Management of hemorrhagic complications associated with antithrombotic therapy

University dissertation from Stockholm : Karolinska Institutet, Dept of Medicine, Solna

Abstract: Anticoagulants are used for the prevention and treatment of arterial and venous thromboembolism. Such therapy is always associated with an increased risk of bleeding. Intracranial hemorrhage is the most serious form of anticoagulantsrelated bleeding and caries a high mortality rate. Many treating physicians and patients are reluctant to restart anticoagulation after such an event. For patients with continued increased risk of cardioembolism, the decision of when to restart anticoagulation is challenging and few small, single-center studies and case reports are the only available source of information. Prothrombin complex concentrates (PCC) have been shown to rapidly correct the prolonged prothrombin time associated with vitamin K antagonist (VKA) therapy. The clinical efficacy of such treatment for controlling bleeding or preventing post-operative hemorrhage in patients treated with VKA is uncertain. Thromboembolism is a potential side effect of PCC and there are limited data on the risk of such complications. New direct oral anticoagulants are approved as alternatives for VKA. These agents lack antidote and the management and outcome of patients with major bleeding on these agents is uncertain. In this thesis different aspects of the management of anticoagulants-associated bleeding complications are studied and discussed. A cohort of 234 patients from Sweden and Canada with VKA-related intracranial hemorrhage was analyzed to establish the optimal time point for resumption of anticoagulation. Early resumption carries a high risk of recurrent bleeding, whereas the risk of thromboembolism increases with delayed start. According to the result of our study, the optimal timing of resuming iii anticoagulation after VKA-associated intracranial hemorrhage seems to be between week 10 and 30 after the event. A subgroup of the above-mentioned cohort was combined with another cohort from the Netherlands. A total of 135 patients were investigated to assess the effect of PCC - compared to plasma - on survival after VKA-related intracerebral hemorrhage. Unadjusted analysis showed a benefit for PCC treatment in reducing the 30-day all-cause mortality. After adjustment for differences in prognostic factors between the two treatment groups, especially the volume of intracerebral hematoma, the beneficial effect of PCC was attenuated and became non-significant. The safety of reversal of VKA with PCC was assessed in a third cohort of 160 Swedish patients with a major bleeding event or before emergency surgery. Six patients (3.8%) developed thromboembolic events (5 arterial and 1 venous) within a week after the administration of PCC. In these patients, the unmasking of the increased risk for thromboembolism by the reversal of anticoagulation and the activation of the coagulation system by the bleeding or surgery, seem to be important contributing factors, while PCC seem to increase the risk only minimally, if any. A cohort of 1121 patients with major bleeding on dabigatran or warfarin, derived from 5 phase III trials, was used to study the management and outcomes of these bleeding events in the absence of an antidote to dabigatran. The patients who suffered major bleeding on dabigatran were older, had more renal impairment and were more frequently treated with antiplatelet agents or non-steroidal antiinflammatory drugs compared to those on warfarin, yet the outcome of dabigatran bleeding was better as reflected by a shorter stay by 1 day in the iv intensive care unit and lower 30-day all-cause mortality compared to the warfarin group. In summary, PCC seem to be safe for the treatment of VKA-associated bleeding, but their beneficial effect over plasma in case of intracerebral hemorrhage is questionable. The absence of a reversal agent for dabigatran does not seem to impact negatively on the outcome after major bleeding. When intracranial hemorrhage occurs on VKA resumption of anticoagulation should be delayed for 10-30 weeks.

  This dissertation MIGHT be available in PDF-format. Check this page to see if it is available for download.