Because of the increased risk of bleeding and/or vascular complications in patients undergoing elective CORO/PCI on aggressive antithrombotic therapy, including dabigatran (or other NOACs) and antiplatelet agents, measures aiming to minimize such risk, while not increasing at the same time the risk of thromboembolic/thrombotic complications, need to be implemented [1–3].

Regarding OAC, timely discontinuation of dabigatran (or other NOACs) is the preferable option [1–3] (Table 2.2). Owing to the rapid offset (as well as onset) of action of dabigatran (or other NOACs) (Table 2.3), withdrawing treatment 24–48 h in advance, depending on the patient’s renal function, would generally be sufficient for performing CORO/PCI during low exposure to the drug (Table 2.4) [3]. Taking into account that the half-life of dabigatran (as well as of other NOACs) is approximately 12 h (Table 2.3), after 24 and 48 h from interruption, the drug concentration, and therefore the pharmacological effect, is expected to be 1/4 and 1/16, respectively, of the initial [3]. Conversely, continuation of dabigatran in elective patients has been shown not to provide sufficient suppression of coagulation activation during and after PCI, as indicated by a consistent increase in the levels of prothrombin fragment 1 + 2 and thrombin-antithrombin III complex plasma levels in comparison to standard unfractionated heparin (UFH) [4]. This might not be true for uninterrupted OAC with rivaroxaban, since the suppression of the above markers of thrombin generation and coagulation activity in a small group of patients undergoing elective PCI was reported comparable to standard UFH [5]. For now, however, pre-procedural NOAC interruption should be regarded as the standard of treatment, regardless of the ongoing NOAC [1–3]. Effective ongoing anticoagulation with dabigatran (or other NOACs) would also preclude the use, if needed, of glycoprotein IIb/IIIa inhibitors, as they have been shown (in patients however on therapeutic OAC with warfarin) to largely increase the occurrence of major bleeding complications [6]. In addition, the lack of the possibility to reliably measure the intensity of OAC with dabigatran (as well as of other NOACs) would also make cumbersome the use of additional anticoagulants, such as unfractionated heparin (UFH), in the event that a thrombotic complication (e.g., acute stent/vessel occlusion) occurs. As an alternative to timely interruption of dabigatran (or other NOACs), consideration may be given to perform CORO/PCI at the drug trough level (i.e., ≥24–48 h from last intake), provided however that the bleeding risk is low and adequate hemostasis is possible (i.e., the procedure is performed by the radial approach) (Table 2.4) [3]. Given the short pre-procedural interruption of OAC with dabigatran (or other NOACs) and therefore the associated low risk of thromboembolic events, bridging with other anticoagulants, generally represented by low-molecular-weight heparins (LMWHs), should be considered virtually in no case [3]. In patients on warfarin either because of AF or other indications, in whom perioperative interruption of anticoagulation is generally rather long (on average 5 days), forgoing bridging anticoagulation with LMWH has been recently shown to be as effective as and safer than perioperative bridging with LMWH [7–9].

Regarding oral antiplatelet therapy, front-loading with aspirin and clopidogrel (if not ongoing) is generally performed in patients referred for elective CORO/PCI. Because of the timely interruption of dabigatran before the procedure, effective OAC should be waning at the time of clopidogrel administration, which therefore can be performed with either 300 or 600 mg loading dose (Table 2.2) [2]. Of note, newer P2Y₁₂-receptor inhibitors, including prasugrel and ticagrelor, are not approved in this clinical setting [10], regardless of whether or not they are going to be combined with OAC and should therefore not be used [1, 2, 10].

Regarding the vascular access site, radial approach should generally be preferred because of the dramatic decrease in bleeding and/or vascular complications reported in general population [11], as well as in a small group of patients on OAC with warfarin [12] (Table 2.2). Given, however, that the effect of dabigatran (or other NOACs) should be minimal when timely interrupted, the choice of the radial as compared to femoral approach may not be as important as for patients undergoing PCI on ongoing therapeutic (i.e., international normalized ratio [INR] ≥2.0) OAC with warfarin.

To prevent thrombosis at the PCI hardware and/or at the atherosclerotic plaque disrupted by balloon traumatism, effective anticoagulation is required throughout elective PCI. Once dabigatran (or other NOACs) has been timely interrupted, intra-procedural UFH administration should be carried out as per usual practice (Table 2.5) [1–3]. Therefore, a standard IV bolus of UFH at the dose of 70–100 IU/kg should be given upon the start of procedure [2]. Even though specific data are not available, also glycoprotein IIb/IIIa inhibitors might likely be used safely, should the indication arise (e.g., high thrombus burden or acute stent/vessel occlusion) [1–3].

Regarding the type of stent to be implanted, it has been extensively proven that DESs are more effective than bare-metal stents (BMSs) in preventing restenosis and associated clinical sequelae, at the price however of a possibly increased incidence of late/very late (i.e., >30 days from implantation) stent thrombosis (Table 2.6) [10, 13]. For this reason, dual antiplatelet therapy (DAPT) with aspirin and a P2Y₁₂-receptor inhibitor, including clopidogrel, ticagrelor, and prasugrel (these latter two only in acute coronary syndromes), has been recommended for 1 month after BMS implantation and 6–12 months after DES implantation [10]. Based on several and increasing data showing that new-generation DESs, with both durable and bioabsorbable polymer coatings (Table 2.7), are associated with minimal, and likely not substantially different, incidence of stent thrombosis compared to BMSs, most recent recommendations reduce the suggested duration of DAPT to 6 months only [10]. Further, even durations as short as 1 month may possibly be considered after the implantation of zotarolimus and everolimus, durable polymer DES, as well as of a polymer-free, biolimus A9-DES, given the apparent lack of an increase in stent thrombosis with the interruption of one of the two antiplatelet agents after the first 4 weeks of treatment [14–16]. At present, however, standard prescription of 6-month DAPT may generally be preferable even with these stents, with the option however of more safely interrupt DAPT in the event that a need arises (e.g., bleeding complication) [1, 2]. Based on the above, the stent to be preferably implanted in an AF patient on dabigatran (or other NOACs) should generally be either a BMS or a new-generation DES, to be chosen taking into account the individual risk of restenosis, stent thrombosis, and bleeding (Table 2.5) [1–3]. The recently introduced bioresorbable vascular scaffolds (BVSs) appear at the moment not to have a specific role in AF patients on OAC given that DAPT is warranted at least until the start of BVS degradation, which occurs not earlier than 6 months from implantation [17]. Similarly, no specific role is currently apparent for drug-eluting balloons (DEBs). Given the indication for a short duration of DAPT (i.e., 1–3 months) [18], they may nonetheless represent a valuable option when treating conditions, like in-stent restenosis and small vessel disease, which represent commonly accepted indications for these devices are present.