Primary PCI coupled with adjuvant antithrombotic therapy is the current standard of care for patients presenting with STEMI (Fig. 5.2) (Table 5.1) [1]. However, managing antithrombotic therapy in patients presenting with both STEMI and AF is a challenge due to the increased risk of bleeding when antiplatelet therapy is combined with OAC [2]. Many trials are available to guide antithrombotic management of patients with either AF or those with STEMI undergoing primary PCI, but there is a limited amount of randomized data focusing on patients with both conditions.

When a patient on warfarin, or other vitamin K antagonist (VKA), is admitted to the cardiac catheterization laboratory for emergency primary PCI, the question arises on how to manage periprocedural anticoagulation and antiplatelet therapy.

Regarding anticoagulation, additional intravenous (IV) unfractionated heparin (UFH), bivalirudin, or, where approved, enoxaparin should generally be administered, regardless of whether or not current INR is known (Table 5.2) [3, 4]. Albeit available, and of possible clinical value in selected emergency patients [5], point-of-care determination of the INR has not an established role in OAC patients undergoing emergency PCI for STEMI. Given that a thrombus has formed in a coronary artery despite ongoing OAC, it can be assumed that either the anticoagulation effect was suboptimal or that the stimulus to thrombus formation was so strong to overcome even effective anticoagulation. Therefore, additional IV UFH should be given, albeit at a reduced dose (i.e., 30–50 IU/kg) in order to limit the risk of bleeding [4] (Table 5.2). The value of monitoring intra-procedural anticoagulation by means of the activated clotting time (ACT), as commonly performed in non-OAC patients, is currently undetermined, as it is also its value in guiding additional intra-procedural administration of UFH. No data are available regarding enoxaparin in this setting. When used, however, an IV bolus dose of 0.3 mg/kg at the beginning of the procedure appears reasonable. IV bivalirudin at the standard dose of 0.75 mg/kg bolus + 1.75 mg/kg/h infusion may be given in alternative (especially when the bleeding risk is deemed particularly high) (Table 5.2), given both the decrease in bleeding complications and mortality reported in large studies on general populations undergoing PCI in the setting of an acute coronary syndrome (ACS), either with or without ST-elevation [6–8], as well as in a small experience in AF patients receiving a VKA [9].

Regarding antiplatelet therapy, aspirin should be given as soon as possible at the standard dose of 150–300 mg orally (or 80–150 mg IV, when ingestion is not possible) as in standard, non-OAC patients [3, 4]. (Table 5.2). At variance, it seems reasonable not to pretreat with P2Y₁₂-receptor inhibitors patients on warfarin or other VKAs [3]. Although current guidelines recommend administration of P2Y₁₂-receptor inhibitors at the first medical contact (1, 10), the evidence supporting pretreatment in STEMI is not conclusive, particularly in the current era of shorter door-to-balloon times, and no pretreatment data exist in patients on warfarin (or other VKAs) [11]. While the newer, more potent, and less safe P2Y₁₂-receptor inhibitors prasugrel and ticagrelor are not recommended in OAC patients [3, 4, 10], and the delayed onset of effect of clopidogrel even at the higher loading dose of 600 mg may be an issue of concern for acute stent thrombosis, waiting until coronary angiography has been performed and indication for PCI has arisen before administering a P2Y₁₂-receptor inhibitor is preferable. Whereas prasugrel and ticagrelor should be preferred to clopidogrel, owing to the superior efficacy observed in their respective TRITON [12] and PLATO [13] trials, the addition of prasugrel and aspirin to VKA has been shown to come at the price of unacceptable bleeding both in a small, single-center experience [14] and a large population of myocardial infarction patients undergoing PCI [15]. Until reassuring data on the safety of combining prasugrel and ticagrelor to warfarin, or other VKAs (with or without aspirin) will be available, it seems more prudent to prefer clopidogrel as the P2Y₁₂-receptor inhibitor to be added on top of warfarin, or other VKAs, even though the delayed onset of effect of clopidogrel (i.e., approximately 2–4 to 6–8 h for 600 and 300 mg, respectively [16]), remains a limitation [3, 4]. Regarding whether a loading dose of 300 or 600 mg of clopidogrel should be given, no specific data comparing the safety and efficacy of the two doses are available for patients on warfarin or other VKAs. The 600 mg dose, however, is associated with a more rapid and intense platelet inhibition and should therefore generally be preferred [4].

On top of the considerations above, the radial vascular access should generally be preferred over femoral access in STEMI patients on warfarin (or other VKAs) due to the established reduction in bleeding and vascular complications reported both in large datasets of ordinary patients, as well as in small groups of patients on warfarin [3, 17–19]. Whenever the radial approach is not feasible, ongoing therapeutic (i.e., with INR ≥ 2.0) OAC should not exclude the femoral approach, provided that the puncture is carefully carried out according to the proper technique (common femoral artery, anterior artery wall only). Fluoroscopic, and likely even more, ultrasonographic guidance may be of value in further enhancing the safety of the femoral approach. Although a clear superiority of vascular closure devices to manual compression has not been established [20, 21], and data on patients on OAC with warfarin (or other VKAs) are lacking, their use is generally recommended after PCI to shorten time to hemostasis.

