From Dr. King: We are encountering with increasing frequency requests from hematologists to perform mixing studies on patients that are known to be on some form of anticoagulation. My question is regarding the utility of doing such testing. Another aspect to the question is whether you can do testing for a prolonged PTT when the patient is known to be on warfarin or evaluate a prolonged PT in a patient known to be on therapeutic doses of heparin. I wonder whether others are encountering this same request and would appreciate yours and their thoughts. Thanks.
Hello, and thank you for your question. The recently released target-specific oral anticoagulants (TSOACs) variably prolong the PT and PTT, rendering their results uninterpretable and hopelessly interfering with mixing studies. See Favaloro EJ, Lippi G. The new oral anticoagulants and the future of haemostasis laboratory testing Biochemia Medica 2012;22:329-41 for a comprehensive review. Likewise, there is little that one can do with warfarin specimens, as the warfarin effect on the vitamin K dependent factors is irreversible. It is possible to neutralize unfractionated heparin using Hepzyme® or polybrene (Hepsorb®) to perform a PTT or a PT mixing study. I am curious about why your hematologists request PT mixes, as the PT mixing study provides limited diagnostic information.
From Dr. Favaloro: Thanks
From Dr. Favaloro: Thanks George for mentioning one of our papers. We have another coming out in Seminars in Thrombosis and Hemostasis soon, entitled “Laboratory testing in the era of direct or non–vitamin K antagonist oral anticoagulants: a practical guide to measuring their activity and avoiding diagnostic errors.” Of relevance to the question at hand: Our organisation also occasionally gets requests to perform mixing studies on patients who are on anticoagulant therapy. I can see some potential clinical interest in doing this perhaps for patients over-anticoagulated with say VKA (e.g. warfarin), to assess whether correction may occur (should there be contemplation of using [say] fresh frozen plasma [FFP] for reversal). Perhaps the potential for reversal of
other anticoagulants is also considered in this context.
On the other hand, mixing studies in the setting of anticoagulant therapy would be of limited use for many reasons–including that mixing tests are done as 1:1 mixtures, and you are not going to give any patient an equal volume of FFP for reversal! Another reason clinicians might ask is that they don’t actually know that (for example) the NOACs actually prolong coagulation tests (since the dogma that they don’t need monitoring by laboratory testing is sometimes misinterpreted as meaning that they don’t affect coagulation tests). In general, one would expect a reasonable correction of abnormal coagulation tests from patients on VKAs, although it is unlikely that complete correction will be observed. This partly depends on the interpretation of a correction, which is quite varied among labs, for example, Kershaw G, Orellana D. Mixing tests: diagnostic aides in the investigation of prolonged prothrombin times and activated partial thromboplastin times. Semin Thromb Hemost. 2013;39:283–90. I recall you ran a survey on this once as well.
Mixing studies with the NOACs are unlikely to show correction unless the initial
coag test is just above the normal reference range, in which case any mix is likely to go into the normal reference range and by definition be reported as a correction. The pattern with heparin might be more complex, and depend on whether the reagent has a heparin neutraliser. Thus, an APTT will not likely correct, but an abnormal PT might correct if the level of heparin is initially well above the therapeutic range and then falls within the ‘neutralisation’ range of the reagent. I can’t recall to having been asked to
perform mixing studies on anticoagulated patients from hematologists, though; the requests typically come from other clinicians.