advertisement

CGMS ISIG and Calibration Guidelines

Discussion in 'Continuous Glucose Sensing' started by Darryl, Jul 29, 2008.

  1. stevecu

    stevecu Approved members

    Joined:
    Jan 14, 2008
    Messages:
    403
    Thanks Darryl, this has all been very helpful.

    One more question for you, and the community.

    What do you concider "stable"?

    I usually look at the current SG and the previous 2 or 3, and look for a change of no more than 2 -4 over any of the 5 minute periods. MM says anything that's within 20% (I guess of the prior SG) is okay. The example of "unstable" they gave me over the phone was rediculous - a change from 120 - 210.

    I described 2 senarios and asked if one was "better" than the other.

    Senario a) 4 SGs in a row of 110, right before a meal, not having eaten for 3 hours

    Senario b) SGs of 110, 120, 130, 132, right before a meal not having eaten for three hours.

    They said both were equally good. What do you think?
     
  2. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313
    Scenario (a) is obviously ideal, but I'd be OK with (b) as well.

    The change of 120-210 is definitely no good.

    I would say as long as the readings are changing 4 points or less in each of the prior two 5-minute periods,
    then you're OK.

    Also keep in mind, if BG is flat but you just ate, the BG is already changing even though not evident on the CGM.
    Always cal before you eat.

    You will notice that after a cal, the new cal number does not take effect for 15-20 minutes. The CGM compares
    your cal with the SG reading 15 minutes later, and tries to correct for any change that occurred during that time
    interval. When the cal has been successfully updated, it will show in the Calibration History.
     
  3. Judith

    Judith Neonatal Diabetes Registry

    Joined:
    Oct 30, 2005
    Messages:
    90
    Cgms

    The whole concept of trying to achieve excessively tight "control" of bg numbers via cgms is a bit ludicrous, since the meters which you use to calibrate are far from precise. This is the standard that meters must meet:

    ISO 15197:2003 Specifies that 95% of glucose results must be within +/- 20% of a reference standard for results greater than or equal to 75mg/dl, and within +/- 15mg/dl for results less than 75mg/dl.

    That bg of 60 might really be 75 - or 45! A reading of 100 may actually be 80 or 120. And 5 tests out of a hundred, the result apparently does not have to have any relationship to the actual bg. The whole thing is just an approximation. Basing treatment decisions on variations of 5 points is just an exercise in futility, at best.
     
  4. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313
    Judith,

    Not sure what your background or CGM experience is, but what you said does not reflect reality.

    We're going on nearly 1000 consecutive days of relying on CGMS readings for BG control, and in that time we've
    never had any bad hypos, never had any BG over 279 (and that was more than a year ago), maintained an A1C
    of 5.3%, and Glycomark of 10.2%.

    Of course, some CGM readings are off by 20-30 points but that is well within the margin of error needed to keep
    BG within a near-normal range. Even the worst CGM reading still would normally preclude the kinds of lows (20's)
    or high's (400's) that people sometimes encounter when managing using BG checks.

    In my experience, BG checks are the least effective means of managing diabetes. Here is a simple example:

    If your target is 120, and you measure a BG of 120, what do you do?
    a) If it was 120 rising, you'd increase insulin.
    b) If it was 120 flat, you'd do nothing
    c) If it was 120 falling, you'd decrease insulin or eat carbs.

    So what do you do?
    The problem is with BG checks you have no idea which is the right choice.
    So 2/3 of the time, you make the wrong decision.
    This is why many people find management by BG check to be futile.

    With a CGM, you always know the direction BG is heading, so your treatment decision is at least in the right direction.

    Another way to look at it - would you rather drive a car with a pair of glasses that is slightly foggy, or with your eyes
    closed except for a quick blink every hour?

    Bottom line, the CGM works. Many people can attest to that.
     
