medwireNews: Among people with type 1 diabetes, adding the sodium-glucose cotransporter (SGLT)2 inhibitor empagliflozin to an automated insulin delivery system may allow replacement of carbohydrate counting with a simple meal announcement, suggest findings from a pilot randomized trial.
However, the addition of empagliflozin did not avoid the need for meal announcements altogether, say Bruce Perkins (University of Toronto, Ontario, Canada) and colleagues.
The researchers explain that at present, users of closed-loop insulin delivery systems “remain required to perform quantitative carbohydrate counting for every meal, in order to achieve acceptable levels of glycemic control,” but this is “challenging for people with diabetes, is associated with an average counting error approximating 20%, and does not eliminate post-prandial hypoglycemia.”
They therefore carried out an open-label noninferiority crossover trial involving 30 people to evaluate glycemic control during treatment with empagliflozin 25 mg/day plus closed-loop insulin delivery on three prandial insulin strategy days: carbohydrate counting; simple meal announcement via pressing of a button; and no meal announcement. These insulin strategy days were separated by a median of 4.5 days, and empagliflozin was continued throughout.
Glucose levels with these strategies were compared with those during control treatment with closed-loop insulin only on two prandial insulin strategy days – carbohydrate counting and simple meal announcement – also separated by a median of 4.5 days.
As reported in Diabetes, Obesity and Metabolism, the study did not meet its primary endpoint of noninferior daytime glucose levels with empagliflozin and no meal announcement versus no empagliflozin and carbohydrate counting, with average levels of 10.0 versus 8.5 mmol/L (180.2 vs 153.1 mg/dL). Therefore, “[e]mpagliflozin did not allow a no meal announcement system,” say the researchers.
However, they found that average glucose levels with empagliflozin and a simple meal announcement were noninferior to those with no empagliflozin and carbohydrate counting (8.5 vs 8.5 mmol/L), indicating that the SGLT2 inhibitor “eliminated the need for carbohydrate counting if a strategy of simple meal announcement [was] implemented.”
Perkins et al note that the proportion of time spent with glucose levels in the target range of 3.9–10.0 mmol/L (70.3–180.2 mg/dL) was comparable with empagliflozin plus meal announcement versus no empagliflozin with carbohydrate counting (68 vs 70%), as was the proportion of time spent in hyperglycemia or hypoglycemia. The average total insulin dose was significantly lower with empagliflozin plus meal announcement, at 18.5 units compared with 26.2 units with no empagliflozin and carbohydrate counting.
The average fasting ketone level was 0.21 mmol/L with empagliflozin plus meal announcements, compared with 0.11 mmol/L for no empagliflozin and carbohydrate counting. There were no episodes of ketoacidosis in the study, but the authors caution that the duration of the pilot trial was limited, which “preclude[d] robust assessment of the diabetic ketoacidosis risk.”
They conclude: “Longer term studies are now warranted to determine the effect on long-term glycemic control and to further investigate the putative risk of ketosis and diabetic ketoacidosis associated with automated insulin delivery and adjunct-to-insulin [SGLT2 inhibitors].”
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2021 Springer Healthcare Ltd, part of the Springer Nature Group