INR management - current European view

[pellicle]

European Society of Cardiology
Guidelines on the management of valvular heart disease (version 2012)
The Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS)

Section 11.2.2.2 Target INR

We recommend a median INR value, rather than a range, to avoid
considering extreme values in the range as a valid target INR, since
values at either end of a range are not as safe and effective as
median values.

I happen to like this view as it seems to settle that old discussion (how many times have we seen this) where somone says "my surgeon says my INR range is 2.5~3.5 so I'm sitting on 2.5" (cos), it also makes it clearer (at least to me) that there is not a 'hard line' where its [LostInSpaceRobot] Danger Will Robinson and you suddenly burst our bleeding or form a clot and seize up.

well, at least that's how I read it.

They go on further to qualify INR and bleeds:
11.2.2.3 Management of overdose of vitamin K antagonists
and bleeding

The risk of major bleeding increases considerably when the INR
exceeds 4.5 and increases exponentially above an INR of 6.0. An
INR ≥6.0 therefore requires rapid reversal of anticoagulation
because of the risk of subsequent bleeding.

so I read this as meaning less than 4.5 is "don't panic"
greater than 4.5 "do something about it"
greater than 6 is "do something about it urgently"


it continues...

In the absence of bleeding, the management depends on the
target INR, the actual INR, and the half-life of the vitamin K antagonist
used. It is possible to stop oral anticoagulation and to allow
the INR to fall gradually or to give oral vitamin K in increments of 1
or 2 mg. If the INR is .10, higher doses of oral vitamin K (5 mg)
should be considered. The oral route should be favoured over the
intravenous route, which may carry a higher risk of anaphylaxis.

They then suggest (relative to self management)

Self-management of anticoagulation has been shown to reduce INR variability and clinical events, although appropriate training is required. Monitoring by an anticoagulant clinic should, however, be considered for patients with unstable INR or anticoagulant-related complications.

in case that should be of benefit to anyone

 

[tom in MO]

A median is "being in the middle" or "relating to or constituting the middle value of an ordered set of values."

You cannot have a median w/o a range. You cannot have a median w/o an existing set of numbers. In other words, a median is not a "target" but a statement about a set of numbers.

My range is 2-2.5. This tells me that outside this range, a dose adjustment may be warranted. If I was just given the median of 2.25, I would not know when to adjust my dose.

Since my range is so small, I do not agree with the European Society of Cardiology that "values at either end of a range are not as safe and effective". I like my range. I have been as low as 1.7 and as high as 3.2 w/o problem and w/o worry on my part. I did adjust though and test the next week, instead of once every 2 weeks.

 

[TheGymGuy]

Awesome read. Can you also post a link to the article. I will archive it in my blog.

 

[pellicle]

Can you also post a link to the article. I will archive it in my blog.

err ... I was given it as a PDF by my sales friend I can put the PDF into my GoogleDocs and put a link to that if that's ok?

PS: I just googled the title and found this link:
http://www.ncbi.nlm.nih.gov/pubmed/22922415

but without access to that service then my google docs copy will be probably the only easilly found public one. Of course it would mean that its an "uncontrolled copy" (meaning not distrubted on an official site)

here you go, link to my google drive

 

[pellicle]

A median is "being in the middle" or "relating to or constituting the middle value of an ordered set of values."

You cannot have a median w/o a range.

correct, pardon me for not discussing that (it could have been left as an exercise for the reader), but essentially I was only interested in commenting on the notion of a median and a range. Rather than a pair of parallel lines on a graph which inside is FINE and outside is BAD I was taken by the idea of a statistical and distributed risk presentation

The guidlines state (more fully):

11.2.2.2 Target INR

In choosing an optimum target INR, one should consider patient
risk factors and the thrombogenicity of the prosthesis, as determined
by reported valve thrombosis rates for that prosthesis in relation
to specific INR levels (Table 20).203,219 Currently available
randomized trials comparing different INR values cannot be used
to determine target INR in all situations and varied methodologies
make them unsuitable for meta-analysis.
Certain caveats apply in selecting the optimum INR:

• Prostheses cannot be conveniently categorized by basic design (e.g. bileaflet, tilting disc, etc.) or date of introduction for the purpose of determining thrombogenicity.
• For many currently available prostheses—particularly newly introduced designs—there is insufficient data on valve thrombosis rates at different levels of INR, which would otherwise allow for categorisation. Until further data become available, they should be placed in the ‘medium thrombogenicity’ category.

