Endocrine Abstracts

Introduction: Glucose homeostasis is central to the understanding of diabetes and is influenced by hormones and by substrate flux. This implies a non-linear system which has been confirmed by time series analysis. A dynamic systems approach is required for describing the inter-relationship of glucose and insulin and we describe a method that measures a mathematical domain of glucose homeostasis termed the observed variability and lability (OVAL).

Methods: Plasma insulin and blood glucose concentration profiles were constructed by drawing blood samples every 2 min for 120 min in 12 patients with T2DM median (range) age 35 (25–47) years, duration of diabetes of 7 (2–9) years, receiving oral hypoglycaemic treatment and 30 controls aged 38 (30–53) years matched for gender and BMI.

Insulin and glucose were plotted as consecutive points and the boundary of the relationship was defined as an ellipse. To avoid bias from outliers we used the 95% confidence intervals (95% CI) of the insulin and glucose to define the height and width respectively of the ellipse or OVAL, and the Log₁₀ insulin and glucose were normalised by weighting.

Results: T2DM patients demonstrated an increase in the variability of mean (S.D.) blood glucose concentration (0.92 (0.55) mmol/l) compared with controls (0.16 (0.08) mmol/l) (P<0.001). Variability of mean (S.D.) Ln plasma insulin concentration was also increased (0.22 (0.67) vs 0.17 (0.07) pmol/l, P=0.02). OVAL showed a degradation of homeostasis in type 2 diabetes by a factor of 4 by comparison to controls (OVAL: T2DM 7.8 (3.8) vs controls 1.9 (1.0), P=0.0003). The data were not degraded by a sampling interval of 4 min or a reduction in sample size down to n=4.

Conclusion: OVAL may enable a complimentary approach towards the examination of the glucose regulatory system that could lead to a better understanding of pathophysiology of abnormal metabolism and as a measure of therapeutic efficacy.