CoaguChek XS - Accuracy verification

[UncleSteve]

The head of my local INR clinic is very anti home/self testing, and discovered I now have my own CoaguChek XS and basically ripped in to me about their accuracy and seems to pour scour on anyone else using them and their methods (inc hospitals not under his control).

This manager wants me to get involved with some sort of external accuracy verification scheme to ensure the machines accuracy. Once I twigged his attitude I stopped trying to defend why I want to use my own machine and let him rant on without saying or agreeing to anything as he's clearly better than anyone else!

Are there any others in the UK subscribed to any schemes for external accuracy verification and what are their costs and who is the scheme run by please?

Thanks

 

[dick0236]

The head of my local INR clinic is very anti home/self testing.......

........This manager wants me to get involved with some sort of external accuracy verification scheme to ensure the machines accuracy.

I had the same problem with a majority of docs where I live and I finally just bought my meter/strips outright. I test weekly and go to the INR clinic in my PCPs office monthly. I guess you could call my monthly visits to INR clinic an "external accuracy scheme". It works for me and my meter(INRatio2) is almost always within +/- .2 with his Coaguchec XS......and last test was "dead on" since we both got 2.8.....so I guess his meter passed the test LOL.

 

[pellicle]

Hi

The head of my local INR clinic is very anti home/self testing, and discovered I now have my own CoaguChek XS and basically ripped in to me about their accuracy and seems to pour scour on anyone else using them and their methods (inc hospitals not under his control).

mmm ... well with his head wedged that far up his **** he's unlikely to smell the coffee even if you do wake him up.

Once I twigged his attitude I stopped trying to defend why I want to use my own machine and let him rant on without saying or agreeing to anything as he's clearly better than anyone else!

fair call really ...

Are there any others in the UK subscribed to any schemes for external accuracy verification and what are their costs and who is the scheme run by please?

sorry that I can't answer your question (I'm a Colonial) but if you wished any 'references' I can send you some papers.

My approach in Australia was to resign from the clinic (QML) and just do it myself out of my own pocket. My Surgeon was comfortable with my decision and only advised a lab draw now and then to compare results ... I can get my GP to arrange that "as needed".

some papers...

Accuracy of Capillary Whole Blood International Normalized Ratio on the CoaguChek S, CoaguChek XS, and i-STAT 1 Point-of-Care Analyzers
Brad S. Karon, MD, PhD,1 Robert D. McBane II, MD,2 Rajeev Chaudhry, MBBS, MPH,3 Lisa K. Beyer,1 and Paula J. Santrach, MD1

Abstract
We evaluated the accuracy of capillary whole blood international normalized ratio (INR) on the CoaguChek S (Roche Diagnostics, Indianapolis, IN), CoaguChek XS (Roche Diagnostics), and i-STAT 1 (i-STAT, East Windsor, NJ) point-of-care (POC) analyzers compared with venous plasma INRs determined by a reference laboratory method. Overall agreement between POC and laboratory plasma INR was very good, with median bias between capillary whole blood and laboratory plasma INRs varying from 0.0 to –0.2 INR units on all devices.
More than 90% of results on the CoaguChek XS ... were within 0.4 INR units of the reference laboratory method.

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

...
The regression line after Bablok-Passing was y = 0.996x - 0.001 (x: mean IRP),
the mean bias for patient data only was -0.01 INR or - 0.04% (Fig. 5 and 6).
The duplicate determinations with the CoaguChek® XS system showed a very low
imprecision (high precision) for the master lot INR results with a CV of 1.1% (Fig. 7).
...
CONCLUSION
The master lot of the new CoaguChek® XS system was successfully
calibrated with high precision according to the WHO guidelines.

 

[UncleSteve]

I'm not going to try and convert the 'guy' as he's clearly on his own mission. Where as my GP is on my side as he's prescribed the test strips, so get them at a reduced cost. My GP did warn me that I would be dropped of all responsibility by the INR masters if I took the route I have.

I have now emailed Roche to see what their view is, or at least to see if they can provide any links or support which I'll pass on if I get a reply (I have asked questions before without getting a reply).

 

[pellicle]

My GP did warn me that I would be dropped of all responsibility by the INR masters if I took the route I have.

what are the implications of that?

 

[UncleSteve]

My GP did warn me that I would be dropped of all responsibility by the INR masters if I took the route I have.

what are the implications of that?

I think he was being careful on what he was saying (I suspect he may have had the same sort of hostile conversion as I had with HQ), but happy to prescribe the test strips on the NHS (not free, but reduced cost).

 

[pellicle]

but happy to prescribe the test strips on the NHS (not free, but reduced cost).

well ... at least you get that. In Oz I just have to reach into my pocket.

