On-X low INR target

[dick0236]

This thread started with the question "How low can your INR go?". I have strong feelings about that since I have lived the last 50 years half-blind as a result of a of a stroke before the INR system was developed. In those early days, 1974, the doctors maintained my PT at 1-1/2 times the normal PT of 1. That equates to an INR of 1.6. I compounded the problem by going a few days without my Coumadin(warfarin) on a fishing trip. I had my one and only stroke that changed my life in several ways.....strokes are never cured. I now maintain an INR of 2.5-3.5 and my target is in the low 3's without any difficulties in my life or lifestyle.........or a serious bleed, internally or externally.

Personally, I think anyone who targets an INR below 2 is asking for trouble.......and remember, STROKE damage is almost always permanent.

To Habana58 who has the friend who has a valve like mine for 41 years, Ask her what her INR target is. Better SAFE than SORRY..........Strokes are FOREVER!!!

 

[pellicle]

but it does increase risk of bleeding, that is why i only take it once a week

and the evidence of research says (drum roll)

https://www.valvereplacement.org/th...ysis-study-focus-on-bleeds-and-cancer.889418/

not much if any unless you have heliobacter (which you should do something about)

 

[pellicle]

we all have different bodies and they operate in different manners under specific personal conditions.

agreed, but you do know that there is a very large intersection set in those Venn diagrams of us as individuals right.

I have no interest in convincing you of doing anything (and as you say, you do take it once a week). Indeed as the lifespan of platelets is about 7 to 10 days you're already doing most of what's needed anyway with that dose.

 

[Habana58]

and the evidence of research says (drum roll)

https://www.valvereplacement.org/th...ysis-study-focus-on-bleeds-and-cancer.889418/

not much if any unless you have heliobacter (which you should do something about)

Actually for me warfarin+aspirin is a no no, besides the 7 doctors in my family, 2 sisters and 5 first cousins, all tell me that, "do not take warfarin and aspirin every day", bleeding happens at many levels, now, i am just a mathematical engineer but those 7 guys spent lots of time reading and doing medicine; so, opinions are opinions , facts are facts. And, i will be ever grateful to you for all your postings and all the time you spend giving answers to questions here; Thank you.

 

[Habana58]

But, when you were at INR of 1.5 to 2.0 you were on aspirin. The FDA has approved the 1.5 to 2.0 range for On-x, but only with low dose daily aspirin. Now that you have INR range of 2.0 to 3.0, the guidelines do not call for daily aspirin, and it is more of a gray area at that point- differs cardiologist to cardiologist and patient to patient.
Hi Chuck, i always like your comments here; now; there are many studies in different parts of the world that show if people test weekly the 1.8-2.8 range is safe for valves like St Jude and Onx and similar designs, and besides the onx story, due to the fact that i come from a family of doctors i happen to know the director of a national cardiology institute in latin america and due to historical facts they have always used Se Jude valves, refuse to use tissue ones AND, all St Jude patients with AVR are managed at 2 -3, and i mean thousands of people..... so for me that is good enough.

 

[pellicle]

"do not take warfarin and aspirin every day",

you'll note that not only did I not instruct you to do that, but that I said specifically

• what you are doing is good
• I do not take it every day

just to be clear

 

[pellicle]

It supports the caution many of us have over that lower target.

actually, to tell the truth, the makers themselves have a much more cautious approach than the clinics seem to. For instance from the following letter from the valve maker we see this

I've added in segments from the references 1 Instructions for use (see attachment), 2 Otto CM et al and 3 Puskas JD et al

Previously other members have posted communications from (previous owners of the patent / IP of On-X and its been like this:

so everyone seems cautious on lower than 2.0 without aspirin including On-X and owners.

Interesting isn't it

 

[carolinemc]

But, when you were at INR of 1.5 to 2.0 you were on aspirin. The FDA has approved the 1.5 to 2.0 range for On-x, but only with low dose daily aspirin. Now that you have INR range of 2.0 to 3.0, the guidelines do not call for daily aspirin, and it is more of a gray area at that point- differs cardiologist to cardiologist and patient to patient.

I not only take warfarin with a St. Jude's valve, I also take a lose dose aspirin due to a stroke I had in 2016, valve replacement was in 2001. And you are correct, there is no call on daily aspirin. But there is the Pro-time, or INR that there is a range for both Aspirin and warfarin use.

 

[carolinemc]

you'll note that not only did I not instruct you to do that, but that I said specifically

• what you are doing is good
• I do not take it every day

just to be clear

I take both due to St. Jude's valve and eye stroke. It can be done this way, pellicle.

 

[danielwoz]

I stand by my Coaguchek in my own testing its been no more than 0.2 INR units away from lab.

This lines up with the median result in the research.
Reference: https://pubmed.ncbi.nlm.nih.gov/11978150/

 

[pellicle]

This lines up with the median result in the research.
Reference: https://pubmed.ncbi.nlm.nih.gov/11978150/

I'm sure you noticed, but that's a study of the old S model, superseded and discontinued. We now use the XS (which recently added a sub type). I believe they are a subtle improvement.

