CGM still lags behind SMBG during aerobic exercise
medwireNews: Even with newer devices, continuous glucose monitoring (CGM) underestimates the fall in glucose levels during exercise, with the drop lagging significantly behind that measured by self-monitoring of blood glucose (SMBG), research shows.
“Due to this clinically important delay in CGM versus SMBG, we suggest patients increase vigilance and perform more frequent fingerstick capillary glucose monitoring around exercise,” Michael Riddell (York University, Toronto, Ontario, Canada) and co-authors write in Diabetes Technology & Therapeutics.
They found that from minutes 15 to 60 of a 60-minute aerobic exercise session, CGM values were significantly higher than SMBG values among the 17 adults (mean age 31 years, 76% women) with type 1 diabetes who each completed three exercise sessions during the study.
All of the participants were using insulin pump therapy with CGM (Dexcom G4® with 505 algorithm or G5®, San Diego, California, USA) calibrated to their own OmniPod® Personal Diabetes Manager (PDM; Insulet, Billerica, Massachusetts, USA). The PDM had a built-in glucose meter (FreeStyle; Abbott Laboratories, Abbott Park, Illinois, USA) that was used for the SMBG measurements.
During the 51 exercise sessions overall, there were 12 cases of hypoglycemia as measured by SMBG, with a mean SMBG value of 60 mg/dL. For the majority (80%) of these cases, the CGM was measuring in the euglycemic range and overestimated glucose by 21 mg/dL, on average.
Therefore, Riddell et al “suggest setting a higher CGM threshold to initiate carbohydrate feeding to treat (i.e., 90 mg/dL) or prevent (i.e., 120 mg/dL) hypoglycemia, if the CGM displays downward trend arrows.”
The researchers calculated that during exercise there was a lag of 11.5 minutes with CGM versus SMBG and a bias of −6.6 mg/dL per minute.
Conversely, during the recovery period, the average CGM value was significantly lower than the SMBG value 30 minutes after a standardized meal challenge.
“These findings may have critical implications for the development of hybrid closed-loop or automated insulin delivery systems for exercise since these devices may inadvertently overdeliver insulin when plasma glucose is dropping, but sensor glucose fails to drop or may even rise. In contrast, for the meal response after exercise, the delayed rise in sensor glucose may result in insulin underdelivery compared with plasma glucose and/or SMBG values,” Riddell and team remark.
Overall, the mean absolute relative difference for CGM versus SMBG was 12.9% during exercise and 7.6% during mealtime.
The investigators note that a limitation of their study was the lack of plasma glucose levels measured using a standardized glucose analyzer, “so it is currently unclear whether the glucose meter was more accurate or if the drop in glucose as measured by CGM was a true underestimation of circulating plasma glucose concentrations.”
Nonetheless, the team concludes that “the accuracy of newer CGM technology remains negatively impacted during prolonged aerobic exercise and patients need to be aware of this potential CGM time delay.”
By Laura Cowen
medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group