Two trials found an association between continuous glucose monitoring (CGM) and improved glucose control for type 1 diabetes in both adolescents and young adults and in older patients.
The first trial randomly assigned U.S. adolescents and young adults to receive CGM or usual care, defined as glucose monitoring with a blood glucose meter. Dexcom Inc. provided the CGM devices and sensors used in the trial. Patients were included if they had had type 1 diabetes for at least a year, were 14 to 24 years of age, and had a screening HbA1c level of 7.5% to 10.9%, among other criteria. The primary outcome was change in HbA1c level from baseline to 24 weeks. Results were published June 16 by JAMA.
Seventy-four patients were assigned to the CGM group, and 79 were assigned to the usual care group. Mean age was 17 years, 50% were female, and the mean duration of diabetes was nine years. One hundred forty-two participants (93%), 71 in each group, completed the study. By week 26, 68% of those in the CGM group were using CGM at least five days per week versus 82% in the 28 days before the six-week visit. Mean HbA1c at baseline was 8.9% in the CGM group and 8.5% at 26 weeks; in the usual care group, mean HbA1c was 8.9% at baseline and did not change over the trial period (adjusted between-group difference, −0.37%; 95% CI, −0.66% to −0.08%; P=0.01).
Among the prespecified secondary outcomes, statistically significant differences were seen for three of seven binary HbA1c outcomes, eight of nine CGM metrics, and one of four patient-reported outcomes. Severe hypoglycemia (three patients in the CGM group vs. two in the usual care group), hyperglycemia/ketosis (one patient vs. four patients), and diabetic ketoacidosis (three patients vs. one patient) were the most common adverse events reported. The researchers noted that their results may not apply to patients with HbA1c levels outside the studied range and that the study period was relatively short, among other limitations. They concluded that CGM was associated with a small but statistically significant improvement in glycemic control over 26 weeks versus usual care with a blood glucose monitor.
The second trial, also published June 16 by JAMA, randomly assigned 203 older U.S. adults with type 1 diabetes to CGM or standard blood glucose monitoring for six months to determine whether CGM could effectively reduce hypoglycemia. Dexcom Inc. also provided the CGM devices and sensors for this trial. Patients were included if they were at least 60 years of age, had not used real-time CGM in the three months before enrollment, and had an HbA1c level below 10.0%. The primary outcome was percentage of time as measured by CGM that sensor glucose values were below 70 mg/dL (3.89 mmol/L).
One hundred three patients were assigned to the CGM group, and 100 were assigned to the standard group. Patients in the standard group were asked to monitor their blood glucose levels at least four times per day, while those in the CGM group were asked to use the monitor daily, to calibrate it twice per day, and to set a “low” alert, recommended as 70 mg/dL (3.89 mmol/L). Fifty-two percent of patients were women, mean age was 68 years, and 53% used a pump versus injections for insulin delivery. Median duration of type 1 diabetes was 36 years, and mean HbA1c level was 7.5%.
One hundred ninety-eight participants (97.5%), 102 in the CGM group and 96 in the standard group, completed the 26-week visit. In the four weeks before the 26-week visit, 81% of patients in the CGM group were wearing the device seven days per week and 89% were wearing it at least five days per week. The CGM group had a median time with glucose levels below 70 mg/dL (3.89 mmol/L) of 5.1% (73 min/d) at baseline and 2.7% (39 min/d) during follow-up versus 4.7% (68 min/d) and 4.9% (70 min/d) for the standard group (adjusted treatment difference, −1.9%; 95% CI, −2.8% to −1.1%; P<0.001).
Among the prespecified secondary outcomes, statistically significant differences were seen for all nine CGM metrics, six of seven HbA1c outcomes, and zero of 15 cognitive and patient-reported outcomes. The CGM group had a decrease in mean HbA1c level versus the standard group (adjusted group difference, −0.3%; 95% CI, −0.4% to −0.1%; P<0.001). Severe hypoglycemia (one patient in the CGM group vs. 10 in the standard group), fractures (five vs. one), falls (four vs. three), and ED visits (six vs. eight) were the most commonly reported adverse events.
As in the first trial, the study period was relatively short, and the study cohort was of relatively high socioeconomic status and were receiving specialized diabetes treatment, among other limitations, the authors noted. They concluded that among older adults with type 1 diabetes, CGM was associated with a small but statistically significant reduction in hypoglycemia over a six-month period.
The authors of both trials called for additional research to determine the potential clinical significance of their findings. An accompanying editorial noted that CGM technology has advanced substantially in recent years and that these two trials show its potential benefit to groups of patients with type 1 diabetes who are at high risk for complications. Remaining barriers to widespread CGM use include insurance companies' strict eligibility criteria, varying levels of state coverage, and lack of training and support for implementing CGM in primary care.
“With CGM innovation happening at a rapid pace and the imminent commercial release of artificial pancreas systems, CGM offers a new outlook for patients with type 1 diabetes and for the clinicians and communities caring for them,” the editorialists wrote. “More effort is needed to overcome current barriers and provide better access to this beneficial technology.”