Purpose: The purposes of this study were to describe a newly developed algorithm that estimates the glucose supplement on a patient- and situation-specific basis and to test whether these amounts would be appropriate for maintaining blood glucose levels within the recommended range in exercising type 1 diabetic patients.
Methods: The algorithm first estimates the overall amount of glucose oxidized during exercise on the basis of the patient's physical fitness, exercise intensity, and duration. The amount of supplemental CHO to be consumed before or during the effort represents a fraction of the burned quantity depending on the patient's usual therapy and insulin sensitivity and on the time of day the exercise is performed. The algorithm was tested in 27 patients by comparing the estimated amounts of supplemental CHO with the actual amounts required to complete 1-h constant-intensity walks. Each patient performed three trials, each of which started at different time intervals after insulin injection (81 walks were performed overall). Glycemia was tested every 15 min.
Results: In 70.4% of the walks, independent of the time of day, the amount of CHO estimated by the algorithm would be adequate to allow the patients to complete the exercise with a glucose level within the selected thresholds (i.e., 3.9-10 mmol·L−1).
Conclusions: The algorithm provided a satisfactory estimate of the CHO needed to complete the exercises. Although the performance of the algorithm still requires testing for different exercise intensities, durations, and modalities, the results indicate its potential usefulness as a tool for preventing immediate exercise-induced glycemic imbalances (i.e., during exercise) in type 1 diabetic patients, in particular for spontaneous physical activities not planned in advance, thus allowing all insulin-dependent patients to safely enjoy the benefits of exercise.
1Department of Biomedical Sciences and Technologies, University of Udine, Udine, ITALY; 2Department of Electrical Engineering, Electronics, and Computer Science, University of Trieste, Trieste, ITALY; 3Department of Clinical, Morphological Sciences and Technologies, University of Trieste, Trieste, ITALY; and 4Diabetes and Metabolism Unit, Public Hospital, Udine, ITALY
Address for correspondence: Maria Pia Francescato, M.D., Department of Biomedical Sciences and Technologies, University of Udine, P.le Kolbe 4, 33100, Udine, Italy; E-mail: email@example.com.
Submitted for publication October 2009.
Accepted for publication May 2010.