Coag-Sense versus CoaguChek xs and labs

[Protimenow]

I just realized that I didn't report on my blood draw and numerous tests with the Coag-Sense (and one test with the XS).

The tests on 7/17, using strips from three lots in the PT1, and two batches in the PT2, yielded an INR ranging from 2.7 to 2.8. For the two meters. and multiple strips, this is pretty impressive consistency. The only outlier was one test on the PT2 that gave me a 2.5.

At the same time (well, within minutes), the XS gave me a 3.6. Again, this is pretty consistent with what I've found when comparing the two meters (and when the INR is above 3 or so.

The lab result was a 3.4. If the lab is to be considered 'gold standard' - both meters are well within the 20% allowable difference between tests.

On 7/17, I reduced my dose to 6.0. Now, 8 days later, the PT1 gave me a 2.3, and the XS gave me a 3.0. There's more than 20% difference between the meter results. I wouldn't be terribly surprised if the lab results were 2.8 or 2.9. Regardless of which meter is closer to correct, I'm comfortable with the results (although a 2.5 would have made me slightly more comfortable. )

 

[ATHENS1964]

I found the following link (my skills in english isn't good ), maybe it is useful to someone.
Comparison of the INR Values Measured by CoaguChek XS Coagulometer and Conventional Laboratory Methods in Patients on VKA Therapy
https://journals.sagepub.com/doi/pdf/10.1177/1076029615595881

 

[ATHENS1964]

On Friday I had a blood test for my operation on Monday , The INR was 1,1. after 2 hours check it with my XS and found it again 1.1

 

[Protimenow]

With an INR that low, the two meters WOULD agree.

 

[Sheenas7]

Hi everyone,
I went to my annual yesterday and asked to include a blood draw at the lab for my INR.
I normally report every 2 weeks with Roche and I use Coagu Check Vantus. On Tuesday I was 3.2.
Next day was my physical and I had to go fasting. I repeated my Vantus check and it was 3.3 and then went to my doctor's office and had him draw blood. Today the lab result on the same day was 2.9. Those of you that know all the mathematical implications, is that significant? I was hoping it would be the same.

 

[dick0236]

I was hoping........

Sounds like typical INR test series using a meter vs lab draw......testing over a couple days.....and fasting vs regular diet prior to testing. I'd take a .4 spread (3.3-2.9) anytime.......especially with all the outside influences between your recent tests.

That's why we are given an INR range with a wide spread of 1.0 (2-3, 2.5-3.5, etc). Staying in a range tighter than 2.5-3.5 is all but impossible for me. I am happy anywhere between the "goal--2.5-3.5---posts". I only follow individual INR results when looking for dosing trends.

 

[pellicle]

Hi
I'm a little uncertain about time lines here...

I went to my annual yesterday and asked to include a blood draw at the lab for my INR.

yep ... yesterday

I normally report every 2 weeks with Roche and I use Coagu Check Vantus. On Tuesday I was 3.2.

so like the lab draw was Thursday (uncertain when yesterday was because >Australia && Timezones && international dateline<

Next day was my physical and I had to go fasting. I repeated my Vantus check and it was 3.3

Today the lab result on the same day was 2.9.

but the lab results are from the draw on which day?

You see my problem here?

Now INR is not static (although there are some who report its always the same that's not actually common) and if its changing it can change about 0.3 INR points per day (based on my personal observations) or less. This means that its possible that for some reason it was climbing and you've got some points in the middle of that climb.

Then there is the fact that no lab will pretty much give you 100% concordance with INR because INR is a rubbery figure at the best of times, its not like measuring a piece of steel with a vernier caliper.

Case in point from a Roche publication:

understanding the role of reagents (or not understanding but accepting it as a key point) is the basis for this issue. Next from this study of 2 quite similar systems (XS and iStat)

so the take out from this is to not obesss over numbers and be careful to regard data from different sources as being "different".

To make a comparison you need to

1. go to the lab and get a draw
2. in the lab or if you can go home and within a few hours take a reading with your XS - document that
3. when the reading comes in from the lab, compare those two which were taken at more or less the same time.

HTH

 

[Sheenas7]

Hi
I'm a little uncertain about time lines here...



yep ... yesterday



so like the lab draw was Thursday (uncertain when yesterday was because >Australia && Timezones && international dateline<






but the lab results are from the draw on which day?

You see my problem here?