 

[Protimenow]

I'm assuming that your clinic relies on the blood draws as the gold standard. This is certainly NOT ALWAYS the case. I once had a blood draw at a clinic and a lab, about two hours apart, and the results were not as close to each other as expected. Some clinics will do the blood draw, spin the blood in the centrifuge, store it until some lab courier picks it up, throws it into a hot car and drives around with it, and eventually the blood is tested. At the lab, it may be properly stored and tested very soon after it's drawn. There are differences between the reagents that labs use. The strips used by the meters are supposedly calibrated so that the reagents will yield the correct INR.

In short, it's not really possible to know which test is 'accurate.'

I've gone through a lot of testing -- and for the 18 months or so, I have been able to compare my meter results to those from a hospital lab (in which I have a fair amount of faith). In my case, I'm looking for a meter that will actually report exact or slightly lower results than those from a lab. Therefore, a 2.0 - 2.5 on my meter would keep me feeling as if my INR is out of stroke range. In my experience the InRatio and CoaguCkek XS were either close to lab results or higher (and sometimes MUCH higher). As the actual INRs get higher, the error in the XS and the InRatio seem to become greater and greater.

After extensive testing of multiple meters, the one that I trust the most is the Coag-Sense, followed by the XS. I've stopped using the InRatio (and 2) because they seem to always return values that are considerably higher than the labs.

The ultimate question is -- if your clinic uses a meter, how sure are they that THEIR meter is accurate? (I recently got a CoaguChek XS Plus, which allows quality control testing using standardized test materials provided by Roche -- this tells whether or not the meter is 'accurate' or has quality control issues).

Comparing your meter's reading to a lab test taken within minutes (or maybe within hours) of your test can help to give you an idea of any bias that your meter may have when it reports your INR.

 

[lance]

Good morning Steveq
Unfortunately your experience with home testing accuracy is all too common. I experienced it and so did my spouse just a few years ago when attending a cardiac rehab following the insertion of a stent for clogged cardiac arteries and Afib.
Spouse was told while attending cardiac classes was told by a pharmacist that INR monitors were not legal in Canada, and are highly inaccurate, etc., etc. so I understand where you are coming from.
There are many Point of care POC monitors being used in the UK and there were a number of users on this forum over the years. Google for International self managing anti-coagulated patients (ISMAAP) based in Europe for more info. I haven't been there for awhile so I'm not sure how active it is.
Maybe if you offered to use your monitor in conjunction with a lab draw for accuracy would help to start with. I don't bother any more because the local lab sends vein draws to the city for testing and the result takes days--useless. Maybe Roche could advise how many of their monitors are being used.
As long as your doctor prescribes warfarin you should be all right. Who how manages your test resulted?
Sandra

 

[tom in MO]

Funny, I just read an article where they said that periodic verification of home testing results should be done in the doctor's office. This was not to check the accuracy of the meter, but to check the efficacy of the patient's use of the meter. Most of the problems with home testing are due to patient's misuse of the meter or failure to test and report...i.e. comprehension or old age related problems. With my 86 yo mother in law, she home tested, but her family did it for her to make sure it was correctly done and on time.

Personally I believe that some people make a living off of testing in clinics, etc., thus they don't want to lose their jobs to home testing...but then I am a cynic today

 

[marc_kowal]

I feel lucky, because after my surgery I simply told my cardiologist that I wanted to home test and this was the machine that I wanted. He had no problem what so ever to write the script for it. About every 4-6 weeks, I'll head to the lab for a veinous draw and then minutes aftewr that, I'll run a finger test in the parking lot. I do this to verify the accuracy of the my InRatio 2. My cardio, the lab and even Allere agree this is a wise thing to do.

 

[tom in MO]

Could you point me to where Allere says routine comparisons are recommended? Thanks.

I use an Allere provided machine and used to use their service and they never indicated that a lab comparison was needed. On my initial training, they said I had to pass a comparison test to the trainer to prove I could use the machine and that it worked properly brand new out of the box. However any recheck was not part of their routine process.

 

[Protimenow]

A year or more ago, when I was using an InRatio (from Alere), my InRatio readings and lab readings were far apart. I called Alere and asked which one to trust. Their words: Trust the Lab. I have a real issue with the accuracy of the InRatio (for me, at least). I don't know that testing methods really impact the accuracy, as long as you do the basic procedures correctly. For me, InRatio seems to have ALWAYS been high.

A monthly blood draw to compare to your meter is not a bad idea -- I do it and compare it to one or more meters.

 

[marc_kowal]

@ tom in MO- It's not written anywhere on the Alere website or pamphlets. It was a discussion I had with their tech. department to discuss discrepancies between the InRation 2 and a veinous draw and why I was doing comparision tests. Sorry if my statement mis-lead you.

