Changes on Serum HDL_{2, 3}-Cholesterol after Improved Glycemic Control in Diabetics with Different Treatments

To determine the effect of diabetic treatment on serum HDL2, 3-cholesterol, thirty-nine diabetic patients with poor glycemic control were divided into three groups according to the type of treatment: insulin treated group (14 patients), sulfonylurea treated (15) and diet treated (10). And in each group, changes of serum lipids, lipoproteins especially HDL2, 3-cholesterols and apolipoproteins were measured before and after 2 months of diabetic treatment.In the insulin treated group, improvement in glycemic control (FBS from 243±57 to 147±59mg/dl, mean±SD) was accompanied by a significant rise in serum HDL-cholesterol level (from 34.7±11.5 to 47.7±12.6mg/dl, p<0.01). This was mainly due to a rise of HDL2-cholesterol level (from 17.5±7.1 to 27.8±11.8mg/dl, p<0.01), but HDL3-cholesterol level was not altered significantly. These changes were followed by an insignificant increase of apolipoprotein A-I.On the other hand, in the sulfonylurea treated group HDL-cholesterol level showed a tendency to decrease (from 42.2±9.1 to 36.4±7.3mg/dl, p<0.1) in spite of the improved glycemic control of same grade (FBS from 236±78 to 148±40mg/dl), but HDL2, HDL3-cholesterol levels were not changed significantly. Also, in the diet treated group in which glycemic control remained poor, significant changes of these lipoproteins were not observed.The raised HDL2-cholesterol levels showed an insignificant correlation with the decrease of serum triglyceride levels either in the insulin treated group or in the patients as a whole, but the raised HDL2-cholesterol levels showed a significant correlation with the doses of daily injected insulin.Whether the increased level of serum HDL2-cholesterol in diabetics treated with insulin is an effect of the corrected lipoprotein metabolism associated by improved glycemic control, or a direct effect of insulin itself is not clear. But from our results, it does not seem only related to the former since the FBS did not differ between insulin and sulfonylurea group, conversely, the positive relation between HDL2-cholesterol and injected insulin dose rather suggests the latter possibility.