Can Anyone Explain this G5 Phenomenon?

Every morning my G5 CGM drops like a stone despite my Contour Next telling me it is not dropping at all.

This morning I woke up at 7:00 AM and my G5 was 104. I checked it with my Contour Next meter and it was 99. Great. An hour later at 8:00 AM my G5 was 63 (it showed it had been dropping since 7:00 AM) whereas my Contour Next was virtually unchanged at 97.

              Next	   G5

7:03 AM 99 104
7:58 AM 97 63

This happens virtually every morning. Before I figured this our I was taking carbs at 8:00 AM when I only looked at my G5. I am happy that my BG is really not dropping but I can’t understand what it happening to the G5 and why. Does anyone have an explanation?


1 Like

I cant explain. How many days in a row did this happen? Does it happen everyday? Does your G5 sound a low warning? If you ignore it, how long before its back on track again?

I’m calling this sensor noise.

Lots of fun new questions on Tu lately.

Yes, it happens every day and has been happening for months or longer. My G5 will sound an alert if it goes lower than the alert. Sometimes when BG is high in the morning it gets lost in the high readings. Its weird but would like to think there is some explanation. If not, that’s ok as long as I now know that the first response is to test BG an not eat.

I can’t trust my G5 either and I would like to know if it is something I am doing. This drives me crazy. My G5 drops too low at different times during the day.

I got by for almost a week with this not happening too often, but some. After I restarted my sensor it worked well for about a day without too much discrepancy, but right after dinner my G5 read 70 and my trusty old glucose monitor read 170. I calibrated the G5 again.

I am having to do a ton of finger sticks. Am I doing something wrong? I suppose I should change sensors and see if it still happens?
In the night I was getting urgent lows when my finger stick read 70. Ok, that is fair I suppose, so I ate something even though I would have preferred to not be woken at 70. I had the alarms shut off. Then the G5 kept bugging me inaccurately for the next 40 minutes. I really dislike this. I had it in another room folded up in a wash cloth and I could still hear it. I guess I need to just turn the whole unit off at night and get by the way i have been for the last 60 yrs. I love this thing when it is accurate. I just wish I could trust it to be more accurate most of the time.

When this happened, was it a precipitous drop over a short period? If so you could be experiencing a compression low which generally happens when you are laying on the sensor. The density of the tissue around the sensor increases making the readings inaccurate. When you get up it generally recovers in 30 minutes or less. This can repeat itself if you lay back down in the same position.

This doesn’t happen all the time (maybe 1 out of 20 sensors for me when I was using the G5, little more often with the G6), but it’s really frustrating. I try to place the sensor closer to my belly or my back so i avoid locations where I’m likely to lie directly on the sensor.

The dropped happened two hrs after I went to bed. Glucose level did fall, but not as much as the G5 indicated.

I am wearing my G5 about 1 1/2 to the left of my belly button. Can’t sleep on my back. :blush:

I have a long history of odd, but consistent CGM behavior in the morning. Almost every day when I wake up, my CGM reads almost exactly 20 points higher than my Contour meter. I don’t calibrate that difference because if I do, the CGM will tank and read much lower than my actual BG for the rest of the morning. So I ignore it and within an hour the two readings merge. Sometimes the CGM starts to drop once I start moving around and sometimes the BG per the meter starts to rise. My guess is that it is a circulation issue although the behavior is the opposite of compression lows.

If I calibrate the G5 when my meter is much lower, I will sometimes get the crazy (and wrong) kind of drop that you mention. Sometimes I will calibrate by splitting the difference and amazingly I get very good results doing that–better than putting in the actual low number.

If I were truly a scientist, I would start using my meter religiously through the night to try to figure out when the divergence in numbers happens. Maybe interstitial fluid and blood glucose are affected difference by dawn phenomenon hormones.

All of this stuff sure is complicated…

1 Like

No. My wife has the identical problem with G5, but she is on acetaminophen every day (a component of her pain meds). I don’t take that stuff and have no problem with bad data in the mornings. I keep telling her it MIGHT be the med, but she argues with me. I have given up trying to figure out why she has trouble in the AM but I don’t.

I love my G5. Have had it for 18 months and can’t even live without it for the two hour warm-up. But it does go haywire at times and it is not compression. I went to bed last night at 96 per the G5 and Contour Link. Within 30 minutes G5 dropped from 96 to 60. That was easy to ignore as I had tested blood 1 hr ago. But this morning it happened again (like most mornings). Although it is good most of the time, I still prick myself almost 8 times a day especially since it is so unreliable at times. I don’t think we’re doing anything wrong. It’s just the way it is.