PS, you may find this interesting reading (attached)
RESULTS OF THE MASTER LOT CALIBRATION OF A NEW COAGULATION MONITORING SYSTEM FOR PATIENT SELF TESTING
Ingrid Leichsenring1, Winfried Plesch1, Volker Unkrig1, Alex Newhart1, Steve Kitchen2, Dyanne P Kitchen2,
Rhona Mclean2, Bert Dikkeschei3, Ton van den Besselaar4
1Roche Diagnostics, Mannheim, Germany, and Indianapolis, USA; 2Royal Hallamshire Hospital, Sheffield, UK; 3Isala Klinieken, Zwolle, The Netherlands;
4Leiden University Medical Centre, Leiden, The Netherlands

also

Performance Evaluation of the CoaguChek XS Plus System
Study LB 157-2005

Results:
For the three test strip lots the maximum mean bias between the CoaguChek XS
Plus and the CoaguChek XS System was 0.03 INR for samples in the range below
an INR of 2.0, and 0.07 INR for samples in the therapeutic range of OAT (INR
2.0 – 4.5).
All regression lines between systems for the low and therapeutic INR range were
equal to the line of identity (y = x). The coefficients of correlation were > 0.97.
The measuring range from 0.8 to 8.0 INR was covered by the samples.

Best Wishes

 

[danielwoz]

Plus and the CoaguChek XS System was 0.03 INR for samples in the range below
an INR of 2.0, and 0.07 INR for samples in the therapeutic range of OAT (INR
2.0 – 4.5).

These numbers in the results are really misleading for users understanding the reality of their use. The SD was 0.33 INR and SD2 seems to be about 0.6, worse outside the therapeutical range. Since they don't provide the raw data, I can only squint at the graph.

 

[pellicle]

These numbers in the results are really misleading for users understanding the reality of their use

How so?

 

[danielwoz]

How so?

Let me expand. The results showed one standard deviation test variance was 0.33 INR and two standard deviations (commonly referred to in medicine as "The Normal Range") seems to be about 0.6 INR. The results were even worse outside the therapeutical range, significantly higher than the weirdly quoted metric of 'bias' at 0.07. Given mechanical valve transplate INR target ranges are usually in a 1.0 bound (2.0-3.0 or 2.5-3.5), it's probably important to understand (even if just for anxiety reasons) that maybe 1 in 20 tests are much as 0.6 INR out, ~1 in 3 are 0.3 INR out.

 

[pellicle]

Morning

The results were even worse outside the therapeutical range,

yes, this is normal (and it was even mentioned in the article) and well established. Because we are not measuring a piece of steel here (and are instead measuring a non linear biochemical process using proxy measurements) we must expect that things in chemistry are often only useful within a range of the process. The area of interest here is "the therapeutic range" and we really don't care what it is exactly when INR is over 3.5 (because that's just simply over). This is a reasonable example of how a section of the curve is used to cover an area of interest

Cited from this article https://edu.rsc.org/resources/presenting-data-non-linear-graphs/2391.article

These things (our Point of Care machines) are measuring the time taken for a clot to form, it therefore involves many aspects and are geared to clinical significance ranges of interest.

From wikipedia

Coagulation, also known as clotting, is the process by which blood changes from a liquid to a gel, forming a blood clot.​

This change is a difficult process to measure and requires a lot of basic assumptions (and has changed over time from a steel ball not falling to the change in electrochemistry of the sample). An excellent article discussing the various approaches and their challenges is found here.

Given mechanical valve transplate INR target ranges are usually in a 1.0 bound (2.0-3.0 or 2.5-3.5),

of course ... and we know that the machine and its system has its greatest accuracy within that range. You really don't want to be pretending its all ok at 1.5 and you may not like what happens (excess bruising to say the least) when you are INR >3 ... thus we need to keep within the therapeutic range.

Some of my own data

date​	Coagu INR​	lab INR​	Difference​
14/12/2012​	3.2​	3.4​	0.2​
17/12/2012​	2.7​	2.8​	0.09​
20/12/2012​	2.6​	​	0​
21/12/2012​	2.8​	2.8​	0​
28/12/2012​	2.5​	2.6​	0.1​
5/01/2013​	2.4​	​	​
11/01/2013​	2.3​	​	​
13/01/2013​	2.4​	​	​
21/06/2014​	2.3​	2.5​	0.2​
31/01/2015​	3.0​	2.9​	0.1​
3/08/2015​	2.4​	2.6​	0.2​
6/02/2016​	3.0​	2.8​	0.2​
8/11/2017​	2.7​	2.6​	0.1​
20/04/2021​	2.3​	2.3​	0​
16/05/2022​	3.2​	3.0​	​
1/02/2022​	2.4​	2.4​	​
​	​	​	​
​	​	​	​
Mean​	2.6​	​	0.109​
StdDev​	0.3​	​	0.083​

You'll see that isn't far away from what they claim and their lab work will be very well handled, meanwhile if you think the nurses here in Australia are the same as the lab workers here in Australia then I can't help your expectation mismatch.

