To investigate the glycemic balance before, during and after the 2016 Paris Marathon using a real-time continuous glucose monitoring (RT-CGM) system in patients with type 1 diabetes mellitus in a prospective single-center observational study.

Inclusion criteria were as follows: type 1 diabetes mellitus; age ≥18 years; HbA1c < 9%. Participants performed two 2h-preparatory races (PR) before the Marathon and were monitored with RT-CGM 24h before, during and 72h after each race. Hypoglycemic events were prevented via carbohydrate intake / insulin dose adjustments. The primary outcome was area under the curve (AUC) < 70 and > 200 mg/dl and percentage of time spent in euglycemia, hypoglycemia, and hyperglycemia during the races.

Twelve patients (2F/10M; median HbA1c=6.8%) were included and completed the study. Median AUC < 70 and time spent in hypoglycemia (< 70 mg/dl) during the PRs and Marathon were equal to 0. However, no hypoglycemic episodes occurred during Marathon, while two patients experienced hypoglycemia during PR1 and PR2. There was a significant increase in AUC > 200 mg/dl during races between PR2 and Marathon (P = 0.009) although the median time spent > 200mg/dl was not statistically different in Marathon versus PR2 (48.4% versus 18.4%; P = 0.09). Median time spent in euglycemia (70-200 mg/dl) was lower in Marathon versus PR2 (51.6 versus 58%; P = 0.03).

Our study proposes a medical support protocol for extreme endurance physical activity in patients with type 1 diabetes mellitus. Our results suggest that RT-CGM, coupled with adjustments in carbohydrate intake and insulin doses, appears to be effective to prevent hypoglycemia during and after exercise.