POC not reading INRs correctly

[Protimenow]

It's indeed troubling to try to figure out why ONE PERSON apparently has this issue with blood draws and POC machine differences being so large. With the tens of thousands (probably) of comparisons between meters and lab blood draws, and the millions (by now) of INR tests made with meters, it's surprising that one person can encounter this degree of error. I wonder if Roche would like to get involved in figuring it out.

My thought, when suggesting the Coag-Sense meter, was that unlike the others, this meter actually times the formation of a real clot, rather than some electrochemical phenomenon. My assumption would be that the Coag-Sense would get it right--whatever 'it' is. The fact that the Coag-Sense was relatively close to the CoaguChek XS result seemed to suggest that maybe the lab was doing something wrong.

The fact that Linda had blood draws from more than one facility, and there was STILL a fairly large distance between lab and meter is a bit confusing. When testing on my own fingers, I found that the Coag-Sense gave me a value a little BELOW the blood draw, and that the CoaguChek XS (or the XS Plus) gave me a result slightly HIGHER than the lab. In many cases, the lab value was almost an average that of the two meters.

It would be great if this can be figured out and the enigma solved.

For now. because Linda has trouble with vein draws (or, actually, the phlebotomist has trouble finding a good vein), doing a finger stick is much preferred. I'm inclined to suggest (in fact, I AM suggesting) that Linda determines the difference between lab and meter. If the two are consistently different -- either by a specific number regardless of the lab results, or by a percentage above the lab's results -- the meter can still be used for reliable testing. It may be as simple as subtracting the meter's value, or multiplying it by a certain percentage, in order to predict what a blood draw would reveal.

A while ago, there was a formula a formula that could predict a lab's value from an InRatio value was developed. Perhaps a similar algorithm can be developed for Linda, so that she could rely on her meter, with occasional blood draws.

Again, it would be great if the question is solved. in the meantime, if a meter (XS or Coag-Sense) could be shown to be reliably different from the lab value, this may be a good option. (Linda -- you ARE tracking all of this on a spreadsheet, aren't you? Keeping this information handy can be helpful in determining which meter is best, and for demonstrating to your doctors that you're on top of your INR and dosing).

 

[pellicle]

Hi Linda

Well you certainly have been struggling with this for a while. I admire your Finnish SISU !

I like your idea of getting a finger stick and a vein draw for testing on the same machine. That would be interesting. The coaguchek actually accepts vein draws according to the manual.

Ihave a question though. What is your warfarin dose?

I am wondering if (going out on a limb here) it is above 7mg per day that you could not just go with the POC results anyway because even if its lower than the lab draw its unlikely that when it reads INR = 3 that you are inadequately anticoagulated.

Just a thoughtt

 

[Protimenow]

I like that limb.
I think that what's been going on here -- and I'm a part of this nearsightness -- is that we were focusing on getting a meter that provided results that were close to the lab results. What really matters is that the INR is above 2.5 or so (yeah, I know, some people are saying above 2.0, but having had a TIA when my meter said 2.6, and the lab said 1.7, I'd rather aim a bit higher than the 'safe' 2.0), and not worry about the other stuff. As long as the INR is ABOVE stroke risk, I think that things should be okay.
Linda's meters gave results in the 4.X range -- even the Coag-Sense gave a 4.1 (if I recall), when the lab said 3.2.
Even if it WAS 4.1, or the higher value that the XS gave, the risks of hemorrhaging or other injury are small, if you don't drop a bowling ball on your toe, or somethng. (This would probably raise a big bruise, but certainly wouldn't be life-threatening).

Using the same venous blood from a single blood draw for testing on the CoaguChek XS and at the lab is a good idea -- it could help to reveal issues that may be related to something happening in the capillary blood sample from the fingerstick, or not; it may show a systematic difference between meter and lab -- in any case, the result should be interesting.

If KNOWING that a 4.1 on the Coag-Sense or a 4.6 on the CoaguChek XS is equivalent to a 3.2 at the lab; or, by extension, a 3.5 on a meter is matched by a 2.5 (or something) at the lab, you may be able to feel safe with the higher numbers on the meters, with relative confidence that your INR - even if not lab accurate - is still in the safe (above 2.5) range. THAT'S what really matters.

