Peri-procedural antithrombotic bridging and the assessment of the associated risk of major bleeding.

Editorial on Omran et al. Thromb Haemost 2012; 108: 65-73. Haemostasis refers to the continuous process of balancing the (sometimes iatrogenic) proand anti-thrombotic forces. The delicacy of this balance is exemplified by conditions, such as atrial fibrillation (AF), which cause a chronic prothrombotic distortion, but the consequent use of antithrombotic therapy increases the risk of major bleeding, especially intracranial haemorrhage (which is the most devastating complication of anticoagulation therapy). Therefore, the everyday antithrombotic management of anticoagulated patients is challenging as the benefits and risks need to be carefully judged for each and every patient. In AF, updated and new risk stratification tools for stroke (CHA2DS2-VASc) (1) and major bleeding (HAS-BLED) (2) improve the guidance on long-term antithrombotic management of AF patients. However, similar degree of guidance regarding bleeding risk assessment in a clinically important ‘temporary’ aspect of antithrombotic therapy management remains lacking, that is, peri-procedural antithrombotic management (3). In this issue of Thrombosis and Haemo stasis, Omran and co-workers (4) present a German, multi-centre prospective cohort study among 1,000 anticoagulated patients undergoing a planned intervention and aimed to study the impact of an antithrombotic therapy bridging regimen on the short term (30 day) risk of thromboembolic events and major bleeding, as well as the clinical predictors of the latter. Their cohort was predominantly male (65%) and had AF (81%), and coronary angiography was the most common reason for bridging (66%). In 6.1% of patients, no bridging therapy was given. Apart from two patients receiving unfractionated heparin (UFH) the remaining patients were treated with weight-adjusted low-molecular-weight heparin (LMWH) (72.7%), fully therapeutic LMWH (18.8%) or a prophylactic LMWH dose (2.2%). Thromboembolic events occurred in four patients and clinically relevant bleeding events in 36. Their novel finding was that a high (>3) HAS-BLED (Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalised ratio, Elderly (> 65 years) score was an independent predictor of major bleeding during antithrombotic therapy bridging, as well as a previous history of mechanical valve replacement (MVR). The patients of the cohort presented by Omran et al. (4) were included because of a variety of reasons to discontinue anticoagulation (3). As the pro-thrombotic mechanisms – and thereby risks – associated with these procedures are likely to differ, perhaps a first question is whether interruption of anticoagulation is always necessary? Jamula et al. performed a systematic review which included eight studies comparing two strategies for perioperative anticoagulation management in patients undergoing pacemaker or implantable cardiac defibrillator (ICD) implantation: interruption of a vitamin K antagonist (VKA) and use of bridging anticoagulation or perioperative continuation of VKA (5). The incidence of pocket haematomas was much higher in the group receiving bridging therapy compared to the group who continued the oral anticoagulation (12–20% vs. 1.9–6.6%). Looking at thromboembolic events, the incidence was only 0.1% irrespective of the used strategy (5). Similar evidence exists that for patients undergoing catheter ablation for AF (6) and elective coronary angiography (7), whereby uninterrupted VKA is safe with respect to bleeding events and may even help prevent peri-procedural (silent) strokes. Thus, periprocedural continuation of anticoagulation can be safe and even associated with less bleeding events compared to antithrombotic bridging, and it appears to be rightful for clinicians to firstly question the necessity to switch from warfarin to some form of antithrombotic bridging based on the type of procedure. Of note, recent consensus recommendations from Europe and North American do not advocate bridging therapy in patients undergoing coronary angiography, which should be performed whilst the International Normalised Ratio on warfarin is within the therapeutic range (8–10). Secondly, given the relative benefit of continuing oral anticoagulation under the circumstances described above and in light of the upcoming new oral anticoagulants (dabigatran, rivaroxaban and apixaban) it is also important to know whether these convenient drugs could convince even more physicians to continue the oral anticoagulants. Unfortunately, the available data are extremely scarce. One study by Lakkireddy et al. analysed patients undergoing ablation and compared patients who continued warfarin periprocedurally with patients on dabigatran who only stopped dabigatran the morning of the procedure and resumed the dabigatran within 3 hours after haemostasis. The dabigatran group showed a significantly increased risk of bleeding or thromboembolic complications compared to the patients with uninterrupted warfarin therapy (11). This is