Dexcom G7 - How it works
This page is not giving you medical advice or treatment advice, and that is very much on purpose! I am not a medical professional, and even if I were, I don’t know your personal circumstances. If you need medical advice, consult your physician! If, at any time, you have reason to doubt your G7 readings, use one or more finger-stick measurements to confirm your G7 readings. Base your treatment decisions (bolus, etc.) on the finger stick reading, and in accordance with your physician’s instructions.
For users with an insulin pump controlled by the G7: Until you have reached a full understanding of your G7’s behavior and when, in your particular case, its readings can be trusted, you may need to double-check, as frequently as necessary to give you confidence, with finger stick readings. Read the page on reading comparisons for important information on interpreting the differences.
Pages
- Introduction
- How it works
- Accuracy and Calibration
- Comparison with finger stick readings
- My Sensor Failed
- Known Issues
- Contacting Dexcom
Dexcom G7 - How it works
A slightly better understanding of how the device works will aid in understanding the other information in this series.
Insertion
The G7 is a sensor that is applied to the body using a spring-loaded applicator. During application, a needle inserts a filament, emanating from the bottom of the sensor, into the subcutaneous tissue (just under the skin). The needle part retracts into the spent applicator, but the filament remains. The bottom of the sensor is covered with a cloth-like tissue that adheres to the sensor body with some adhesive. There is a small hole through which the filament passes. The other side of the “cloth” also contains adhesive to help the sensor stick to the skin.
Dexcom wants you to cover the whole application with what they call an “overpatch”. The one provided in the box looks like a rounded corner green “tape” with a hole in the middle. That hole is for the sensor. The bottom of the overpatch has adhesive that makes it stick to the skin and to the small margin of cloth (mentioned above) around the sensor (about 1/8 inch wide), and white-ish. This is meant for water protection.
That is the theory of application. In practice, quite a few users have problems with it, and there are also frequent reports of the overpatch failing to stay in place. I’ll address these issues later.
Measuring
The sensor’s filament, embedded in the subcutaneous tissue, is surrounded by what is called interstitial fluid (ISF), a fluid inside the interstitium. This fluid contains glucose that comes from the blood through a process called “glucose transfer.” Glucose passes through the permeable capillary walls (capillary = tiny blood vessels) from areas of high concentration (blood) to lower concentration (ISF) by a process called “passive diffusion.”
Diffusion occurs when molecules flow from an area of high concentration to an area of low concentration, and it essentially stops when concentrations are equal. The diffusion process takes time: diffusion rates can be higher when the concentration difference between the two sides is greater, but there are physical and chemical limits to how fast it can proceed. Consequently, there is a 5-25 minute delay (time lag) between changes in blood glucose levels and their appearance in the interstitial fluid.
Under stable conditions, interstitial fluid glucose levels are very similar to blood glucose levels, which is the principal reason why a device like the G7 can work: the chemical interaction between the ISF and a specialized electrochemical enzyme coating on the filament produces an electrical signal. This signal is measured inside the sensor body and “transformed” into a digital measurement.
As mentioned, when blood glucose levels are (rapidly) changing, the changes do not show up in the ISF right away. This does not have to be a problem, as long as you, or your pump if you use one, understand this time lag.
Sensitivity to interfering substances
Prior generations of Dexcom sensors were quite sensitive to Tylenol (acetaminophen) and ascorbic acid (Vitamin C), and their readings could fairly easily be affected. This has improved with the G7:12
- Doses of Tylenol (acetaminophen) lower than 1,000 mg every 6 hours should have no effect. Larger doses (which are questionable from a health perspective anyway) may cause artificially higher readings.
- Hydroxyurea is a medication used in the treatment of diseases, including cancer and blood disorders; it is known to interfere with sensor readings. Sensor readings will be artificially high. How much depends on your medication dosage. Dexcom recommends not using G7 measurements for treatment decisions in this case.
- Ascorbic acid (Vitamin C) does not significantly affect G7 readings, unlike some other CGM systems.
All cases of falsely high readings may lead to incorrect insulin dosing; in extreme cases, it can cause (severe) hypoglycemia!
Communicating
During insertion, the sensor is effectively separated from a powerful magnet embedded in the applicator. Contact with the magnet disables the sensor’s circuitry, preventing it from draining the battery during storage and transport. Right after application, with the magnetic contact now removed, this circuitry is activated, and the initial power-up puts the sensor in “startup” mode.
The startup mode does two things that I know of. The first is to start measurements and internally calibrate to the signal. Dexcom calls this process “warm up,” and during this period, measurements are not available to you. After the 30-minute warm-up, measurements are available, but the internal calibration may not be complete.
The second is to establish communication with an outside and separate receiver. Measurements, taken inside the sensor, need to be communicated to a device that can display them. To this end, the sensor communicates via Bluetooth. This is a short-range radio signal for communicating digital information, and requires “pairing” between the sender (the sensor) and the receiver (a Dexcom display or receiver, your smartphone, or your smart watch).
Pairing
The pairing process involves the receiver agreeing to a specific, unique code assigned to the sensor (found on the outside of the applicator). Just in case multiple sensors offer to pair at the same time, this lets your receiver or phone immediately select the correct one.