 

[Aggie85]

Hi Guys,
Thanks Protimenow for setting the record straight in that the POCs are reading higher than the labs not lower, but both your points about using the data to at least determine that I'm not low is possibly sound. I'm not sure my clinic would go for it well enough to take responsibility for my monitoring without still at least twice monthly checks, but it would at least give me some idea of where I am. The first time I checked on NewMitral's machines, they were reading I the 3.5range and we thought we had it. But, it turned out my INR was low and below range. But it would still give me an idea of where I am if not an exact number. A percentage of the difference might work, but a direct differential wouldn't be too accurate. The amounts over have not been consistent. I could probably get an average of what that differential is to give a rough guess if I'm any where near in range. Then if any extreme, go get tested to verify real level. BTW my dose of Warfarin is currently 7mg 4 days and 6mg 3days, alternating doses. 6 straight out kept me too low and 7 every day made me too high.

I'm still going to pursue getting both methods on the same machine and see what that shows. It's the one variable not being delt with.

As to spread sheets, nope. I can get the numbers because my ACC clinic has kept all of them in the computer, but I haven't been. I HATE math and anything to do with numbers. so after the first couple months, I've only kept rough track. Good at thinking scientifically, but not statistically. I'm always amazed and overwhelmed with Pellicle's statistics charts and diagrams. I can understand them but not break them down. I could figure out the average differential between my results but no more than that. My husband or son probably could do more if i could tell them what I need. They'd roll their eyes that I couldn't do it, but know it's me and not be surprised. I'm the practical common sense person in the family, they're the budget and math members.

I still don't buy that I'm the only person this sort of issue would happen to, there's got to be a reason that is being missed somewhere. But, I'll be on this the rest of my life. I've only been at it for six months. I've got LOTS more time to figure out what's going on. I refuse to believe there's not an answer.

Linda

 

[Protimenow]

Linda:
A couple things:
I keep a running spreadsheet in part to be able to convince a new doctor or clinic that I know what I'm doing. I record the date, the time, the machine that I'm using, my weekly dose, and any changes (change in medications, diet, health) that I think may have had some impact on my INR. I record tests done with a meter at home, and tests done at the lab or the anticoagulation clinic. If you can track what the meters are saying, and relate that to the result of blood draws, it should give you some idea, historically, of what your actual INR is (when you look at earlier, similar results from meter and lab).

You shouldn't have to do any data massaging. If your lab wants to draw blood twice a month, and you still have veins that can be tapped, this may not be a bad idea -- you can test weekly, just to follow any changes in your INR.

The other thing:
It's been well documented that the CoaguChek XS displays a higher variance from lab values the higher the actual INR goes. Thus, an INR of 2.5 may show up on the meter as 3.1, while an INR of 3.0 may show up as 4.1. The higher the actual INR, the higher the difference between lab and meter may become. I don't think that the Coag-Sense displays that kind of variation.

---

Your goal should still be assuring yourself that your INR is above the dangerous range (to my mind, below 2.0 is truly dangerous if your INR is there for more than a few days, but 2.5 and higher is usually quite safe). If you can establish, with relative certainty, that an INR of 2.5 is AT LEAST a certain number on your meter, your target would be to stay at or above that number on the meter. You may not really need to adjust the value. (If, for example, a 2.5 at the lab is ALWAYS a 3.3 or above, just stay above 3.3 on your meter).

You SHOULD be able to keep your INR in range regardless of how close to the lab's values your meter reports your INR is, as long as you are confident that a particular minimum value on your meter represents a minimum lab value. The goal here should be keeping your INR within range -- not necessarily finding a meter that matches the lab's values.

 

[pellicle]

Hi

As to spread sheets, nope. I can get the numbers because my ACC clinic has kept all of them in the computer, but I haven't been.

there's two ways to view this:
1) whats past is past and knowing it does not change what is - so keeping records is unimportant
2) knowing the past can sometimes be helpful. Not keeping records makes knowing that impossible.

