This article is more than 5 years old. Information may no longer be current.
Real-time CGM beats SMBG regardless of insulin pump use
Click Here to Manage Email Alerts
Among adults with type 1 diabetes, those who used real-time continuous glucose monitoring with any form of insulin delivery experienced significantly lower HbA1c and spent more time in range and less time in hypoglycemia at 3 years compared with those who used self-monitoring blood glucose, according to real-world data published in Diabetes Care.
“Our understanding of the real-world use and benefits of real-time CGM has been limited. Findings from randomized controlled trials often fail to reflect actual participant behaviors and resultant outcomes in real-world clinical practice,” Jan Šoupal, MD, PhD, of the 3rd department of internal medicine and 1st faculty of medicine at Charles University in Prague, and colleagues wrote. “Because diabetes management is primarily dependent on participant behavior, different research approaches are needed to more definitively assess these behavior-based interventions.”
Šoupal and colleagues evaluated 3-year follow-up data from 94 participants from the 1-year Comparison of Sensor Augmented Insulin Regimens (COMISAIR) trial: 22 used real-time CGM with multiple daily injections of insulin (mean age, 32.6 years; 41% women; mean HbA1c, 8.2%), 26 used real-time CGM with continuous subcutaneous insulin infusion from an insulin pump (mean age, 32.3 years; 50% women; mean HbA1c, 8.2%), 21 used SMBG with multiple daily injections of insulin (mean age, 33 years; 52% women; mean HbA1c, 8.3%) and 25 used SMBG with an insulin pump (mean age, 35 years; 48% women; mean HbA1c, 8.3%). The researchers assessed HbA1c quarterly for 3 years.
During the study, “all sensor-augmented insulin regimen participants wore their sensors > 70% of the time,” the researchers wrote.
“The consistently high percentage of time that participants wore their sensors during the 3-year study period suggests that real-time CGM was perceived to be a valuable tool in their self-management regimens,” the researchers wrote.
Glycemic measures
The average HbA1c level at 3 years was 7% for those using real-time CGM with multiple daily injections compared with 8.2% at baseline (P = .0002), 6.9% for those using real-time CGM with an insulin pump compared with 8.2% at baseline (P < .0001), 8% for those using SMBG with multiple daily injections compared with 8.3% at baseline and 7.7% for those using SMBG with a pump compared with 8.3% at baseline. Only the decreases from baseline experienced by those using real-time CGM reached statistical significance, according to the researchers, and the differences in reductions from baseline between the two real-time CGM groups were not statistically significant. In addition, an HbA1c of less than 7% was reached by 48% of those using real-time CGM and by 9% of those using SMBG.
Among those using multiple daily injections, there was a –0.87% difference in HbA1c favoring the real-time CGM group vs. the SMBG group (P = .0016). There was a –0.78% difference in HbA1c favoring the real-time CGM with multiple daily injections group vs. the SMBG with insulin pump group (P = .0011). There was a –0.99% difference in HbA1c favoring the real-time CGM with insulin pump group vs. the SMBG with multiple daily injections group (P < .0001). Lastly, among those using an insulin pump, there was a –0.9% difference in HbA1c favoring the real-time CGM group vs. the SMBG group (P < .0001).
Among those using multiple daily injections, the average sensor glucose level was –27 mg/dL lower for those using real-time CGM vs. SMBG (P = .0007). The average sensor glucose was –21.1 mg/dL lower for those using real-time CGM with multiple daily injections vs. those using SMBG with an insulin pump (P = .0009). The average sensor glucose was –29.2 mg/dL lower for those using real-time CGM with an insulin pump vs. those using SMBG with multiple daily injections (P < .0001). Among those using an insulin pump, the average sensor glucose was –23 mg/dL lower for those using real-time CGM vs. those using SMBG (P < .0001).
Among those using multiple daily injections, there was a –12.24 mg/dL difference in glycemic variability favoring the real-time CGM group vs. the SMBG group (P = .0003). There was a –12.06 mg/dL difference favoring the real-time CGM with insulin pump group vs. the SMBG with multiple daily injections group (P = .0006). Additionally, there was a –7.56 mg/dL difference in glycemic variability between the two SMBG groups, favoring the SMBG with insulin pump (P = .02). Glycemic variability did not differ by a statistically significant degree in other comparisons.
Time in range, hypoglycemia
The researchers also found that participants who used real-time CGM with multiple daily injections spent 69% of time with HbA1c between 70 mg/dL and 180 mg/dL at 3 years compared with a rate of 48.7% at baseline (P < .0001). Participants who used real-time CGM with an insulin pump spent 72.3% of time in range at 3 years compared with 50.9% at baseline (P < .0001). The researchers noted that the amount of time in range did not change significantly for those using SMBG, with those using it in combination with multiple daily injections posting rates of 51.8% at baseline and 54.7% at 3 years and those using SMBG in combination with an insulin pump posting rates of 50.6% at baseline and 57.8% at 3 years.
Similarly, participants who used real-time CGM with multiple daily injections spent 5.5% of time in hypoglycemia with HbA1c less than 70 mg/dL at 3 years compared with 9.4% at baseline (P = .0387). Participants who used real-time CGM with a pump spent 5.3% of time in hypoglycemia at 3 years compared with 9% at baseline (P = .0235). In comparison, those using SMBG with multiple daily injections spent 7.2% of time in hypoglycemia at 3 years compared with 8.3% at baseline and those using SMBG with a pump spent 6.4% of time in hypoglycemia at 3 years compared with 8.9% at baseline, but these differences did not reach statistical significance, according to the researchers.
Participants who used either insulin delivery method with real-time CGM averaged 6.9 daily insulin boluses while those using either delivery method with SMBG averaged 4.5 daily insulin boluses (P < .0001). In addition, those using real-time CGM had a mean of three daily finger-stick tests while those who used SMBG had a mean of 3.8 daily finger-stick tests (P = .001), although the researchers stated that “the real-time CGM devices required twice-daily calibration with finger-stick testing.”
“Although similar improvements in glycemic control have been shown in previous randomized controlled trials, our findings demonstrate the long-term sustainability of real-time CGM use, its clinical benefits, and its implications regarding medication adherence within the context of real-world diabetes self-management,” the researchers wrote. “Our findings may provide guidance to clinicians when discussing treatment/monitoring options with their participants.” – by Phil Neuffer
Disclosures: Šoupal reports receiving speaker honoraria and consulting for Abbott, Dexcom, Eli Lilly, Medtronic, Novo Nordisk and Roche. Please see the study for all other authors’ relevant financial disclosures.
Click Here to Manage Email Alerts