In case you might be wondering why or how more than one sensor could be offering to pair: One scenario is two persons in the same household, on very similar sensor schedules, both inserting a new one about the same time. Another, and perhaps more likely, scenario I have encountered myself is when you remove an almost-expired sensor just before applying a new one. It is possible that the sensor just removed is now offering pairing, as it has not quite reached the end of its life. It may do that to allow you to re-pair if, for some reason, pairing has been lost. I am not sure this will always happen, but it sure happened to me twice in two years. The code on the applicator solves the problem.
Your smartphone can recognize this unique code by scanning the QR code on the applicator or by entering the numerical code. I do not have a Dexcom receiver, but I imagine you need to provide the code for it as well. The Bluetooth connection supports up to 2 connections, allowing both your smartphone and your smartwatch to connect. If you configure your smart watch connection, called “Direct to Watch”, you can see readings on your watch, even when your phone is not nearby. If not configured, your watch will display readings by getting them directly from your phone.
Bluetooth
While the Bluetooth connection enables continuous communication, it would drain the sensor’s battery too quickly, making a 10-day life expectancy impossible. During the initial “pairing” phase, the sensor communicates somewhat frequently, but once paired, it provides sensor readings to the receiver only every 5 minutes or so. For the periods in between, the Bluetooth radio circuitry is powered down or in a low-power mode. The sensor itself may, or may not, go to “sleep” for the entire period. I do not have precise information on that.
Additionally, Bluetooth radio signals (2.4 GHz) have limited penetration through the human body because water strongly absorbs that signal, and the body contains a lot of water. Dexcom advises placing the sensor on the back of the arm, but your phone and watch will typically be in front of you or to the side. That means that the signal may have to travel through the arm, or possibly other parts of your torso, in order to reach the receiver. This explains why people experience more reception problems when the receiver is on the opposite side of the body from the sensor.
Some WiFi radio bands also operate on the 2.4 GHz frequency, but the communication protocol is entirely different from Bluetooth. The signals generally do not interfere.
Energy efficiency and lifespan
Making and keeping the communication as energy-efficient as possible is key to the sensor’s lifespan. The sensor’s small form factor also limits the battery size and antenna design (a radio signal requires an antenna to transmit). The result of all that is a limited effective communication range, often much shorter than what you might be used to with other Bluetooth devices, such as headphones, and, worse, it can be even shorter depending on positioning, as described above.
Dexcom recently introduced a 15-day version of the G7. Not much, if anything, is known about internal design differences, but it stands to reason that the extended lifetime was achieved by a combination of internal space optimization (allowing a physically larger battery) and firmware code optimization within the sensor.
Receiver functions
The sensor is a simple device, and once a measurement has been transmitted, it is kept inside the sensor for 24 hours. If the receiver or mobile phone application has a data gap (was out of range for a while), it can request up to the last 24 hours’ worth of measurements to fill it.
That same 24 hours’ worth of data can also be used by the sensor for data smoothing and trend analysis (as shown by the arrow in your mobile application).
It is the receiver’s function to capture and store measurements over time, present graphs, and communicate measurements to the Cloud so they can be shared with others or your physician. Of course, there are also functions for setup and calibration.
Mobile application
A mobile application is available for both Android and iOS devices. The application is used for:
- Pairing a newly applied sensor
- Receiving readings and storing them in the app, and optionally in the Cloud
- Displaying a graph of readings, available in 3, 6, 12, and 24 hr overviews. The graph also indicates (with a gray band) the range between configured low and high target glucose levels
- Displaying alerts and controlling their settings, with separate options for “primary” (daytime) and “night” periods. Night mode has a configurable schedule, and anytime this schedule is not active is considered “primary” mode
- Displaying “Clarity” data
- Available insights for the last 3, 7, 14, 30, and 90 days
- Shows the average glucose level over the period
- Shows GMI (Glucose Management Indicator) over the period
- Shows a stacked bar graph for “time in range”
- Provide an optional “Clarity Clinic” feature, where you can configure your data to be shared with your physician
Glucose Management Indicator
Dexcom has invented a measure called the Glucose Management Indicator (GMI), which is meant to give you an idea of how well your sugar levels are managed. It is calculated using your sensor data over the period selected in the Clarity section.
It seems that, in many ways, this is Dexcom’s attempt to come up with a number similar to your HA1C level, but, unlike HA1C levels, which “capture” about 3 months’ worth of information, this is computed over a shorter (chosen) period. Its value, like HA1C, is expressed in a percentage.
Many people confuse the GMI and the HA1C. Since HA1C reflects a 3-month period, you should only look at the 90-day GMI if you want to compare. Don’t be surprised if the HA1C that’s measured isn’t the same as the 90-day GMI you saw on that same day. My results routinely show my HA1C about 0.7 lower than the GMI!
With that understanding, you should learn how your GMI numbers behave for you when “things are good” and only compare current readings to determine how well you are doing compared to a desired outcome. Over time, as you get your HA1C measured, you might notice a correlation between GMI over 90 days, taken just before your lab work, and your HA1C. This correlation is somewhat tenuous at best and is probably highly individual for you, so take that for whatever it is worth.