Abstract
Continuous glucose monitoring (CGM) is an emerging tool for dietary counseling in athletes. This study aimed to explore blood glucose profiles in Para cyclists and evaluate CGM accuracy at rest and during exercise. Thirteen Para cyclists, comprising eight hand bikers and five cyclists, wore a CGM sensor (Abbott) for 2 weeks. Participants recorded the timing of meals and regular training sessions and executed one standardized training session. Fifteen capillary blood glucose reference values (seven at rest and eight during the standardized training) were obtained by finger pricks. Mean glucose concentrations and time spent in hypoglycemia (<3.9 mmol/L), euglycemia (3.9-7.8 mmol/L), and hyperglycemia (>7.8 mmol/L) were calculated over 24 hrs and during daytime, nighttime, exercise, and 2 hrs postprandial periods. Mean absolute relative differences (MARD) were calculated between the CGM and capillary blood glucose. The mean glucose concentration over the 24 hr-period was 5.7 (5.6-5.8) mmol/L. Athletes were in the euglycemia range 91% of the time. Hyperglycemia was almost exclusively observed postprandially and during exercise. Hypoglycemia was restricted to the night and was particularly observed in athletes with a spinal cord injury. CGM accuracy was acceptable at rest (MARD: 12%) but markedly lower during exercise (MARD: 34%; p = 0.01), especially for hand bikers (MARD: 41%) compared with cyclists (MARD: 24%; p = 0.01). Para cyclists generally do not display signs of disturbed glucose regulation. However, the increased risk for nocturnal hypoglycemia in athletes with a spinal cord injury warrants attention. Furthermore, CGM accuracy is compromised during exercise, especially if the sensor is in proximity to highly active muscles.