Did Dexcom make the right G6 choice?


Not everyone is planning a quick or even medium speed transition to the G6. I took delivery of a new G4 Share system just two months ago. I’m good with the 9% G4 MARD and the 505 algorithm, at least until the G5/G6/new receiver/Medicare drama settles down.


@Shaw1 Okay this is great news! My G6 anxiety has gone down…a bit. :joy: Where did you come across this info? I’d like to read more. Please post a link if you have one!!


An automatic upgrade to the G6 would make me very unhappy.

I’m not looking forward to the bulky inserter or figuring out a 10 day work around.

Supposedly the G6 will be more accurate on the 1st day, but I don’t see any other real benefits to this system. I’m a little doubtful that no calibrations will be required, and even if they aren’t, I’ll probably still test my blood a couple times a day.

I’m not impressed, and I’m going to be quite annoyed if I’m forced to upgrade. Will there be an option to continue ordering the G5? The link you provided doesn’t discuss this.


I guess a call to Dexcom would be in order. I just got mine a few weeks ago, so I will be “in warranty” for a while before I am fore fed the upgrade. From what I read - anyone out of warranty will be auto-enrolled and will be forced to upgrade. All new people will get the G6, and anyone who goes out of warranty after the launch within 90 days. Looks like Dexcom does not like us using the sensors for more than the 7 days, and will force feed the update to maximize their margins.


This info is based upon discussions with Dexcom staff, and some of it is inferred. Obviously, no Dexcom employee would specifically indicate a way to bypass the intended use of their product. That said, the specifics about was is and isn’t necessary for proper function was easy to parse. The real test will be when folks get their hands on pre-release samples, which should happen within the next few weeks…


Dexcom may not intend it to happen, but it will happen. Just like there’s a way to hack transmitters to work beyond 3 months (with a battery replacement). The sensors themselves are too small (and cheap) to include a “countdown timer” to tell them to shutdown in 10 days. That’s why they use the SN pairing to do it. The transmitter/app will do the heavy lifting of blocking that sensor’s SN from working.

Yes, all sensor usage extensions are against the FDA guidelines, but that doesn’t bother me as long as the FDA isn’t paying for my medical supplies and the tech still functions accurately. My whole artificial pancreas system isn’t FDA approved, but it’s giving me the best control and management ease I’ve ever had, far better than any commercially available, FDA approved system I’ve used in the past.

Pre-release samples are going out for user testing in the next few weeks. We’ll know the answer before June.


I’m sure they will, and before June. Pre-release samples are going out soon. I’m not willing to give up my “Loop”.


I am British and so for the sake of balance I would point out that Dexcom need as much cash from you all as it can get.

My recollection of their trading statement to NYSE for last year was
Tuurnover 780 million
R & D 100 million


Sorry I will repeat always for getting about return key on fb Turnover 780 million dollars
R & D 100 million expenditure
Loss 64 million

Burning cash like a drunken sailor to get to be no 1


The financial groups obviously believe in them so far and stump up more cash when asked. However if Abbott or Medtronic manage to get ahead they are vulnerable. Personally I am with Libre/Blucon/xdrip+/Nightscout. If only Blucon were a bit better (was less heavy on my current batteries) life would be perfect. What might hit Dexcom would be if Abbot made its own transmitter with FDA approval. Apparently in their co-operation with Bigfoot they have the technology. When I asked an Abbott rep recently he said he thought they did not see it as a big enough market (only a rep not a spokesman). Only 400,000 type 1s in UK! And would not type 2s not be interested in seeing how they react to their meds and perhaps adjust their timing and dosage? Here in UK I understand most type 2s have to buy their own fingersticks if they want to check.


Aside from the hard cut-off, I am concerned that factory calibration won’t prove accurate for everyone. We all have seen a huge variance in how many days we can get from a sensor. I am one of the unlucky ones whose body chemistry won’t allow more than 12 good days, if that, and 3 weeks is impossible. How can factory calibration account for the huge variance in performance due to body chemistry differences? I really hope that calibration-free CGM (Dexcom or Libre or whoever) doesn’t turn out to be unusable due to body chemistry.


