Assessment of basal insulin requirement using fasting tests in insulin-treated patients with type 2 diabetes mellitus

Aims: Characterizing the time course of the rise of blood glucose concentrations in the fasting state during the day and night in patients with type 2 diabetes. Methods: 40 consecutive insulin-treated patients with type 2 diabetes underwent fasting tests on two different days with either no breakfast and lunch (fasting time of 20 hours) OF no dinner (fasting time of 21 hours). Glucose-lowering medication was stopped prior to the test according to the half-life of the medication prescribed. At the start of the fasting tests, blood glucose concentrations were lowered to below 7 mmol/L using an insulin infusion. Results: 26 men and 14 women were included in the study. Mean (+/- SD) age was 61 +/- 10 years, BMI 31 +/- 7 kg/m(2), and HbA1c 7.5 +/- 1%. Diabetes duration was 14 +/- 8 years and duration of insulin therapy had been prescribed for a mean of 6 +/- 6 years. During the daytime fast, plasma glucose concentrations rose by a mean of 0.8 +/- 1.6 mmol/L. During the nighttime fast, plasma glucose concentrations increased particularly after midnight, by 4.3 +/- 2.1 mmol/L, i.e. significantly more than during the daytime fast. Conclusions: Fasting blood glucose concentrations in the majority of insulin-treated patients with type 2 diabetes increase markedly after midnight. No similar increase is observed during the day. Thus, for most patients with type 2 diabetes, an intermediate- or long-acting insulin injected at bedtime with a peak action six to eight hours after injection should be appropriate.