Table 20 Target international normalized ratio (INR) for mechanical prostheses

Prosthesis thrombogenicity [sup]a[/sup]	No risk factor[sup]b[/sup]	Risk factor ≥1[sup]b[/sup]
Low	2.5	3
Medium	3	3.5
High	3.5	4

[sup]a[/sup] Prosthesis thrombogenicity: Low ¼ Carbomedics, Medtronic Hall, St Jude Medical, ON-X; Medium ¼ other bileaflet valves; High ¼ Lillehei-Kaster, Omniscience, Starr-Edwards, Bjork-Shiley and other tilting-disc valves.
[sup]b[/sup] Patient-related risk factors: mitral or tricuspid valve replacement; previous thromboembolism; atrial fibrillation; mitral stenosis of any degree; left ventricular ejection fraction ,35%.

if you don't mind me using this as a working example

My range is 2-2.5. This tells me that outside this range, a dose adjustment may be warranted. If I was just given the median of 2.25, I would not know when to adjust my dose.

So if you were following a trend in the INR and it was continuing to trend away from your median then dosage adjustment (or dietary adjustment) may be warranted. The seriousness of the trend would be how rapidly (slope) and how long (duration) it was going away from your median.

Risk of event is related extremes of range and duration. So as long as you kept close to your INR median and outside the danger zones all should be fine. Yhey didn't discuss a lower danger zone, only upper ones as identified, perhaps because they feel that this overcomplicates things and that if you are keeping near your target median you won't be down low for long.

As they say:

High variability of the INR is a strong independent predictor of reduced survival after valve replacement.

Clearly this isn't a problem for you because your range is so tight. Further the supplimentary numbers (the ones I gave) and (probably different) training would help you to understand that.

Since my range is so small

actually it is very small, can I ask how often you are within that range? Say as a percentage? My surgeon gave me a range of 2.2~3 and I have been within that range 95% of the duration of 2013 (for instance).

I have been as low as 1.7 and as high as 3.2 w/o problem and w/o worry on my part.

which fits within the guidelines I cited above. As you've stated your data it seems that you pretty much conform to their guidelines. I suspect that the problems you have is with the wording and that its a different description to what you're used to

Anyway, I'm not concerned with your methods, for you are already comfortable with them and happily managing your INR. I posted this for others who may not be yet comfortable with their self management and may be still struggling with "what to do and how to do it" (you know, newbies).

PS: for interest my statistical data on mylast years data (graph available here) is mean: 2.5 std Dev: 0.3
(although I doubt anyone was interested in that ...)

 

[tom in MO]

I can't say how often I am in the range. I don't track that because I do not feel it is relevant. I measure success by life w/o clots or bleeds; not by time in range.

For my valve, the table states the European guideline would be 2.5, not my range of 2-2.5. If I was a nervous patient, I'd get concerned with values around 2.0, and maybe think I need an adjustment. However, with my range, I'd still be OK.

I wonder why they did not set it at 2.2 to match the US work. Did they set it high with the thought that if one is going to have a single value for a target make it the high level...

I take a more existential viewpoint of INR management rather than an emprical one. To me the INR is not a "real" thing. The number is the clotting time normalized to an international standard, so to me its a measurement of the kinetics of a reaction in relation to a standardized reaction. This is pretty esoteric compared to most blood tests.

There are a host of possible sources of error in home or laboratory testing, thus, to me, an INR value of 2.5 is not a number but really a range of 2.2 - 2.8; my meter just gives a single number for convenience. For most people comparing a range to a range is even more problematic than comparing a number to a range. So for me, one number with a given range is a much simpler approach than one number compared to a reference number with no way to tell how close you need to be.

 

[dick0236]

There are a host of possible sources of error in home or laboratory testing, thus, to me, an INR value of 2.5 is not a number but really a range of 2.2 - 2.8; my meter just gives a single number for convenience. For most people comparing a range to a range is even more problematic than comparing a number to a range. So for me, one number with a given range is a much simpler approach .

I like "simpler approach". I did my weekly test this am and it was 3.2. All that really tells me is that I am in the mid/upper part of my range(2.5-3.5)......last Wednesday it was 2.7. Either of these numbers is OK with me. So long as I stay within, or close to, 2.5-3.5.....I don't sweat it.

 

[pellicle]

Hi

being a bit 'gun shy' about "its only conversation" around here ... I'll engage in conversation

I can't say how often I am in the range.... not by time in range.

fair enough ... its only relevant to people who look at trying to understand why problems happen. so if you don't have problems ... well then no problems

For my valve, the table states the European guideline would be 2.5, not my range of 2-2.5.

true, but in practice (for me and others I've conversed with) such a narrow range is as good as saying your target INR is 2.25 (as you mentioned).

personally I'd think of 2 as a minium ... which valve do you have?

