Endogenous Glucose Production Variation Assessment for Malaysian ICU Patients Based on Diabetic Status

No Thumbnail Available
Razak A.A.
Abu-Samah A.
Razak N.N.
Baharudin S.
Suhaimi F.M.
Jamaludin U.
Journal Title
Journal ISSN
Volume Title
Springer Science and Business Media Deutschland GmbH
Research Projects
Organizational Units
Journal Issue
Intensive Care Insulin-Nutrition-Glucose (ICING) model is used in Stochastic TARgeted (STAR) protocol to personalize glucose control in critically- ill patients. One of the important ICING parameters included in this physiological mathematical model is endogenous glucose production (EGP) which is defined as a constant value. EGP however may vary in individual patients and vary differently in critically-ill diabetic patients. This paper studies this aspect specifically to identify if certain EGP values will improve the estimation of insulin sensitivity (SI) through the reduction of unlikely SI estimation; SI = 0, blood glucose fit errors and simulated STAR glycaemic control performance. Analysis on 151 patients from two Malaysian hospitals, divided into 54 diabetic and 97 non-diabetic were done using 5 EGP values (1.16, 1.50, 2.00, 2.50, and 3.00) mmol/min to see the effect of EGP variations on both type of patients. The results indicate that the frequency of SI = 0 was improved with reduction, from 25.3% to 0.01% in diabetic and 13.4% to 0.008% in non-diabetic patients when EGP is raised from 1.16�mmol/min to 3.00�mmol/min. BG fit errors varied but with small variation and lower than 1%. The highest performance results of % blood glucose time in target range 6.0�10.0�mmol/L was obtained for EGP at 2.50�mmol/min, at 70.8% (diabetic) and EGP = 2.00�mmol/min, with 72.2% (non-diabetic). Overall results showed that choice of EGP values can have an impact on SI estimation and glycaemic control performance. Furthermore, certain EGP values have been identified to be beneficial to distinguish based on diabetic status. � 2021, Springer Nature Switzerland AG.
Biomedical engineering; Blood; Estimation; Insulin; Intensive care units; Physiological models; Stars; Stochastic models; Stochastic systems; Constant values; Control performance; Critically-ill patients; Diabetic patient; Endogenous glucose; Icing parameters; Insulin sensitivity; Small variations; Glucose