- advertisement -

To calibrate or not to calibrate...that is the question

Discussion in 'Continuous Glucose Sensing' started by iluvmhp, Apr 24, 2009.

  1. iluvmhp

    iluvmhp Approved members

    Joined:
    Mar 10, 2008
    Messages:
    1,582
    I need some calibrating tips. Dexi alarmed at 78, she was 76. Had tabs, 15 minutes later was 115. Dex said 79, so she calibrated. Next bs test at lunch was way off. So, next time ..dont calibrate???
     
  2. moco89

    moco89 Approved members

    Joined:
    Mar 1, 2008
    Messages:
    2,430
    Don't calibrate when food and insulin are playing a role with blood sugars. I know it is harder to deal with, since you are not the one connected to the device or the one that is solely dealing with this disease.

    I would try to calibrate as conservatively as possible. I am sure it will pay off.
     
  3. ErikB

    ErikB Approved members

    Joined:
    Oct 16, 2007
    Messages:
    53
    We do not calibrate with every test we take during the day. We try only to calibrate before meals when we check. I don't think you needed to calibrate the second time when the finger poke was 115 and the receiver was 79.

    There is a 15 to 20 minute delay between the blood glucose level and the interstitial fluid glucose level (in the skin where the sensor is inserted). I bet if you would have left the receiver alone it would have been at that 115 number in about 15 or 20 minutes. Especially since you know the bg is going up with the tabs.

    I'd say with the first check 78 compared to 76 is jut excellent!!! :cool:
     
  4. hawkeyegirl

    hawkeyegirl Approved members

    Joined:
    Nov 15, 2007
    Messages:
    13,157
    But I thought with the Dex you didn't have to wait until BG was stable to calibrate?
     
  5. betty6333

    betty6333 Approved members

    Joined:
    Aug 20, 2008
    Messages:
    2,018
    You can calibrate when changing, but when you get a reading that is spot on like the first one, and you are low you want to wait 15 minutes and restest, well if the 15 minute retest shows you have gone up you will will want to wait another 5-10 minutes to see if the sensor catches that upward trend. IF the sensor misses the trend you want to calibrate, but if it catches it there is no need to recalibrate.

    There are no forced calibrations and you can calibrate whenever you want. There is an algorithm in there to calibrate rate of change but it is not perfect and there is no need to calibrate unless it misses the swing , but you want to give it the 5-10 minute lag time.
    Every system is different so Susan you are doing great and Keep up the good work and learning!
     
  6. Nancy in VA

    Nancy in VA Approved members

    Joined:
    Jul 16, 2007
    Messages:
    7,308
    I'm guessing that difference 15 minutes later was really the 20 minute lag between interstitial fluid and BG readings.
     
  7. ecs1516

    ecs1516 Approved members

    Joined:
    Dec 11, 2007
    Messages:
    7,028

    Yea, I agree. I could just be the lag. Not 'really off'
     
  8. iluvmhp

    iluvmhp Approved members

    Joined:
    Mar 10, 2008
    Messages:
    1,582
    Thank you all. Live and learn with CGMS!!! So, now I know, dont calibrate when just treating a lowish blood sugar. It really "confused" it and it was spot on before that. Betty, I may need advice on how to make it cry again!

    Thanks again folks. I get frustrated with other posts, but these are so informative to me. Thank you again!
    susan
     
  9. stevecu

    stevecu Approved members

    Joined:
    Jan 14, 2008
    Messages:
    403
    We’re using the MM and have learned that it is far, far more accurate if we limit calibration to times when Sean’s BG is stable and in range. I try to calibrate when he wakes up, and before meals and that generally works out pretty well.
    We’ve also found that the sensor is not a very good indicator of recovery from a hypo excursion. It takes too long for Interstitial Glucose to catch up with Blood Glucose.
    I’d be really interested to hear anything others know about the difference between the Dexcom and MM calibration algorithms. We love the MM CGM, but they are neither forthcoming, nor very knowledgeable (the support line) about how calibration works. We’ve had to figure it out on our own. They seem more concerned about protecting trade secrets.
     
  10. Jacque471

    Jacque471 Approved members

    Joined:
    Nov 7, 2008
    Messages:
    995
    For us the Dex is generally a very good indicator after treating a low. If it is a sever low we will still do a fingerstick, but usually we start to see a rise within the 15-20 minutes we would normally wait before doing a recheck anyways.

    Sunday we had a low at 54 and treated. Checked again because Dex still showed going down and he was 38, so treated again. Then Dex started to rise and when we did another finger check 30 minutes after second treatment the Dex and fingerstick were within 10 points of each other.
     
  11. rickst29

    rickst29 Approved members

    Joined:
    Jun 2, 2006
    Messages:
    578
    I *KNOW* this one.

    Dexcom follows bG very closely in the downwards direction, but it is very slow to follow rising bG.

    And so, you can make a decent calibration when you your bG is "flat" OR "declining", the ISF readings do establish valid calibration for those fingerstick bG readings. Dexcom is reading ISF-- and during declining bG, sugar is being sucked into cells from ISF, the drop occurs in ISF *before* it actually occurs in blood. (Liver-related situations excepted.) This "ISF precedes blood" effect counteracts the internal delays of the Sensor in getting "freshened" ISF readings, and often results in an almost perfect match of ISF reading versus blood-- for many people, just as good as you get when bG and Dexcom's ISF readings are both flat.

    But the bG upwards direction is always badly delayed: bG gets to the bloodstream first (from stomach or liver), then it starts soaking into ISF, and then it starts soaking into the Sensor. There's no "counteracting" on this path, it's all delay, delay, and more delay.

    Thus Dexcom nearly always trails bG readings during rising bG. (For me, this is 20+ minutes). Calibration readings always attempt to match the Dexcom raw data against the current bG reading, with no consideration of rise or fall-- and the Dexcom raw data will almost always be way too low, the slope of your calibration curve will be to steep.... and basically, you messed up. ;) If you absolutely have to type in a number to make the graph re-appear, and you'll accept an inaccurate graph just to see the trend line, then type in a GUESS for bG from 20 minutes ago. Does this sound like a risky, bad, unsound idea?

    Yeah. You got it. :D Only in a desperate emergency, where you HAVE to have a trend line. It's a mis-calibration, and the resulting numbers will be bad. In post #1, when it was already showing a graph of SOME kind (and actually showing about the delay I'd expect), you should not have entered that calibration figure.
     
    Last edited: Apr 30, 2009
  12. stevecu

    stevecu Approved members

    Joined:
    Jan 14, 2008
    Messages:
    403
    Great explaination Rick! Thank you.

    This matches what we see with the MM CGM. I had not concidered the reason for the difference in delay for going up v. going down.

    Are you familiar with how the dex treats previous calibrations when calculating a new calibration factor? I know the MM does take them into account, but that the current calibration gets more "weight". MM is very tight lipped about their calibration algorithm.
     
  13. ecs1516

    ecs1516 Approved members

    Joined:
    Dec 11, 2007
    Messages:
    7,028
    I thought I read the same thing.:confused:
     

Share This Page

- advertisement -

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice