Advancements in PCI technologies have allowed extending this treatment option to high-risk patients who were traditionally managed conservatively. These patients typically have extensive CAD and multiple comorbidities that, not only increase their risk for thromboembolic events, but also for bleeding complications. Indeed, a study of an all-comer population undergoing DES implantation has shown that as high as 1 in every 15 patients experienced post-discharge bleeding at a median of 300 days after the procedure [5]. Interestingly, the impact of bleeding on two-year mortality was significantly larger compared with post-discharge MI [5]. This study, together with other observational studies, shed light on the prognostic relevance of post-discharge bleeding after PCI [10].

Finding patients at HBR is of highest importance for the management of antithrombotic therapy after PCI. Nonetheless, a lack of standardization in defining this population limits the generalizability of trial results as well as clinical decision-making. Based on review of the literature, the Academic Research Consortium (ARC) recently published an agreement definition of patients at HBR based on fulfillment of specific criteria (Fig. 1, Ref. [11]) [12]. Several registry-based studies have validated the ARC-HBR definition by showing an incidence of Bleeding Academic Research Consortium (BARC) 3 or 5 bleeding risk of $\ge$4% in HBR patients at one year after PCI [13, 14, 15]. Moreover, these studies revealed that HBR patients account for around a third of all PCI patients, reiterating the need for tailored antiplatelet therapy strategies that mitigate the bleeding risk.

Fig. 1.

ARC-HBR definition of HBR. Major and minor risk factors used in the definition for HBR [11]. 1 major or $\ge$2 minor criteria qualify as HBR. DAPT, dual anti-platelet therapy; eGFR, estimated glomerular filtration rate; Hb, hemoglobin; NSAID, non-steroidal anti-inflammatory drugs; OAC, oral anti-coagulant.

Over the past years, numerous risk scores have been designed to inform and guide decision making on DAPT duration and intensity after PCI. The European Society of Cardiology guidelines recommend using the DAPT and PRECISE-DAPT risk scores [16, 17]. The DAPT scoring system was developed using predictors of both ischemic and bleeding events to identify patients who derive the greatest benefit over harm from prolonging DAPT beyond 12 months of PCI [18]. Conversely, the PRECISE-DAPT was developed to assess the risk of out of hospital bleeding up to 2 years post-PCI [19]. The PARIS score encompasses two separate prediction models to evaluate ischemic and bleeding risks after PCI [20]. Although these scores share similar components, each has its own features (Table 1). The DAPT score included somewhat lower risk patients who were event-free at 12 months post-PCI, while both PRECISE-DAPT and PARIS included patients immediately after discharge of index PCI. While HBR patients composed approximately 25% of all subjects considered, each scoring system identified different rates of bleeding ranging from 1% to 10% [11]. Therefore, it has been difficult to confidently use such scoring systems in HBR patients undergoing PCI. Recently, Urban et al. [12] developed the ARC-HBR trade off model, which predicts the risk of non-periprocedural major bleeding and thrombotic events at one year among HBR patients who have undergone PCI. Although this was the first risk score especially dedicated to HBR patients, it should be noted that this tool was derived from studies using different DAPT durations (i.e., driven by the protocol of the study or guideline-based) and recommendations should not be solely made based on its risk predictions.

The paradigm in interventional cardiology research has shifted over the past few years into testing strategies that unite modern DES platforms with short DAPT durations (Table 2, Ref. [21, 22, 23, 24, 25]). LEADERS-FREE was a randomized double blind trial comparing outcomes of HBR patients receiving the polymer-free biolimus-eluting BioFreedom stent vs. a similar bare metal stent (BMS); patients in both arms were maintained on DAPT for one month after PCI [21]. The BioFreedom stent was found to be superior to BMS with regards to the composite of cardiac death, MI, or stent thrombosis, largely driven by decreased rates of MI. Conversely, the ONYX ONE trial examined the same BioFreedom stent in comparison to the durable-polymer zotralimus-eluting Resolute Onyx stent in a similar HBR population [22]. The Resolute Onyx stent was found to be non-inferior to the BioFreedom stent with respect to the same primary outcome as above. However, since both the LEADERS-FREE and ONYX ONE trials preceded the ARC-HBR consensus, definitions of HBR differed, making comparisons among studies difficult to interpret. Most notably, LEADERS-FREE and ONYX ONE considered age $\ge$75 alone as HBR criterion for study inclusion. Despite this, improvements in stent technologies have clear benefits in patients at HBR, most notably in the ability to decrease DAPT duration. The SENIOR trial examined outcomes of elderly patients $\ge$75 years old by randomizing the bioabsorbable-polymer everolimus-eluting Synergy stent vs. BMS followed by shortened DAPT (1 month in stable patients and 6 months if unstable). DES was superior regarding the primary outcome of all-cause death, MI, stroke, or ischemia driven target vessel revascularization at one year, mostly driven by the latter [23]. Additionally, the DEBUT trial examined whether drug-coated stents were non-inferior to BMS in HBR patients [26]. Not only did they determine that drug-eluting stents were non-inferior to BMS with respect to major adverse cardiac events (MACE) after 9 months, they found superiority. Although these 4 studies showed that shortening DAPT duration is a safe and effective strategy in HBR patients who undergo PCI with new-generation DES, optimal therapy durations cannot be determined based on these trial designs.

