Continuous Glucose Monitoring Underreports Blood Glucose During a Simulated Ultraendurance Run in Eumenorrheic Female Runners

Purpose: Continuous-glucose-monitoring (CGM) sensors provide near-real-time glucose data and have been introduced commercially as a tool to inform nutrition decisions. The aim of this pilot study was to explore how factors such as the menstrual phase, extended running duration, and carbohydrates affect CGM outcomes among trained eumenorrheic females in an outdoor simulated ultraendurance running event. Methods: Twelve experienced female ultrarunners (age 39 [6] y) participated in this crossover study. Participants completed an ultraendurance simulation run of 4 hours in the midfollicular and midluteal phases of their menstrual cycle, which consisted of a 3-hour fasted outdoor run (FASTED) followed by a 1-hour treadmill run (TREAD), where 3 standardized 20-g oral glucose doses were provided. Results: Using a mixed linear model, the menstrual phase was statistically significant for differences in glucose measurements from CGM compared with capillary glucose sampling during TREAD (P = .02) but not FASTED. Additionally, the CGM sensor reported glucose levels with an average of -0.43 mmol<middle dot>L-1 (95% CI, - 0.86 to -0.005) and -1.02 mmol<middle dot>L-1 (95% CI, -1.63 to -0.42) lower in fasted and fed scenarios, respectively, when compared with capillary glucose. Conclusion: CGM underreports capillary glucose during fasted and fed exercise. Factors contributing to this underreporting between the sampling methods (CGM vs capillary) were dependent on a combination of exogenous glucose availability, individual biological differences, and the menstrual phase.