
• Control Systems

Background:

• Propofol pharmacodynamics exhibit common delays in total intravenous anesthesia (TIVA).
• Hypnosis control during TIVA is challenging due to these delayed effects.
• A proportional-integral (PI) controller with a Smith predictor (PI+Smith) was developed to address this challenge.

Purpose of the Study:

• To evaluate the feasibility of a closed-loop PI+Smith controller for propofol administration.
• To compare the performance of the PI+Smith controller against manual control (MC) guided by Bispectral Index (BIS).

Main Methods:

• Fifty-seven adult patients undergoing TIVA with propofol and remifentanil were randomized into PI+Smith and MC groups.
• The target Bispectral Index (BIS) was set to 50.
• Performance was assessed using global score (GS), median performance error (MDPE), median absolute performance error (MDAPE), offset, and Wobble.

Main Results:

• The PI+Smith group (25 patients) demonstrated significantly better performance than the MC group (29 patients).
• Key metrics like GS, MDPE, MDAPE, and offset were significantly improved in the PI+Smith group (P<0.001).
• The percentage of time BIS remained within the 40-60 range was higher in the PI+Smith group (80.8%) compared to MC (59.1%) (P<0.001).

Conclusions:

• The PI controller with a Smith predictor effectively managed propofol administration delays.
• This automated system outperformed manual practice in maintaining BIS within the target range.
• The controller successfully regulated propofol, maintained desired BIS levels, and managed oscillations during TIVA.