When dealing with a patient on oral anticoagulation (OAC) with a non-vitamin K-antagonist oral anticoagulant (NOAC) referred for emergency primary CORO/PCI, the immediate challenge is how to manage anticoagulant therapy. In accordance with most recent guidelines on the management of STEMI patients, parenteral anticoagulation with either bivalirudin or unfractionated heparin (UFH) must be routinely used during primary PCI [3] (Table 6.3). In alternative, intravenous (IV) enoxaparin may also be considered [3] (Table 6.3). Whether and how a patient on chronic NOAC therapy should be further anticoagulated when undergoing emergency primary CORO/PCI is currently incompletely defined. While directly inhibiting thrombin (dabigatran) or factor Xa (rivaroxaban apixaban, edoxaban) and effectively preventing clinical events associated with spontaneous thrombosis (and thromboembolism) both at the arterial and venous site [4–11], NOACs have yielded equivocal results when evaluated regarding the activation of coagulation induced by PCI. In stable patients undergoing elective PCI, dabigatran as compared to standard UFH has been shown ineffective in inhibiting the generation and activation of thrombin [12], which in contrast appears to be effectively antagonized by rivaroxaban in the same setting [13]. Given this disparity, for which a clear explanation is lacking, and in the absence of a laboratory test allowing for rapid and reliable information on the presence and magnitude of the anticoagulant effect of ongoing NOAC [14], it is currently suggested to further anticoagulate a patient on chronic NOAC treatment undergoing PCI [15] (Table 6.4). This appears even more applicable in patients submitted to PCI in the context of an acute coronary syndrome (ACS) because of the highly thrombogenic milieu and the evident inability of ongoing anticoagulation with NOAC to prevent coronary thrombosis responsible for the current coronary event. Based on previous observations of preserved efficacy [16] and aiming at increasing the overall safety, a reduced dose of IV UFH, however, should preferably be given [15] (Table 6.4). When deciding, however, on the dose of additional IV UFH, the time of last NOAC intake, as well as the renal function (if known), should be taken into account: in the presence of normal renal function, the elimination half-life of NOACs is approximately 12 h [17] (Table 6.5), and therefore the exposure to the drug (and to its pharmacological effect) will be reducing by approximately 50 % every 12 h passing by (Fig. 6.2). Given that after 24 h from last administration the drug concentration (and pharmacological effect) is expected to be about 25 % of initial, full-dose UFH (i.e., 70–100 IU/kg) should be added upon start of PCI [15]. A reduced dose of UFH (i.e., 50 IU/kg) should on the contrary be administered when the last NOAC intake has occurred within the previous 12–24 h [15]. Adjustments of the above doses should be carried out based on the presence and degree of renal insufficiency, with reduced UFH dose to be given for creatinine clearance < 50 ml/min [15].

Whether additional UFH may be of value to prevent radial artery occlusion (when this approach is used for CORO/PCI), as it has been shown for patients on warfarin [18], is currently unknown. Given that NOACs have a direct inhibitory effect on activated coagulation factors, similarly to UFH and differently from warfarin (which acts by promoting the hepatic synthesis of inactive coagulation factors), further anticoagulation to prevent radial artery occlusion might not be needed.

As an alternative to UFH, IV bivalirudin may be considered, especially when the bleeding risk is deemed particularly high [14, 15, 19] (Table 6.4). In large studies on general populations undergoing primary PCI [20], as well as in a small experience in patients with AF on warfarin [21], bivalirudin at the standard dose of 0.75 mg/kg bolus followed by 1.75 mg/kg/h infusion has been shown significantly more effective and safer than standard IV UFH. The very limited evidence in this regard, as well the lack of information regarding the optimal dose (standard vs. reduced) and administration scheme (standard vs. shorter infusion duration), does not strongly support at present the routine use of bivalirudin in place of standard IV UFH in NOAC patients undergoing emergency primary PCI.

