Diabetes and Continuous Glucose Monitoring
Are You Sure the Patient Requires Continuous Glucose Monitoring?
Continuous glucose monitoring has the potential to offer several benefits, enabling patients and their health care providers to:
Identify large excursions of hyperglycemia or hypoglycemia that the patient may not be aware of.
Recognize real-time trends in glucose levels that permit therapeutic intervention to minimize glucose fluctuations.
Closely monitor glucose levels while changing their therapeutic regimens (e.g., starting a new insulin).
Frequent testing of glucose readings by glucometer give only a limited number of data points to represent the glucose excursions over 24 hours. Multiple measurements by capillary blood glucose monitoring may not be practical due to expense, discomfort, hypoglycemic unawareness failing to trigger the patient to check glucose levels when low, and sleeping.
Most studies have focused on patients with type I diabetes, but there are studies in patients with type 2 diabetes and diabetes during pregnancy or gestational diabetes. Studies have looked at children/ adolescents, and adults.
Current continuous glucose monitor systems (CGMS) measure glucose in the interstitial fluid. A sensor is inserted under the skin in the abdomen, arm or back, and a transmitter sends each glucose measurement every 5 minutes to a receiver where the patient can see the glucose level; in professional continuous glucose monitoring (see below) they are blinded to the value. The values can be subsequently downloaded for review. CGMS will alarm at both a low reading and a high reading. Sensors can provide accurate data from 3-7 days depending on the model. CGMS values range +/- 15% compared to simultaneous laboratory glucose readings.
There are two basic types of continuous glucose monitoring:
Professional use, in which the provider owns the devices and the patient wears it intermittently for 3-7 days at a time;
Personal use, in which the patient owns the CGMS device.
CGMS can be carried out in real-time or retrospectively:
Retrospective CGMS does not allow the patient wearing the device to see the actual glucose values until the CGMS is returned to the health care provider's office where the data are downloaded for review.
Real-time CGMS allows the patient wearing the device to see the glucose changes while wearing the device as well as review the glucose readings with the health care provider when the device is returned and the data downloaded.
Retrospective or real-time CGMS can be used with professional use of CGMS, but personal use of CGMS requires real-time CGMS.
Specific patient populations for whom CGMS may be recommended include:
Patients with type 1 or 2 DM who are willing to wear a sensor regularly and often enough to obtain data to use in managing diabetes.
Patients with hypoglycemic unawareness or nocturnal hypoglycemia.
Patients who have Hgb A1c measurements within target to assist in maintaining that level of control.
Patients with Hgb A1c measurements above target range who are trying to achieve better control.
Patients for whom there is an unresolved difference between their Hgb A1c and capillary blood glucose monitoring
Women with diabetes who are pregnant.
Prior to seeking approval for CGMS for personal use for a patient, it is prudent to make sure the patient tries and wears a CGMS to assess sufficient utilization of the CGMS to improve glucose control. This can be done by having the patient wear a CGMS for professional use and assessing its utilization
CGMS is not recommended for inpatient use, especially in the intensive care setting. CGMS should also not be used in individuals who cannot follow through with all the required steps, including calibration to obtain useful and meaningful clinical data.
Sensor augmented pump therapy (integration of CGMS with insulin pump therapy)
The goal: A closed loop system that would manage a patient's glucose levels with limited or no input from the patient. Although this goal is getting closer, it is still not possible at this time. However, data do support the use of CGMS to assist patients with insulin pumps in the management of their diabetes. In the STAR 3 trial, sensor augmented pump therapy proved superior in lowering Hgb A1c at 1 year to a regimen of multiple insulin injections guided by capillary glucose blood monitoring. There were no differences in the rates of hypoglycemia (note, however, that the rates of severe hypoglycemia were low compared to rates seen in previous studies). Advances in both sensor technology and integration with insulin pumps have greatly advanced the potential for a closed loop system. The first hybrid closed loop system has been approved for use in the USA by the Food and Drug Administration. Please see the section regarding insulin pumps for further information.