Now INR is not static (although there are some who report its always the same that's not actually common) and if its changing it can change about 0.3 INR points per day (based on my personal observations) or less. This means that its possible that for some reason it was climbing and you've got some points in the middle of that climb.

Then there is the fact that no lab will pretty much give you 100% concordance with INR because INR is a rubbery figure at the best of times, its not like measuring a piece of steel with a vernier caliper.

Case in point from a Roche publication:

understanding the role of reagents (or not understanding but accepting it as a key point) is the basis for this issue. Next from this study of 2 quite similar systems (XS and iStat)

so the take out from this is to not obesss over numbers and be careful to regard data from different sources as being "different".

To make a comparison you need to

1. go to the lab and get a draw
2. in the lab or if you can go home and within a few hours take a reading with your XS - document that
3. when the reading comes in from the lab, compare those two which were taken at more or less the same time.

HTH

Hi and thank you for your comments.
I’ll be more clear.
August 4 using Vantus my result at 7:50 a.m. after breakfast was 3.2.
August 5 at 7:15 a.m. and fasting my INR with the Vantus machine read 3.3. At 8 am same day and still fasting, I had a blood draw which resulted in 2.9.
I was trying to establish how accurate my Vantus is. What do you think?

 

[pellicle]

Hi

August 4 using Vantus my result at 7:50 a.m. after breakfast was 3.2.
August 5 at 7:15 a.m. and fasting my INR with the Vantus machine read 3.3.
At 8 am same day and still fasting, I had a blood draw which resulted in 2.9.

I'd call those ranges within expectable ranges and well within the 0.3 units change per day that I've seen. Further I'd say that the "clinical significance" (meaning would a clinic react to that) is negligible. I'm talking of the difference between the 3.3 and 2.9 on the same day. An experienced clinician might say that 3.3 is high enough to adjust, but equally they may say lets do a mid week test and see where its going. Thats what I'd do, and I would not adjust or change dose in the mean time. Fasting may itself be sufficient to alter your INR, which will return of its own accord later.

I was trying to establish how accurate my Vantus is. What do you think?

ok, so did you read and look at all the data in the colums above? Did you notice the variances? They are all (for each row) from the same sample. If you did not, please take a moment to open that table full screen and trace across the lines and see what 2.1 resulted in on other systems, then 2.5 ...

This difference is explained in the text.

Next we come to the word accuracy; accuracy can be measured in a watch but can not be measured in INR. Accuracy is not even a logical term when you understand how vague and cooked up an INR is. It is a rubbery thing. What you want is repeatability. You can get repeatability with a single test type, but as soon as you start introducing different tests getting agreement is nearly impossible (as every lab technician knows).

First lets go back to basics, why do you have an "INR" that you are interested in? Its because we are trying to reduce the rate of clotting speed and thus the formation of clots caused by a valve in the blood flow. Exactly how much you need to slow it is unknown without out specific testing specific to you.

The range of INR that's ideal can't be known without that, thus we have a range to be within and we also know that its less worrying to be a tiny out of range bit high than a bit low out of range. See this blog post of mine:
http://cjeastwd.blogspot.com/2017/01/2016-inr-data.html
We know also that the XS reports slightly higher in the high range (although Roche tried to address that with recent (2017) strip changes). But as we know that if INR is a little high its not of significance. The trap to fall into is attempting to micro manage and obessing about 0.1 or 0.3 difference.

Think of it like this, if shoe sizes were in 1 decimal place fittings (instead of 7, 8, 9 ..) would it make a significant difference if you were to buy a 8.3 or an 8.5? Probably not.

The Vantus btw is just another handset for the XS system (you'll see that written on your strips).

HTH

 

[MdaPA]

Think of it like this, if shoe sizes were in 1 decimal place fittings (instead of 7, 8, 9 ..) would it make a significant difference if you were to buy a 8.3 or an 8.5? Probably not.

I agree, probably not. To take this analogy a bit further, I would add the foot measuring device and that one's foot size does not (normally) change from one store's device to another. However, I think some of us get hung-up when we experience INR results that vary when comparing from lab/device to another. As you well explained, measuring INR is a "rubbery thing" so the lab/device measuring the INR is not comparable to the foot measuring device.

