Direct oral anticoagulant use in patients with antiphospholipid syndrome and unprovoked venous thromboembolism: a single centre experience

Abstract

The recommendations on the use of long-term anticoagulation in antiphospholipid syndrome (APS) changed following two randomised-control trials (RCT) by Pengo et al. and Ordi-Ros et al. They showed that patients with APS had higher rates of thromboembolism recurrence (19% and 12% respectively), and major bleeding (7% and 6%) with the direct oral anticoagulant (DOAC) rivaroxaban. In comparison, there was a 3–6% rate of recurrent thrombosis and 3– 7% major bleeding rate with vitamin K antagonists (VKA). This risk was greatest in those with ‘triple-positive’ disease (positive for lupus anticoagulant, anticardiolipin and antibeta-2 glycoprotein-1 antibodies) and/or preceding arterial thromboembolism and extravascular manifestations. As a result, the Medicine and Healthcare products Regulatory Agencies (MHRA) advised that DOACs should avoided in patients with APS, with agreement from the British Society for Haematology (BSH) and International Society on Thrombosis and Haemostasis. However, the role of DOACs in APS presenting with unprovoked venous thromboembolism (VTE) without arterial thromboembolic events was not clearly established following these studies. A meta-analysis by Elsbaie et al. showed rates of VTE recurrence (RR 0 88; 95% CI: 0 26– 2 98) and bleeding (RR 0 80; 95% CI: 0 37–1 73) were similar between DOACs and VKAs following VTE due to APS. Recurrence rates in APS were comparable to those following VTE without identifiable thrombophilias. They found 50% of those with recurrence were triple-positive for antiphospholipid antibodies. Current BSH guidance advises patients with APS with previous VTE be changed from DOAC to VKA if they are triple-positive. It also advises that DOAC should be continued or changed to VKA if the patient is not triple-positive, following patient counselling. However, it recognises that there is a lack of data to fully support this statement. We, therefore, retrospectively reviewed patients with APS having unprovoked VTE who were prescribed DOACs and attending our centre between 1st June 2019 and 1st December 2020. Our centre is a large university hospital providing both local and specialist services for patients with APS. We discussed the outcomes of recent studies with each patient and repeated antiphospholipid antibody testing, if not recently checked, to assess for triple-positive status, in accordance with BSH guidance to ensure ‘high-risk’ patients were identified. Our methods for identifying antiphospholipid antibodies have previously been described with cessation of anticoagulants a minimum of 24-h prior to testing. Anticoagulation plans were agreed with the patients and documented to their general practitioners. Within the discussion, we considered the patients’ duration of DOAC use, the nature of previous thromboembolic events and/or bleeding and any issues with previous VKA use and monitoring. We avoided dose reduction of DOACs in extended thromboprophylaxis unless clinically indicated, such as renal impairment, low body weight or bleeding history. We identified 82 patients with APS, according to the Sydney classification criteria for APS, taking DOAC after unprovoked VTE. Their clinical features are shown in Table I. The median follow-up period from initial VTE diagnosis was 9 0 years (range, 0 9–32 0). Forty-eight (59%) patients used VKA preceding change to DOAC (median duration 5 6 years, range, 0 1–28 0 years) and 34 (41%) were initiated on a DOAC after the initial VTE. The median duration of DOAC treatment was 5 0 years (range, 0 1–7 8 years) with a total of 389 patient years. During the use of DOACs, four (5%) patients experienced recurrent events with an incidence of 1 0 events per 100 patient-years. Two of four patients were triple-positive, having unprovoked recurrent VTE despite good compliance. The incidence of recurrent thrombosis in triple-positive patients was 1 7 per 100-patient years and 0 7 per 100-patient years in non-triple-positive patients (RR 2 72; 95% CI 0 41–18 2). Two had provoking factors: anatomical venous compression and inadequate thromboprophylaxis peri-operatively. If excluded, the rate of unprovoked thrombosis recurrence was 0 5 events per 100 patient-years. During DOAC use, one patient (1%) had an episode of major haemorrhage (lower gastrointestinal) and 13 (16%) had clinically relevant nonmajor bleeding (CRNMB: 11 menorrhagia, 1 epistaxis and 1 following trauma) with an event rate for clinically relevant bleeding of 3 6 events per 100 patient-years. Two patients with CRNMB had symptomatic improvement after changing DOAC. Despite counselling about recommendations for VKA use, 70 patients (85%) decided to continue the DOAC, one (1%) chose to stop anticoagulation, eight (10%) changed to VKA out of preference and three (4%) changed to VKA following thrombosis recurrence. correspondence