Interference of direct oral anticoagulants in haemostasis assays: high potential for diagnostic false positives and false negatives.

Many patients have a variety of prothrombotic risk factors or conditions, and these patients are often given anticoagulant/antithrombotic treatment. Such treatments for thromboembolic disorders are given to patients with atrial fibrillation or acute coronary syndrome, for prevention of venous thromboembolism in patients undergoing major, orthopaedic or cancer surgery or therapy and for prevention of recurrence in patients with previous episodes of venous thromboembolism1. Anticoagulant therapy has for decades been based on the administration of two main classes of drugs: vitamin K antagonists (e.g., warfarin) and heparinoids (unfractionated heparin, low molecular weight heparin, and other heparin-like molecules such as fondaparinux). These "classical anticoagulants" have proven effective for both primary and secondary prevention of systemic thromboembolism, but nonetheless have numerous drawbacks and limitations1-3, thereby paving the way for the introduction of new classes of anticoagulant/ antithrombotic agents. These were originally termed new (or novel) oral anticoagulants (NOAC), and later (no longer being either new or novel) direct oral anticoagulants (DOAC); however, other terms have also been proposed, such as non-vitamin K antagonist oral anticoagulants (NOAC), direct specific oral anticoagulants (DSOAC) and target-specific oral anticoagulants (TSOAC)4,5. DOAC have been designed specifically to inhibit directly either activated factor II (FIIa; dabigatran), or activated factor X (FXa; rivaroxaban, apixaban, edoxaban and betrixaban)4-7. DOAC have several important advantages over classical anticoagulants, making them favoured by clinicians and patients alike6-8. An unfortunate downside to their popularity is under-recognition of their effects on laboratory tests by general clinicians. Indeed, the pervading "mantra" that laboratory monitoring is not required for patients taking DOAC may cause some clinicians to assume that the drugs do not affect haemostasis tests. They may thereby request a range of laboratory investigations on their "prothrombotic" patients, without even considering that these tests may be adversely affected by the drugs, which many of their patients may be taking at the time of blood collection. DOAC, like vitamin K antagonists and heparins before them, are particularly problematic in laboratory test practice, because of the timeline of causal events, given that: (i) patients are clinically assessed (e.g., in an emergency department) as having suffered some sort of "prothrombotic" event requiring anticoagulation, and are then (ii) quickly placed on some sort of anticoagulant therapy, and (iii) only afterwards "when the situation is considered more stable", is an investigation into why the patient suffered the initial event initiated4,5,9. However, once patients are on anticoagulant therapy, any ensuing attempt at laboratory testing for investigation of likely thrombophilia markers is earmarked for likely disastrous consequences. This is because in reality, contrary to the perception that DOAC do not affect haemostasis testing, most tests of haemostasis, including those commonly used to assess the prothrombotic status of a patient who has had a thrombotic event, are significantly affected by DOAC (Table I)5. In another potential scenario, patients placed on DOAC therapy may have a bleeding event while on the drug, and may subsequently be investigated for laboratory markers of bleeding. This may occur either because of a disconnection between the patient and the clinician investigating the bleeding event (e.g., a patient presenting with trauma to an emergency physician), or because some preliminary investigation has uncovered an "unexpected" prolonged routine coagulation test result (e.g. activated partial thromboplastin time [APTT], or prothrombin time [PT]), and then more detailed investigations are undertaken (e.g., clotting factor assays) to help to explain the "initially unexpected" test result. Table I provides a summary of the effects of DOAC on a wide range of haemostasis tests9. In brief, all DOAC affect routine coagulation tests such as PT and/or APTT, with dabigatran affecting the APTT more than the PT, and rivaroxaban affecting the PT more than the APTT. The thrombin time is very sensitive to dabigatran, but insensitive to the anti-FXa agents, and fibrinogen is usually not affected by any DOAC except for occasional