Glargine Versus NPH in Patients With Chronic Kidney Disease

This study consists of a randomized, cross-over, open-label controlled clinical trial. Randomized patients will be allocated alternately into two groups to receive the following therapies: GROUP 1 - insulin analog glargine once a day associated to insulin lispro at mealtime and GROUP 2 - NPH human insulin, three applications per day ( breakfast, lunch and bedtime) and insulin lispro at mealtime. Patients receiving insulin NPH plus insulin lispro will be oriented to mix both of them in the same syringe at breakfast and lunchtime. The laboratory tests will be performed at baseline and 12, 24, 36 and 48 weeks after the study start. During routine medical appointments the patient should bring the self- monitoring of capillary blood glucose (SMBG), eight points per day once a week, and hypoglycemia score. After 24 weeks of insulin therapy, a continuous glucose monitoring system (CGMS) will be implemented for three days, and after that, the basal insulin changed i.e. patients using NPH insulin will receive insulin glargine and patients on insulin glargine will be changed to NPH insulin, both groups will keep insulin lispro before meals. A new CGMS will be carried out 24 weeks after therapy has been changed. Methodology: The metabolic profile will be evaluated throughout SMBG; biochemical, hormonal and hematological measurements; hypoglycemia score and CGMS (Medtronic/Northridge, CA). All randomized patients who use at least one dose of any study treatment will be considered in the Intent-to-treat (ITT) population. The initial plan is to randomize 40 patients, assuming a drop-out rate of 15%, to obtain a sample size of at least 34 randomized patients. .Statistical analysis will be performed using comparative descriptive analyzes, such as chi-square distribution, t-test and non-parametric tests. Analyze of data CGMS will include the area under the curve and the related statistic. Finally, logistic regression models will be adopted to evaluate the effect of the treatment on the several variables in question.