Study Finds Best Use Of Insulin As Diabetes Progresses And Impact Factors Of Insulin Treatment Of Diabetic Patients

Objective: The deficiency of insulin secretion is the main characteristic of diabetes.This subject finds the optimization program of insulin therapy in patients with diabetes and related impact factors of insulin treatmentof diabetic patients.Methods: We reviewed the data of 606 type 2 diabetic patients who received insulin therapy.The insulin dose and clinical characteristics were taken down and analyzed.All subjects started injecting insulin intensive treatment on day of admission, during which not used for any oral hypoglycemic agents. The time spent and insulin dose at target blood glucose controlling and insulin doses at discharge were observated. We analyzed the circumstances of hypoglycemia.Statistics by SPSS16.0 software. The primary endpoint objective statistical methods, comparative discharge baseline fasting blood glucose and blood glucose (the median of fasting plasma glucose in the first day of admission) difference, analysis of covariance estimation with glucose difference (between groups), the 95% CI, and the P value were obtained from an ANCOVA model,with insulin injection method and patient compliance as factors, baseline blood glucose value as a covariate, 3 times a day if the performance of non-inferiority, the evaluation of its advantages. Secondary endpoint of statistical methods, the end of fasting and 2h postprandial blood glucose by the percentage of logistic regression analysis to insulin treatment, the number of cases to explanatory variables and baseline blood glucose; insulin dose, HbA1c, and BMI on admission changes between treatment groups Covariance analysis to insulin treatment and the number of cases as factors, baseline blood glucose as covariates; Poisson regression model with low blood sugar.Results: All patients treated with insulin control target blood glucose that time spent for 6.6±1.4 days, its average total insulin were 38.99±7.87U / d, and the amount of insulin needed per body weight 0.59±0.07U/kg.Insulin delivered per body weight in each group as follow(QD: 0.54±0.06 U / kg, BID: 0.56±0.06 U / kg, TID: 0.60±0.08 U / kg, QID: 0.68±0.09 U / kg). Treatment of hypoglycemia occurred in 1.2±0.38 times / person, asymptomatic hypoglycemia in 0.8±0.32 times / person, low blood sugar reactions 1.98±0.46 times / person. The blood glucose value of QD and BID group at the discharge compared with the baseline that the two groups were statistically significant decreased at the primary end point target analysis.The value(mean±standard deviation) were 4.00±0.92 and 4.02±0.96 mmol / L. Difference: 0.02 [95% confidence interval from 0.62 to 0.16]; BID and TID groups of ideal blood glucose and fasting blood glucose declined 4.02±0.96 and 4.22±0.93 mmol / L. Difference: 0.20 [95% confidence interval from 0.42 to 0.12]; TID and QID groups of ideal blood glucose and fasting blood glucose declined 4.22±0.93 and 4.24±0.97 mmol / L. Difference: 0.02 [95% confidence interval from 0.62 to 0.16]. According to records at discharge in patients with insulin on the total amount of insulin per body weight compared pairwise differences were statistically significant at the secondary endpoint target analysis. Hospital intensive insulin therapy in each group the number of days required for blood glucose difference was not statistically significant. Correlation analysis showed that the amount and duration of insulin per body weight, glycosylated hemoglobin was positively correlated with age, BMI was a negative correlation. To achieve the target blood glucose in the number of days as the dependent variable, age, disease duration, BMI, HbA1c, FBG and 2hBG before treatment as the independent variable multiple regression analysis, age and FBG first day of admission into the equation, standardized regression coefficientβwere offset 0.206 and 0.228, P values were 0.033 and 0.036, the difference was statistically significant. To achieve the target blood glucose using insulin before sleep based on the maximum amount of volume as the dependent variable, age, disease duration, BMI, HbA1 c, FBG and the first day of admission as the independent variables 2hBG multiple regression analysis, course and treatment of FBG into the equation before , standardized partial regression coefficientβwere 0.220 and 0.240, P values were 0.036 and 0.012, and the difference was statistically significant.Conclusions: Insulin in each group the average total dose at discharge were as follows: QD group were 30.75±8.12 U / d, BID group were 36.70±8.68 U / d,TID group were 42.20±7.09 U / d, of which QID group were 46.30±6.76 U / d, the dose of insulin per body weight were 0.68±0.09U/kg, based on the effect of insulin or long-acting insulin analogue dose 14.21U/d, accounting for 32.30% of total insulin.Short-acting insulin before meals or the amount of short acting insulin analogues 29.78U/d, accounting for 67.70% of total insulin. Multiple Linear Regression Analysis showed that the insulin dosage is positively correlated with duration of diabetes and HbA1c, while negatively correlated with BMI the highest body weight level. From the view of the insulin dose, patient compliance and the number of hypoglycemia, once daily injections of human insulin or long-acting human insulin analogue for the best choice.