• Forty-two obese subjects with non—insulin-dependent diabetes mellitus had their plasma insulin, C peptide, and glucose levels measured after an overnight fast and in response to a 75-g oral glucose loading. Subjects were then prospectively followed up with dietary treatment, and the same measurements were repeated at 1 year. Although insulin values tended to be lower with greater fasting hyperglycemia at baseline, no correlation was observed among three parameters. However, near-normalization of glycemia (measured as the level of hemoglobin A1) was associated with significantly higher fasting and stimulated plasma insulin concentrations. Sixteen subjects were matched to each other for equivalent baseline hyperglycemia (by glycosylated hemoglobin) and divided into group 1 (normalization of the hemoglobin A1 value to 7.0% ± 0.3% [mean ± SE]) and group 2 (persistent hyperglycemia) (hemoglobin A1 value, 10.7%±0.7% [mean ± SE]). Before dietary therapy, the plasma insulin concentrations were twofold to threefold higher in group 1, and despite similar degrees of weight loss, group 2 failed to demonstrate improved glycemia. We concluded that the outcome of diet therapy for non—insulin-dependent diabetes mellitus is dependent on the duration of diabetes and endogenous insulin secretory reserve. There is a subgroup of patients with non—insulin-dependent diabetes mellitus in whom delayed dietary intervention may have a beneficial effect.
(Arch Intern Med. 1990;150:169-172)