Continuous Glucose Monitoring in Type 1 Diabetes Marathon Study

Race Performance and Blood Sugar Stability

Amateur runners with type 1 diabetes (T1D) completed a 42-kilometer race at speeds comparable to peers without diabetes. The observational study tracked 20 participants during the 2025 Poznań Marathon. Ten runners had type 1 diabetes, and 10 served as controls.

The group with type 1 diabetes finished in a median time of 228 minutes. Control runners finished in 248 minutes. Researchers from Poznan University of Medical Sciences and Raszeja City Hospital presented the data at the European Association for the Study of Diabetes (EASD) Annual Meeting.

Sensor Divergence During Endurance Exercise

Continuous glucose monitors overestimated blood sugar levels during the race and showed reduced measurement accuracy. Investigators tested one intermittently scanned sensor and one real-time sensor against capillary fingerstick measurements from a glucometer. Checks took place at the start line, 10 kilometers, 19 kilometers, 30 kilometers, and the finish line.

Accuracy was evaluated using mean absolute relative difference (MARD). The intermittently scanned sensors diverged from capillary values by an average of 43 percent. Real-time monitors differed by 37 percent.

Both device types ran higher than fingerstick values throughout the event. Intermittent sensors overestimated capillary glucose by an average of 32.2 mg/dL. Real-time devices overestimated capillary glucose by 50.4 mg/dL.

Fueling Protocols and Insulin Reductions

Participants managed glucose levels through structured carbohydrate intake and pre-race basal insulin reductions. Runners consumed a median of 53.5 grams of carbohydrate each hour. Total intake reached 2.61 grams per kilogram of body weight across the race.

Basal insulin reductions depended on the delivery device. Five runners used multiple daily injections (MDI) and reduced basal doses by 25 percent. Three runners used continuous subcutaneous insulin infusion (CSII) pumps with basal rates cut by half.

Two runners used automated insulin delivery (AID) systems set to a target glucose level of 150 mg/dL. Median capillary blood sugar started at 183.5 mg/dL and finished at 108.5 mg/dL. The target pre-race range for all runners was 140 to 200 mg/dL.

Hypoglycemia Risk and Meter Checks

Capillary blood tests detected two episodes of low blood sugar during the event, both occurring in a single participant. That runner began the marathon below the recommended 140 mg/dL pre-race threshold. The runner consumed 49.5 grams of carbohydrate per hour and finished the race.

Fear of hypoglycemia affects up to 45 percent of people living with type 1 diabetes. This concern remains the primary barrier preventing regular physical activity. Sensor overestimation makes relying entirely on automated readings risky during endurance exercise.

Study Limitations and Practical Guidance

The observational design and small group size mean these findings cannot guarantee outcomes for every endurance athlete. The trial evaluated 10 runners with type 1 diabetes who had lived with the condition for a median of 16.5 years. Their baseline glycated hemoglobin (HbA1c) averaged 6.4 percent, indicating strict baseline management.

Athletes managing type 1 diabetes should review race nutrition and insulin strategies with an endocrinologist. Runners should carry a traditional fingerstick glucometer during long events to verify any continuous glucose monitoring reading that does not match physical symptoms.