More recently, the EVOLVE Short DAPT registry enrolled n = 1437 HBR patients treated with the Synergy stent followed by 3-month DAPT [25]. Such short DAPT duration was found to be non-inferior to a historical cohort of patients treated with 12-month DAPT with respect to death or MI but failed to show and advantage in terms of bleeding. Notably, the study was non-randomized and the control group was not uniform as it included multiple different stent types, potentially limiting the generalizability of the results. The XIENCE Short DAPT program included 3 registries (XIENCE 28 Global and 28 USA, and XIENCE 90) for a total of n = 3652 HBR patients undergoing PCI with the fluoropolymer-based cobalt-chromium Everolimus-eluting Xience stent who discontinued DAPT at 1 or 3 months post-PCI if event-free and treatment-adherent [27]. Both short DAPT regimens (1 and 3 months) were non-inferior to standard DAPT (6 to 12 months) with respect to death or MI and superior with respect to major bleeding, after propensity score-stratification vs. an historical group of patients receiving the same stent [24]. In a subsequent exploratory analysis from the XIENCE data, 1 month of DAPT was shown to have comparable ischemic outcomes and lower bleeding risk compared with 3-month DAPT [28]. MASTER DAPT was the first RCT testing different DAPT durations in HBR patients treated with a new-generation DES. The trial included 4434 HBR patients who underwent placement of the biodegrable-polymer sirolimus-eluting Ultimaster stent [29]. Subjects who were event free after 1 month of index PCI were either randomized to DAPT discontinuation followed by either aspirin or a P2Y${}_{12}$ inhibitory monotherapy or continuation of DAPT for at least 5 additional months. The short DAPT (1 month) regimen was shown to be non-inferior to prolonged DAPT with regard to net adverse clinical events and MACE [29]. Specifically, the composite of major or clinically relevant bleeding was observed in 6.5% in the experimental group as compared to 9.4% in the control without tradeoff in ischemic events. Therefore, shortening DAPT duration to 1 to 3 months may be a reasonable approach in selected HBR patients, pending additional data from large clinical trials.

