Indication for and timing of CORO/PCI in patients with non-ST-elevation acute coronary syndrome (NSTE-ACS) are driven by the risk of adverse prognosis, as estimated by clinical, instrumental, and laboratory findings and/or by the application of the GRACE score [1] (Tables 3.1 and 3.2). With the exception of patients at low risk, in whom a noninvasive stress test (preferably with imaging) for inducible ischemia is recommended before deciding on an invasive strategy, all other NSTE-ACS patients should undergo CORO/PCI because of the superiority of a routine compared to a selective invasive strategy on clinical outcomes and recurrent ACS episodes, subsequent rehospitalization, and revascularization [1]. Whereas all NSTE-ACS patients should undergo invasive evaluation/treatment within 72 h of admission, those at higher risk should undergo CORO/PCI within 24 h [1] (Table 3.1). NSTE-ACS patients at highest risk should enter a fast-track management program allowing for emergency CORO/PCI as in ST-elevation myocardial infarction (STEMI), i.e., within 2 h [1] (Table 3.1). Therefore, the periprocedural management of both anticoagulation and antiplatelet therapy is an issue of relevance in patients who are on chronic OAC. Such condition, in fact, likely increases the risk of in-hospital bleeding complications, which in turn are associated with increased mortality, especially in the event that invasive evaluation/treatment is carried out. The CRUSADE score currently recommended for stratification of in-hospital risk of bleeding in NSTE-ACS patients has not been validated in the subset on OAC, and therefore its predictive value in these patients has not been established (Table 3.3) [1]. Nonetheless, application of the CRUSADE score may be of value also in patients on OAC to identify, and potentially correct, established factors for increased risk of bleeding and to estimate, if not the absolute, at least the relative risk of bleeding.

In the acute phase of an NSTE-ACS, effective anticoagulation is required to inhibit thrombin generation and/or activity, thereby reducing thrombus-related events [1]. Effective anticoagulation is also required throughout PCI to avoid thrombus formation, both at the PCI hardware and the vessel plaque injured by the balloon or stent trauma. Several anticoagulant strategies are currently approved for NSTE-ACS patients (Table 3.4).

In patients on OAC with warfarin (or other VKAs) undergoing CORO/PCI in the context of an NSTE-ACS, the periprocedural anticoagulation strategies may include (1) continuation of ongoing OAC or (2) interruption of ongoing OAC, associated or not with heparin bridging, and perform CORO/PCI after the INR has reached < 1.8–2.0. While this latter option would delay performance of CORO/PCI of several days, therefore hampering the benefits of early revascularization, as recommended in most recent guidelines [1], available data suggest that uninterrupted OAC with warfarin is an option as safe and effective [2, 3] and should therefore generally selected (Table 3.5) [1, 4–6]. Besides allowing to provide the optimal invasive management within the recommended time frame, this strategy would also avoid the common INR fluctuations associated with warfarin interruption, and subsequent re-initiation, and the prothrombotic state induced by inhibition of the anticoagulant proteins C and S occurring upon initiation of warfarin, which are in turn associated with increased incidence of bleeding and risk of thromboembolic complications, respectively. Finally, the uninterrupted OAC strategy would reduce the duration of hospitalization. Concern regarding the reversal of OAC with warfarin in the event that the need arises (e.g., severe periprocedural bleeding and/or procedural complication, such as coronary perforation) should not represent a barrier for the adoption of the uninterrupted strategy given that administration of nonspecific (e.g., prothrombin complex concentrates, fresh frozen plasma, recombinant factor VII) and/or specific (i.e., vitamin K) reversal agents are generally, and rather rapidly, effective in antagonizing the anticoagulant effect [7]. To further increase the safety of COR/PCI during uninterrupted OAC with warfarin, however, targeting the preprocedural INR between 2.0 and 2.5 may be preferable.

