medwireNews: Flash glucose monitoring is associated with small but sustained improvements in glycemic control and a reduction in severe hypoglycemia relative to self-monitoring of blood glucose (SMBG) in type 1 diabetes, observational data show.
The study, by Jan Bolinder (Karolinska Institute, Stockholm, Sweden) and colleagues, was based on Swedish National Diabetes Registry records for 14,372 adults (mean age 45 years, mean diabetes duration 24 years, 45% men) with type 1 diabetes who initiated flash monitoring between 2016 and 2017 and maintained its use for 2 consecutive years.
Glycated hemoglobin (HbA1c) and hypoglycemia rates in the flash monitoring group were compared with those of 7691 sensor-naïve individuals (mean age 55 years, mean diabetes duration 26 years, 41% men) using conventional SMBG during the same observation period.
The majority of participants were using multiple daily insulin injections, with just 21% in the flash monitoring group and 9% in the SMBG group using insulin pumps.
The researchers report in Diabetologia that HbA1c levels decreased with time in both groups, but after adjustment for confounding using inverse probability of treatment weighting based on propensity scores, the decrease was significantly greater with flash monitoring than with SMBG.
Specifically, there was a significant estimated mean absolute HbA1c reduction of 0.11% (1.2 mmol/mol) after 15 to 24 months.
The greatest weighted difference between flash monitoring users and controls occurred among individuals with a baseline HbA1c of 8.5% (70 mmol/mol) or higher, with a significant mean absolute reduction of 0.23% (2.5 mmol/mol) at 15 to 24 months.
For those with a baseline HbA1c of 7.0–8.5% (53–69 mmol/mol), the HbA1c reduction was a significant 0.07% (0.7 mmol/mol) greater with versus without flash monitoring, and for people with optimal HbA1c at baseline (≤6.9%; ≤52 mmol/mol) the difference between the two groups was a significant 0.12% (1.3 mmol/mol).
However, Bolinder et al note that “the clinical relevance [of their findings] is uncertain” because the difference between the two groups did not pass the noninferiority margin of 0.3% to 0.4% in HbA1c that is typically considered clinically meaningful by regulatory authorities.
Nonetheless, the team also found that individuals in the flash monitoring group had a significantly lower severe hypoglycemia event rate than those in the sensor-naïve group, at 4.1% versus 5.2%, corresponding to a significant 21% lower risk.
The authors note that the increasing use of glucose sensor-based technologies in type 1 diabetes “will soon make it practically and ethically difficult to perform long-term randomised controlled trials with conventional SMBG as comparator.”
Instead, they suggest that “a practical alternative may be to perform large well-balanced population-based studies comparing different technologies and outcomes,” which they believe they have shown is feasible with the current analysis.
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2021 Springer Healthcare Ltd, part of the Springer Nature Group
Diabetologia 2021; doi:10.1007/s00125-021-05437-z