I had a clinic (not INR clinic) lose all my data in an accident. So I like to keep my own data on things which I manage (like my INR) now.

I HATE math and anything to do with numbers.

I get that ... but if you put it on a spread sheet then you don't need to do any maths ... its just a record. Then if later you need it, and someone else is wanting to help you, then they can. For instance:

My husband or son probably could do more if i could tell them what I need.

I still don't buy that I'm the only person this sort of issue would happen to, there's got to be a reason that is being missed somewhere.

I agree with you. The problem is that getting to the bottom of it may be quite time consuming (and require investigations). Without trying to dissuade you from that task I'd also suggest that rather than just banging your head against that particular wall, take another approach.

The goal is not having clots yes? Thus INR is just a route to an indicator for that. It is not the grail in itself. I would wonder if you could also have d-dimer tests done when your INR is at 2.5 by your lowest testing route and see if there is any evidence of thrombosis?


However before I go further on this topic, what was your dose when you were slipping under and what was your INR that you called as "under"

I refuse to believe there's not an answer.

There may be an answer or there may not be an answer thats available to you (or us), but there are alternative ways to manage the 'blackbox'

anyway if you can let me know those answers (INR and dose) I'll chew on it and see if I can present something

 

[pellicle]

Hi

not to disagree with the concepts, but to clarify an issue:

...(to my mind, below 2.0 is truly dangerous if your INR is there for more than a few days, but 2.5 and higher is usually quite safe). If you can establish, with relative certainty, that an INR of 2.5 is AT LEAST a certain number on your meter, your target would be to stay at or above that number on the meter. You may not really need to adjust the value. (If, for example, a 2.5 at the lab is ALWAYS a 3.3 or above, just stay above 3.3 on your meter).

the INR is a number which is rubbery in many ways. Its not like a count of a molecule present (such as with blood glucose or CRP). The actual base lines (that form the divisor in the ratio) are varient between people (perhaps even within a given uncoagulated person from week to week) and subject also to reagents variation. So mussing about with small variations is problematic.

To re-iterate this point I again mention such data as this:

for the benefit of Linda, please look at the graph with a ruler (or anything straight) in your hand and see that going straight up from 3.5 on the Lab chart we see two dots, one (looking across at the Coaguchek XS INR) one dot at 4 one dot at 3

So this means in this testing they found for a Lab result of 3.5 Coagucheck XS reported 4 and 3 ... spending time looking around this graph we also see that they discovered that for Lab INR = 2 Coaguchek XS readings were between 2.5 and 1.5

Am I wrong Linda in suggesting this fits within (or not far away from) what you are experiencing?

Note also that as the INR increases the variation increases.

Its important now to go back to the point above that "we don't know which is the truth" and "we don't know how that result fits with your ratio of uncoagulated time to clot (which is what INR is formed on) and your coagulated time to clot (which would be your INR not an INR generated from a 'baseline")

 

[Aggie85]

Hi Guys, I've just returned from a long day trip so don't have much energy to carefully analyze your posts but wanted to acknowledge them. I think before I fully answer them, let me get all my results from my ACC clinic so I'm dealing with real numbers and not remembered ones. BUT I can do a rough answer by going back to my original post in this thread and a few I remember I know with accuracy.

COPIED: For example: POC check read 6.5 & 6.7 (tested twice with new finger stick to make sure) and lab INR read 3.7. Slightly lower numbers on a different POC machine (above 5) this week, but still way out from lab reading of 3.3.

The above was from my first two visits to the ACC clinic. You can see the variance was quite extreme. Now for a couple more data points:
FIRST visi (just to have all grouped here): CCXS : 6.5 and 6.7 Lab 3.7 Second visit CCXS = 5.+? Lab = 3.3
Fast forward to August when I was trying NewMitral's machines: CCXS = 3.5 INRatio = 3.3 lab= 1.9
Fast forward again to this week just for consistency : CCXS =4.6 CS=4.2 lab=3.2
What is a shame is I did not get a POC reading the week my lab had shown a 6.0. It would have been a good data point to see how off the CCXS would be when a level was truly high. If there is (when there is) a next time with a high lab, I'll be sure to get a CCXS reading. But i ddint that time, so moot.