I wonder why they did not set it at 2.2 to match the US work.

dunno .. I wasn't part of the panel and nothing is written in the document (if you read it) to suggest how they came to their conclusions. My surgeon is of the view that 2.5 is better than 2.2 (even though he gave me the range of 2.2~3) because he has said to me (and I did ask btw) that evidence he's read suggests that pannus growth is reduced with INR above 2.2 ... in reality the possibility of accurately measuring 2 VS 2.2 VS 2.7 is unlikely with any reliability, so I guess (going out on a limb here) that the panel picked a target of 2.5 to mitigate against error on the under side.

Protimenow has had a TIA with a meter reading in the 2's IIRC ...

There are a host of possible sources of error in home or laboratory testing, thus, to me, an INR value of 2.5 is not a number but really a range of 2.2 - 2.8;

this is sounding to me like zeroing in on the 2.5 target (taking into account error) ... but its perhaps just because I'm a stats kind of person....

my meter just gives a single number for convenience.

so does my meter and so did my lab ... they didn't give me an error range in their measument

anyway ... I found it worthwhile, sorry that you didn't ... but it didn't cost much to post it

 

[pellicle]

So long as I stay within, or close to, 2.5-3.5.....I don't sweat it.

this attitude is of course perfectly fine.

Again, the only reason I thought it was a worthwhile post is because I have read people say thing to the effect of:
"my Dr said my range is x to y so I'm going to sit on x because I want to be lower"

To me this type of (mis)interpretation is removed with a target INR (and it being a median) rather than a range.

 

[Ola Thoresen]

My experience is that my INR varies so much, even from day to day, that having an extremely narrow range, or trying to stay spot on any single value is simply not possible. All it might lead to is to many and to big adjustments of your medication doses.

I agree with pellicle that one should therefore try to aim for the median or middle of the suggested range.

On the other hand, my range is 2-3, but I would be much more worried if I measured a 1.5 than a 3.5. This is also the reason I recently increased my dosing slightly, after 3 weeks in a row with < 2.5 (but still "in range")
My experience from my "NovembINR"-project (read about it in another thread) showed that the INR could vary with as much as +/-0.8 from one day to the next. So if you measure your INR once a week, and get a 2.0 one day, you could potentially be at < 1.5 the next day, and you would not know it. And if you measure a 1.5, you could actually have been a lot lower for several days already.

I wonder why your range is as narrow as a 0.5. It must be almost impossible to stay in that range, and for me at least that would be really frustrating. Like a goal I could never really reach. At least with my current range I stay within the margins most of the time.

 

[Protimenow]

Tom-I'm worried about a range of 2.0 - 2.5. As I've said in other posts, if your meter is reporting values that are higher than your ACTUAL INR, being satisfied with a meter value of 2.0 COULD mean that your actual INR is more like 1.7 or 1.8. As Ola just said, the variance from day to day could also mean that your ACTUAL INR may be lower than 2.0 for the day (or week, perhaps) between tests.

My meter put me around 2.0 -- and after a week of this, I had a TIA. The hospital values were 1.7 and 1.8.

I don't know that it matters WHICH valve you've got ticking in your chest, I would be concerned with having an ACTUAL INR below 2.0 for more than a day or two. To me, I am much more comfortable with an INR in the middle of my range (which happens to be 2.5-3.5, in spite of my clinic's recent recommendation of 2.0 - 3.0 - clearly, they don't know that there's a difference between valves). I am also more comfortable with an INR at or above my range than I am with an INR at or below the bottom end of the range.

Last year, I conducted a search for the meter that I had most trust in. I tested with Coag-Sense, InRatio and InRatio 2, CoaguChek XS and ProTime (and ProTime 3), and monthly lab tests. I found that the InRatio meters were reporting too high, too often for me to be confident in their values. The CoaguChek also often reported a bit higher than the labs, but not so much that it was much of a worry. The CoagSense reported equal to, or somewhat lower than the lab (and often lower than the other meters). To me, because I am much more concerned about being BELOW 2.0 than I am about being ABOVE 3.5, I chose the CoagSense as my meter for most of my testing. It's the one that I am trusting with my life.

(That said, I just got a CoaguChek XS Pro on eBay and will probably try that one, too, comparing it to the CoagSense, the lab, and probably the other CoaguChek XS that I own. I may, soon, put some meters on sale -- maybe even on eBay).

 

[pellicle]

(That said, I just got a CoaguChek XS Pro on eBay and will probably try that one, too, comparing it to the CoagSense, the lab, and probably the other CoaguChek XS that I own. I may, soon, put some meters on sale -- maybe even on eBay).

In camera land we have a term "Gear Acquisition Syndrome" or GAS for short.

I think you may have a touch of GAS

Some camera humor http://cjeastwd.blogspot.com/2014/01/camera-related-illnesses.html

 

[Protimenow]

GAS. Probably so. In regards to INR testing, I'm still looking for the 'perfect' meter -- and in my testing, I have (for myself, at least) had enough experience to eliminate one from my list of usable (accurate) meters. You should see my power tools -- can't these manufacturers ever sell a saw without throwing in a drill? I must have 6 or 7 fully usable, but completely unneeded, electric drills.

 

[gerrychuck]

In camera land we have a term "Gear Acquisition Syndrome" or GAS for short.

I think you may have a touch of GAS

Some camera humor http://cjeastwd.blogspot.com/2014/01/camera-related-illnesses.html

Funny; in the music world we have GAS too, but it's "guitar acquisition syndrome" , not to be confused with an even more serious, but related malady in the aviation world: AIDS, or "aircraft induced divorce syndrome". I definitely suffer from the first, but have so far avoided the second!

 

[gerrychuck]

More seriously, I really like the "median" versus "range" paradigm. My prescribed range is 2.5-3.5, or a median of 3 (which coincides perfectly with my recommended median from the Euro guidelines). I just went back 20 weeks and calculated my median INR from that period: 3.025. Can't do much better than that, and the logic of using that way of looking at INR really makes sense to me.

 

[TheGymGuy]

Us gear-heads say that G.A.S means - Gadget Acquisition System.

 

[river-wear]

You guys are arguing classic "old school" quality control methods. In manufacturing, you have control charts with the average process output (~your INR) and control limits based on statistical variation of historical results (usually 3-sigma). Then you have to make sure that those limits are within your drawing specifications (INR range from the doctor). The key in either scenario is to target the optimal value, but not make adjustments unless you fall outside of the allowable range. And as pellicle says, you don't want to try to stay at the upper or lower limit of a reasonable range because statistically, you will fall outside of the therapeutic range far too often.

Hope that's not too obscure... I like to read your discussion. It helps me get a grip on this new hassle that will probably be a part of my life soon.

Do you guys really trust meters (and test strips) purchased on eBay? I remember buying some razors for my hubby on eBay once and the packaging was in another language... and they just weren't up to the standard he was used to. I don't know if they were counterfeit or just made to a lower standard for another market. Either way, I don't buy certain things like that on eBay anymore because of it.

 

[Agian]

You guys are arguing classic "old school" quality control methods. In manufacturing, you have control charts with the average process output (~your INR) and control limits based on statistical variation of historical results (usually 3-sigma).

I've always said that. But, who listens?

 

[Protimenow]

Michele:

Average is good -- but 'average' can still happen when the INRs are wildly out of range (as you noted). Personally, I don't EVER want my INR below 2.3 or 2.4 -- even though some guidelines (probably ignoring the older valves, like the one in my chest, that probably still need a higher range) say 2.0-3.0.

As far as trusting meters -- I've bought many meters on eBay and, aside from those that were 'as is', have had few problems with them - other than inherent inaccuracies of the meters. In some cases, these meters were previously used in clinics, hospitals or doctor's offices and were 'retired' for tax reasons (the capital equipment was amortized so there were no further deductions available), or owned by grandpa, who didn't use it, so the kids are selling it off. Most are probably fully functional and still have hundreds of tests left.

I ran a large series of tests comparing meters to blood draws (I think the thread was called 'A New Meter - Coagusense', I determined that, for me at least, the InRatio and InRatio2 (the InRatio 2 came directly from the manufacturer) were always so much higher than labs that I wouldn't trust my life to it. The CoaguChek XS results were often close to the lab results, but usually somewhat higher than the lab. I found that the Coag-Sense was close to my lab results, and often just slightly below. For me, avoiding an INR that was too low was more important than having an INR that was at the high end of my range. Although they're not as easy to find, I decided that the Coag-Sense is my preferred meter.

I don't know that I would be particularly worried about buying a meter on eBay. I buy my strips on eBay. I don't think that I've had a problem. (The only bad batch of strips that I've encountered came from a medical supply company -- and the manufacturer issued a recall on them, after I reported that they seemed to be giving results that didn't match the lab results). If you buy a meter, see if you can make a return if you don't like it.

 

[big_L]

Sorry, but 30 years in quality control here...

It's the old "we're in specification, so we're fine" argument vs. the approach to "manufacture to target". If you're in range, but not at the mean do you say "I'm OK" or do you twiddle with the process (dosage). Of course, to make this decision, you need knowledge of the process and measurement variability. To much twiddling can lead to increased variability. Of course, you have to balance that against the risk you're taking leaving it alone when it's barely in specification.

I think we've all seen enough variability in our results to decide (anecdotal evidence) that we don't fiddle with dosage if it's in range or maybe a hair out of range.