    Last edited: Aug 6, 2009
  5. hawkeyegirl

    hawkeyegirl Approved members

    Joined:
    Nov 15, 2007
    Messages:
    13,156
    Some meters are more accurate than others. Our Aviva is almost always within one or two points in a recheck, even when he's high. Our One Touch appears to display numbers at random, rather than based on any blood glucose levels. :rolleyes:

    Regardless, Darryl's original post is aimed at when to calibrate, not what the actual numbers are. He's looking at the movement of the numbers, not the numbers themselves, which we find to be very, very accurate (once you compensate for the delay). We calibrate the same way Darryl does - we scroll back through to see if there has been more than a 3-4 point change in BG in any of the past few readings. If there has been, we don't cal. If there hasn't been, we do.

    Still, I'm not sure if you dredged up this thread just to get another poke in at Darryl or what, but if you want to attack his tight control, there are many other more appropriate threads in which to do it in. This thread is about his calibration technique, which is spot on, and has been supremely helpful to many, many posters on this board. It's a bit absurd of you to pop in to a year old thread to derail it, when it has nothing to do with your issues with him. This is really getting ridiculous.
     
  6. Judith

    Judith Neonatal Diabetes Registry

    Joined:
    Oct 30, 2005
    Messages:
    90
    ????

    My post was not personal re: Darryl. I don't know Darryl. I was simply pointing out that the whole monitoring situation is "only approximate". The tone of your post is very defensive.

    As far as I know, ALL forums (with the exception of neonatal monogenic), ALL threads, are open to everyone in the cwd community.
     
  7. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313
    Judith,

    No offense taken - your question was valid.
    Yes, monitoring is approximate, as is any available method of D management.

    The difference is:

    Management by BG is more precise in terms of accuracy, but hugely imprecise in terms of not knowing the trend,
    and therefore often making the wrong treatment decision. Giving insulin on a 120 dropping is a huge problem; whereas
    giving carbs based on a CGM-displayed 120 dropping that is really 100 dropping is just fine.

    CGM monitoring is sufficiently accurate in terms of accuracy, and very accurate in terms of knowing the trend.

    Both methods involce approximation, but management by CGM is generally much better (MUCH better) in terms of the outcome.
    On a side note, reactions to people's posts are generally more favorable if the person has some type of personal identification in their signature.
     
  8. betty6333

    betty6333 Approved members

    Joined:
    Aug 20, 2008
    Messages:
    2,018
    Thank-you Darryl, we don't use the MM CGM, so I can't comment on it in particular, but we too find the trending information from our CGM to be incredibly useful in keeping our son safe. We are well aware of meter inaccuracies, and I find that the CGM actually HELPS with reducing the problems that the meters can allow. On more than one occasion I have tested my son and got a reading that differs from the CGM. Then went and tested on a different meter, only to find that the first meter gave a "bad' reading. Sometimes it was a dirty finger or it was still a little wet, or the alcohol swab hadn't dried up, finger had been squeezed, ect...

    BUT I think that the CGM has been a fantastic tool to augment fingersticks, as most people who use a CGM become aware VERY quickly to just how inaccurate some meters are.

    I am thankful , so very very thankful, for our CGM. It makes both my sons life and my life immeasurably better. It gives additional insight to the fingersticks that we take and makes tight control possible... fingersticks are inadequate by them selves for us, we have used the CGM for a year, and day in and day out it has proved its reliability and has been nothing short of AMAZING.
     
  9. hawkeyegirl

    hawkeyegirl Approved members

    Joined:
    Nov 15, 2007
    Messages:
    13,156
    They are, but relevance is always appreciated. This thread is about calibration techniques. Your post about "tight control" was really almost completely irrelevant.

    I know you don't know Darryl, but you were previously very aggressive in posting your opinion on his daughter's care. (And have had almost no posts since then.) That's why I was bemused to find you popping up in a thread where he was the OP, a thread that no one has posted in in almost a year, a thread that has nothing to do with tight control, but relates to the proper way to calibrate the MM CGM to offer your opinion (out of nowhere, as it's not the subject of this thread, or wasn't until you piped up) on "tight control."

    If you'd like to rehash that argument, more power to you, but to bring it up in this thread is absurd and a total distraction from the helpfulness of this particular thread.
     
