Glycemic Control and Hypoglycemia in Continuous Glucose Monitoring

Where glycemic control and hypoglycemia are concerned, does continuous monitoring mean continuous improvement?

Unsurprisingly, lowering HbA1c has numerous health benefits, but lowering blood glucose too quickly also comes with risks, namely hypoglycemia. Improvements in blood glucose levels have been shown to increase the number of hypoglycemic events, some of which may require assistance or even hospitalization. Real-time continuous glucose monitoring has been used alongside alerts and alarms to decrease the number of hypoglycemic events successfully, as well as to reduce the fear associated with hypoglycemia. Studies have also shown that real-time continuous glucose monitoring has reduced the amount of work missed as well as hospital admissions for severe hypoglycemic events. While past studies had shown an increased risk of hypoglycemia when HbA1c was closer to target, these studies had not seen that to be true when continuous glucose monitoring was involved.

This analysis sought to look at the relationship between HbA1c, glucose levels, and hypoglycemia in people with type 1 diabetes using real-time continuous glucose monitoring. Information from two previous randomized controlled studies, Multiple Daily Injections in Diabetes (DIAMOND) and HypoDE studies, were used in this analysis. Of note, in the HypoDE study, all recruits were at high risk of hypoglycemia, as all of them reported having had severe hypoglycemia in the past year or as having decreased hypoglycemia awareness. Both studies included a run-in period with continuous blood glucose monitoring to establish a baseline, with the DIAMOND study having a two-week run-in and HypoDE having a four-week run-in. Both studies used participants that received multiple daily injections of insulin who were randomized into two different arms: one using continuous glucose monitoring, and one self-monitoring glucose levels. In both studies, time with blood glucose below 54 mg/dL was the reported primary hypoglycemia metric. Linear regression was used to assess the relationship between HbA1c, mean glucose, and hypoglycemia. The significance level was set at P < 0.05, and the statistical tests were two-tailed.

Between the two study results analyzed, there was data from 307 participants at baseline and 270 at follow-up. The mean age at baseline was 47 years old, baseline HbA1c was 8.1%, and 42% of participants were female. At baseline, the participants had blood glucose below 54 mg/dL 1.5% of the time and below 70 mg/dL 5.1% of the time. There was no clear relationship between HbA1c and the amount of time spent in hypoglycemia at baseline. In the self-monitoring arm, a relationship was seen between mean glucose and the amount of time that blood glucose was below 70 or 54 mg/dL (P < 0.001 in both cases) with the relationship appearing greatest at lower blood glucose values. Likewise, a relationship was seen between HbA1c and time spent below 54 mg/dL (­P = 0.006) and time spent below 70 mg/dL (P < 0.001). This relationship leveled off at an HbA1c of ~8.5%. In the real-time continuous blood glucose monitoring, there was a relationship seen between the time spent with blood glucose below 70 or 54 mg/dL and the mean blood glucose levels (P < 0.001 for both). Time spent with blood glucose below 70 mg/dL was decreased for 79% of participants that were randomized into the real-time continuous glucose monitoring arms, with those participants with the highest HbA1c having a slightly lower risk of hypoglycemia. There was an association between time spent with blood glucose below 70 mg/dL (P < 0.001) and HbA1c, but there was no statistically significant relationship seen between HbA1c and time spent with blood glucose below 54 mg/dL (P = 0.10).

This analysis of two randomized controlled studies sought to analyze the benefit of real-time continuous blood glucose monitoring in terms of helping decrease the number of times patients spent in hypoglycemia, which it ultimately found support for. This support was most apparent when the patient’s HbA1c was lower. Typically, the lower the HbA1c, the higher the chance of hypoglycemic events, but these changes were decreased with the real-time continuous blood glucose monitoring implemented in both the DIAMOND and HypoDe studies. This study only sought to evaluate the potential usefulness of continuous glucose monitoring and did not try to assess the usefulness of real-time alerts or reminders. It did include many patients who have a history of being at high risk for hypoglycemia, which some past studies excluded. Going forward, it would be good to see studies include evaluations on when and what type of alerts patients find most useful to help further decrease the risk of hypoglycemia.

Practice Pearls:

• Real-time continuous blood glucose monitoring decreases the number of times patients have low blood glucose.
• Patients using real-time continuous blood glucose monitoring saw a massive decrease in the amount of time spent in hypoglycemia, especially among patients with lower HbA1c.
• The decreased risk of hypoglycemia may allow stricter HbA1c goals for patients that are at an increased risk for hypoglycemia.

Reference for “Glycemic Control and Hypoglycemia in Continuous Glucose Monitoring”:
Oliver, Nick, et al. “Continuous Glucose Monitoring in People With Type 1 Diabetes on Multiple Dose Injection Therapy: The Relationship Between Glycemic Control and Hypoglycemia.” Diabetes Care, American Diabetes Association, 1 Jan. 2020, care.diabetesjournals.org/content/43/1/53.

 George McConnell, PharmD. Candidate, LECOM School of Pharmacy