 

[Sheenas7]

Hi


I'd call those ranges within expectable ranges and well within the 0.3 units change per day that I've seen. Further I'd say that the "clinical significance" (meaning would a clinic react to that) is negligible. I'm talking of the difference between the 3.3 and 2.9 on the same day. An experienced clinician might say that 3.3 is high enough to adjust, but equally they may say lets do a mid week test and see where its going. Thats what I'd do, and I would not adjust or change dose in the mean time. Fasting may itself be sufficient to alter your INR, which will return of its own accord later.


ok, so did you read and look at all the data in the colums above? Did you notice the variances? They are all (for each row) from the same sample. If you did not, please take a moment to open that table full screen and trace across the lines and see what 2.1 resulted in on other systems, then 2.5 ...

This difference is explained in the text.

Next we come to the word accuracy; accuracy can be measured in a watch but can not be measured in INR. Accuracy is not even a logical term when you understand how vague and cooked up an INR is. It is a rubbery thing. What you want is repeatability. You can get repeatability with a single test type, but as soon as you start introducing different tests getting agreement is nearly impossible (as every lab technician knows).

First lets go back to basics, why do you have an "INR" that you are interested in? Its because we are trying to reduce the rate of clotting speed and thus the formation of clots caused by a valve in the blood flow. Exactly how much you need to slow it is unknown without out specific testing specific to you.

The range of INR that's ideal can't be known without that, thus we have a range to be within and we also know that its less worrying to be a tiny out of range bit high than a bit low out of range. See this blog post of mine:
http://cjeastwd.blogspot.com/2017/01/2016-inr-data.html
We know also that the XS reports slightly higher in the high range (although Roche tried to address that with recent (2017) strip changes). But as we know that if INR is a little high its not of significance. The trap to fall into is attempting to micro manage and obessing about 0.1 or 0.3 difference.

Think of it like this, if shoe sizes were in 1 decimal place fittings (instead of 7, 8, 9 ..) would it make a significant difference if you were to buy a 8.3 or an 8.5? Probably not.

The Vantus btw is just another handset for the XS system (you'll see that written on your strips).

HTH

Thank you! This makes me feel good. I love all the detail you gave. I’m glad I’m doing my tests at home.

 

[Sheenas7]

Thank you! This makes me feel good. I love all the detail you gave. I’m glad I’m doing my tests at home.

So do you think doing a lab draw is not necessary or do it once a year?

 

[pellicle]

So do you think doing a lab draw is not necessary or do it once a year?

For a number of years I did one every six months, now it's been a couple of years since doing a lab draw comparison.

 

[pellicle]

As you well explained, measuring INR is a "rubbery thing" so the lab/device measuring the INR is not comparable to the foot measuring device.

well every analogy breaks down when examined from different perspectives. An analogy is a simplification to facilitate exploring a point, its not holistic.

A cone is a circle from one view, and a triangle from another ... it is neither in 3D.

 

[Sheenas7]

For a number of years I did one every six months, now it's been a couple of years since doing a lab draw comparison.

One more question. Do you ever wander if our home machine can break down / become inaccurate over the passing of years? I use Roche service and wonder if they replace after a number of years. I know a lot is based on the strips but what about the actual machine we use at home?

 

[pellicle]

Hi

One more question. Do you ever wander if our home machine can break down / become inaccurate over the passing of years?

I know it will break down one day, but I have an amount of confidence that it will not become inaccurate (or drift from its existing levels). This is because of the following:

1. the XS system is two parts, one part of the technology is in the strips, which are QC'd by the factory and it has a >lot< to lose reputation wise if that attention was to falter.
2. The strips are replaced regularly (single use disposable) with new ones
3. it is not in the nature of digital electronics to "drift", it either fails or operates. So one may see segments failing on the display (on mine, you have a more complelx model which will fail differently) and we may see it not turn on or just give error codes (recall it has internal self diagnosis modes
4. fundamentally the electronics comes down to a timer, timers are quartz driven which relies on the physics of a crystal and its oscillaiton, the electronic version of a clock pendulum. As long as its mass remains the same and gravity does not change significantly the pendulum continues to function in the same way for centuries. I have a seiko quartz watch which has remained accurate to seconds per month after 38 years
5. lastly I'd know that something was wrong if (although not triggering an error code) if I got an INR of (say) 1.4 or (say) 7 on a test suddenly. I would almost immediately go to a lab for a draw for a comparison. (there are a few stories here like that).
6. My INR has been in range for 98% of the time in the last 8 years with a dose that is typically around 7mg, if that changed I'd also be going to a lab, so any significant drifts would arouse my interest.

