Comparison of glucose variability assessed by a continuous glucose-monitoring system in patients with type 2 diabetes mellitus switched from NPH insulin to insulin glargine: the COBIN2 study.

BACKGROUND: Glucose variability combined with glycosylated hemoglobin (HbA1c) assessments more reliably represents the level of glycemic control. The study was aimed to compare blood glucose variability with insulin glargine vs. neutral protamine Hagedorn (NPH) in patients with type 2 diabetes mellitus using a continuous glucose-monitoring system (CGMS), in patients treated with basal insulin using stable dose of oral antidiabetic agents and HbA1c in the range of 4.5-8.0 % International Federation of Clinical Chemistry (IFCC) units. [6.2-9.4 % Diabetes Control and Complications Trial (DCCT) units].

METHODS: This was a multicenter, prospective, open-label, single-arm study in patients (N = 116) treated for ≥ 2 months with NPH and metformin combined with sulfonylurea or glinide. Glucose variability was measured after a 4-week NPH treatment phase and after a subsequent 12-week glargine treatment phase using CGMS. Based on 72-hour CGMS, glucose variability was assessed by area under the curve [AUC (mmol/L · h)]. Differences (glargine-NPH) in AUC within 24 h in the glucose ranges of ≤ 3.3, ≤ 3.9, 7.5-3.9 (margins excluding), ≥ 7.5, ≥ 10, and ≥ 15 mmol/L were evaluated. Circadian fluctuation of glucose was assessed by M-value (log-transformation of the deviation from an arbitrary standard).

RESULTS: AUCs of glucose in the lowest ranges (≤ 3.3 and ≤ 3.9 mmol/L) did not change significantly after treatment with glargine. Those in the higher ranges (≥ 7.5, ≥ 10, and ≥ 15 mmol/L) were significantly lower (p < 0.001 for all ranges), whereas AUC of glucose in the normal range (3.9-7.5 mmol/L) was significantly higher (p < 0.001) at the end of glargine treatment phase. Circadian fluctuation of glucose assessed by M-value showed a significant decrease after glargine treatment (p < 0.003). No significant differences in hypoglycemia confirmed by glucose value ≤ 3.3 mmol/L were found between treatment phases. This trial is registered at ClinicalTrials.gov, NCT00659477.

CONCLUSIONS: As monitored by CGMS, switching from NPH to glargine with active titration shifted glucose from abnormally high to normal levels with reduced fluctuation and without increased risk of hypoglycemia.