The FreeStyle Libre continuous flash glucose monitoring system (FGM) provides a unique opportunity to evaluate glucose trends to optimize insulin therapy compared to the glycemic self-control by glucometer. However, working with trends requires individualization.
The main purpose of this study is the development of customized decision-making algorithms based on FGM trend arrows for T1D patients.
T1D patients on insulin pump therapy used FGM over the period of 10-14 days. At the first stage, endocrinologists used standard algorithms of action on trend arrows, evaluated their effectiveness and in case of unsatisfactory results, began to select individualized algorithm rules. At the second stage, the effectiveness of the developed individual algorithms was checked, and further corrections were made if necessary.
For 5 patients out of 15 (33.3%), the standard (developed by the company) algorithms were effective. For the rest 10 patient (66.6%), the modification of standard algorithms was required as follows: 5 patients (33.3%) needed the development of almost completely individualized algorithms (75-80% - changes), and for the other 5 patients (33.3%) some corrections (15-30% - changes) were needed. For 10 patients the insulin doses correction in the standard algorithms was 59%, and bread unit correction - was 2%. In patients trained to assess trends, the average value of TIR was 40.9%, for untrained patients - 34.6%.
Standard trending algorithms are suitable for only one-third of T1D patients who receive insulin pump therapy, indicating the actuality of the problem of diabetes treatment based on trend arrows.