INR Discrepancy

[RobNDenver]

I have been testing about once a week with my INRatio results in the 2.3 to 3.0 range for the past several weeks. Last night I tested and my INR was 2.4. A little low but not bad . . . So today I had to have a lipid panel done and the lab tech asked if I wanted a PT/INR done too. Sure, I said and the results came back 1.9??? WTH is up with this?

I went back and looked at my spreadsheets and realized that the lab's INR is typically about 20% lower than the INRatio shows.

At this point, I am really concerned that I have been under dosing my self, and the INRatio is way off? How can I determine that it is functioning properly. I don't have anyone I can stick who is not on warfarin.

 

[Freddie]

Rob, I have the same concern and I've been told from the good people here to "trust your monitor".

The difference I get from the lab vs monitor is anywhere from .2 to .4, and I've comparing the two for the past year. Problem is my doc goes by the lab results - my doc believes the monitor is not accurate enough the higher your range is - BUT THAT's my Doc.

I've also read (someone correct me if I'm wrong) that a acceptable difference lab to monitor can be from a .2 to a .9. Personally a .9 is a little high for my liking, and the lab has a acceptable variance of +/- of .2

I wouldn't worry, trust your monitor.

 

[Ross]

Rob your comparing apples to oranges. The lab could/is using a different reagent which will give a different result. INR was supposed to level the playing field, but I have not found that to be true when comparing labs. Trust your machine. Any time you doubt your machine, test twice and if the results are within .2 of each other, it's accurate or stab someone not on Coumadin and test theirs. It should be .9 to 1.2.

Also, you'd have to test at the exact same time they drew the blood with your machine to even try to compare. INR changes from minute to minute.

I have a chart posted around here somewhere explaining the difference in reagents and results. Let me see if I can find the link.

Your strips have an ISI of 1.0

 

[olefin]

I too, once doubted my INRatio2. It wasn't long after I started home testing. It was showing 6.8, retested a finger on the my other hand, 7.1. My wife is not on warfarin so I tested her, 1.1. The high INR was because I had increase too much warfarin. Since then, I've kept my control right on the money by testing weekly and making 2% changes in warfarin when needed.

I haven't had a lab test in over a year. Back then the cardiologist was requiring once a month lab test so he could collect pay from Medicare. His bookkeeper wasn't aware he could have been collecting pay from Medicare for reading INR results from QAS. I then switched cardiologist.

 

[dick0236]

Rob your comparing apples to oranges. The lab could/is using a different reagent which will give a different result. INR was supposed to level the playing field, but I have not found that to be true when comparing labs. Trust your machine. Any time you doubt your machine, test twice and if the results are within .2 of each other, it's accurate or stab someone not on Coumadin and test theirs. It should be .9 to 1.2.
0

I agree with Ross. You really can't compare results taken with two different machines and you cannot compare results from machine vs lab draw. Taking blood from a normal (nonACT) person, who will be around 1.0, is the better way to calm your fears. You must have ONE friend who would "walk thru fire" for you Offer to buy him/her a beer AFTER you stick 'em.

 

[Bina]

I've been using my Coaguchek XS for almost 3 years and whenever I have done any comparisons my results are usually exactly the same as the hospital lab.
No difference has been more than .2
Rob, stick a normal person if u can, or try another lab.... Just for the sake of knowing where u stand.

 

[HemosenseTech]

So I thought I'd dabble into this a bit since I deal with this on a daily basis. There are 3 ranges allowed by The World Health Organization (WHO) and International Organization of Standards (ISO) when comparing a Point-of-Care (POC) meter and a lab:

Lab INR: 0.0 - 2.0: POC meter shall be within +/- 0.5 units of the Lab.
Lab INR: 2.0 - 4.5 (Most important range): POC meter shall be within +/- 30% units of the Lab. (e.g. Lab 3.0, meter COULD HYPOTHETICALLY be as low as 2.1 or as high as a 3.9 (3.0 x 30% = 0.9)
Lab INR: > 4.5: No criteria established for acceptable correlation standards.

I've posted this before on the old Valvereplacement and thought it would be good to re-emphasize. We, as a company, have a summarized document that describes this and there is a wonderful study out there by Dr. Alan Jacobson that describes why these ranges were set. Here are the two summarizing documents:

INR + ISI and the Jacobson study:
http://www.hemosense.com/docs/5500258_TechBull103_INR_ISI_revC.pdf

Clinical Expectations according to WHO and ISO:
http://www.hemosense.com/docs/5500330vC_TechBull109_ClinicalExpPT.pdf

If there are anymore questions or need for further explanation of these ranges and why they might exist feel free to post here or PM me

 

[ALCapshaw2]

So I thought I'd dabble into this a bit since I deal with this on a daily basis. There are 3 ranges allowed by The World Health Organization (WHO) and International Organization of Standards (ISO) when comparing a Point-of-Care (POC) meter and a lab:

Lab INR: 0.0 - 2.0: POC meter shall be within +/- 0.5 units of the Lab.
Lab INR: 2.0 - 4.5 (Most important range): POC meter shall be within +/- 30% units of the Lab. (e.g. Lab 3.0, meter COULD HYPOTHETICALLY be as low as 2.1 or as high as a 3.9 (3.0 x 30% = 0.9)
Lab INR: > 4.5: No criteria established for acceptable correlation standards.

