Warfarin reversal in patients with antiphospholipid syndrome: Caution required, but clear guidance not available

Two cases of antiphospholipid syndrome (APS)1 had serious outcomes, resulting in death after warfarin reversal for urgent interventional procedures. The cases highlight a lack of appropriate advice in manufacturers' and professional guidelines for warfarin reversal in APS and other significant prothrombotic states. The first case was a 40-year-old female with ‘double-positive’ (lupus anticoagulant and anti-cardiolipin) APS diagnosed following recurrent lower limb deep vein thrombosis (DVT) while on warfarin (target INR chosen was 3.0–3.5). Many years later, she presented with a ruptured ovarian cyst and 2 L intraperitoneal haemorrhage (INR 3.7). To cover urgent laparotomy, she was given 50 IU/kg of Prothrombinex-VF (a 3-factor Prothrombin Complex Concentrate, PCC), and 10 mg IV vitamin K. She died after developing multiple intracerebral venous sinus thromboses with secondary cerebral haemorrhages 5 h after the PCC infusion. The second case was a 57-year-old man with multiple DVTs and ‘triple-positive’ APS who had been on long-term warfarin and low-dose aspirin. He was scheduled for gastroscopy to investigate gastrointestinal bleeding (INR 4.5) and received 50 IU/kg of Prothrombinex-VF, 5 mg IV vitamin K and FFP (two units). Extensive lower limb DVTs and bilateral renal vein thromboses followed a few hours later. ‘Triple therapy’ (steroids, warfarin and plasma exchanges (PEX) was started for what was considered to be catastrophic antiphospholipid syndrome (CAPS). Three years later, the patient was being prepared for a major ABO-incompatible kidney transplant with PEX to reduce ABO antibody titres. PEX was covered with LMWH anticoagulation, and anti-thrombin (AT) post-PEX to offset predictable AT loss through PEX. Baseline (pre-PEX) AT was 120% (normal 75%–135%), and, empirically, 1000 units of AT was given after each PEX. LMWH was withdrawn for 36 h for a central line insertion and the procedure was followed by an ischaemic stroke. Accidental AT dose omission on the day prior may also have contributed. The transplant was cancelled, and the patient died of an immunosuppression-related atypical pneumonia. ED was aware of the first, but not the second, patient's APS diagnosis. Per current guidelines, maintaining INR between 2.0 and 3.0 would have been sufficient for both patients.2 Both were deemed to need urgent invasive procedures (arguable with the second patient) and urgent warfarin reversal. Procedural bleeding risk in the first patient was high, but in the second, probably only moderate.3 The first had bridging LMWH prescribed but not commenced before untreatable cerebral venous thromboses occurred. Both patients received 50 IU/kg of Prothrombinex-VF and vitamin K for warfarin reversal, which created a high risk of thromboembolism with the additional prothrombotic effect of lupus anticoagulant.4 For both, given procoagulant effects of lupus anticoagulant, lower doses of Prothrombinex-VF may have been sufficient and safer. Indeed, for the second patient who did not have life-threatening bleeding, just low-dose vitamin K may have been sufficient. Urgent warfarin reversal in the two patients with 3F-PCC followed the current regional guideline.5 We reviewed similar guidelines from Canada, UK and USA and, like the Australasian guideline, they have no warnings against high PCC doses in patients with significant prothrombotic states.3, 6, 7 The PCC doses used in our patients likely quickly raised FII, IX and X levels significantly contributing to catastrophic thromboses. The Prothrombinex-VF manufacturer quotes FII increments of 0.02 IU/mL following a 1 IU/kg dose; recovery data for the other factors are provided. Thus, a 50 IU/kg dose would increase FII, IX and X levels by about 100%, 50% and 85% respectively.8 The 192 IU of unfractionated heparin included in each vial of Prothrombinex-VF might have delayed the onset of catastrophic thromboses in both patients. It should also be noted that FII and X have less effect on traditional INR than FVII, but FII and X may better predict bleeding and thrombosis risk.9, 10 A 50 IU/kg dose of a 3F-PCC for prothrombotic patients will likely result in FII, and X levels that exceed the minimum requirements for surgical haemostasis and create thrombotic risk. The optimal dose of 3F- or 4F-PCC for urgent warfarin reversal in patients with significant prothrombotic states is undefined, but is likely to be much lower. We suggest 20–25 IU/kg may achieve sufficient but not excessive FII and X levels—approximately 50% (the lower end of the reference range) in those with the added procoagulant effect of lupus anticoagulant. The approach of Kasthuri and Roubey is worth considering.11 In APS, they recommend keeping FII levels at 20%–30%. Thus, if a patient had a ‘therapeutic’ level INR (e.g. 2.0–3.0) but a FII level >30%, they suggest increasing warfarin dosage (and INR) until FII levels of 20–30% are achieved. Alternatively, new, modified tests (e.g. Fiix-Prothrombin time and Fiix-normalised ratio that measure only the effect of FII and X reductions during warfarin monitoring) may be useful.10 As the frequency of APS is low, international collaborative or registry-based studies may be needed to define FII and X levels after urgent PCC reversal of warfarin for safe haemostasis without unacceptable thrombosis risks. We suggest that a post-reversal FII not exceeding 50%, and 20–25 IU/kg PCC will be satisfactory in those with APS or other prothrombotic risk. Each author contributed a case, and both authors were involved with drafting, reviewing and approving the manuscript for submission.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.