What I’ve Learned about my CGM

I’ve been using a Medtronic Minimed continuous glucose monitor (CGM) for a little more than a year and a half now, and here’s what I’ve learned about it.

  • Having CGM is the greatest thing. It’s a decision-support tool, a dashboard, and a security blanket. I’ve learned so much about my diabetes from it, and it’s provided the data I needed to make some important changes to my self-management. I wish everyone had it.
  • The BG values it reports are delayed by 15 minutes. This is a fact of life. The CGM looks at “interstitial glucose” between cells (ISIG) instead of blood glucose, and there’s a lag time between the two. Don’t expect the BG numbers from your meter and your CGM match unless your having very stable BGs.
  • The most important factor in getting believable CGM readings is calibration. Wait until your interstitial glucose (ISIG) is the most stable before calibrating.
  • If my BG is on the move but the system says I need to calibrate it, I’ll let it go past the “Meter BG Now” deadline and look at the “Sensor ISIG” values on the CGM information screen. When this value isn’t changing much, that’s when I’ll calibrate.
  • Lots of things affect ISIG values and therefore CGM accuracy: the rate of change of BGs, the level of hydration, and the wetness of the sensor (both too wet and not wet enough).
  • I find that if I keep my sensor and transmitter covered, I get a longer sensor life and better readings.
  • Tegaderm film works the best for me. Unlike IV-3000, it doesn’t come off when I’m swimming or sweating a lot from exercise; it doesn’t leave behind a lot of residue; and it’s stretchy enough to cling nicely and create a good barrier.
  • Hydration matters a whole lot to accuracy. If you’re dehydrated, the sensor will read incorrectly, or perhaps not at all.
  • Don’t calibrate first thing in the morning. There’s too much going on. Almost everyone is slightly dehydrated when they wake up, and I’ve noticed that the very act of waking up and getting moving causes my CGM sensor to give inaccurate readings for about a half hour.
  • Don’t calibrate right before or after exercise. For me, my blood sugar usually is on the move during the first bit of a workout and immediately after it. Plus, the chance of BG swings is greater while exercising, which is bad news for calibration.
  • I find that sometimes I can count on my CGM values when I exercise, and other times I can’t. The lag is a little less during exercise, but my hydration is also different so it’s less accurate. Sometimes it catches big BG swings, and sometimes it doesn’t.
  • Swimming: I find that I can swim for at least an hour without affecting the sensor or transmitter. But I also find when I swim three or more times in a week, the off-label lifespan of each sensor is noticeably shorter.
  • Sensor life: The longest-lasting sensor I’ve used that produced reliable BG estimates is 12 days, which is not too bad for something that’s only supposed to work for three. (Just tell the CGM that you’re starting a “new” sensor when it says “Sensor End,” even though it’s the same one and not new at all.) Because my transmitter battery requires a quick recharge after 6 days, I have to very carefully peel off the Tegaderm I put over the sensor and transmitter, give the transmitter a charge for 15-30 minutes, and then reattach it and apply more Tegaderm.
  • Occasionally, the sensor values will just drop out after the three days; they might come back to match my BG, and they might not.
  • When the CGM graph gets really noisy, the sensor is dying.
  • Sensor freshness matters. A sensor getting close to its expiration date won’t last as long and seems to give less accurate readings.
  • Treat the number after finding a “Lost Sensor” with some skepticism. I find I often need to recalibrate 30-60 minutes after reacquiring the sensor.
  • To avoid “CAL ERROR” messages, I don’t calibrate if the CGM and BG meter values are very different.
  • The Medtronic CGM does some curve fitting, so it often doesn’t “snap” the CGM values to the calibration value when they’re very different. I hate that. Entering another, new BG value usually gets them closer, though.
  • Post meal, a period of flat BG values followed by a rise 1.5-2 hours later almost always means that the meal bolus was too small. (There are a lot of other trends you can spot, but this one was the most consistent.)

Do you use CGM? What have you noticed?

This entry was posted in Data-betes, Diabetes, Life Lessons. Bookmark the permalink.

14 Responses to What I’ve Learned about my CGM

  1. Sara says:

    That is very interesting information.

    We are waiting for a pump from the local hospital however, I was worried that the sensor for the CGM have to be changed every three days, that is a lot of money.

    I will have to keep these in mind for when we get the pump and CGM (which might come later since my kid is only 6).

  2. Jeff Mather says:

    I typically get 5-6 days. Anything over six is very atypical.