The other main challenge in dealing with patients on NOAC undergoing emergency primary CORO/PCI for STEMI is the management of antiplatelet therapy. In accordance with the most recent guidelines on the management of STEMI patients, dual antiplatelet therapy (DAPT) with aspirin and a P2Y₁₂-receptor inhibitor should be given as early as possible [3] (Table 6.3). The newer, more potent, and more effective P2Y₁₂-receptor inhibitors ticagrelor and prasugrel (this latter when not contraindicated, i.e., in patients with previous stroke or transient ischemic attack) should generally be preferred over clopidogrel [3]. In patients with ACS, however, both ticagrelor and prasugrel are associated with increased noncoronary bypass-related major bleeding events compared to clopidogrel [22, 23]. Accordingly, the use of the newer, more potent, P2Y₁₂-receptor inhibitors ticagrelor and prasugrel should generally be avoided in patients who are at increased risk of bleeding at baseline, such as those on chronic OAC who also receive aspirin [14, 15, 19, 24]. Initial experiences on both mixed and myocardial infarction populations undergoing PCI [25, 26] indeed show the significantly increased risk of bleeding when prasugrel is added to OAC (with vitamin K antagonist) and aspirin in triple combination. Together with aspirin (at the standard recommended doses), which has a nearly immediate antiplatelet effect [27], clopidogrel only should then be considered for patients on NOAC undergoing emergency PCI for STEMI [14, 15, 19, 24] (Table 6.4). A loading dose of 600 mg should be given as early as possible, taking however into account that effective antiplatelet inhibition may not be reached before at least 2 h [28]. Clopidogrel loading dose of 300 mg should generally not be considered, given the required time of at least 9 h for effective platelet inhibition [28]. Similarly to non-OAC patients, prasugrel and ticagrelor may likely be considered when STEMI is related to stent thrombosis during ongoing clopidogrel therapy, provided that technical problems (e.g., incomplete stent expansion and/or lesion coverage) have been excluded (possibly also based on intracoronary imaging) and/or nonresponsiveness to clopidogrel has been documented (by using platelet reactivity testing) [19].

As a final periprocedural issue, selection of the vascular approach associated with the lowest incidence of bleeding is of great importance. Both in large, general populations not on OAC [29] and in a small experience in OAC patients on warfarin [30], the radial access has been shown to decrease the rate of bleeding and/or vascular complications compared to the femoral approach and should therefore be considered routinely [19]. In the event that the radial approach has failed or is not feasible, ongoing therapeutic OAC (i.e., last NOAC intake within 12–24 h) should not exclude the femoral approach, provided that the puncture is carefully carried out according to the proper technique (i.e., puncture of common femoral artery and of the anterior artery wall only) and possibly guided by fluoroscopy or, even better, by ultrasonography. Whereas no specific data are available regarding hemostasis in patients on NOAC undergoing emergency primary PCI via the femoral route in STEMI, it appears reasonable to routinely consider the use of hemostatic devices in this setting. In general populations, they have been consistently shown comparably effective and safe to manual compression, while allowing for more rapid ambulation (and discharge) [31, 32].

A first issue to be considered in the presence of acute thrombotic occlusion of a coronary artery in the context of STEMI is whether thrombus aspiration may be of value. While having shown not to impact significantly on hard cardiac outcomes of patients undergoing emergency primary PCI [33, 34], manual thrombus aspiration may nonetheless be considered (Table 6.5). Given the proven ability to remove thrombus, thereby reducing the risk of slow flow/no reflow, manual thrombus aspiration might be especially helpful in the context of STEMI in a patient on NOAC where aggressive antithrombotic therapy with additional anticoagulant and antiplatelet agents, namely, IV glycoprotein IIb/IIIa inhibitors (GPIs), is hindered by the increased risk of bleeding.

Regarding the use of parenteral platelet inhibition, glycoprotein IIb/IIIa inhibitors (GPIs) should generally be avoided when ongoing anticoagulation is presumed effective, that is, within 24 h from last drug intake (or longer in the presence of renal dysfunction) (Table 6.6), or may be considered only as a bailout therapy in patients with large thrombus burden [24]. Data from patients on effective OAC with warfarin (i.e., INR > 2.0) show, in fact, a substantial increase in the risk of bleeding with apparent no benefit on the occurrence of major adverse cardiac and cerebrovascular events [35]. Owing to the reported lack of differences in terms of major bleeding, as well as of clinical benefits, with the intracoronary administration of abciximab as a bolus compared to standard IV infusion [36], the limitations in the use of GPIs should be confirmed regardless of the route of administration.