After reading more here, I think it was time to change my sensor. I assumed I would be able to restart a couple of times but I can’t.
The G5 isn’t as accurate as I hoped it would be, so I hope that the G6 will be more perfected. When it works well it is great, but without several finger pokes a day, it just can’t be trusted. Small differences in glucose levels are of course expected, but large ones aren’t helpful.

1 Like

There are times when the sensor data won’t match, such as when bg’s are moving just moderately and you expect the data to match a finger stick. Try checking the CGM reading 20 minutes later and see if it more closely matches the finger stick from earlier.

If you calibrate when bg’s are moving, even if just moderately, you will introduce errors. If you calibrate more than a couple times a day, chances are extremely high that you will make the G5 inaccurate. The exception to that is when a sensor needs to be “beat” into submission, which is to calibrate 3 times, 15 minutes apart, or “double calibrate” with from a single finger stick reading.

I’ve been using the system for 2 years, am well hydrated, calibrate seldom, have it in my abdomen, and have great results, except on day one. that’s why I suggest “soaking a sensor” for a day, as I’ve mentioned a million times.

  1. Never calibrate unless the the arrow next to BG has been in horizontal position for at least 3 readings (Dots) That would mean a horizontal arrow for at least 15 minutes.

  2. Don’t calibrate more than 2 times in a 24 hour period.

  3. Don’t restart sensor after 7 days - The latest batch of sensors tend to go wonky by day 8 or 9.

  4. If side sleeping, apply sensor between belly button and xiphoid process, no more than 3/4 inch off of that center line to avoid compression lows.

  5. Try to keep BG’s above 70. I believe that there is something in the algorithm that is built in to ask pump to shut off glucose when often below 70 and then CGM tanks reading to like 48 even if you fingerstick and show 78. The problem here is that often individuals will re-calibrate when that happens and that is the worst thing you can do. It just exacerbates the problem. The quickest way to combat this is to eat a set number of carbs and bolus 5 minutes after eating those carbs. Within 15 minutes number will climb back to normal and then within 1 hour insulin will hold you at normal. There are other ways but that is the easiest. If you just ignore it, eventually the CGM will correct itself after a few hours but the low alarm will drive you nuts.

I believe this is just all part of the learning process of how to best use Dexcom CGM and dealing with the hardware.


For me, I almost never got beyond day 9 (as far back as 2 years ago) so I stopped restarting. I’ve got plenty of sensors so there is no rational reason for me to do restarts. But it was nice in the beginning, to create a bit of a stash.

1 Like

Totally agree - I used to be able to get 3 weeks, but for the past several months day 9 has been the max and just not worth it as I am on Medicare and have built up my stash.

It is interesting I see very few comments on the fact that if you try to micro-manage the Dexcom CGM before you really understand it and all of its quirks and your body quirks as well, the system will quickly drive you nuts to the point you just want to throw it out. I think a lot of patients discontinue use of the Dexcom because they don’t trust it, which is the result of lack of patience and learning and trying to micro-manage the system thinking they will get better results when actually the opposite is what occurs.

And that would be a shame, as it works very well. Heck I couldn’t get the Enlites to work well in a year’s vain attempt, but the G5 is remarkable accurate, so I just figure if my body is that temperamental towards sensors, than just about anyone should also be able to get great results from a G5. I think the jury is still out on the G6. Too many people report it doesn’t go 10 days, or it has no data, or it is inaccurate, or it falls off. I’m glad I’ve been advised that I can stay on the G5 when the Medicare G6 roll-out begins later this year. (and “yes”, to all you here who have great luck with the G6, I know there are also patients who state the G6 is fantastic)

I have built up a substantial reserve of G5 sensors. For me they last an average of 3 weeks and have gone to six weeks. Mine seem to be more accurate after weeks of use so I wouldn’t throw them out unless there is a sign they are becoming inaccurate. I used a brand new sensor by mistake the last time and it only lasted a week. Why do you believe they new G5 sensors only last 7-10 days? I thought the last one was an abnormally but since it was a new G5 sensor I wonder if they are manufacturing them diffidently?

1 Like

Reread what I said. I didn’t say what you thought.

Every time I’ve had this problem it is usually a haywire transmitter or receiver. The Dexcom g5 is Notorious for this. I have since changed to the Dexcom g6 wich is a lot better. I would call Dexcom and get a replacement system as this shouldn’t be happening.

I’ve had one receiver that stopped receiving data, out of the last 2 years. But I asked for quite a few receiver replacements for other issues such as funky buttons, or failure of the screen to turn on for roughly 15 minutes after pressing a button.

For the first 7 days I never recalibrated more than every 12 hrs, but I didn’t know I was supposed to pay attention to the arrows.

I have never thought of 70 as low, so it will be hard to me to adjust to never going below 70, but I will try.