Lower Basal Insulin Dose is Associated with Better Control in Type 1 Diabetes

Abstract

To test the hypothesis that lower basal insulin doses may be paradoxically associated with better diabetic control, we assessed the association between the basal insulin dose and hemoglobin A1c (HbA1c) level in a group of children and young adults with type 1 diabetes. This was a retrospective study of 89 patients with type 1 diabetes (mean age, 14.67 ± 4.8 years; range, 3-29 years) treated in a single outpatient clinic. Forty-six of the 89 patients were treated with continuous subcutaneous insulin infusion, and the other 43 were treated with multiple daily injections (glargine as basal insulin). The daily basal insulin dose was taken either as downloaded from the insulin pump or as registered in the chart at the most recent clinic visit. Glucose data were taken from computerized registration of downloaded patient glucometers. The mean time between data download and HbA1c determination was 0.9 ± 0.78 months. HbA1c level and basal insulin dose were entered with other variables in a multivariable linear regression model. There was a significant correlation between injection of less total daily basal insulin and lower HbA1c level (Pearson correlation, 0.441; P < .001). Optimal HbA1c level was associated with use of 0.28 ± 0.08 U/kg/day of basal insulin (35 ± 10% basal/total). With lower basal insulin levels, lower HbA1C was achieved despite the same total bolus dose. The optimal basal dose as determined by this study is similar to that found in fasting individuals of similar age.