Model-based glycaemic control: methodology and initial results from neonatal intensive care

Area of Science:

• Neonatalogy
• Pediatric Endocrinology
• Clinical Informatics

Background:

• Premature infants frequently experience blood glucose dysregulation (hyperglycemia/hypoglycemia) due to physiological stress.
• Current insulin protocols may not adequately address individual patient variability in glucose metabolism.

Purpose of the Study:

• To evaluate the efficacy of the STAR-GRYPHON computerised protocol for glycemic control in very/extremely premature infants.
• To compare STAR-GRYPHON's performance against retrospective data for hyperglycemic episodes.

Main Methods:

• Utilized a model-based insulin sensitivity parameter to personalize therapy.
• Analyzed 16 hyperglycemic control episodes in 13 premature infants from January 2013 to January 2015.
• Compared outcomes with a retrospective cohort using standard glycemic control methods.

Main Results:

• Achieved a 26% absolute improvement in blood glucose within the target range.
• Significantly reduced hyperglycemia (%BG>10.0 mmol/l: 5.6% vs. 17.7%, p<0.001).
• Lowered median per-patient blood glucose (6.9 vs. 7.8 mmol/l, p<0.001) with no reported hypoglycemia.

Conclusions:

• STAR-GRYPHON demonstrates improved glycemic control and safety in premature infants compared to traditional methods.
• High intra-patient variability in insulin sensitivity necessitates adaptive treatment strategies.
• The protocol highlights the balance between treatment safety and efficacy in this fragile population.