At best, it is pretty inexact. In fact, users are cautioned not to use it as the basis for deciding to take extra insulin to move the BG lower, or to consume some glucose to raise it. As a result of that inability to accurately measure BG by itself, the user is required to calibrate the CGM 3 to 4 times a day with a traditional blood glucose meter. Normally (and ideally) the numbers you get from the BG meter and the CGM are fairly close. If not, trying to get the CGM to work right is going to be a problem. For the last couple of days, I've had problems calibrating the CGM unlike anything I've seen in the 4+ years that I've been using them.
On Monday, I inserted a new CGM sensor. I did the calibration and things seemed fine. A few hours later, my insulin pump issued an alarm telling me to calibrate the CGM. I did that. That is when things got ugly.
Before I calibrate, I take a look at the CGM reading. Then I run the BG test on my meter. Before using the meter reading to calibrate the CGM, they need to be fairly close. Long story short - the sensor that I had died after barely 7 hours. It is advertised to last for 6 days. I called Medtronic to figure out what I was doing wrong and how to not have this happen again.
The representative was really helpful. If I had one problem, it is that I might be to thin to use the CGM. It is designed to work where the sensing filament resides in fatty tissue. I don't have very much of that. In fact, when I told the gentleman about that, he thought that might be the culprit. Just a week earlier, my Endocrinologist and a Certified Diabetes Educator she works with went over this very issue with me. We tried to figure ways to make the CGM work with what little fat I had. When I put in the sensor in question, I followed their guidance. But here I was with a sensor that could not be properly calibrated, and as a result the insulin pump software decided it must be defective and advised me to change it.
One thing the representative advised was to do a calculation to determine if the data from the meter and the CGM are close enough for a calibration to be accepted by the algorithm. I was to navigate through the pump menus and find the CGM sensor's ISIG. According to the user manual, ISIG is, "Sensor values that indicate whether the transmitter electronics in a sensor-augmented system are working properly. If you see an ISIG value that is less than 24.00 nA or more than 29.00 nA, call for product assistance." Well, I'd already been doing that before calibration.
But here is a new wrinkle. The representative advised me to write down the ISIG. Then I was to run the BG test on the meter. After I did that, I was to divide the BG by the ISIG and if the result was between 3 and 8.33, then I could calibrate.
Fantastic. Now I have more arithmetic to do.
Fine. That is what I would do. The call ended, I went home and that night I replaced the sensor. I let it stew in my juices overnight (one of the informal tips that some folks are convince are helpful). The next morning I fired it up and then calibrated it at the appropriate time. So far, so good.
Nope. Later that day, I saw this when I went to calibrate the CGM.
I restarted insulin delivery and again called Medtronic. This time the experience wasn't quite as positive. The customer service representative (CSR) gave me guidance that was frankly baffling.
For one, she said that even though my CGM and BG values differed by a factor of four, I should have calibrated it then. She then looked through my calibration and CGM data and came to an interesting conclusion: to get accurate calibrations, I should calibrate 1 to 2 hours before each meal. OK, fine. But let's put that into practical terms.
It is mandatory to do a BG test every time prior to having a meal. It is advisable to do another one 2 hours after a meal. Three meals equals 6 needle sticks. It is also advisable to do another one before bedtime. That is 7. It is also highly recommended to run a BG each time I get behind the wheel to drive. That is minimum of twice a day for me. We now up to 9 (NINE) finger sticks a day. And now this person is telling me that to get my CGM to work properly I need to add another 3 BG tests. So that adds up to 12 blood tests a day. But I only have 10 fingers. I'm usually up 16 hours a day, so now I have to test my blood every 1 hour and 20 minutes. At this point I was pretty furious - both at her lack of useful help and her adamant advice that was utterly devoid of any sense of reality.
In fact, what made me hot is that this device is supposed to make life better by helping me to avoid trouble caused by diabetes. What it was doing was not helping me to regain something similar to a normal life, but was in fact drawing me deeper into a psychological slavery to this disease. The devices are designed to help me, not the other way around.
I'm still using the CGM, but a large part of me wonders why. I don't trust it. Even when it was not working well, it would alert me to the need to take a BG reading. Now I am not sure I can even rely on it to do that. Below is what I got on that first day the sensor went bad. Not good. Not good at all.