My new Tandem Tslim uses the Dexcom G5.


Looking at the history of the Tandem t:slim G4, it came out in September of 2015. The Dexcom G5 came out at about that time, too. I wonder when the last t:slim G4 was sold. If it was 2016, then a four-year warranty period would make it likely that Dexcom would support the G4 at least until 2020.

The Animas Vibe also used the G4 and didn’t exit the pump market until October 2017. The four year warranty on the last Vibe sold would run to 2021.

Of course there are no guarantees that Dexcom will produce the G4 for that duration, but Animas & Dexcom would need to accommodate those customers somehow.


I would not TRUST a system that is NO CALIBRATION to bring erroneous results back to near-TRUTH. But then I am extremely hesitant to allow a CGM to dictate insulin delivery. Last week’s example: On day 1 of a sensor I noticed the extremely RARE and unexpected Double Up and Down arrows. 10 hours after initial calibration my evening calibration showed a 73 point difference (sensor 199, meter 126). Studio showed that 3 hours earlier 2 sensor reading jumps upwards resulted in (what I believe was) sensor readings 60 points too high. WHAT would a closed loop system have done to me under a similar situation, too much insulin? And what control would a pump result in during 1 to 3+ hour periods of ??? on days 5, 6, 7?


Yes - especially in light of Bigfoot moving away from Dexcom because they wanted the no-calibration option NOW (as they get with the Libre…).


The G6 does allow for calibrations, should you choose to. I watched a video about that last nite…


I would not TRUST a system that is NO CALIBRATION to bring erroneous results back to near-TRUTH.

Bigfoot went with the Libre instead of the Dex because poor calibration is a risk they can’t control.

Studio showed that 3 hours earlier 2 sensor reading jumps upwards resulted in (what I believe was) sensor readings 60 points too high. WHAT would a closed loop system have done to me under a similar situation, too much insulin?

Do you ever take too big a bolus for your meal? Do you ever experience unexpected lows? The main protection I have from these kinds of events is that I wear a CGM and I have my low alert set at 85. As you point out, given a large-enough error in the CGM reading we can still get into trouble. So we’ll need two layers of protection for our closed-loop solutions. One is stronger requirements for CGM accuracy: it’s not good enough that 80% of the readings are within 20% of the true value if the other CGM readings are allowed to be arbitrarily bad. The FDA has taken steps to strengthen the accuracy requirements with their new category of medical device intended for closed looping, called “Integrated CGM.” https://www.accessdata.fda.gov/cdrh_docs/pdf17/DEN170088.pdf The Dex G6 has been approved as an iCGM device. The second layer of defense is the closed loop’s critical examination of CGM data. If the algorithm sees too drastic a change in CGM values it should become skeptical and involve a human. To me that doesn’t mean silently reverting to default basals, it means raising an alert.

And what control would a pump result in during 1 to 3+ hour periods of ??? on days 5, 6, 7?

Again, the closed-loop algorithm should handle this gracefully. If your BG was smoothly trending they may be able to coast through the blind spot. If the BG was moving fast or showing signs of trend change, or if you ask for a bolus to cover carbs, the algorithm should involve you so that you can compensate for the CGM’s failure. And maybe the CGM helpline folks would tell you to put in an new sensor and they’d replace the failed one under warranty if you are using an FDA-approved closed loop system.


My wife works for Dexcom and I’m a very happy G5 user. You are correct. I’m fine with calibrating, so this info told me that Dexcom is making all potential customers happy. We’ll soon see, but it seems that the no calibration feature is targeted to young kids.


I totally agree. I can restart 2-3 times.
Im using dexcom and the vibe. My warrantly expires this month… I’ve been told to wait for the tech developments that will happen this summer. I know I don’t want Medtronic. I’m looking at Tslimx2.
I just don’t know. I don’t think I would “not calibrate” whether it requires is or not.


Other than the Vibe warranty running out, is there any reason you need to make a switch now? The t:slim X2 looks like an attractive option to me but if you wait a year you will likely have more choices.

Does the Vibe permit you to restart your G4 sensor like G4/traditional receiver can?