ֱ

Pros and Cons of CGM Alarm Sensitivity

— Researchers identified optimal alarm thresholds for hypo- and hyperglycemia

MedpageToday
Radiating red circles around a glucose monitor showing an alert in a persons hand

This article is a collaboration between ֱ and:

Alarm settings on continuous glucose monitors (CGMs) can play a big role in time spent in range, a small observational study suggested.

In an analysis of nearly 100 individuals with type 1 diabetes, those that had a higher glucose threshold for alerting them of hypoglycemia -- 73 mg/dL or higher -- spent about 51% less time in hypoglycemia with a blood glucose below 70 mg/dL, reported Yu Kuei Lin, MD, of the University of Michigan Medical School in Ann Arbor, and colleagues in the .

Also, compared with people with a hypoglycemia alarm set at under 73 mg/dL -- the average threshold for a hypoglycemia alarm -- those with a higher glucose threshold for alerts spent about 65% less time in severe hypoglycemia with a glucose level below 54 mg/dL.

However, these individuals with a higher threshold set for hypoglycemia alarms on their CGM also had significantly higher average blood glucose levels (175 mg/dL vs 153 mg/dL). Although their average glucose levels were still in range, those with a higher alarm threshold also spent a longer proportion of time out of target range of 70-180 mg/dL (57.3% vs 67.6%) versus those with their hypoglycemia alarm set to only go off after they dropped below 73 mg/dL.

Lin's group suggested the "optimal" glucose threshold setting for a CGM alarm in order to spend less than 1% of time in hypoglycemia was a higher threshold, set at 75 mg/dL.

On the reverse, similar patterns were seen among patients with a lower glucose threshold for alerting hyperglycemia on their CGM.

As for the 83 individuals with their CGM hyperglycemia alarms on, those with a lower average threshold with less than 205 mg/dL had significantly lower average blood glucose levels versus those with a higher alarm threshold over 205 mg/dL (155 mg/dL vs 172 mg/dL).

Not surprisingly, people with a lower hyperglycemia alarm threshold also spent less time in hyperglycemia or in severe hyperglycemia:

  • Time spent with glucose over 180 mg/dL: 30.0% vs 37.1%
  • Time spent over 250 mg/dL: 8.36% vs 14.5%
  • Time spent over 320 mg/dL: 2.02% vs 5.13%

In order to spend less than 5% of the time in a hyperglycemic range, and have an HbA1c level at or below 7%, Lin's group determined that the ideal alarm upper limit threshold was 170 mg/dL.

Patterns reported among those with a higher hypoglycemia and lower hyperglycemia alarm thresholds were maintained throughout the night, defined as the hours from 10:01 p.m to 5:59 a.m.

This observational analysis included 95 patients with diagnosed type 1 diabetes, average age about 45 years, with a 22-year duration of diabetes. All participants used a CGM for longer than 3 months prior to the analysis, and used the CGM 86% of the time or more. Also, all used Dexcom CGMs with either a G4, G5, or G6 sensor, although the vast majority used a G5.

Some drawbacks to having a more restricted window for CGM alarms -- a higher limit for hypoglycemia and lower limit for hyperglycemia -- included alarm fatigue in the user, such as enduring a greater number of alarms while sleeping. Experiencing an overwhelming number of CGM alarms throughout the day can also lead to patients becoming less likely to respond to alarms, the researchers pointed out.

In a comparison with a small percentage of patients who chose to turn off their hypoglycemia and hyperglycemia alarms, there were no differences in average glucose levels or the percentage of time spent out of range. However, the researchers noted that while this finding is certainly "counterintuitive," these individuals who turned off their alarms may "be a sub-cohort with stronger mechanisms to prevent or recover from hypo/hyperglycemia."

"Identifying the behavioral advantages of these participants may be of clinical interest to further develop hypo/hyperglycemia prevention/management programs," they suggested.

  • author['full_name']

    Kristen Monaco is a senior staff writer, focusing on endocrinology, psychiatry, and nephrology news. Based out of the New York City office, she’s worked at the company since 2015.

Disclosures

The study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases, the Diabetes Research Center at Washington University in St. Louis, the University of Utah Diabetes and Metabolism Research Center, and the National Center for Advancing Translational Sciences.

Lin and co-authors disclosed no relevant relationships with industry.

Primary Source

Journal of the Endocrine Society

Lin, YK "Alarm settings of continuous glucose monitoring systems and associations to glucose outcomes in type 1 diabetes" JES 2019; DOI: 10.1210/jendso/bvz005.