Aspirin has been the mainstay therapy for long-term secondary prevention of ischemic events for decades. Recently, its undisputed benefits have been challenged for several reasons: (1) increased risk of intracranial and extracranial bleeding, especially in HBR patients, (2) widespread use of optimal medical therapy including disease-modifying drugs (i.e., angiotensin converting enzyme-inhibitors, angiotensin receptor blockers, statins, etc.), and (3) the introduction of more potent antiplatelet agents. However, the introduction of new antiplatelet agents in PCI practice has always requested these new agents to prove their benefits on a background of aspirin therapy such that their individual effects have never been truly assessed. The PLATO trial showed that the more potent P2Y${}_{12}$ inhibitor, ticagrelor, is superior to clopidogrel in reducing ischemic events at 12 months among ACS patients, although at the cost of increased bleeding [30]. To note, PLATO also suggested that low dose (300 mg) aspirin was more effective than a high dose ($\ge$300 mg) in preventing ischemic events when combined with ticagrelor [31]. This raised the question as to whether aspirin is at all needed in presence of potent P2Y${}_{12}$ inhibitors [32, 33]. The GLOBAL LEADERS trial addressed this question in an all-comer population of patients undergoing PCI for stable CAD or ACS [34]. It randomized over 15,000 patients to either ticagrelor monotherapy for 23 months after 1 month of DAPT or 12 months of DAPT followed by aspirin monotherapy. Ticagrelor monotherapy was not superior to 12-month DAPT for the primary endpoint of all-cause death and new Q-wave MI (3.81% in experimental vs. 4.37% in control; p = 0.073) and was associated with similar rates of bleeding events [34]. The study had several limitations including an open label design [35]. GLASSY, a GLOBAL LEADERS adjudication substudy conducted at the top-10 enrolling sites (n = 7585) yielded similar conclusions as the parent trial but for the first time suggested a reduction in thrombotic events (MI and ST) with ticagrelor monotherapy vs. aspirin between 1 and 2 years post-PCI [36]. The SMART CHOICE study showed P2Y${}_{12}$ inhibitor monotherapy after 3-month DAPT was non-inferior to the standard treatment of 12 months with respect to major adverse cardiac and cerebrovascular events (MACCE) a composite of all cause death, stroke, MI [37]. In line with previous studies, they found a decrease in bleeding rates, specifically bleeding BARC 2-5, but did not find any differences in major bleeding [37]. Unlike the GLOBAL LEADERS trial, SMART CHOICE included multiple P2Y${}_{12}$ inhibitors with clopidogrel the most frequently used, giving evidence to its benefit as a monotherapy, similar to STOPDAPT2. Although this study was conducted in a low-risk Asian population, decreasing the bleeding risks would most certainly be beneficial to those at HBR. Similar results were found in the STOPDAPT-2 trial where they found that 1 month of DAPT followed by clopidogrel monotherapy was superior to 12 months of DAPT in a composite of cardiovascular and bleeding events, largely driven by a reduction in bleeding [38]. However, the trial included mainly low-risk Japanese patients with very high rates of intravascular imaging use, therefore its generalizability has been questioned. More recently, the STOPDAPT-2 ACS trial, an extension of STOPDAPT-2, enrolling only ACS patients, showed that 1-month DAPT followed by clopidogrel monotherapy was not non-inferior with respect to net adverse events (including ischemic and bleeding endpoints) when compared to standard DAPT [39]. These results were driven by a significant decrease in the occurrence of major bleeding events, which was offset by a concomitant increase in ischemic events. The TICO trial further investigated ticagrelor monotherapy compared to DAPT among patients with ACS. Although event rates were lower than expected, ticagrelor monotherapy after 3 months of DAPT decreased the incidence of net adverse clinical events (composite of major bleeding and adverse cardiac and cerebrovascular events) [40]. Again, this difference was mostly driven by a decrease in bleeding complications, with no tradeoff in ischemic events. Even though this study excluded patients at HBR, it further supported the use of P2Y${}_{12}$ inhibitor monotherapy, reducing overall risks of bleeding.

The TWILIGHT study examined ticagrelor monotherapy following 3-month DAPT in high-risk patients undergoing PCI. Patients were considered at high risk for ischemic and bleeding events if they fulfilled at least one clinical and one angiographic high-risk feature. This double-blinded placebo-controlled study randomized patients to receive either ticagrelor monotherapy or ticagrelor plus aspirin for 12 months after being event free for 3 months post PCI. Ticagrelor monotherapy was shown to reduce the incidence of the primary endpoint of BARC 2, 3, or 5 bleeding without an increase in ischemic events [41]. A sub-analysis of the TWILIGHT trial looking at patients who qualify as HBR based on ARC-HBR criteria showed consistent results, with larger absolute risk reduction in major bleeding observed with ticagrelor monotherapy in HBR versus non-HBR patients [42]. Several meta-analyses of the above studies showed decreased risks of bleeding, while no concomitant increase in events [43, 44, 45]. Although many of these studies do not specifically examine HBR patients, a short DAPT duration followed by P2Y${}_{12}$ inhibitor monotherapy emerged as a safe and effective bleeding-avoidance strategy, although extra caution might be needed in those presenting with ACS (Table 3, Figs. 2,3, Ref. [29, 34, 37, 38, 40, 41, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61]).

Fig. 2.

Trial designs of shortening DAPT after PCI regardless of stent design. Comparison of studies of short DAPT duration regardless of specific stent designs [34, 37, 41, 46, 47, 48, 49, 50, 51]. Blue arrows indicate duration of DAPT, red arrows indicate duration of monotherapy. DAPT, dual antiplatelet therapy.

Fig. 3.

Short DAPT trial with specific stent designs. Comparison of studies with short DAPT designs with specific stent designs [29, 38, 40, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61]. Blue arrows indicate duration of DAPT, red arrows indicate duration of monotherapy. BES, biolimus eluting stent; BP-BES, bioresorbable polymer biolimus-eluting stent; BP-DES, bioresorbable polymer drug-eluting stent; BP-EES, bioresorbable polymer everolimus-eluting stent; BP-SES, bioresorbable-polymer sirolimus-eluting stent; CoCr-EES, cobalt chromium everolimus-eluting stent; DAPT, dual antiplatelet therapy; DP-EES, durable polymer everolimus-eluting stent; E-ZES, endeavor zotarolimus-eluting stent; PF-DCS, polymer free drug coated stent; ZES, zotarolimus-eluting stent. * indicates experimental group. † indicates control group.