Manual thrombus aspiration has not been shown to impact significantly on hard cardiac outcomes of patients undergoing primary PCI [22, 23]. However, it remains a valuable tool to clear the vessel from large thrombi, reduce the likelihood of slow flow/no reflow, and take appropriate decisions about balloon and stent sizes [10]. This technique may therefore be of special value in a setting like that of STEMI in a patient on warfarin, or other VKAs, where the thrombus burden is generally large, and aggressive antithrombotic therapy with additional anticoagulants, and antiplatelet agents, namely, IV glycoprotein IIb/IIIa inhibitors (GPIs), is hindered by the increased risk of bleeding (Table 5.3). To minimize the risk of stroke, special care should be devoted to ensure that the guiding catheter is engaged in the coronary while retrieving the thrombus aspiration device. Except for patients on warfarin, or other VKAs, with known INR < 2.0, IV GPIs may generally be considered only as a bailout therapy in patients with large thrombus burden [10] (Table 5.3), but their use increases bleeding in patients on warfarin [24]. In the different scenario of patients with no AF, the presence of a large thrombotic burden and the lack of preloading with any P2Y₁₂-receptor inhibitor would suggest a lower threshold for the use of GPIs at this stage, with the goal of addressing the expected gap in platelet inhibition. However, the impact of this “blocking and bridging” strategy (with intracoronary 0.25 mg/kg bolus-only administration of abciximab plus oral loading with a P2Y₁₂-receptor inhibitor) [25] is uncertain and potentially harmful in patients on warfarin, or other VKAs, and should therefore not be used.

Regarding the choice of the stent to be implanted (Table 5.4), the consideration made for the general patient not on OAC should at present be generally made also for patients on warfarin or other VKAs. Whereas bare-metal stents (BMSs) allow for a duration of dual antiplatelet therapy (DAPT) of only 1 month, current new-generation drug-eluting stents (DESs) are associated with similar or even lower rates of stent thrombosis compared with BMSs which adds to their established benefit in reducing restenosis [26]. In elective PCI, the package insert of everolimus- and zotarolimus-eluting stents demands for a minimum of 1-month DAPT, but whether this is also valid for STEMI remains hypothetical and unproven. Of note, a polymer-free umirolimus-coated DES versus BMSs (with 1 month of DAPT in each group) has been recently shown significantly more effective and safer in patients at high bleeding risk (including patients on VKAs) undergoing PCI [27]. The limited proportion of ACS patients evaluated needs to be acknowledged as a possible limitation of the study [27]. Indeed, in patients with ACS, the benefit of extended duration of DAPT goes beyond the reduction of stent thrombosis [28]. Patients on warfarin, or other VKAs, who receive stents will generally necessitate a period of triple therapy (TT), including OAC plus DAPT, whose duration essentially depends on the bleeding risk [3, 4]. For this reason, the choice of the stent should take the bleeding issue into consideration: whereas new-generation DESs should generally be preferred over BMSs in patients who according to clinical judgment cannot safely tolerate more than 4 weeks, either polymer-free umirolimus-coated DESs or BMSs should generally be prioritized.

Prescribing the correct antithrombotic regimen in patients undergoing primary PCI in the context of STEMI and AF requires a clear definition of the individual stroke and bleeding risk. While this assessment is difficult before and during primary PCI, it can be easily accomplished in the hours after a successful and uncomplicated procedure. Defining the stroke and bleeding risks in a patient with AF requires the calculation of the CHA₂DS2₂-VASc (Table 5.5) and the HAS-BLED (Table 5.6) scores, respectively [29]. In STEMI patients at high stroke risk (i.e., CHA₂DS₂-VASc score ≥ 2), the duration of TT should be guided by the individual risk of bleeding, namely, up to 12 months, and for a minimum duration of 6 months, when the risk of bleeding is low (i.e., HAS-BLED score 0–2), as opposed to 1–3 months only when the risk of bleeding is high (i.e., HAS-BLED score ≥ 3) [3, 10] (Table 5.7). In these latter patients, considerations on the drop of aspirin could be even cautiously taken into account and dual therapy (DT) of OAC with warfarin (or other VKAs) and clopidogrel therefore prescribed, based on the results of a small, randomized, prospective trial [30] where, however, ACS were poorly represented and efficacy data were not conclusive [11, 31, 32] (Table 5.7). When the risk of stroke is low (i.e., CHA₂DS₂-VASc score 1, because of coronary artery disease as the only risk factor) (Table 5.8), especially in association with high bleeding risk (i.e., HAS-BLED ≥ 3), temporary withdrawal (i.e., 3–6 months) of OAC and prescription of DAPT only (in this case also possible with prasugrel or ticagrelor as P2Y₁₂-receptor inhibitor) may be considered [3]. Given the low risk of stroke associated to a CHA₂DS₂-VASc score of 1, in fact, the potential benefit on stroke reduction of adding OAC to DAPT may be easily offset by the increase in bleeding associated with TT.