CGM is reading low values at night

Summary

Continuous glucose monitors (CGMs) provide important metrics about the health of study participants. However, CGMs may falsely read low glucose values and issue "low glucose" alerts when they are being depressed, commonly known as "compression lows." This articles explores some of the common causes and solutions.

 

Details

Problem

When pressure is applied to a CGM sensor, such as when a participant sleeps on their side and lies on the sensor, the CGM may read a false low value. This is often evident by a steep, sudden drop in the continuous glucose graph. This occurs because CGM sensors do not check blood, but rather interstitial fluids (fluids between cells), and the volume of such fluids decreases when pressure is applied to that area, causing the sensor's needle to detect less fluid and therefore less glucose.

Uploaded Image (Thumbnail)

 

Solution

  1. Participants should be made aware of the potential for compression lows and asked to avoid sleeping on the sensor. If possible - and depending on the manufacturer's recommendations - apply the sensor in a place where the participant is unlikely to lie on it.
  2. Clinical guidelines for managing hypoglycemia, or for when to verify an alert with a finger stick, are outside the scope of this article and are the responsibility of the study team and/or the participant's health care providers.

 

Notes

  • Researchers should apply data analysis techniques to differentiate between compression lows, true fast lows, and reactive hypoglycemia. Generally speaking, a fast sudden drop when values are otherwise steady or slowly changing, in the absence of exercise and fast-acting insulin, would point to a compression low. In contrast, reactive hypoglycemia would show a fast increase, high peak, fast drop, and slow rise over a period of approximately one hour. Confirmed compression lows should be excluded from aggregate measures such as average, minimum, and standard deviation, but should not be excluded from total or average wear time (since participants are still wearing the device in compliance with study protocols).

 

Resources

 

About the Author

Gabriel Mongefranco is a Mobile Data Architect at the University of Michigan Eisenberg Family Depression Center. Gabriel has over a decade of experience in data analytics, dashboard design, automation, back end software development, database design, middleware and API architecture, and technical writing.

 |  | 

 

Print Article

Related Articles (3)

How do you record naps in an Apple watch? Possible solutions to Apple Watch not tracking sleep outside of a sleep schedule.
This article summarizes how to hide the sleep tile within the Fitbit app.
This article lists some of the common glucose meters being utilized by U-M researchers, as well as upcoming devices, as of the date of this writing.