Effects of insulin vs. glibenclamide in recently diagnosed patients with type 2 diabetes: a 4-year follow-up

Aim: To compare effects of early insulin vs. glibenclamide treatment on beta‐cell function, metabolic control and quality of life (QL) in recently diagnosed patients with type 2 diabetes. Methods: Forty‐nine patients with type 2 diabetes diagnosed 0–2 years before inclusion were randomized to two daily injections of premixed 30% soluble and 70% NPH insulin or glibenclamide at six diabetic clinics in Sweden. C‐peptide–glucagon tests were performed yearly after 3 days of withdrawal of treatment. Results: Thirty‐four patients completed 4 years of study. Daily dose of insulin was increased from 20.4 ± 1.8 U at year 1 to 26.1 ± 2.9 U at year 4 (p = 0.005). Glibenclamide dosage increased from 2.7 ± 0.4 mg at year 1 to 4.5 ± 0.8 mg at year 4 (p = 0.02). Weight increased more in insulin than in glibenclamide treated (+4.4 ± 0.8 vs. +0.3 ± 1.0 kg, p < 0.005). Following short‐term withdrawal of treatment, the C‐peptide responses to glucagon were significantly higher in the insulin vs. glibenclamide group at years 1 (p < 0.01) and 2 (p < 0.02). HbA1c improved identical during the first year but thereafter deteriorated in the glibenclamide group (p < 0.005 for difference at year 4). Ratios of proinsulin to insulin were higher during treatment in glibenclamide‐ vs. insulin‐treated patients after year 2. QL after 4 years as measured by the MOS 36‐item Short‐Form Health Survey (SF‐36) form was not significantly altered. Conclusions: In a 4‐year perspective, beta‐cell function deteriorated in both groups. However, deterioration occurred faster in the glibenclamide group, indicating that alleviating demands on secretion by insulin treatment is beneficial.