I've posted this before on the old Valvereplacement and thought it would be good to re-emphasize. We, as a company, have a summarized document that describes this and there is a wonderful study out there by Dr. Alan Jacobson that describes why these ranges were set. Here are the two summarizing documents:

INR + ISI and the Jacobson study:
http://www.hemosense.com/docs/5500258_TechBull103_INR_ISI_revC.pdf

Clinical Expectations according to WHO and ISO:
http://www.hemosense.com/docs/5500330vC_TechBull109_ClinicalExpPT.pdf

If there are anymore questions or need for further explanation of these ranges and why they might exist feel free to post here or PM me

HemoSense Tech -

Thank you for this input which clearly shows the differences between different ranges.

As I have stated several times, it takes 2 Points to define a straight line and multiple points to define a curved line.
(i.e. confirming INR at 1.0 is a Necessary But NJT SUFFICIENT condition in proving INR Measurement Accuracy over the FULL Range.)

My Coumadin Clinic uses Finger Stick Instruments. They were well aware that INR measurements over 5.0 were 'questionable' or at least unreliable. Their solution was to require a Vein Draw for INR over 5.0 (which has now been revised to require a Lab Draw for readings over 4.5. FWIW, they recently changed from a Finger stick instrument that could read Bar Codes, i.e. Patient I.D.s, to Coaguchek XS instruments. Their experience is that the XS is 'better' than their previous instruments but still exhibits sufficient variability above 4.5 to warrant Lab Test confirmation. My CRNP told me that she had seen Identical Results at 5.5 (XS vs Lab) and another patient read 4.7 on the XS but 3.8 from the Lab.
Their experience seems to confirm that In-Range Readings are sufficiently accurate to NOT require Lab Confirmation.

'AL Capshaw'

 

[HemosenseTech]

The toughest thing for me to explain to patients or nurses or even doctors is that there is no gold standard for INR testing. This is a case where so many factors play a part in determination of INR that it would be almost as amazing if you got perfect correlation, rather than the expectation that there is. Sample type, methodology, reagent sensitivity all play major roles in determining an "accurate" INR but ultimately because there is nothing to compare a result to, as far as a standard goes, it's all about how well a group of labs or POC meters agree with each other. INR is a situation where accuracy by consensus should be the correct method but that'll be a generation away...

 

[Cris N]

The toughest thing for me to explain to patients or nurses or even doctors is that there is no gold standard for INR testing. This is a case where so many factors play a part in determination of INR that it would be almost as amazing if you got perfect correlation, rather than the expectation that there is. Sample type, methodology, reagent sensitivity all play major roles in determining an "accurate" INR but ultimately because there is nothing to compare a result to, as far as a standard goes, it's all about how well a group of labs or POC meters agree with each other. INR is a situation where accuracy by consensus should be the correct method but that'll be a generation away...

I agree... there is no gold standard. The unfortunate thing is that many medical professionals seem to believe that the lab's test is it, when in reality there are so many reasons for a lab's results to be off. As long as my test strips are fresh and haven't been recalled - I know that my actual testing is consistent from week to week. It doesn't sit on some counter waiting for someone to get around to running the test.

I trust my monitor.

 

[Philip B]

Differences

Differences

I agree with what others have said regarding differences between your home tests and the results your lab is getting. The differences used to bother me when I first began home testing.

The problem with the lab results is that there's little consistency. Results can vary due to how the lab techs do the testing with samples. An accident last June really confirmed this.

I had home tested my INR on the morning of the accident and determined that I was in range. Later in the day, I was a victim of a hit and run driver who rammed my bicycle from behind. The accident landed me in our local hospital's ER. Since I had sustained a major ead injury, the ER doc ordered an INR test; the test came back with my INR well within range. I was transported by ambulance to a larger hospital in another town 45 minutes away. When I arrived another INR test was ordered; this second lab test came back with extremely high results. Based upon the results of the test the ER doc wanted to admit me to the hospital. I told him the results of the second lab test were wrong and to run the test again. My response angered him, but he ordered another lab test. This test, which was processed by the same techs who had done the previous one, came back with results that placed me well within my range.

I trust my monitor. Week to week, I use the same procedures when I test. Since I'm consistent, I trust the results.

-Philip