Alere also informed me that the acceptable variance between a lab draw and a home testing unit as established by the WHO (World Health Organization) is 30% of the lab result, which could be the diffenrence from being "in range" to being "way out of range". Hence the reason I personally like the check my machines accuracy.

 

[tom in MO]

Thanks for the reply.

 

[Protimenow]

It's the 30% that scares the hell out of me -- especially when the new 'wisdom' seems to put the range for most valvers at 2.0 - 3.0. A 2.0 on a meter could actually mean that the INR is 1.4 -- and I had a situation very much like that when my InRatio gave me an INR that was at the low end of my old range, and the hospital lab got a 1.7 after I went in with a TIA. I have a serious problem accepting that new range for ANY self-tester (or even doctor's offices that use some of these meters) because an 'in range' result could actually overlook a potentially dangerous INR.

(My InRatio seemed to always report high, according to testing that I did comparing it to lab results). My XS was also a bit higher than labs. The meter that I have the most trust in is the Coag-Sense that, if anything, reports a bit below my hospital lab. I'd rather have a 2.0 on the Coag-Sense, and a hospital result of 2.1 or 2.2 than a 2.0 on an InRatio and an actual INR of 1.6.

 

[UncleSteve]

Well after some wild readings with my original Coaguchek meter, it was replaced. The first reading with my new meter was exactly the same as the clinic, yipee However... I'm losing faith in my new one as I've never achieved the same reading again as the clinic, even doing finger pricks side by side (different fingers).

I'm wondering where I'm going wrong?

Here's the readings so far...

Date	Clinic INR	Clinic Venous	Clinic Finger Prick	Coaguchek Finger	ERROR
13/06/2014	4.6	4.6		3	1.6
12/06/2014				3.8
10/06/2014				3.1
06/06/2014	Finger pricks all tested @ clinic
06/06/2014	4.1	4.2	4.1	3.8	0.3
05/06/2014				3.4
30/05/2014				2.1
23/05/2014	3.3			2.3	1
20/05/2014				2.1
16/05/2014	4.7			3.9	0.8
02/05/2014	3.4			3.1	0.3
25/04/2014	2.6			2.6	0
24/04/2014	new meter

With regards to the clinic results, I can't remember if some where done with a finger prick test or a venous sample (taken with a hypodermic syringe) taken to the clinic by my wife.

The whole point of self testing was to avoid visiting the clinic, but with unreliable results, this is out of the question

 

[pellicle]

those results seem all close enough to me (although are greater than I get). Given the scatter plots I have published showing variances grater than what you have reported, and given that it only matters if the reading difference results in a dose change. For instance 4.7 vs 3.9 would probably not result in a dose change unless there was a trend..

Remember we are not talking measuring thichness of steel with a micrometer here, its more like measuring a sponge.

Also, the variance gets a little higher with INR too.

Do you do your test with identical method?
- always the same time from warmup ready alarm to blood application?
- always a good full hanging drop?
- always within 15 seconds of lancing?

If you want the consistency in results you must use consistency of testing. Actually that is the only point which causes technicians to be sceptical of the usefulness of POC machines in users hands.

As we say in IT ... Remove PEBKAC before testing the software.

 

[pellicle]

those results seem all close enough to me (although are greater than I get). Given the scatter plots I have published showing variances grater than what you have reported, and given that it only matters if the reading difference results in a dose change. For instance 4.7 vs 3.9 would probably not result in a dose change unless there was a trend..

Just thought (now that I'm near a computer, not just on the phone)

Scatter plots from testing of commercial grade lab machines vs coaguchek

which suggests to me that if the authors don't think that's worryingly different (and conclude they're accurate enough) then perhaps its not worryingly different and accurate enough?

with respect to the data presented

clinic	coagu	difference	%age difference
4.7	3.9	0.8	17.0
4.1	3.8	0.3	7.3
3.3	2.3	1	30.3
4.7	3.9	0.8	17.0
3.4	3.1	0.3	8.8
4.6	3	1.6	34.8

by looking at the percentage difference we get an idea of the scale, so a drop of 0.4 when you are 2 is far more worrying than when the INR is 4. Which sort of indicates that the differences aren't as scary as when you just look at the absolute amounts of variance.

As I mentioned above, the instrument is not foolproof and so just make sure your technique is accurate and repeatable. This is what lab workers need to do to make their results consistent and that they are not 100% machine-like may explain some of the variance that you would see in lab on lab comparisons (and yes there is some). They are regularly checked and cross checked, so make sure you apply some rigor to your own testing.

I'm not saying you don't test consistently with rigor but since you have not mentioned it I'm just bringing it up, as I know that it made a difference to me when I was starting out.

I also know that getting a good "full hanging drop" (as specified in the manual) makes a difference to the testing.