    Last edited: Aug 6, 2009
  10. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313
    Judith,

    I did not realize we had conversed before on the topic of CGM usage until Karla pointed it out above. Just as you had asked in that thread, the data I posted shows a "week in the life" of a child with T1 using the latest technology:
    http://forums.childrenwithdiabetes.com/showpost.php?p=455863&postcount=74

    Some weeks are not this good, but most are. It rarely gets much worse than what you see in that graph, with use of a pump, CGM, Apidra, tight alarms, willingness to awaken overnight to adjust blood sugar whenver needed, and a reasonable diet. Before using these tools, every day was 50's and 300's...

    I saw in one of your other posts that you are a pediatric nurse who raised a T1 child from age 8 weeks. However, the tools now available for D management are completely different now vs. 3 years ago. CGMS-based control does work, it works beautifully, it has worked for us for nearly 1,000 days straight without incident.

    What might seem ludicrous in light of experience of the past decades needs to be viewed with an open mind by anyone who has not tried it, especially by the medical community. Unfortunately, many medical professionals in the D industry believe likewise that ISO-standard meter accuracy is all that matters, that +/- 5 points is crucial, but that routine daily BG's of 400 and average BG's of 200 are perfectly acceptable (and the best an infant or small child can achieve). That, IMO, is what I would consider "ludicrous" in the year 2009.
     
  11. Judith

    Judith Neonatal Diabetes Registry

    Joined:
    Oct 30, 2005
    Messages:
    90

    I saw in one of your other posts that you are a pediatric nurse who raised a T1 child from age 8 weeks. However, the tools now available for D management are completely different now vs. 3 years ago.


    I find your above statement condescending and dismissive. You seem to assume that my knowledge of diabetes is outdated. You also seem to have missed among my credentials that I am a nurse practitioner, and have current board certification in advanced diabetes management. I have worked with nearly a thousand kids with diabetes over the years. Your sample of one child, for three years, is not terribly impressive by comparison. What works for one may not for another.

    I remain concerned about the long-term effects on the psychosocial development of any child who is subjected to compulsive diabetes management.
     
  12. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313
    I edited the OP today with the following addition:

    Edited 8-10-2009 - A tip to get good initial calibrations with the MM CGM - install the sensor at least
    2 hours before attaching the transmitter. We do this by installing the new sensor at night, then switching
    over the transmitter in the morning. This has helped get more consistetly good initial calibrations.

    This is after doing this ourselves for the past year, and also hearing from others who do the same thing.

    The MM sensor usually works just fine without doing this, but in our experience the first cal's are more consistently accurate if the new sensor has been in place overnight prior to the transmitter switchover.
     
  13. Rachel

    Rachel Approved members

    Joined:
    Oct 16, 2006
    Messages:
    834
    Do you cover the sensor with something? I've always just assumed that the pointy end that goes into the transmitter is sensitive and haven't wanted to get lint, tape or bandaid adhesive, dog hair (!), etc in it. And I worry that the sensor would fall off overnight, particularly with a 5 year old pulling shirts on and off. What do people do to keep it fresh?
     
  14. hawkeyegirl

    hawkeyegirl Approved members

    Joined:
    Nov 15, 2007
    Messages:
    13,156
    Thank you for getting this bizarrely derailed thread back on track. Who would have thought that someone could singlehandedly make a thread on calibration techniques for the MM CGM controversial? :rolleyes:

    I put a small piece of Opsite over the "flat" purple part of the sensor. Jack wears his on his bottom, and his underwear protects it too. But the pointy end is unprotected, and we've never had an issue with it. I do believe that it is not waterproof unless the transmitter is connected, so they can't take a bath or anything this way.
     
    Last edited: Aug 11, 2009
  15. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313

    We use the upper butt area, and it has not been necessary to cover/protect the sensor so far. We do, however tape both the plastic and adhesive tabs down with tegaderm as soon as the sensor is installed.

    Depending on the location, it might be good to tape the sensor down with a band-aid or something that would not adhere.

