Relationship between mean blood glucose and glycated haemoglobin in Type 2 diabetic patients

Aims: To correlate the values of MBG to HbA1c in Greek patients with Type 2 diabetes and/or metabolic syndrome. Methods: We followed up 140 Greek adult patients: 92 patients with Type 2 diabetes treated with insulin or oral glucose-lowering medication, and 48 patients with newly diagnosed Type 2 diabetes or metabolic syndrome not receiving any treatment. MBG was calculated for each patient from self-measurements of blood glucose using a portable glucometer, made six times a day (before eating and 2 h after a meal), three times a week for 1 month. HbA1c was determined by HPLC at 0 and 12 weeks. Results: HbA1c at 0 (x) and 12 weeks (y) correlated strongly (y = 0.790x + 1.115, r = 0.92), confirming that the patient's glycaemic status remained stable during the whole period of follow-up. Linear regression was performed on MBG values; HbA1c at 12 weeks, sex, age, body mass index (BMI) and patient status (Type 2 diabetes treated or not) were used as independent variables. None of the independent variables reached statistical significance in the model, with the exception of HbA1c at 12 weeks. The final model was: MBG (mg/dl) = (34.74 × HbA1c) - 79.21, r = 0.93; or MBG (mmol/l) = 1.91 × HbA1c - 4.36, r = 0.93. Conclusions: Our results establish for the first time a strong correlation between MBG and HbA1c in Type 2 diabetic patients and support the idea of expressing HbA1c results as MBG. This will help patients to gain a clearer interpretation of the result, with less confusion. This simplification will allow every person with diabetes using home glucose-monitoring to understand his or her own target level. © 2008 The Authors.