How is your coagulation laboratory at diagnosing patients with factor deficiencies? What impacts your laboratories capacity for detecting levels that correlate with the patients' clinical diagnosis? Testing factors involve capturing the result of an individual component using a complex mixture of relatively unstable protein that are difficult to purify in which inactivated factors must be activated. The test requires activated cofactors, phospholipid and calcium. As well as deficient plasma, buffer and a calibrator. Not to mention, variability in reconstitution of reagents, instability of not only reagents but also factor themselves. Let's also not forget to mention the impact of errors from pre-analytical variables!
The coagulation cascade allows logical effective laboratory based screening through the PT and aPTT and does play a role in the laboratory evaluation of a potential clotting disorder. It doesn't reflect clotting physiologically. The in-vivo cascade better correlates with how thrombin generation actually occurs in 2 phases as opposed to thrombin being a onetime conversion of fibrinogen to fibrin. Initiation results in a small amount of thrombin being produced, which then becomes amplified. Propagation where the bulk of thrombin is formed to promote normal hemostasis
Everyone knows the importance of keeping that 9:1 ratio of blood to anticoagulant, as well as the impact of having a hematocrit >55% and the prolongation of your results. Hemolysis can effect results, and can cause activation of factors, which can falsely shorten results. One of the issues laboratories need to be aware of is stability of your factors- let's look at this case:
A patient who was known to have an acquired Factor VIII inhibitor comes in for FVIII level testing. Her initial FVIII was <1%, with 12.1 BU. After treatment, her FVIII levels rose to 100%. Two weeks later, the coagulation laboratory receives a sample and her level is 16%. Is the antibody back? What happened? The hematologist noticed her aPTT was 34%, which was in the normal range- but with a 16% FVIII level something was wrong. After investigation, it was found that the sample was drawn at 4pm, received in the lab at 10pm and run at 11pm, and at that time the plasma for the FVIII was frozen. There was the reason for the 16% FVIIII, as this factor is labile causing an inaccurate result. The patient was redrawn and the FVIII assay was performed within 4 hours, and the level was 89%. Stability of factors is very important and needs to be considered when diagnosing a factor deficiency.
When testing for factor deficiencies it is important to understand your reagents and know the activity level where a factor deficiency will prolong your screening assays. Sensitivity can depend on concentration and type of phospholipid, or even the activator.
So what happens when you have a patient with a positive family history with totally normal screening tests? Okay you rule out any platelet disorders- but how do you know that you reagent picks up a low level of factors? CLSI defines a sensitive reagent as a reagent that is will prolong your PT and or aPTT when the level of factor are 30% or lower. So let's look at this case.
A 13 year old male with a positive family history and some bleeding history himself presents with a normal PT and aPTT. The pediatric hematologist requests that factor assays be performed. Results are as follows:
VIII = 102% (50-150%)
Abnormal level of factor XI, patient is deficient in factor XI. This reagent appears to have a poor sensitivity to factor XI. You can determine your reagent sensitivity by diluting a pooled normal plasma with a known concentration of factor - in this case FXI with factor deficient plasma (in this case FXI) at different levels and run an APTT on the sample. Compare the results to the upper limit of the normal range
Factor XI Sensitivity: APTT= (29.5-33.5)
|Normal plasma +||Deficient plasma =||% Activity||APTT|
This reagent does not reflect an abnormality until a factor falls within the range of < 25%, so there is poor reagent sensitivity to factor XI, resulting in a patient with a normal aPTT and an abnormal XI level. Reagents can be your worst nightmare, you should know as much as possible about them to provide the best information when looking for a factor deficiency.
The next thing up for discussion is your calibration curve. What do you use to calibrate? Do you use the manufacturer's assay value or do you establish your own value? What type of curve, is it linear or is it polynomial? How many points? Do you have more than one curve? How far apart are the seconds in relation to the points on your curve, does your curve flatten out? Do you use manufacturer's control, or do you use independent controls? What is your acceptance criteria from the previous curve? How often do you calibrate? Whew that is a lot of questions! All important ones that should be evaluated and understood by each laboratory performing factor assays.
It is important to understand that if the analyzer you are using is FDA approved for factor assays using a linear curve, if you choose to change that curve to a polynomial curve, you must perform a full validation versus a predicate that uses polynomial curves. So choose wisely. The seconds between the points on the curve should be at least 6 seconds or more apart so that the curve has a good slope.
Best case scenario is to use a single standard curve as long as it can represent the reportable range that you need for your patient population. One curve has less variability in that the answer is read off of one curve, as opposed to using 2 curves and deciding what result to use when the activity overlaps both curves. What is your acceptability between dilutions? Make sure they you have standard operating procedures and everyone uses the same acceptance criteria.
Using manufacturer's controls is easier since they are assayed for your specific reagents, but having third party controls really help when you are troubleshooting. Fresh frozen controls have the same matrix as patients and are a good to use as an additional check when you are investigating a problem.
How often do you calibrate depends on several things, the stability of your reagents and the acceptance of your quality control PT based assays seem to be more stable and not require calibration. APTT based assays are more problematic and sometimes require more frequent and in some cases daily calibration. Laboratories should determine what works best for their instrument reagent combination.
Diagnosing factor deficiencies is a complicated process that requires not only an understanding of what is happening with the patient, but how your reagents and your system perform. The better the understanding that you have of all the components and sources of variability, the more confidence you will have in the results you provide to your patients.