From Joyce Low, St Vincent’s Hospital, Sydney NSW: For those who use the Roche CoaguChek XS INR meters…
- At what INR would they repeat the point of care (POC) prothrombin time and international normalized ratio (PT/INR)?
- At what INR would they do a central laboratory (plasma-based) INR to confirm the POC INR?
- Regarding the correlation between CoaguChek INR and central lab (plasma-based) INR how much does the reagent source of lab INR affect the results. Roche use a recombinant thromboplastin.
Hello, Joyce, and thank you for this important question. We had some discussion of when to compare POC PT/INRs to the central laboratory plasma-based PT/INRs in 2012, referenced in these links:
Our discussions centered on whether to use INR 4.0 or 5.0 as the trigger point. The results of our Quick Question survey seems to strongly favor 4.0. Consideration was also given to thromboplastin reactivity, with preference going to the reagents whose international sensitivity index (ISI) is near 1.0. There was some discussion about whether it was necessary to make systematic comparisons with plasma-based INRs at all, though it seems most make that their policy.
The Clinical and Laboratory Standards Institute approved guideline Point of Care Monitoring of Anticoagulation Therapy, POCT 14-A, (formerly H 49-A) does not address comparison to plasma-based PT/INR reference results. For this reason, I’m contacting one or two guideline authors and other experts for opinions and references. Check back soon, as we may have additional input. Geo.
Hi Joyce in response to your questions:
– We recommend repe
Hi Joyce in response to your questions:
– We recommend repeating POCT INR if result not consistent with previous history or unexplained (Essentially if any doubt with result check it)
– We previosuly had a INR value which above we would check with lab, but we now apply logic if machine performing and no intereference (High Hct LA etc – these patients excluded), then act on the results. In our situation may be > 12hrs before blood sample could be tested and result available. Potentially more harm to not act, than have a minor difference.
– Re thromboplastin’s, in my experience yes source does effect correlation, especially at supratherapeutic levels – See previous information.
Alan
I recently looked at two articles that dealt with POC device
I recently looked at two articles that dealt with POC devices compared to a automated coagulation analyzer. Solvik UO, et al,in “Discrepancies in INR results between instruments: a model to split the variation into subcomponents.” Clin Chem 2010;56:1618-26 found that the mean difference between instruments varied between 1.0-14.3%; between-subject variation (CV) was 3.3-7.4%. The analytical imprecision of the differences varied between 3.8-8.6%. The conclusion was that the differences could probably be ascribed to different patient-specific effects in the patient plasma. To minimize this variation in a monitoring situation, each site and patients should use results from only 1 type of instrument.
The same investigators later published on the Effect of coagulation factors on discrepancies in INR results between instruments. Clin Chem Lab Med, 2012;50:1611-20. They used the same instruments cited in the previous article. They found that coagulation factors, especially fibrinogen and factors II and VII could explain between 16-45% of the total variance of the differences in INR between instruments dependent on instruments compared. They then concluded that different levels of coagulation factors in the plasma of the patients played an important role in explaining discrepancies between INR instruments.
These articles could explain a lot of the clinically significant differences in results that we see between POC devices and the standard automated coagulation laboratory instrumentation.
Regardless of whether or not you are regulated to correlate
Regardless of whether or not you are regulated to correlate POC with the Lab for a particular test, I believe you will want some correlations run, if only to have an answer for the occasional clinician when they ask “how does that compare to regular Lab results”.
In the case of INRs, the whole point of using them, if I am not mistaken, is so that clinicians can depend on reproducibility across platforms when monitoring warfarin therapy. Therefor it makes sense to decide at which point an INR is significantly deviant from a regular Lab plasma value.
Scott