To compare the effectiveness of alternative combined treatments in patients with non-insulin-dependent diabetes mellitus (NIDDM) with secondary failure to sulfonylureas.

Research Design and Methods

A crossover study was carried out by randomly assigning 16 NIDDM patients to a combined treatment with the addition of either a single low-dose bedtime injection of 0.2 U/kg body wt NPH insulin or an oral three times a day administration of 1.5 g/day metformin to the previously ineffective glyburide treatment.


Both combined therapies significantly (P <0.01) reduced fasting plasma glucose (FPG), postprandial plasma glucose (PPPG) and percentage of HbA1. The addition of metformin was more effective than the addition of insulin (P <0.01) in improving PPPG in the 8 patients with higher post-glucagon C-peptide levels. In contrast, the efficacy of neither combined therapy was related to patient age, age of diabetes onset, duration of the disease, percentage of ideal body weight, and FPG. The addition of insulin but not metformin caused a significant (P <0.01) increase of mean body weight. Neither combined treatment caused changes in serum cholesterol and triglyceride levels. No symptomatic hypoglycemic episode was reported in any of the 16 patients.


The addition of bedtime NPH insulin or metformin was effective in improving the glycemic control in most NIDDM patients with secondary failure to glyburide. The combination of metformin and sulfonylurea was more effective in reducing PPPG and did not induce any increase of body weight.

This content is only available via PDF.