What I see at least right off is there is not consistency between what my lab shows and what the CCXS shows. One week 3.3/ 5+ then 3.2/4.6 I think one of you guys had suggested that if at least the misreadings were consistent, I could just use that number to stand for what i need to reach. But alas, it's not very consistent. depending on fudge factor you accept as normal variance anyway. Problem is still also limited data points since I'm still relatively new at this whole needing to monitor thing. After a year or two, we'll have more data points to get a better picture of variations. BUT, let me get all the data I do have availble from the clinic before I make a definitive statement about my trends. It will still be limited though because once we discovered there was a problem, we quit double dipping (as it were) into clinic time by having me do both testing methods and just stuck with the lab testing unless I asked for a POC as part of my exploring possible causes.

Pellicle, briefly the dosage question: my dosage has varied a bit as I recovered, got more active, increased my diet, changed meds. Some of the dips were with various changes in metabolism and didn't last long. We are still in the process of tweaking the maintance dosage and me getting better at being consistent in my K intake. I'm not sure which "Under" youre referring to, so can't address your question accurately. HOWEVER, i'm not so worried about the dosing regime right now as we are consistently using the lab's (vs. the CCXS) values to determine dosages. As that is a consistent measurement source, it's the best to guide off of and I'm confident in it. The INITIAL point of this whole drawn out thread was not so much what to use to determine the dosing (although eventually it would be nice to) but to try to discover what was going on with these abberant POC values and what I could look into as to would have been the cause. The goal being to find that grail issue, resolving it and then being able to switch to home monitoring. I think we have explored almost everything we can think of to check as to what might be causing the extreme values and can't come to any solutions other than to figure out a fudge solution. Using fudging for lose monitoring might be fine but I wouldn't want to use it as a means to actually adjust dosage by. Maybe still the newbie factor here in that I'm not very sanguine about all this yet. Anyway...

Let me cogitate and humm on what you two have brought up some more when I'm not so tired and when I've gotten access to all my data points. Then I might be able to discuss your points more intelligently.

Linda

 

[pellicle]

Hi

a few quick points

The above was from my first two visits to the ACC clinic.
You can see the variance was quite extreme. Now for a couple
more data points:
...
FIRST visi
CCXS : 6.5 and 6.7 Lab 3.7

...
Fastforward again to this week just for consistency :
CCXS =4.6 CS=4.2 lab=3.2

What I see at least right off is there is not consistency
between what my lab shows and what the CCXS shows.

what I see seems to be showing a trend towards them (lab and poc) trending to be equal ... from a difference of nearly double (1.77) to a lesser difference of 1.4

One week 3.3/ 5+ then 3.2/4.6

which isn't as big a big difference

Pellicle, briefly the dosage question: my dosage has varied
a bit as I recovered, got more active, increased my diet,

I think that's something which happens to all of us ...

The INITIAL point of this whole drawn out thread was not so much what to use to determine the dosing (although
eventually it would be nice to) but to try to discover what was going on with these abberant POC values

understood ... but I had the dose in mind with relation to my suggestion as to it not being as critical, and that the actual thing of interest is not clotting, INR is just a metric that points to that.... see my suggestion on the d-dimer test ... after all the goal of anti-coagulation is to not have thrombosis ... INR is just an indicator we are familiar with that relates to the goal..

The goal being to find that grail issue, resolving it and then being able to

as I see it: the goal being to get your anticoagulation home measured conveniently and reliably

I think I said it right at the start, perhaps not on this thread but on another, that so soon after surgery expect lots of things to be unstable. Especially if there is other medical issues and medications compounding things.

My own situation is that there is no "maintenance dose" ... that's a "simplification" instead I vary my dose by small margains as needed over the months (not over the days). I would be curious to see what happened if you took a consistent daily dose. I split my 1mg tablets into quarters so that I can have the same dose every day. Still, your daily variation is not so big ... 1mg between cycles.

