The treatment of NIDDM patients with secondary failure to sulfonylureas is still a debated problem. In this study we compared in NIDDM patients with secondary failure to glyburide, the effect of adding a single, low-dose bed time either NPH or ultralent insulin injection (0.15-0.2 U/kg) to the previously ineffective sulfonylurea treatment. Both NPH and ultralent insulin therapy have been demonstrated to be effective in ameliorating metabolic control in NIDDM patients with secondary failure to sulfonylureas. However, the addition of bed-time ultralent insulin caused a greater and significant decrease in post prandial plasma glucose. In contrast, the average fasting plasma glucose decrease was significantly greater after NPH insulin administration. These results indicate that in NIDDM patients with secondary failure to glyburide bed-time ultralent insulin administration is a better tool to improve the post prandial plasma glucose.