Provided that OAC with warfarin with an INR > 2.0 has been continued throughout CORO/PCI, the next issue is whether or not additional anticoagulation should be given at the time of procedure. Whereas heparin administration is required upon initiation of PCI to avoid thrombus formation at the angioplasty hardware and/or at the coronary lesion site traumatized by balloon inflation and/or stent implantation in non-OAC patients, whether such strategy is necessary also in patients on effective OAC with warfarin is uncertain. On the one hand, warfarin has been shown to prolong the activated coagulation time (ACT), which is commonly used to monitor the degree of intra-procedural anticoagulation with heparin, in a predictable manner [8], thereby suggesting that no additional anticoagulation is needed. On the other hand, the need for operating in a highly prothrombotic milieu, such as that of an NSTE-ACS, as well as the mechanism of anticoagulation exerted by warfarin (which does not directly inhibit the coagulation factors that have been activated, like heparins and bivalirudin do, but rather leads to the presence in the circulation of inactive coagulation factors) may suggest that additional intra-procedural anticoagulation is advisable [4, 6]. Also, recent data suggest that additional UFH in patients on effective OAC with VKAs may benefit of additional UFH for the prevention of radial artery occlusion, when such access is used for CORO/PCI [10]. Therefore, it appears prudent to provide further anticoagulation, irrespective of the current INR level: a reduced dose of 50–70 IU/kg of intravenous (IV) unfractionated heparin (UFH) should be given in the presence of an INR > 2.0, whereas a standard 70–100 IU/kg dose should be administered for an INR < 2.0 [6]. Whereas enoxaparin may be used for intra-procedural anticoagulation in non-OAC patients, it should preferably be avoided in those on effective OAC with warfarin undergoing CORO/PCI, given the lack of data and the uncertainty regarding the optimal dose and timing of administration. Regarding bivalirudin, which has been shown to be associated to a reduced incidence of both bleeding and adverse ischemic events, both in general ACS populations [10] and a small dataset of AF patients on OAC [11], again there is uncertainty regarding the optimal dose and the infusion scheme, therefore making this option of uncertain value.

Regarding the type, dose, and timing of antiplatelet therapy to be given to patients on OAC with warfarin undergoing CORO/PCI because of NSTE-ACS, again there is no specific evidence. As in non-OAC patients [1], dual antiplatelet therapy (DAPT) with aspirin and a P2Y₁₂-receptor inhibitor is generally recommended [4–6]. Therefore, the patient should be loaded upon presentation with 150–300 mg of aspirin orally or IV [4, 6] (Table 3.6). Whereas newer P2Y₁₂-receptor inhibitors ticagrelor and prasugrel should not be used in association with warfarin and aspirin because of the reported increase in bleeding with prasugrel in combination [12, 13], as well as because of the significant increase in bleeding compared to clopidogrel reported in non-OAC populations of NSTE-ACS patients [14, 15], loading with clopidogrel may either be considered at presentation or after CORO has been performed and indication for PCI has emerged [4]. Withholding triple therapy (TT) until performance PCI may, in fact, reduce the risk of bleeding associated with an invasive procedure performed during aggressive antithrombotic therapy. With the same intent, it may be considered an alternative strategy, recently proposed but not properly validated, consisting of withholding aspirin (which has almost an immediate antiplatelet effect) until performance of PCI, and pretreat the patient with clopidogrel instead [16]. Given, in fact, the need for several hours (approximately 6 and 2, respectively, depending on whether 600 or 300 mg of clopidogrel are given) before effective inhibition of platelet aggregation is reached [17], as opposed to the nearly immediate antiplatelet effect of aspirin, such strategy may be preferable [16]. The higher loading dose of 600 mg of clopidogrel, however, should generally be preferred [5, 6]. The strategy of pretreating patients on warfarin with clopidogrel may also be valuable in the selected cases at high hemorrhagic risk and concomitant low risk of atherothrombotic and thromboembolic events in whom dual therapy (DT) with warfarin and clopidogrel may be considered [4–6]. Even in these cases, however, it needs to be remarked that additional intra-procedural aspirin at the standard dose of 150–300 mg is warranted, because of the likely insufficient protection against periprocedural ischemic complications of the antiplatelet effect of clopidogrel only.

As, and even more than, in non-OAC patients referred for CORO/PCI because of NSTE-ACS, there is currently no indication for preprocedural initiation of glycoprotein IIb/IIIa inhibitors given the lack of significant clinical benefit [4–6].

Finally, as an additional strategy to limit the risk of periprocedural bleeding in patients undergoing CORO/PCI while on effective OAC with warfarin, the radial vascular access site should routinely be preferred, when expertise is available [4–6]. Both in large populations not on OAC [18] and in a small observational experience of warfarin patients [19], the radial approach has been shown to dramatically decrease the incidence of access-site complications. Should the radial access not be feasible, the conventional femoral approach may nonetheless be used also when the INR is > 2.0, provided, however, that is meticulously carried out according to optimal technique (i.e., puncture below the inguinal ligament and of anterior wall only) and possibly with the support of ultrasonographic guidance.