    ETA - As Rachel said, we do not allow it to get wet (bath, etc.) the next morning until after it is connected to the transmitter.
     
    Last edited: Aug 11, 2009
  16. Snowbound

    Snowbound Approved members

    Joined:
    Jan 14, 2009
    Messages:
    869
    My spouse was speaking to Medtronic about calibrations errors and she was given the secret formula to determine when the pump would accept a calibration vs. giving a cal error.

    BG divided by ISIG needs to be in the range of 1.5 - 20. Outside of this range and you'll get the calibration error.

    We're in Canada and need to include the mmol/mg conversion, so for us the formula is BG * 18 / ISIG.

    We had been having low ISIG numbers, so BG's that we thought should work were giving us errors.
     
  17. MikailasMom

    MikailasMom Approved members

    Joined:
    Dec 13, 2009
    Messages:
    383
    Darryl,

    So i think I have been doing this wrong for awhile, can you please help me? We usually do sensor inserts about 2 hours after dinner, no snack that night so that would give us like 4 hours after bolus/ food before it requests the 2 cal #'s. Im a night owl so I thought since this is a fairly stable time for her it would be ok, as long as its done by 12 am as she tends to get wonky some nights after this. Can you explain to an idiot what the ISIG is? We didnt get any training with our dex, it was more like call us with any problems kind of thing :confused:. Thank you SO much for sharing your wisdom! I am in aww of your success with CGMS and D management. I dont always trust the number but do tend to rely alot on the trending arrows. Thanks alot!:D
     
  18. Darryl

    Darryl Approved members

    Joined:
    May 8, 2008
    Messages:
    4,313
    The timing of your insertions sounds good.

    To explain the ISIG: The sensor puts out an electric signal which goes up as BG goes up. That signal is a number (the ISIG, or Input Signal) which is typically around "10" for a BG of 100, but varies from sensor to sensor.

    The sensor sends its ISIG to the receiver every 5 minutes. The ISIG alone is not enough for the receiver to know the BG. However, when you calibrate, this is what happens:

    Let's say the receiver is reading an ISIG of 10 from the sensor.
    You do a BG check and see that BG is 100, then calibrate the receiver.
    Now, the receiver knows that an ISIG of 10 lines up with a BG of 100.

    Next, let's say the ISIG doubles from from 10 to 20
    The receiver then knows that the BG doubled, and is now 200

    So, the ISIG is just a number. The receiver uses the ISIG, along with the last calibration, to compute the BG.

    The ISIG is useful because it tells you how your sensor is doing. If a new sensor has an ISIG of 10 for a BG of 100, then on day 7 the ISIG is only
    5 for a BG of 100, you know that your sensor has lost half it's strength on day 7.

    Or, let's say on day 2, you measure a BG of 200 and ISIG is 10, that is a clue that the fingers may have had sugar on them, causing an erroneous high BG reading. In this case, the ISIG could give a hint to repeat the BG check, and save you from a bad low.
     
  19. Diana

    Diana Approved members

    Joined:
    Aug 23, 2006
    Messages:
    1,565
    ISIG is a MiniMed thing. The Dexcom doesn't give you access to this information.

    Also, we found that nighttime startups on the Dex really didn't work that well. We learned quickly not to calibrate after bed time - it led to inaccurate numbers.
     
  20. SarahKelly

    SarahKelly Approved members

    Joined:
    Nov 14, 2009
    Messages:
    1,147
    newbie with CGMing here...have a few questions:
    1. When to calibrate with a toddler whom is eating almost ALL the time and nursing at night? Ideas, tips?
    2. Do you calibrate only when you're not also giving insulin?
    3. How do you know when the sensor is dying? I ask this because our endo said Isaac's readings from the CGM are fabulous compared to what she's seen with other kiddos his age, however due to how quickly his BG rises or falls it can be off by almost 100pts. I was told that the sensor is toast when it's off frequently, but this is often. We really are more so utilizing the CGM for finding trends, not for specific numbers KWIM.
    4. Do you store your sensors in the fridge? Somebody mentioned this and I wanted to know if this really helped.
     

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