Coaguchek Strip Recall - alternative meters

[Protimenow]

I've been away from this forum for quite a few months - much longer than I should have. Every few months, I check out information about the various meters used for INR self-testing. Today I found that Roche has recalled MANY lots of the strips for the CoaguChek XS meter. This was an FDA mandated recall. Perhaps this was discussed already on this site. Perhaps not.

A few years ago, I did extensive testing of a much larger list of meters than are available today. I tested the Coaguchek S and Coaguchek XS and CoaguChek XS Pro, the Protime Classic and a newer model, InRatio and InRatio 2, and Coag-Sense meters. I checked them against each other, and used a lab test at a trusted medical facility or a hospital as a control.

In my testing, it became clear to me that the InRatio and CoaguChek XS often reported INR values that were higher than the lab - sometimes significantly higher. In 2013, I had a stroke because my InRatio meter showed that my INR was 2.6, when it was actually 1.7 (according to the hospital lab). This issue prompted my quest to find the most trustworthy meter.

Over the years, I detected an error in Protime strips. I reported it to the manufacturers of the Protime strips and they issued a recall and replaced the strips.

I was gratified to see that Alere discontinued the InRatio and InRatio 2 because of errors in reported INRs. I only wish they had done this a few years earlier.

In my testing, the CoaguChek XS sometimes reported a value near to that of my Coag-Sense meter, and similar to the hospital lab. In most cases, the results were higher (sometimes substantially higher) than either my Coag-Sense meter or the lab.

The Coag-Sense meter that I was testing often reported INR .1 or .2 below the lab results. In most cases, the prothrombin time was very close to (within 10%) the time reported by the lab. The only difference in INR seemed to be related to the value of the reagent used to compute the INR.

I felt (and still feel) that the consistency of the Coag-Sense in reporting the results is a strength. I believe that reporting an INR that is slightly lower than the lab results is a good thing -- I'd rather have a result that may require a slight increase in warfarin dose (and usually requires NO modification) than to have one that reports HIGHER than actual values. I feel that I'm at much higher risk when my INR drops below 2.0 than I am if it's above 3.5.

I don't work for Coag-Sense. I'm not in the business of supporting or promoting any particular meter or manufacturer.

In the light of the CoaguChek XS strip recall, I'm just suggesting that this should make Coag-Sense - still relatively unknown - worth considering if you're looking for a meter - or a replacement meter.

 

[Jamey T]

Well of course I am going to mostly agree with you. My Inratio 2 was much more accurate than my Coagucheck, when I could find the strips. On your recommendation, I bought a Coag-Sense, and am amazed at the accuracy compared to my lab. My finger stick test here at home is within .1 of my lab. This was just verified 4 days ago, with the readings being the same.
It is nice to be able to trust my personal meter again.

 

[pellicle]

Hi

... I bought a Coag-Sense, and am amazed at the accuracy compared to my lab. My finger stick test here at home is within .1 of my lab..

for what its worth I normally am about that between my lab (as a company, for I use different collection points) and my Coaguchek, which could just mean that my personal chemistry, my lab and my coagucheck are aligned (and of course doesn't reflect upon you and yours ;-)

none the less its important to recognize that these are reasons why we should never think in terms of certainty, but in terms of zones. Thus I usually counsel against sitting on a low number (like 2 or 1.7) because "that's inside my range" ... we aren't measuring steel with a micrometer at 20C, we're dealing with a much more rubbery figure.

I encourage all self testers to read the Wikipedia page: https://en.wikipedia.org/wiki/Prothr...rmalized_ratio

and observe all the possibilities for variance (such as what is a "normal" individual, which is a statistical identifier):
International normalized ratio

The result (in seconds) for a prothrombin time performed on a normal individual will vary according to the type of analytical system employed. This is due to the variations between different types and batches of manufacturer's tissue factor used in the reagent to perform the test. The INR was devised to standardize the results. Each manufacturer assigns an ISI value (International Sensitivity Index) for any tissue factor they manufacture. The ISI value indicates how a particular batch of tissue factor compares to an international reference tissue factor. The ISI is usually between 0.94 and 1.4 for more sensitive and 2.0-3.0 for less sensitive thromboplastins.

The INR is the ratio of a patient's prothrombin time to a normal (control) sample, raised to the power of the ISI value for the analytical system being used.​

 

[LondonAndy]

I have been using a CoaguChek XS for about 4 years now, since a few weeks after my valve op. I do weekly tests, and go every six months to my hospital's anticoagulation clinic for a "calibration check". My meter is never more than 0.2 different to their lab result, though I appreciate I don't have a high number of comparison points. My meter is usually showing a slightly higher reading than the lab result, but has shown 0.1 lower too. I am not bothered by these, which I consider well within the margin of error.