What matters is that we are in a range, and that we remain in that range for the majority of the time. You'll see that I remain in that range greater than 90% of the time. That's whats actually "clinically significant"


HTH

 

[danielwoz]

of course ... and we know that the machine and its system has its greatest accuracy within that range. You really don't want to be pretending its all ok at 1.5 and you may not like what happens (excess bruising to say the least) when you are INR >3 ... thus we need to keep within the therapeutic range.

I don't understand this comment, I was suggesting quite the opposite, it's quite possible to get a result of 2.3 and actually be at 1.8 (a data point from your paper). I think after understanding the variance better, the smart thing to do is to aim for 0.6 above the bottom of your range to be safe. Also large dosage adjustments should only be done in response to lab test data or after recieving a high coagu result 2 days in a row. Why? In the paper you provided, one of coagu's data points was 5 INR when it was 3 INR in the lab, if you responded to that 5 INR by skipping a day you could be well outside of therapeutic range in the following days.

The mean on your personal data table seems wonky, it should either be 0.119 or 0.116 if you include the last two data points. I guess thats why Queensland is #2 in Math

Thankyou for your data, it roughly aligns with the paper you provided, that in therapeutic range, that in 1/3 samples there is a ~0.3 INR variance. You have way too little data to know if it confirms, for you as an individual, that 1/20 samples could be as high as 0.6 in variance, since this may be individual specific.

 

[pellicle]

I don't understand this comment,

seemed pretty simple and clear to me

I was suggesting quite the opposite, it's quite possible to get a result of 2.3 and actually be at 1.8 (a data point from your paper).

firstly I think you misunderstand statistics, secondly I didn't write any paper (so I'll assume you mean "the paper I provided"

If you have some specific point in that then its considered good form (and required in some areas) to actually cite that (gone are the says of just mentioning the book and author; page and quotation are simple courtesy)

I think after understanding the variance better, the smart thing to do is to aim for 0.6 above the bottom of your range to be safe

I wouldn't have an issue with doing that ... but you'll need to be aware that doing that will expose you to more bruising (like when I walked into my tow bar)

yes, that one really hurt..

Also large dosage adjustments should only be done in response to lab test data or after recieving a high coagu result 2 days in a row.

I'm not sure what this is in relation to ... normally I'd fully agree, but without any context its meaningless

In the paper you provided, one of coagu's data points was 5 INR when it was 3 INR in the lab,

that's more like it ... "the paper I provided" ... but where? Please reference that because I can't find that

if you responded to that 5 INR by skipping a day you could be well outside of therapeutic range in the following days.

I would doubt that ... some evidence would be nice

The mean on your personal data table seems wonky, it should either be 0.119 or 0.116 if you include the last two data points.

what are you talking about?

I guess thats why Queensland is #2 in Math

I guess ... nice stereotyping there ...

Thankyou for your data, it roughly aligns with the paper you provided, that in therapeutic range, that in 1/3 samples there is a ~0.3 INR variance. You have way too little data to know if it confirms, for you as an individual, that 1/20 samples could be as high as 0.6 in variance, since this may be individual specific.

again I think I don't know that we are talking about the same thing ... I suspect you have misread what I've put up and conflated my charts (weekly readings per year dose given as daily dose) with the table I put up of the times I've directly compared my coaguchek to my lab.

My own data readings data is more akin to this:

​	​	INR​	daily dose​
​	average​	2.6​	6.8365​
​	std dev​	0.4​	0.3​
​	max​	3.3​	7.5​
​	min​	1.8​	6.0​
​	over event​	2​	​
​	under event​	1​	​
​	inRange %​	94.2​	​
​	​	​	​

which is the data for the 2021 graph

This discussion reminds me of a comedy sketch with you now about to ask me for payment.



Your replies so far have appeared to me to be disingenuous and your resort to stereotyping me based on my State of residence shows that you are not actually discussing science and maths (and maybe its also beyond you? ¯\_(ツ)_/¯ ... I can't be sure). Either way I'm done here

I can't see how you support this

You have way too little data to know if it confirms, for you as an individual, that 1/20 samples could be as high as 0.6 in variance, since this may be individual specific.

I'm not even going to speculate, as it feels like you've made a lot of conflated assumptions. I wonder if you properly read and understood the papers I attached.

bye

 

[dick0236]

........and 1+1-2=0

 

[danielwoz]

bye

Based on other hostile replies I see you write regulary on this forum, I wish this was true.

 

[Protimenow]

The study makes me wonder how many people were hurt by following On-X recommendations for low INR ranges. The suggestion that 1.5 - 2.0 is safe, and convincing gullible doctors that there's some benefit to patients when they maintain an INR of 1.5, versus 2.2 - 2.5 (or slightly higher) seemed to be irresponsible and, perhaps even worse. Marketing is fine - but not when it puts patients at real risk.