The first 24 hours or so of a new Dexcom G6 sensor are expected to be a little off, or sometimes pretty far off. I’m guessing all users have experienced this endearing habit of the G6. So far the sensors seem to settle in after 24 hours. I’ve also had good luck with a 12-24 hour presoak of the sensor before starting it.
This morning for various reasons I was not able to do a presoak on a new G6 sensor. When I’ve checked against a fingerstick the readings are not too far off considering this is still the first 12 hours. However The sensor readings are much jumpier than I’ve seen on previous sensors. Has anybody had a graph like this that jumps all over but (hopefully) then settles down and becomes a good sensor lasting 10 days? Or is this jumpiness evidence of a sensor that is doomed from the start?
My current sensor, on my abdomen, was pretty jumpy in the first 24 hours but mostly has settled down. I only was able to do a 3-hour presoak and am in no code mode. I seem to have better results when I don’t use my abdomen, go figure.
Yeah, the first 12 hours ca be pretty bad, but for me I will say recently I found the back of the arm is much more accurate out of the gate than my stomach. So I’ll probably use that location more.
Thanks for the feedback… the jumpiness got me a little spooked as I had not seen that before on starting. Lucky me I guess, now the sensor seems to be settling in OK.
I am regularly using upper arm sensor placements but have still had all kinds of inconsistencies with G6, ranging from a batch of sensors that only lasted 8-9 days to sensors that were stellar for 15-20 days.
Mine have been real shoddy for the same lot number. Some require calibrations every 12 hours and some don’t. It’s pretty hit or miss for me, which I only share not to complain but to say that I, too, get inconsistent experiences within the same box.
I’m on my second G6 sensor, and I have to say I am not impressed. They’ve been accurate for a few hours each day or night and spend the rest of their time in la-la land. Calibrating – and I try not to do it too much but temper gets the better of me – provides only a brief experience of reality.
This has been one of the wonkiest first day of a sensor for me, and I started on Dexcom back in June. I have no idea why it’s acting like this - I have always coded and never calibrated and never had problems. Also, it’s still about 40-45 points away from my finger sticks… which is troublesome when you think you’re healthy at 115 and dropping but really you’re at 152.
Yes, same for me. What I usually experience is the first hour is not bad, less than 20 mg/dL off. Then it may steadily get worse throughout the day. If it is far off, I have no choice but to calibrate, but I try to hold off on that based on what most experience, that is, that calibrating can make it more inaccurate! By the 2nd day it is pretty good. I start to get inaccurate readings again on the last couple of days, but not as bad as the first. Again, sometimes first and last few are totally within reasonable distance. I do usually soak it for a few hours prior to changing sensors. Then after I apply the transmitter to the new sensor, I wait 10-20 minutes before actually tapping to start new sensor. This was recommended by Dexcom and I find it to be the most helpful of all tips, including soaking the sensor.
I may have to try these hacks, as this last sensor took roughly 14 hours to snap into rhythm with me. It’s finally within the accuracy ranges, thank god. But last night was the worst - kept getting woken up by false urgent low alarms, and the darn thing just wouldn’t calibrate itself. Even still this morning after breakfast I had to keep finger sticking because even the trend arrows were inaccurate. (I.E. I went from 115 to 120 with a downward arrow lol)
I am going to try and restart this sensor after my 10 days are up, because I’m pretty sure ADS hasn’t notified me I owe money and I ordered my sensor refill a little over a week or so ago.
Those false urgent lows could be “compression” lows which are common if you happen to lay on the sensor at night! I get them frequently. I don’t think there is a hack for it other than don’t lie on the sensor!!! They usually resolve once you turn or are no longer compressing the sensor!
I purposely put my pod and my sensor on the same side of my body so I wouldn’t lay on them (lmao) and so far I think I’ve been successful. Pretty sure these were just calibration false lows, I get these often when my sensor exits warmups.
I always code and never calibrate, I probably chose the wrong word there.
I meant the new sensor was probably just getting used to my IF like it usually does… and the first day I get wonkiness and inaccurate urgent lows lol.
I found it wasn’t the sensor after all on this one, I got a notification a week ago that my transmitter battery was getting low. So I figured I’m due to change it anyway - popped out the old one with a credit card, inserted the new one, restarted my sensor and I’m doing much better now with accuracy.
It’s possible. Or maybe if you put the old transmitter back in and restarted it would have read ok because the sensor was just settling down right around that time. I once had a sensor that took 3 days to settle, then it gave me good readings to the end.
I read an explanation of the bogus urgent lows: inserting the sensor wire causes tissue damage, and the subsequent tissue repair is powered by glucose; glucose that is taken directly from the interstitial fluid right there around the sensor wire. Which leads to a local deficit of glucose, i.e., a false low reading. I vaguely think that the dex algorithm tries to compensate for this, but they would rather say you are low when you are not than saying that you are high when you are not. Because if they falsely say you are 140 when you are 100, you may take enough insulin to drive you down to a falsely indicated 100 which would be an actual 60. It’s not a big problem if it reads 100 when you actually are 140, because hanging around at 140 for several hours without knowing it won’t hurt you. But hanging around at 60 for several hours without knowing it could lead to a car crash or other accident.
I sometimes get good G6 readings from the getgo, but my current sensors are the worst I have had in years. Lot number ends in 9790. They are all failing in four to six days, either on arm or abdomen. Tandem is replacing them, and the online form is great for that, but here is the jumpy trace from tonight, with one low at 44 when my meter said 72! I am using the Bluejay watch as the collector. RandysScreen|314x500
How old is the sensor that made that graph? That is how mine look before they settle, and before they fail. Fortunately I can still see those jumps because of loop. The Dex app rewrites history to show a nice smooth line, which improves their MARD statistics for the benefit of their marketing department, but at the expense of removing important information that helps the person wearing the sensor to evaluate the trustworthiness of the current CGM readings.
That’s a big error, but in my experience they pretty consistently make the error go in the right direction, i.e., it would lead you to raise your BG, which is the safe direction. I’m guessing that when their algorithm sees the jitter, they add a proportional safety margin and show us a BG number lower than the raw measurement.
@bkh, is this a fact? I have not seen my older measurements change when they get into Clarity, even when I compare my screenshots, or my dad’s Nightwatch.
I seen that same ‘wacky’ plot just this morning… But in my case it was after mowing the lawn for a couple hours in a warm and humid climate. Now 6 hours after that initial wacky pattern, it is still moving up and down at will.
I believe that I have seen the pattern you posted on a few CGM’s over time, but does not seem to be any type of a regular pattern and ‘usually’ straightens itself out over time.
