To compare the effect of bedtime NPH insulin or preprandial regular insulin combined with glibenclamide on metabolic control in non-insulin-dependent diabetes mellitus (NIDDM) patients with secondary failure to sulfonylurea therapy.


Eighty NIDDM patients were randomized to treatment with either three preprandial doses of regular insulin (daytime group D) or a bedtime dose of NPH insulin (nocturnal insulinization, group N), both regimens being combined with 10.5 mg of glibenclamide. Metabolic profiles were obtained at 0, 6, 16 weeks.


Glycemic control had improved significantly in both groups after 4 months. Fasting blood glucose was significantly lower compared with baseline in both groups. The mean change ± SD in group D was −2.8 ± 3.5 mmol/l and in group N −6.4 ± 3.0 mmol/L, the reduction being more pronounced in group N compared with group D (P < 0.0001). HbA1c was lowered similarly, from 9.2 ± 1.4 to 7.1 ± 1.2% in group D (P < 0.0001) and from 9.1 to 1.1 to 7.5 ± 1.5% in group N (P < 0.0001). The total daily insulin doses were similar, 29 ± 11 U in group D and 26 ± 9 U in group N, and the circulating insulin levels during daytime were higher in group D than in group N. Total serum cholesterol and triglycerides were similarly and significantly lowered compared with baseline in both groups. Weight gain was more pronounced in group D (3.4 ± 0.3 kg) than in group N (1.9 ± 1.9 kg; D vs. N, P < 0.002), and the change was inversely correlated with initial eight but not with the improvement in HbA1c.


The two insulin regimens exert similar effect on glucose metabolism and serum lipids in NIDDM patients on combination therapy. Weight gain is more pronounced in patients given insulin during the daytime when preprandial doses of short-acting insulin are used.

This content is only available via PDF.