Best Wishes

 

[newmitral]

Hi Linda,

It's a small point in the overall discussion, but just to be completely accurate, when you checked on my two meters, the Intraio2 showed INR=3.5 and the Coaguchek-XS showed INR=3.6 (vs the 3.3/3.5 numbers you quoted above. I still have the photo of the 2 machine readings that I took with my smartphone.
There was also an additional reading that you took on my Coaguchek machine after you returned from your vacation and had a lab reading that same day, although I don't recall that particular reading value. If you do try to create some formula to translate between meter and lab reading, then having the most accurate numbers will help.

Regards,

Newmitral

 

[Aggie85]

Hi All, HAPPY NEW YEAR! Sorry I've been off so long. Now that I'm back up and functioning spare time for hanging on the site dropped, especially as the holidays approached. I just saw NewMitrals comment, I must have mis typed or mis read some of my figures. I'm not sure where I got the numbers, NM, as I have the same photo we took that day too. Still, the numbers are way off from the labs.

I've continued to have a significant variance between the lab results and POCS results. With multiple labs being checked, different types of machines and different methods (chemical reagent reaction or mechanical clotting time reaction) machines still showing a marked difference. We can't figure out why; why would I be the only person on the earth, who doesn't have Lupus or something, that gets out of whack readings just doesn't make sense, but that's how it is. I'm pretty much out of recovery, on normal diet, normal activities, confident in lab results and technical competence of testing and still getting abbherrent readings. So my clinicians and I have decided to ignore POCs and use the lab's results. I'm becoming good friends with the phlebotomy techs!

I did buy a CoagUCheck and strips on Ebay to have for my own use. I'm using them to continue comparing with the lab and to give rough monitoring of values for myself. More just for my own confidence level than clinical use. If values are showing normal range I assume I'm a bit low, if around 4s, probably in real range and if upper 4s or higher, maybe actually out of range high. If either extreme, I'll go get lab tested. I have a standing order for lab tests whenever I want one. We are continuing to see that there is not a consistent differential between lab and POC. So I won't rely on POCS for medication changes, just rough monitoring. It does let me get an idea when I'm suspicious or have been inconsistent with my diet or when on vacation away from testing facilities. Which is what I wanted my own POC mainly for anyway. I'd like to be able to rely on a fingerstick value instead of the one to three, to six, sticks it takes at the lab, but oh well.

Thanks ALL for helping me think of possible causes and factors to look into. Especially thanks to NewMitral and ProTimeNow who let me use their machines to further test outside the clinic setting!

BTW the lab head chemist was TOTALLY uncooperative with my idea to do Veinous and capillary testing on their machine and the clinic was unable to do so because of lack of facilities for handling the veinous draw. So, so much for that idea.

Take care all and thanks for all the input.
Linda

 

[Protimenow]

After self-testing on a variety of meters for more than five years, I've gotten to the point where I don't expect the POC results to match the lab results. I've found that even two labs get different values, even if the blood is drawn the same day, just minutes apart.
As with Linda, my main goal is to stay within (and perhaps, if necessary) slightly above my range. My Coag-Sense often tests a bit below a hospital lab that I have a fair amount of trust in, and my CoaguChek XS tests slightly above the same hospital lab. The lab results often fall between the Coag-Sense and the CoaguChek XS. I'm most comfortable with my Coag-Sense, because if it gives me a reading of 2.0, I can be quite confident that it is AT LEAST 2.0 (and, for me, probably more like 2.2 or 2.3). I'm also not particularly uncomfortable with a 2.6 or higher from my CoaguChek XS, which probably means that the hospital's result will probably be around 2.3 or 2.4.

It's learning your meter, and getting a feel for how a meter's results will correlate to a lab's results, that can help you to get a sense of where the lab's results will be, and give the assurance that you're in range.

The goal, of course, is to be in range. Even if a meter will CONSISTENTLY give a result that doesn't match labs, being confident that a 'bad' result means that the lab result will say that your value is in range, should be enough to give you the comfort of knowing (or strongly suspecting) that your INR IS in range.