Sensor lag in glucose readings
Glucose readings from the sensor can lag capillary blood glucose monitoring by-20 minutes, an issue of greatest concern during periods of rapid change in glucose levels. When there is a concern about sensor lag in a clinically significant glucose range, capillary blood glucose monitoring is recommended. For example, if a patient sees that several readings in a row are stable, there is not a pressing need to perform capillary blood glucose monitoring. However, if it seems the glucose level is falling or rising rapidly, capillary blood glucose monitoring should be performed, permitting therapeutic intervention even before the CGMS would be able to detect the change. Hypo- or hyperglycemia documented on the sensor should ideally be confirmed by capillary blood glucose monitoring before treatment of the out of range glucose.
Management and Treatment of the Disease
Data Supporting the use of CGMS in the management of diabetes
In some, but not all studies, CGMS has significantly lowered Hgb A1c levels. The benefit occurs primarily - and in some reports solely - in subjects who actually use the sensor the most time. In the Juvenile Diabetes Research Foundation study, age also proved a factor. Adults 25 years of age and older saw a significant improvement in hgb A1c and other measures of glycemic control compared to the control group. Children from ages 8-14 did not have a significant difference in Hgb A1c compared to the control group, but did see significant improvement in some other measures of glycemic control. No differences were noted between children and adults between the ages of 15-24 compared to the control group. There was no difference between the groups in terms of hypoglycemia. In particular, in the age group of 25 and above there was no increase in hypoglycemia seen with the significant lowering of the HgbA1c.
Negatives associated with the use of CGMS
Not all evidence demonstrates an improvement in glucose control or hypoglycemia.
Patient's adherence to use of the sensor is rarely ideal.
Current sensors are not as accurate as desirable, especially at the extremes of glucose measurement, and particularly in the hypoglycemic range.
What to look at in CGMS Readings
The first thing to look at in any interpretation or reading of CGMS is the validity and reliability of the data obtained.
The first day of the sensor requires a period of calibration, which may range from 2-10 hours. Interpretation of data should occur after the calibration time is over. In addition, most CGMS requires calibration with 2-4 capillary blood glucose readings a day. With fewer than this number of readings showing calibration with the capillary blood glucose reading, the rest of the readings for those 24 hours are suspect.
Next, it is important to see how many total readings occur during a day. With glucose measurements every 5 minutes by the CGMS, if there are no issues with the sensor there should be as many as 288 readings in a single 24 hours if there are no issues with the sensor. If the number of readings is much lower (less than 50%), it will impact the usability of the results. A patient log book of diet, activity, and events can be very helpful in interpretation of the CGMS data.
There are several ways to look at the actual data for interpretation. One approach is to assess and document the number of days that are available for review. Then:
Look at the percent time that is above and below the target range. Look for patterns in the data that occur on different days. Use of the actual tracing is helpful in doing this.
Next, look at all the tracings and glucose values overnight for determination of hypo- or hyperglycemia. This will assist in any need adjustments to the basal insulin or overnight basal insulin rates in the pump.
Now look at the glucose readings before and after meals. This will help manage mealtime boluses of insulin as well as basal rates of insulin pump during the day as well as help assess the impact of correction boluses of insulin.
If a logbook is available, compare the events listed in the logbook to the glucose tracings. A summary of the overall glucose patterns should be made. After that, treatment suggestions should be provided.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
Sign Up for Free e-newsletters
Psychiatry Advisor Articles
- Adjunctive Therapies for Bipolar Disorder Show Promise, Need More Evidence
- Predicting Treatment-Emergent Mania to Tailor Pharmacotherapy in Bipolar Disorder
- Abnormalities of Cortical Thickness in Bipolar Disorder With Auditory Hallucinations
- Prevalence of ADHD Relatively Stable Over Time Despite Increase in Diagnoses
- Prevalence of Major Depressive Disorder Remains High In US Population
- The Way to the Head May Be Through the Gut: Probiotics for Depression
- Suicide-Screening Toolkit Can Help Identify Youths at High Risk for Suicide
- Agoraphobia: An Evolving Understanding of Definitions and Treatment
- Parental Pressure to Diet Linked With Long-term Harm in Adolescents
- Does Access to Medical Cannabis Reduce Risk for Opioid Abuse?
- Peer-to-Peer Depression Awareness Program May Be Beneficial
- Examining Rates of Long-term Opioid Use in Youth With Psychiatric Disorders
- Mortality Rates for Substance Use Disorders, Intentional Injuries Vary Widely By Country
- Facial Emotion Recognition Differentiates Behavioral Variant Frontotemporal Dementia From MDD
- The Challenge of Helping Uninsured Patients While Protecting Practice Finances