Biological Variation in Coagulation


All stars laboratory week: teaming up for patient care!

Biological variation is the fluctuation of in coagulation parameters. It looks at healthy individuals during steady state conditions. It has also been demonstrated that variations can be influenced by stress, acute phase reactants, circadian rhythm and seasonal variations.

Biological variation looks at the within and between subject variation. Why should we be concerned with this variation? How does this impact our testing, our results and even our reference ranges. Is the variation significant? And in what tests can this impact where a cutoff point (such as d-dimer or lupus) is used to define the presence or absence of disease. Biological variation can also help to determine allowable imprecision and bias.

A study was conducted to determine the biological variation in 40 healthy individuals over the course of a year. Tests that were looked at were: PT, aPTT, fibrinogen, antithrombin, Factor VIII, IX, von Willebrand factor, Protein C, and Protein S. Biological variation was different for coagulation testing. The PT had a low within subject and between subject variation at 2.6% and 4.1% versus a 31.2% vWF ristocetin cofactor activity. If you compare this to variability in proficiency testing, or even within your own laboratory, the differences are expected. The PT with a low variability, even when you look at inter-laboratory variability and you will see the same issue with vW activity testing.

Biological variation can have an impact in the variation of the International normalized ratio (INR). Samples were tested on warfarin patients including fibrinogen, factors II, V, VII and X. The results showed the CV of the INR were method dependent and varied between 18 and 24%. The in-treatment biological variation of INR was higher than reported for healthy individuals as well as patients in a steady-state condition, but by correcting for appropriate coagulation factors it was reduced. The association between INR and coagulation factors was different for the different PT methods mainly due to different sensitivity towards FII and FVII. Now we have added the confounding variable of the level of factors (the range can be in some factors 50-150%) as well as the sensitivity of reagents. Factor II tends to be problematic, reagents can be insensitive- that is a normal PT level when Factor II is </= 30% versus hypersensitivity of Factor VII, that is an abnormal PT despite normal levels of factor VII.

Reference interval determination in a laboratory is an important process. It is the basis of deciding if a patient is normal or requires additional testing or intervention. BV can become an important consideration when establishing RI and its outcome on clinical decisions. A study by Chen looked at 16 coagulation testing including: PT, fibrinogen, aPTT, thrombin time, INR, prothrombin time activity, activated partial thromboplastin time ratio, fibrin(-ogen) degradation products, factors II, V, VII, VIII, IX, and X). All intra-individual coefficients of variation (CV) of the screening tests (except Fbg) were less than 5%. While CVs for all factor activities were > 5%. Again, if you look at your own internal studies, they will most likely mimic these findings, making the impact minimal on ranges. Additional studies looked at 25 healthy subjects for lupus anticoagulant (LA), antithrombin (AT), protein C (PC), and protein S (PS) testing, and the values of von Willebrand factor antigen (VWF:Ag). The analytical biological variation values of all the parameters were less than 3% and no significant differences were observed in the intraday and interday biological variations of LA tests, or in AT, PC, PS, and VWF:Ag values. This information minimizes the impact of when you collect samples for your RI. Most laboratories tend to collect samples in the morning to either verify or validate a RI and it appears that has minimal impact on outcomes.

There are so many variables in coagulation testing - reagents having different activators, phospholipids and sensitivities, analyzer variation, different deficient plasmas, different methodology- adding biological variation to the mix can further complicate coagulation results. It appears that the impact of BV in coagulation testing is minimal, with the exception of vW testing, which in itself is problematic. Reviewing any proficiency survey will demonstrate the high CV among laboratories in von Willebrand activity testing. Laboratories should understand that adding BV to that mix may help to explain additional previously unidentified variability in patient results.

Now that I have given you more to think about, I wanted to again wish you all a Happy National Medical Laboratory Week. We have been called unsung heros, the people who provide information for which about 70% of clinical decisions are made, and yet most people still don't know what we do- but we know what we do. So tell someone what you do, be proud of how much you know, teach what you know, lead someone in the right direction. As automated as laboratories are, issues are more and more complex and we need to have technologists with even better problem solving skills. I love meeting technologists, talking to them and always learning from them. So celebrate what you do, whether it be with food (we are big on that) a continuing education lecture or teaching a student- be proud of your contribution in patient health care!