Targeted glycemic reduction in critical care using closed-loop control

Area of Science:

• Physiological modeling
• Biomedical engineering
• Critical care medicine

Background:

• Critically ill patients frequently experience hyperglycemia, posing challenges for standard glucose management protocols.
• Existing fixed protocols and sliding scales often lead to errors and suboptimal glycemic control in diverse patient populations.
• Tight glucose control is crucial for reducing mortality in critical care settings.

Observation:

• A physiologically based two-compartment model was developed, incorporating time-varying insulin sensitivity and glucose removal.
• An adaptive, bolus-based control protocol was designed to monitor patient status for precise glycemic regulation.
• Preliminary proof-of-concept clinical trials were conducted over 5 hours each.

Findings:

• The adaptive protocol achieved preset glycemic targets with an average absolute error of 9%.
• 75% of glycemic targets were met within the 7% measurement error margin.
• Observed errors exceeding 7% ranged from 17% to 21%.

Implications:

• The adaptive system demonstrated tight stepwise glycemic control, aligning with patient dynamics.
• Errors were often linked to external factors like drug therapies, suggesting areas for protocol refinement.
• This approach offers targeted, stepwise tight glycemic regulation through insulin boluses in critical care.