The CoaguCheck XS is the only meter approved by the UK's NHS. I don't think I would have any concerns if I moved to the newer models by Roche, but I have no plans to do so unless my meter dies or something.

 

[LondonAndy]

PS! Regarding the CoaguCheck strip recall. You are correct that there have been previous posts, and indeed I have posted a copy of the Roche letter sent in the UK - see this thread for some comments and that letter, which I believe affected the same batches as elsewhere in the world. Since the problem affected only readings over 4.5, and my range is 2.5 to 3.5, I decided to continue using mine.

 

[Jamey T]

Pellicle, I believe I have posted in the past of my thinking that for some reason my blood chemistry didn't like the Coagucheck. The difference between finger sticks and lab draws was fairly predictable, and that was also over several different lots.

You wrote a very good reminder, especially for people newer to warfarin, in my opinion.

By the way, I have kept all my old chips, in case I ever read about someone losing or needing one, and I happened to have the correct one.

 

[pellicle]

Hi

Pellicle, I believe I have posted in the past of my thinking that for some reason my blood chemistry didn't like the Coagucheck. The difference between finger sticks and lab draws was fairly predictable, and that was also over several different lots.

to refresh my memory, what was that difference (between finger sticks and lab draws? (in INR units)

 

[Jamey T]

Hi



to refresh my memory, what was that difference (between finger sticks and lab draws? (in INR units)

For me it was typically +.6 to +.8, but it was consistent, so it worked for me for a couple of years.

 

[pellicle]

For me it was typically +.6 to +.8, but it was consistent, so it worked for me for a couple of years.

Gee that's too much for my comfort. Sorry if this is rehashing old conversation, but did you ever get any reasons from Roche?

That's more than I've ever seen in any comparisons.

 

[Protimenow]

When I was testing the various meters, I found that the CoaguChek XS was sometimes considerably higher (by as much as .6 - .8 points) than the lab results, and higher than the Coag-Sense meter. The InRatio had a similar issue.

As Pellicle said, these measurements aren't exact - the accepted standard for meter accuracy is, if I recall correctly, within 30% of the lab value. The lab isn't already right, either - I've had blood drawn at clinics that was MISHANDLED by the clinic and resulted in wildly inaccurate values.

For me, I'm much more comfortable with a meter that routinely gives a value that is equal to or below the lab values - this way, I'm comfortable that a 2.3 on my meter may actually be a 2.5 or so at the lab (and not drift in the the dangerous sub-2.0 range), than I am with a meter that sometimes reports values that are HIGHER than the labs.

In terms of my Coag-Sense, although the INR values are a ratio between the actual clotting time and the reagents used on the strips or at the lab, my prothrombin (clotting) time is often almost an exact match (within less than a second) to the lab. The difference between the INR reported by my meter and the one reported by the lab (assuming similar clotting times) is an artifact related to the different 'values' of the reagent used for the test.

I've had one lab that had a consistent difference in reported INR - although the clotting times were always almost identical. Both INRs were within range - the only difference was the values used for the reagents.

An important point worth repeating - INRs are not exact measurements.

 

[sweetmarie]

All this is interesting material. Lately my coaguchek XS has been reading slightly higher than my local lab results. My health clinic told me that the tolerance range is only +/- 1.38%. My inr range is 2.0-3.0. So if the lab results are say 2.50, my meter must be between 2.47 and 2.53. As you all know, the coaguchek xs only measures one decimal point. So they're telling me that 2.4 or 2.6 is out of range and I should not consider my meter as accurate. Unless my math is failing me.

Anybody else experiencing this at their clinics? Have your clinics ever told you what your meter tolerance range is compared to their lab results?
tks

 

[pellicle]

Anybody else experiencing this at their clinics?

myself I use a clinic about once a year, even then that's simply to do a comparison between my Coagucheck INR reading and what they say. Usually its within 0.2

I'd phrase your description of tolerance as "baffle you with bullshit".
I don't use a clinic anymore because I don't wish to let those circus clowns near me.

my maths goes like this from INR = 2.5 an increase of 1.5% would be 2.5375 (where 1.5/100 = 0.015 then multiplying that by 2.5 gives 0.0.0375 and adding that adding that to 2.5 gives 2.5375 ... which is so small an increase as to not even be within believable limits)

I look at it this way:

• if my INR is between 2.3 and 2.7 I don't even blink
• if my INR goes over 3 I will probably just wait to see what next week brings (or if I'm super scratching my self silly anxious I'll make a mid week reading
• if my INR goes under 2.2 I'll simply add 10% to my does and move on

there are amounts of accuracy with numbers that maths gives but which in reality no blood test between differing clinics with differing methodlogies will give. Thus to me its non-sensical to fret about what amounts to nothing.

