CGM decreased HbA1c levels, hypoglycemia in patients with type 2 diabetes

Two industry-supported studies of continuous glucose monitoring (CGM) found encouraging results in patients with type 2 diabetes, leading editorialists to call for broadened access to the technology in this population, particularly in primary care.


Two studies published online on June 2 by JAMA looked at the effects of continuous glucose monitoring (CGM) in patients with type 2 diabetes.

The first study, an open-label randomized controlled trial of patients with type 2 diabetes, found that those who received CGM had a significantly greater decrease in HbA1c level at eight months compared to those who received traditional blood glucose meter monitoring. The industry-funded trial enrolled adults with type 2 diabetes treated with one or two daily injections of long- or intermediate-acting basal insulin without prandial insulin at 15 centers in the U.S. from July 30, 2018, to Oct. 30, 2019, with follow-up completed on July 7, 2020. Researchers randomly assigned patients from primary care practices who were not under the care of an endocrinologist for their diabetes to CGM (n=116) or blood glucose meter monitoring (n=59). Study funding and devices were provided by Dexcom Inc. All participants received a Bluetooth-enabled blood glucose meter and test strips. Those in the CGM group also received a Dexcom G6 device and were instructed to use CGM continuously and to perform blood glucose meter testing as needed, particularly if CGM readings did not match symptoms. Those in the control group were instructed to perform blood glucose meter testing (fasting and postprandial) one to three times daily. Both groups had a clinic visit at three months and a final visit at eight months for HbA1c level measurement, the primary outcome. Secondary outcomes included CGM-measured time in target glucose range (70 to 180 mg/dL or 3.9 to 10.0 mmol/L), time with glucose level greater than 250 mg/dL (13.9 mmol/L), and mean glucose level at eight months.

Of 175 randomized participants (mean age, 57 years; 50% women), 165 completed the trial. At eight months, the mean HbA1c level decreased from 9.1% to 8.0% in the CGM group and from 9.0% to 8.4% in the control group (adjusted difference, −0.4%; 95% CI, −0.8% to −0.1%). In the CGM group, the mean percentage of time in the target glucose range was 59%, compared to 43% in the control group (adjusted difference, 15% [95% CI, 8% to 23%]; P<0.001). The mean percentage of time with glucose level greater than 250 mg/dL (13.9 mmol/L) was 11% in the CGM group compared to 27% in the control group (adjusted difference, −16% [95% CI, −21% to −11%]; P<0.001). The means of mean glucose level were 179 mg/dL (9.9 mmol/L) in the CGM group and 206 mg/dL (11.4 mmol/L) in the control group. The adjusted difference was −26 mg/dL (−1.4 mmol/L) (95% CI, −41 to −12 mg/dL [−2.3 to −0.7 mmol/L]; P<0.001). Severe hypoglycemic events occurred in one participant in each group.

Among other limitations, the duration of follow-up was only eight months, and some of the eight-month visits needed to be completed virtually due to the COVID-19 pandemic, resulting in some participants not having eight-month HbA1c or CGM data. Participants also had greater contact with clinic staff than they would as part of usual care, potentially limiting generalizability to most routine clinical practice settings, the authors said.

The second study, a retrospective cohort study supported in part by an independent investigator award from Dexcom, found that use of real-time CGM was associated with significantly lower HbA1c levels and lower rates of ED visits or hospitalizations for hypoglycemia. Researchers looked at 41,753 participants with insulin-treated diabetes (type 1, n=5,673; type 2, n=36,080) receiving care from a Northern California integrated health care delivery system from 2014 to 2019. All were being treated with insulin, were self-monitoring their blood glucose levels, and had not used CGM between Jan. 1, 2014, and baseline. The study identified patients who initiated real-time CGM (based on a durable medical equipment vendor claim for real-time CGM supplies from 2015 through 2019) and compared their outcomes with those who didn't start CGM on 10 end points.

Overall, 3,806 patients (mean age, 42.4 years; 51% women; 91% with type 1 diabetes) initiated real-time CGM. The reference group included the remaining 37,947 patients (mean age, 63.4 years; 49% women; 94% with type 2 diabetes). Mean HbA1c level decreased among real-time CGM initiators from 8.17% to 7.76% versus from 8.28% to 8.19% in controls (adjusted difference-in-differences estimate, −0.40% [95% CI, −0.48% to −0.32%]; P<0.001). Rates of hypoglycemia requiring a hospital or ED visit decreased among real-time CGM initiators from 5.1% to 3.0% and increased among controls from 1.9% to 2.3% (difference-in-differences estimate, −2.7% [95% CI, −4.4% to −1.1%]; P=0.001). The mean HbA1c level decreased significantly more with CGM among participants with type 2 diabetes than in those with type 1 diabetes (P=0.003 for interaction). There were also statistically significant differences between groups in the adjusted net changes in the proportion of patients with an HbA1c level lower than 7% (adjusted difference-in-differences estimate, 9.6% [95% CI, 7.1% to 12.2%]; P<0.001), lower than 8% (adjusted difference-in-differences estimate, 13.1% [95% CI, 10.2% to 16.1%]; P<0.001), and higher than 9% (adjusted difference-in-differences estimate, −7.1% [95% CI, −9.5% to −4.6%]; P<0.001) and in the number of outpatient visits (adjusted difference-in-differences estimate, −0.4 [95% CI, −0.6 to −0.2]; P<0.001) and telephone visits (adjusted difference-in-differences estimate, 1.1 [95% CI, 0.8 to 1.4]; P<0.001). However, initiation of real-time CGM was not associated with statistically significant changes in rates of hyperglycemia requiring an ED or hospital visit, ED visits for any reason, or hospitalizations for any reason.

The study authors noted the possibility of unmeasured residual confounders and cautioned that the findings should be interpreted as exploratory because of the potential for type I error due to multiple comparisons. They added that the study was unable to account for advancements in real-time CGM technology that occurred during the study period, such as predictive low-glucose alerts and no requirement for calibration.

These studies confirm that CGM is a technology that can be effectively used by patients with type 2 diabetes to improve glycemic control, an accompanying editorial said. In addition, the randomized trial demonstrated the promise of the technology in primary care settings, where most patients with type 2 diabetes receive their care, according to the editorialists.

“Important policy changes in Medicare eligibility to CGM for type 2 diabetes and institutional changes that promote its use in primary care will go a long way to improving diabetes control and reducing complications, particularly among the populations most in need,” they wrote. “The time has come to broaden access to CGM for patients with type 2 diabetes.”