medwireNews: Findings from a modeling study suggest that short-term glycemic variability is at least as important as average glucose concentration for predicting hypoglycemia risk, and maintaining both variables at target levels could minimize the risk.
Louis Monnier (University of Montpellier, France) and fellow researchers used continuous glucose monitoring (CGM) data measured over 2 consecutive days from 100 individuals with type 1 diabetes to generate 200 24-hour glycemic profiles.
As reported in Diabetes Care, mean daily glucose (MDG) levels were significantly lower, and glycemic variability was significantly higher, among people who did versus did not experience hypoglycemia, regardless of the glucose threshold used.
Specifically, when hypoglycemia was defined as spending at least 15 minutes with glucose levels below 3.0 mmol/L, MDG levels were 7.9 mmol/L in the 80 hypoglycemic profiles compared with 9.8 mmol/L in the 120 non-hypoglycemic profiles. The corresponding MDG levels were 8.2 versus 10.4 mmol/L when a threshold of 3.9 mmol/L was used.
And within-day glycemic variability – measured by the coefficient of variation for glucose (%CV) – was 44% for hypoglycemic profiles versus 33% for nonhypoglycemic profiles using the threshold of 3.0 mmol/L and 42% versus 31% using the threshold of 3.9 mmol/L.
Monnier and team found “strong negative relationships” between MDG concentration and hypoglycemia risk and “strong positive relationships” between %CV and hypoglycemia risk using univariate mixed regression analysis, whereas daily insulin dose, age, and BMI were not associated with hypoglycemia risk.
The researchers then used a classification and regression tree model to assess the relative contributions of MDG and %CV, finding that %CV “appeared as the primary factor associated with a higher risk of hypoglycemia” when the glucose threshold of 3.0 mmol/L, but not 3.9 mmol/L, was used.
“[A]ll these results reinforce the opinion that, in addition to achieving near-normal glycemia for preventing the development or progression of micro- and macrovascular complications, it is crucial to try to reduce as much as possible the magnitude of glucose fluctuations in order to limit the risk of hypoglycemia,” say Monnier et al.
The team also divided glycemic variability into tertiles for the 65 glycemic profiles with near-normal MDG (≤7.8 mmol/L) in order to establish a target level of acceptable variability. In this analysis, the amount of time spent in hypoglycemia according to a threshold of 3.0 mmol/L was significantly higher for the highest tertile of glycemic variability (%CV >44.1%) compared with the middle (34–44.1%) or lowest tertiles (<34%), at 15.6% versus 3.1% and 0.0%, respectively.
Because the amount of time spent in hypoglycemia “was rarely positive when the %CV was less than 34%,” Monnier and colleagues recommend that “striving to achieve a %CV below 34% in patients who already have satisfactory glycemic control in terms of chronic glucose exposure […] should be one of the main objectives in the management of [type 1 diabetes].”
They note that few people achieve both targets with conventional insulin treatments, but “the expanded use of novel devices for CGM and ceaseless progression toward the implementation of more sophisticated systems for closed-loop insulin delivery both raise promising expectations for improving glucose homeostasis.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2020 Springer Healthcare part of the Springer Nature group