  3. Karmel says:

    Word. This aligns extremely closely to my experience. A few additions:

    • During the life of the sensor, the ISIG will take periodic drops, establishing a new mean. Don’t be fooled and assume you’re 40 all of the sudden, and, when the mean drops below 5, it’s time for a new sensor.
    • It’s okay to lie to the sensor. The pump is essentially using a weighted average of the last 4 calibrations to compute BG from ISIG; if you need to calibrate, but are heading up or down, or know that the sensor is picking up a dampened version of your actual BG, calibrate with a BG that is slightly closer to the currently reported value than the actual value reported by your BG meter. This may lead to greater inaccuracy in the short term, but you’ll be better off an hour out once BGs have stabilized again.
    • Try to clean the transmitter on day 6 when you take it off to charge it. It gets pretty nasty under that Tegaderm…
    • Buy Tegaderm from Amazon. Cheaper and better than IV-3000s.

    Thanks, Jeff, for an excellent collection of tips!

  4. Jeff Mather says:

    Karmel, thanks for the additions! I’ve noticed most of these myself. The first one and the last one are especially true.

    I’ll have to try the “lie to the sensor” trick.

    The rubbing alcohol swabs, which I rarely use anymore before testing my BG, come in handy for cleaning the transmitter. They’re fast drying and antiseptic, and they make it less likely that you’ll get water into the guts of the transmitter when it’s detached from the sensor.

  5. Karmel says:

    Yes! The alcohol pads are exactly what I use. Two and a half years into this sensor, sometimes it doesn’t charge correctly– the little light on the charger just goes steady green. Cleaning the connectors of the sensor and the little glove of the charger seems to help, and alcohol pads are perfect for that too.

  6. Mark says:

    Hi Jeff, what’s the deal with the ISIG value? I realise it’s affected by alot, but what range is a “good” range? Also – are you pumping or on pens? I’m currently on pens and want to use the CGMS as a monitoring tool for a period – do you know if the fact that I’m not bolusing from the pump affect it’s measurements?

    Thanks!

  7. Jeff, this post is full of some awesome information. Thank you!

  8. Pingback: MySentry Trial: Wearing the Sensor | Scott's Diabetes

  9. jill says:

    I just received and started using MM cgm today. I’ve used MM pumps for 21 years but this thing is amazing! I am stunned and intrigued by the accuracy of the sensor and meter…moslyt 5pts off..in less than 8 hours I was warned of and avoided a low while shopping and stopped a post meal high. Already noting what basals will most likely be adjusting with a few days more consistency. I think I will go for a record A1c change in the shortest period!
    Thanks for the tips!

    Jill

  10. Lori Rogers says:

    Jeff and Karmel,

    This is THE most reliable and affirming info I have seen on the nuances of using the CGM. It’s nice to meet you. Let’s keep in touch. The sharing of info like this is invaluable.

  11. Ned says:

    I’ve been using the MM 530G pump with Enlite Sensor for a few weeks. Really like it. One problem, though. The morning after I insert the sensor, the results are often way off: anywhere from 60-110 points. (I’ll calibrate before bed, not having any swings during night.) After a day, it works itself out and usually very accurate from days 3-7. Happened several times–always the morning after insertion. I’m calling MM tomorrow. Just curious if anyone else has experienced this.

  12. Laura says:

    I recently started the MM530G with Enlite Sensor. I have changed it 3 times in only a little over a week due to inaccurate readings and low ISIG less than 10. I have a lot of difficulty with tape due to exercising. Also noticed that my ISIG is low when I am standing vs. sitting. When my ISIG is low then my reading on cgm is low even though my bg is fine (stating less than 40 on cgm and bg on meter stating 118). Feeling very frustrated. Any insight into things would be fantastic and greatly appreciated.

  13. Jessica says:

    Laura, I am going through the same thing. I work nights and it gets frustrating to be ‘alerted’ several times each hour of a low glucose, only to check and find out it is 120. This is the down side to this machine, otherwise I like it. My MM530G seems to do this any time day or night and always wants to suspend my insulin when my glucose is normal.

  14. Tammy says:

    Now that I have read this, I know why my sensor has been so off of my BG, I am drinking wine and I am not hydrated. Even when I don’t have wine I haven’t been in the mood to drink water because it is so cold. So, I am on a mission to see if I can get my isig value between 3 and 5 before I calibrate. Lately its been high teens. Going to look at the standing vs. sitting thing too mentioned here. The 530 g can be very annoying when not working well. Have any of you Enlite folks tried to insert by hand?? I found it much easier than with serter, nerve racking since I pay out of pocket.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>