Whether the potent, rapid-acting, reversible, parenteral P2Y₁₂-receptor inhibitor cangrelor, which on top of standard antithrombotic therapy has been shown effective in reducing ischemic events without increasing bleeding in elective or ACS patients undergoing PCI [37–39], may be of value in OAC patients remains to be determined. As yet therefore, cangrelor has no specific indication in patients on NOAC undergoing emergency PCI in STEMI.

A key decision during the procedure is the choice between drug-eluting stents (DESs) and bare-metal stents (BMSs). Even though the main determinant of the duration of dual antiplatelet therapy (DAPT) is the clinical syndrome in which the stent has been implanted (i.e., ACS as compared to elective) and not the type of stent implanted (as is on the contrary the case for PCI performed in elective patients) [24], the type of device implanted may nonetheless have an impact in the management of subsequent antithrombotic therapy. This is particularly true in OAC patients in whom the increased risk of bleeding associated with the unavoidable combination of OAC and antiplatelet therapy after stenting may induce to tailor the duration of DAPT more on the time required for prevention of stent thrombosis (i.e., between 1 month for BMSs and 1–6 months for new-generation DESs) rather than on that required for prevention of recurrent cardiac events after an ACS (i.e., 12 or more months). Currently available new-generation DESs, either with durable or resorbable polymer coating or polymer-free (Table 6.7), are associated with similar or even lower rates of thrombosis than BMSs, thereby generally allowing for a duration of DAPT not longer than 6 months, and possibly as short as 1 month only. Because of that, as well as of the superior efficacy compared to BMSs in preventing restenosis, and associated subsequent re-revascularization, DESs should be generally preferred also in NOAC patients, including in the context of the STEMI [15, 19, 24, 40]. At present, no individual DES appears preferable to others, although those approved for a DAPT duration of 1 month only may be of value, especially in patients at increased risk of bleeding in whom premature discontinuation (i.e., earlier than 3–6 months) may be required if a bleeding occurs. In selected patients deemed at particularly high risk of bleeding, such as those with a HAS-BLED score ≥ 3 [41] (Table 6.8), BMSs may be considered instead [15, 19, 24]. In this respect, it is, however, noteworthy that a recent trial showed that among patients at high risk for bleeding, a polymer-free umirolimus-coated stent was superior to a bare-metal stent with respect to the primary safety and efficacy end points when used with a 1-month course of dual antiplatelet therapy after PCI [42].

At present, it is unknown whether drug-eluting balloons (DEBs), which allow for a duration of DAPT as short as 1 month only [43], or bioresorbable scaffolds (BVSs), which also may allow limited duration of DAPT (apparently, however, not less than at least 6 months) [44] may be of value in patients at increased risk of bleeding, such as those on NOAC undergoing emergency primary PCI for STEMI, and should therefore not be routinely considered.

Given that DAPT of aspirin and clopidogrel has been shown significantly less effective than OAC (with warfarin) for stroke prevention in AF patients with at least one risk factor for stroke [45] and that OAC (with warfarin) has been shown significantly less effective than DAPT with aspirin and a thienopyridine in preventing subacute stent thrombosis [46], a combination therapy of OAC and antiplatelet(s) is indicated after PCI in AF patients at increased risk of stroke and therefore requiring OAC. Early stratification of the risk of stroke should be performed by using the CHA₂DS₂-VASc score [41] (Table 6.9): while individuals scoring 0 should be treated with DAPT only (as they have no indication for long-term OAC) and those scoring ≥ 2 should receive combination OAC and antiplatelet(s) therapy, male patients scoring 1 and female patients scoring 2 should be considered for either DAPT or combination therapy of OAC and antiplatelet(s). Because of the limited risk of stroke, especially in the short term, associated with CHA₂DS₂-VASc score 1 (Tables 6.9 and 6.10), the net benefit (i.e., combined incidence of ischemic events, including death, myocardial infarction, repeat revascularization, stent thrombosis and stroke, and major bleeding events) of triple therapy (TT) may be lower than that of temporary DAPT, especially in patients at increased risk of bleeding. Early stratification of the risk of (major) bleeding should therefore also be performed by using the HAS-BLED score (Table 6.8) [19, 41].