Monitoring oxidative stress enables predictions of personalised glucose targets in diabetic patients

Abstract

The principal cause of hyperglycemia-mediated post-diabetic complications (PDCs) is - oxidative stress (OS). Therefore, establishing a quantitative relationship between OS and glycemic status (GS) of a diabetic individual could help in deciding how much and how long OS should be controlled via external anti-glycemic treatment. To monitor serial changes in OS (as measured by glutathione or GSH, an OS marker), a group of newly diagnosed type 2 diabetic patients kept on anti-diabetic treatment were followed for the period of 8 weeks. A cluster analysis performed on the GSH values pooled from non-diabetics and diabetics before and after therapy (0 and 8 weeks) show that GSH can be used to classify individuals based upon their diabetic status, independently of glucose. That is, GSH can be an excellent anti-oxidant to monitor along with glucose in defining diabetes status. Further, GSH levels are found to be inversely correlated with the GS of diabetic individuals. We propose a physiological minimal mathematical model to capture a quantal dose-response relationship between GSH and glucose for each diabetic patient. Individualised diabetic GSH-glucose curves are parameterised by: maximal glutathione level (Gtot), glucose concentration when GSH is half maximal (v) and slope of the curve (k). Finally, to relax the assumptions imposed in the physiological model, a statistical phenomenological model is proposed to capture OS-GS trajectories in diabetic patients. We show that a phenomenological model is a statistically better and simple alternative to the physiological minimal model. We propose that individually parameterised GSH-glucose curves can be helpful in deciding optimal glucose control strategies through which OS is maximally controlled. Thus, glucose targets can be personalised based upon the OS state of an individual. I