Register Login

Yet More on Reference Intervals

This is a follow-up comment from Bob Gosselin on Griselle Font-Bonet’s December 31 post about reference intervals. The original post drew many thoughtful and provocative comments from Dave McGlasson, and Drs. Ali Sadeghi-Khomami and Emmanuel Favaloro; consequently, it became so long that an add-on comment would make it unwieldly. Bob says these “last thoughts”…

Lest we all forget, factor assays are typically ordered for two reasons: prolonged PT and/or APTT or for monitoring therapy. As such, sometimes we get all wrapped up in determining factor reference ranges, but we should be comparing the threshold of the PT/APTT to factor sensitivity–how low is the factor activity that prolongs the PT/APTT to outside the reference range. Otherwise we generate angry clients (AKA clinicians) who are perplexed by prolonged PT or APTT, but normal factor levels, no LAC, and no inhibitor. This should especially be vetted if the screening test uses different reagents than the factor activity test–we used different reagents: Innovin and SynthasIL vs factor assays with Innovin and Actin FS. The irony is that it is likely the same population that was used for factor activity RI was also used for PT/APTT RI…at least that was usually true for our lab, as the donors were readily available in the lab, especially if inducement was offered (Geo: we pay $5 a tube).

I don’t think there are too many people that get that worked up over factor levels between 40–60%, unless the level is used for Rx monitoring. When evaluating factor activity, I was more concerned about 1) limit of quantification (LOQ), because of diagnostic implications, 2) comparison with a predicate between <1 to 30–35% factor activity; 3) monitoring replacement therapy, and 4) any potential reagent synergistic effect (like a synergistic prolongation of Innovin PT with low normal factors VII and X.  All the rest of the verification of method stuff is local window dressing. In this day and age of #3, and the different long acting replacement therapies and the potential bias noted with same, this should be more of a concern than RI determination and potential population bias, if this site has these hemophilia patients.

Dave McGlasson adds, “On a project years ago I was using an APTT reagent that would prolong for isolated FXI and FXII deficiency when one or the other got below 15% activity. Bob is right in checking the APTT against specific factor levels.

Comments (2)
Bleeding Disorders
Jan 5, 2018 10:20am

Thanks for all your comments
Thanks for all your comments and recommendations.

Jan 3, 2018 1:14am

Always keen to have the final
Always keen to have the final say, I did neglect to mention before that we should not forget that reference intervals (RIs) are meant to capture only a proportion (albeit high; usually ~95%) of the normal population. This means that a small subset (~5%) of normal individuals will always fall outside the RI ranges. For more, see: Favaloro EJ, Lippi G. Reference ranges in hemostasis testing: necessary but imperfect. J Lab Precis Med 2017;2:18.

Leave a Reply