Fig. 1Glycemic variability in three hypothetical patients who have the same mean blood glucose concentration. Patient B has relatively small variations during the day and on different days; this patient should have little difficulty in lowering daily mean blood glucose concentrations without inducing hypoglycemia. In comparison, patient A has marked blood glucose variations on the same day and patient C has marked blood glucose variations on different days.
Fig. 2Calculation of mean amplitude of glucose excursion (MAGE). In the first step, all the local maximum/minimum values are determined. The next step is an assessment of maximum/minimum pairs against the standard deviation (SD). If the difference from minimum to maximum is greater than the SD, this variation from mean measure is retained. If the local maximum/minimum is less than 1 SD it is excluded from further calculations. These troughs are retained and summed to achieve the MAGE.
Table 1Glycemic variability indices
Continuous glucose monitoring |
Mean (average)±standard deviation |
J index |
Coefficient of variance |
Low blood glucose index, high blood glucose index |
Average daily risk range |
Mean amplitude of glucose excursion |
Mean of daily differences |
Continuous overall net glycemic action |
Serum |
Glycated albumin |
1,5-anhydroglucitol |
Glycated albumin/glycosylated hemoglobin ratio |
Table 2Indications for continuous glucose monitoring
Patients with T1DM not meeting HbA1c targets or recurrent diabetic ketoacidosis |
Patient with repeated hypoglycemic episodes or hypoglycemia unawareness |
Subjects requiring better glycemic control while avoiding hypoglycemia |
Before or during pregnancy in women with T1DM or T2DM |
Need for improving brittle diabetes |