Back in the days that **** started on warfarin the idea was more simple; take this dose, if you see blood in your urine consult a doctor (who'll change your dose and take some bloods).

My machine has done me 8 years of daily service and (while I was not charged for it) cost AU$500 when new so less than $70 a year. I could not pay for the trip to the lab and parking fees in a city (not to mention times being late for work because of lab crowds) were I go to a lab for the strips .

HTH

 

[Protimenow]

A few years ago, a rheumatologist told me that the lab said my INR was 5.2. I didn't believe it. I tested at home after leaving the doctor's office. My meter said 2.8. I had two different labs draw my blood - one came back with 3.5, the other with 3.6. The labs were close, but not exact.

I confirmed to my doctor that the lab results were wrong and he thanked me - he had another patient whose INR was ALWAYS in range, and her results were similarly inaccurate. I told him before he changed her dosage of warfarin.

Sometimes it's good to be somewhat skeptical.

As fare as machines drifting or failing, consider the fact that some clinics and doctor's offices may use these meters for many tests each day, possibly as many in one week or less that you'd probably do in two or three years using your meter.

Meters are made to do thousands of tests without failing. They MUST give accurate (or fairly accurate) results, both for reputation's sake and for protection from lawsuits that may arise if patients wind up with strokes or bleeding incidents. They HAVE to be accurate.

Both the XS meters and the Coag-Sense use codes that either come with strips, or that are printed in bar codes on each strip, that is used to calculate INR.

Meters have to be accurate.

(Look on eBay, though, and you'll see some liquidators selling meters 'as is' or with cracked screens, or otherwise not working -- so some DO fail.)

And, with all this talk about accuracy, I've done a lot of comparison between my and the Coag-Sense, with an occasional blood draw. For some reason that I still can't figure out, at levels above 2.5 or so, the two meters diverge -- the XS goes higher faster, and the Coag-Sense doesn't report INR increases as high as those of the XS. Labs seem to be closer to the XS than to the Coag-Sense. (Years ago, when I was testing many meters in the hope of figuring out the one that I trust with my life, all three - XS, Coag-Sense, and labs were all quite close.

With two Coag-Sense and one XS, it's still a bit of a challenge to figure out why there's a growing difference between the reported INRs on the two different meters -- when both manufacturers insist that their meters (and strips) are the most accurate.

I'm almost out of XS strips, with no funds to get new ones, so I think I'll be comfortable with Coag-Sense results above around 2.3 and up to 3.0.

 

[Sheenas7]

Very important points. I had no thought that doctors use this so many times more than I would. Has traveling out of the country change anything? I would guess not but I thought I would ask. Thank you all for the comments. Very helpful.

 

[tom in MO]

Hi everyone,
I went to my annual yesterday and asked to include a blood draw at the lab for my INR.
I normally report every 2 weeks with Roche and I use Coagu Check Vantus. On Tuesday I was 3.2.
Next day was my physical and I had to go fasting. I repeated my Vantus check and it was 3.3 and then went to my doctor's office and had him draw blood. Today the lab result on the same day was 2.9. Those of you that know all the mathematical implications, is that significant? I was hoping it would be the same.

I would suggest you should ask your doctor if the difference is significant for your valve type and your INR range. You should ask your meter manufacturer if that difference is "typical" or within product tolerances.

...Next we come to the word accuracy; accuracy can be measured in a watch but can not be measured in INR. Accuracy is not even a logical term when you understand how vague and cooked up an INR is. It is a rubbery thing. What you want is repeatability....

I have a target INR range of 2-2.5. We all have a target range. We are all concerned about the accuracy of our INR results with accuracy being defined as "precise and on target ( the degree to which the result of a measurement, calculation, or specification conforms to the correct value or a standard)."

There are many definitions of accuracy. In the case of INRs, where there is a reference method, not a reference blood sample, we care about how close our meter's measurements are to the measurement system that set up the INR range. If the laboratory analysis method used for Sheenas7's blood draw is not the one used to calibrate the Vantus, her laboratory vs. Vantus results may not be a valid measurement of accuracy.

 

[ATHENS1964]

https://www.medscape.com/viewarticl...MIPM1PcgndMD-u3ZsEiQz9oIuVMmL_d-0EaaqMZfZPp8E