Characteristics And Determinants Of Daily Insulin Dosage In Type 2 Diabetes

Objectives:To discuss the characteristics of type 2 diabetes using insulin, and the determinants of daily insulin dose in those patients.Methods:A total of 150 patients with type 2 diabetes (62 men,88 women) who were intended to be hospitalized for glucose control were included after excluding accompanying factors such as emergent conditions and other endocrine disorders. Clinical parameters including sex, age, diabetic duration (DD), waist circumference (WC), weight, body mass index (BMI), dosage of insulin before admission (bDOI) and at discharge (dcDOI), duration of insulin use (DIU), existence of chronic diabetic complications, diet, physical activity (PA), alcohol consumption (AC), smoking, and oral hypoglycemic agents before admission (bOHA) and at discharge (dcOHA) were collected. Laboratory parameters including glycosylated hemoglobin (HBA1C), fasting serum glucose (FBG), postprandial serum glucose (PBG), fasting serum C-peptide (FCP), postprandial serum C-peptide (PCP), fasting serum insulin (FIns), postprandial serum insulin (PIns), total serum cholesterol (TC), serum triglyceride (TG), high density lipoprotein cholesterol (HDL-C), and low density lipoprotein cholesterol (LDL-C) were determined. Multiple linear regression analysis was use to study the determinants of daily insulin use in type 2 diabetes. Paired-sample t test and Chi square test were used to analysis the difference of mean blood glucose, mean insulin using dosage and percents of the use of insulin sensitizer between admission and discharge.Results:The initiation of insulin use in this group began at around ten years of the diabetic duration, the bDOI was at about of 0.5U/Kg. WC, weight, BMI and DIU were positively related to bDOI (0.4U/cm, 95%CI=0.2-0.6,0.4U/Kg,95%CI=0.2-0.6,1.5U/kg/m2,95%CI=0.9-2.1, 1.4U/year,95%CI=0.7-2.1, respectively, adjusted for sex and age). After adjusted sex, age and weight, BMI, DD, DIU, HBA1C were positively related to bDOI, there was an increase of 1.7U of insulin per Kg/m2 BMI(95%CI= 0.4-2.9), an increase of 0.5U of insulin per year DD (95%CI= 0.2-0.8), an increase of 1.6U of insulin per year DIU(95%CI=0.9-2.2), an increase of 1.1U of insulin per%HBA1C (95%CI=-0.2-2.4). AC, HDL-C were inversely related to bDOI. A bDOI 8.8U (95%CI=-15.5--1.8) decreased was found for AC compared to subjects without AC intake, and a decrease of 3.3U bDOI (95%CI=-5.8--0.7) per mmol/L HDL-C. The results were similar as for the dcDOI with the above mentioned indexes after adjusted for sex and age. WC, Weight, BMI, DIU, HBA1C, FPG, PPG were positively related to dcDOI, PA was inversely related to dcDOI. There an increase of 0.5U of dcDOI per cm WC (95%CI=0.3-0.7), an increase of 0.3U of dcDOI per Kg weight (95%CI=0.1-0.5), an increase of 1.1 U of dcDOI per Kg/m2 BMI (95%CI=0.5-1.7), an increase of 1.2U of dcDOI per year DIU (95%CI=0.5-1.8), an increase of 2.2U per%HBA1C (95%CI=1.0-3.4), FBG and PBG were positively related to dcDOI (1.1U per mmol/L,95%CI= 0.5-1.7 and 0.7U per mmol/L,95%CI=0.2-1.2, respectively). PA was inversely related to dcDOI, an decrease of 2.1U/day per activity score unit (95%CI=-4.2--0.04). There was no difference between mean bDOI and mean dcDOI (35.8U vs 36.5U, t=-0.646,P=5.19) although the mean capillary blood glucose at discharge decreased significantly from the mean capillary blood glucose before admission (9.4 vs 13.2,t=14.32, P=0.000). Although the rate of metformin and thiazolidinediones intake at discharge was 57%, there was no statistical significance compared to admission intake (35%, P= 0.335).Conclusion:Patients with type 2 diabetes start insulin therapy relatively late. The lower insulin sensitivity, the longer duration of diabetes, the longer duration of insulin use, and the higher HBA1C level predict the much more daily insulin dosage. PA, Moderate alcohol consumption, and higher HDL-C were predictors of lower daily insulin dosage. Although mean glucose level decreased significantly at discharge, but the daily insulin doses did not differ between at before admission and at discharge, other factors needs to be studied for the improvement in glucose control.