When dealing with a bleeding event potentially having relevant clinical consequences, like hypovolemia and/or hemodynamic impairment, permanent damage in close spaces (e.g., skull, spinal cord, or joints), or ultimately death, the initial question is how rapidly the hemorrhagic event needs to be controlled. Whereas measures aiming to preserve organ perfusion, including fluid and/or plasma expander administration and vasopressors, together with specific maneuvers directed to the treatment of the cause of bleeding, such as surgical or endoscopic hemostasis and/or removal of the bleeding lesion, are used in a standard patient, the management is more complex when bleeding occurs in a patient on OAC with warfarin or on dual antiplatelet therapy (DAPT) with aspirin and a P2Y12-receptor inhibitor or, even more, on triple therapy (TT) of warfarin, aspirin, and clopidogrel [1]. In these latter cases, careful balance needs to be applied in order to effectively manage bleeding while not exposing the patient to an increased risk of thrombotic and/or thromboembolic events.

In a patient experiencing a clinically relevant bleeding [2–5] (Table 11.2), while on TT of warfarin, aspirin, and clopidogrel, the immediate question to be addressed is what is the risk of adverse outcome when continuing as opposed to interrupting one or more of, or even all, the components of combination antithrombotic therapy. Namely, the risk of permanent sequelae and/or death associated with the continuation of antithrombotic therapy needs to be balanced with that of stroke and/or stent thrombosis and/or recurrent coronary events associated with the interruption of OAC and/or antiplatelet agent (s). With this purpose, stratification of the above risks needs to be performed.

Although not standardized in a specific scoring system, the risk of (in-hospital) adverse outcome associated with bleeding appears dependent on several variables, including age, systolic blood pressure, impairment of consciousness, and pre-existing comorbidities (including heart failure, cancer, peripheral artery disease, chronic obstructive lung disease), as well as type and location of bleeding and degree of over-anticoagulation with warfarin (when ongoing) [6, 7].

The risk of stroke should be evaluated by means of the CHA₂DS₂-VASc score (Table 11.3) [8]. It gives an estimation of the annual risk of stroke and/or thromboembolism in the absence of antithrombotic treatment. While acknowledging that the predictive value of CHA₂DS₂-VASc score is modest (i.e., C-statistic range 0.64–0.79, median 0.673) [9], it nonetheless should be regarded as the standard tool for stroke risk stratification.

The risk of stent thrombosis is related to the type (i.e., bare-metal vs. drug-eluting) of stent, the generation (i.e., early vs. new) of drug-eluting stent, the clinical setting (elective vs. acute) where the stent has been implanted, and the time elapsed from stent implantation. Premature discontinuation of DAPT, however, should be regarded as the strongest predictor of stent thrombosis [10]. In this regard, it is considered safe to interrupt DAPT and continue with single antiplatelet therapy with aspirin (or clopidogrel) 1 month and, respectively, 6 months after bare-metal stent and new-generation drug-eluting stent implantation in an elective setting. A period of 12 months should be considered after an early-generation drug-eluting stent implantation, irrespective of the elective or acute clinical setting, and either bare-metal, early-, or new-generation drug-eluting stent implantation in an acute coronary syndrome (ACS) [11]. If needed, such as in the event of bleeding, a shorter duration of DAPT, that is, 3 or even 1 month, may be considered for (some) new-generation drug-eluting stents (Table 11.1). Care, however, should be put in not interrupting both antiplatelet agents at the same time, given that this has been associated with an increased risk of stent thrombosis (Fig. 11.1) [12]. The location where the stent was implanted (i.e., proximal in a main and/or largely distributed vessel vs. mid-distal in secondary and/or poorly distributed vessel) should also be taken into account when trying to foresee the potential consequences of stent thrombosis.

The risk of recurrent coronary events is higher after ACS compared to the elective setting. Over the first year after the former, an incidence of < 1 % may be expected, as compared to up to 10 % after the latter. Of note, however, the risk of adverse events (namely, death) after and ACS tends progressively to decrease as long as time passes from index event (Fig. 11.2) [13]. After 1 year (with no recurrence) from ACS, a patient is then considered stable, thereby generally requiring low-intensity (i.e., single-agent) antiplatelet therapy. In selected cases at high risk of recurrences, especially because of previous myocardial infarction, prolonged (i.e., up to 3 years) DAPT with aspirin and P₂Y₁₂-receptor inhibitor may be considered [14–16].

In the absence of hemodynamic impairment and/or harmful location (e.g., intracranial and/or intraspinal and possibly also intra-articular) of bleeding, no fluid or plasma expander administration nor blood transfusion nor inotropic support nor reversal of OAC appears warranted. Oral or intravenous administration of vitamin K and/or prothrombotic agents, in the form of prothrombin complex concentrates or fresh frozen plasma (Tables 11.4 and 11.5), should generally be reserved to the emergency management of life-threatening bleeding events (Table 11.6). With the exception of these circumstances, immediate warfarin withdrawal should generally be considered as the only measure to adopt. Even in the presence of an increased risk of stroke, which may range from about 0 % to 15 % per year based on the range of CHA₂DS₂-VASc scores between 0 and 9 (Table 11.3) [8], it is unlikely that such event will develop during a short (i.e., some days to a week or so) interruption of warfarin OAC. A possible exception may be represented by a very high (i.e., 7–9) risk of stroke, especially when a history of stroke is present as a variable contributing to the total CHA₂DS₂-VASc score. In this case, bridging warfarin interruption with (possibly reduced dose) subcutaneous low-molecular-weight heparin may be considered once bleeding has ceased in the light of the superior convenience of management in the event that the need for subsequent invasive or surgical intervention arises [17]. Indeed, interruption of warfarin performed as initial management strategy for bleeding also allows safe endoscopic (with biopsy) or surgical intervention when needed.