There is an important term: Clinically Significant Variation
2.47 and 2.53 are not clinically significant results

INR = 3.5 is not harmful or risky ... and so as long as you sit between 2 and 3 you're sweet.

 

[pellicle]

.... Lately my coaguchek XS has been reading slightly higher than my local lab results.

PS: look this table over carefully

 

[sweetmarie]

BS is the first thing that came to mind also. Of course the clinic doesn't want to lose customers, it's their jobs on the line. tks

 

[Jamey T]

I test myself weekly, and go to the lab every 6 weeks, but my GP wants me to go every 4 weeks.

I had good results with the Inratio 2, but not the Coagucheck. With the Coag Sense, I am always within .1 of the lab. I always have a blood draw done at the lab.

To your point, my Coagucheck was always higher than the blood draw, but it was fairly consistently higher. With that meter, I always used it as an "indicator" as opposed to an actual reading, if that makes sense. If I was very high or very low, I would get tested, otherwise I just guestimated at my actual reading.

I don't worry about being over my high of 3.5, but i get very nervous of going low.

 

[sweetmarie]

I test myself weekly, and go to the lab every 6 weeks, but my GP wants me to go every 4 weeks.

I had good results with the Inratio 2, but not the Coagucheck. With the Coag Sense, I am always within .1 of the lab. I always have a blood draw done at the lab.

To your point, my Coagucheck was always higher than the blood draw, but it was fairly consistently higher. With that meter, I always used it as an "indicator" as opposed to an actual reading, if that makes sense. If I was very high or very low, I would get tested, otherwise I just guestimated at my actual reading.

I don't worry about being over my high of 3.5, but i get very nervous of going low.

I've only compared my Coaguchek XS with the lab four times over the 7 years I've owned it. But as you, my home results have always been higher than my local lab test. I didn't buy the meter to really compare to the lab, it was more for a convenience for me when we're at the cottage in the summer, on an "away from home" vacation, etc.
tks

 

[tom in MO]

...INR = 3.5 is not harmful or risky ...

Might be if you have an accident and suffer trauma to somewhere vital, for example the brain. I've known people who have died from internal bleeding with normal INRs. Could also be important if you get an ulcer; the higher your INR the more severe the complications and the harder it is to heal.

For an operation, I went off warfarin for about 5-days. Quickly after the surgery, I healed fine, the blood in my urine disappeared in less than a day.. However, as soon as my INR got to "normal" for me 2-2.5, I started bleeding again. Per my urologist that's normal when you are on warfarin. He said it always takes longer for patients on warfarin to heal from any operation that requires scabbing for wound healing. He said the scabs heal and break open and heal again and you don't even notice it when your INR is low. But once your INR gets up there, they bleed longer and take longer to scab and it's noticeable and you need to restrict activity.

 

[pellicle]

Might be if you have an accident and suffer trauma to somewhere vital, for example the brain. I've known people who have died from internal bleeding with normal INRs.

maybe ... but like you say ... you know people who've died from bleeding with normal INR ... so (as I was saying) its a statistical thing not a "I know a guy" thing and the risk is always represented in this way. You know, like calculating the chances of being in a car accident.

 

[Protimenow]

Knowing how much a meter is than the lab results are can help you to predict the actual lab results. Before I got the Coag-Sense, I was pretty comfortable in the knowledge that the CoaguChek XS was usually .2 - .4 higher than my lab. That way, as long as the INR reported by my meter was above 2.7 (on the meter) and below about 3.7, I wasn't all that concerned about my INR - I felt that I was in range.
A few years ago, someone on this forum came up with a formula for converting CoaguChek results to lab results (but this undoubtedly had to do with his lab, and lab results from one lab to another aren't always consistent, either.
Learning the variance between meter and lab makes it pretty easy to know when your INR is in a safe range.
It's kind of like knowing a clock is ten minutes fast -- just subtract ten minutes to get the correct time.

For myself, I use the CoagSense, and its results are usually 0 to .2 below the lab's results. The prothrombin time is almost exactly the time reported by the lab - so the calculated INR varies because the value for the reagent on the strip is different from the one at the lab.

The point here - getting comfortable with the variance from meter to lab (choose a reliable lab) should be all you need to comfortably know that you're in range. Occasionally comparing meter results to blood draw results can be useful for confirming the relationship between meter and lab. (I made the mistake years ago of trusting my InRatio to be correct, and didn't compare it to a blood draw -- and paid for trusting an incorrect meter)