Article Text

Download PDFPDF

Bleeding associated with the management of acute coronary syndromes
  1. Kalpa De Silva1,
  2. Aung Myat2,3,
  3. James Cotton4,
  4. Stefan James5,
  5. Anthony Gershlick6,7,
  6. Gregg W Stone8
  1. 1Department of Cardiology, King's College Hospital, London, UK
  2. 2Sussex Cardiac Centre, Brighton and Sussex University Hospitals NHS Trust, Brighton, UK
  3. 3Department of Clinical and Experimental Medicine, Brighton and Sussex Medical School, Brighton, UK
  4. 4Department of Cardiology, Heart and Lung Centre, The Royal Wolverhampton NHS Trust, Wolverhampton, UK
  5. 5Department of Medical Sciences and Uppsala Clinical Research Centre, Uppsala University, Uppsala, Sweden
  6. 6Department of Cardiology, University of Leicester, Leicester, UK
  7. 7NIHR Leicester Cardiovascular Biomedical Research Centre, Leicester, UK
  8. 8Department of Cardiology, Columbia University Medical Center, New York Presbyterian Hospital, New York City, New York, USA
  1. Correspondence to Dr Gregg W Stone, Department of Cardiology, Columbia University Medical Centre, New York Presbyterian Hospital, 161 First Washington Avenue, Herbert Irving Pavilion, 6th Floor, New York, NY 10032, USA; gs2184{at}

Statistics from

Learning objectives

  • Enhance understanding with regard to modes of risk stratification alongside consideration of ischaemic versus bleeding sequalae.

  • To demonstrate the prognostic importance of bleeding complications

  • Synthesis of strategies to minimise bleeding complications.


Rupture or erosion of a coronary artery atheroma exposes flowing blood to the prothrombotic contents of the plaque, resulting in platelet activation and subsequent thrombus formation. If this process results in reduced coronary blood flow, the patient may present with an acute coronary syndrome (ACS). Total thrombotic occlusion generally results in ST-segment elevation myocardial infarction (STEMI), whereas incomplete occlusion (or extensive collateralisation) is more likely to present as non-STEMI or unstable angina without evidence of myonecrosis (collectively non-ST-segment elevation ACS (NSTE-ACS)). Revascularisation, most commonly with percutaneous coronary intervention (PCI) is standard of care in ACS, as it restores myocardial perfusion by addressing both the thrombotic obstruction and the underlying coronary stenosis. However, adjunctive pharmacological treatment after revascularisation, or in patients managed conservatively, may be of equal importance in influencing prognosis.1–3

Contemporary adjunctive antithrombotic therapy in ACS includes potent antiplatelet and anticoagulant agents, each of which carries the risk of bleeding. The frequency and implications of haemorrhagic complications must be factored into the risk-benefit analysis for each patient since PCI is increasingly performed in complex subgroups such as those with renal dysfunction, underlying anaemia and the elderly, cohorts with inherently increased bleeding risk.4 ,5 Furthermore, although the absolute bleeding risk will vary according to individual patient characteristics, the overall relative bleeding risk increases with the number, potency and duration of agents co-administered. For example, those patients with ACS, already taking chronic oral anticoagulation (OAC) for stroke protection in atrial fibrillation, are then treated with dual antiplatelet therapy (DAPT) (so-called ‘triple therapy’).6 ,7

There is extensive evidence in the published literature that demonstrates major bleeding to be an independent risk factor for morbidity and mortality.8 This review will appraise the current pharmacological approaches available for the management of ACS, and their impact on bleeding and overall outcomes.

Prognostic implications of bleeding in ACS

It has become increasingly recognised that major bleeding is strongly associated with a subsequent increase in mortality, potentially offsetting the long-term benefits of potent therapies capable of preventing ischaemic complications in ACS.9 Bleeding has been shown to be an independent predictor of both early and late mortality, although unmeasured confounders cannot be excluded, precluding a declaration of causality.10 ,11 One possible explanation for the relationship between bleeding and mortality is given the overlap between the comorbidities associated with bleeding and ischaemia, bleeding is merely a marker for increased ischaemic risk. A second possibility is that bleeding has direct adverse effects (eg, hypoxaemia and hypotension), and may set in motion adaptive changes, which in turn lead to adverse outcomes (figure 1). For example, acute bleeding within the first 1–3 months after PCI may lead to DAPT interruption, increasing the risk of stent thrombosis and myocardial infarction (MI). Red blood cell transfusions have also been reported to be independently associated with mortality, although the mechanisms(s) underlying this effect are still debated.12–14 Finally, specific bleeding diathesis, such as intracranial bleeding will directly affect mortality. Nonetheless, it is generally agreed that bleeding avoidance strategies should be implemented when possible in all patients and especially in those at heightened bleeding risk.

Figure 1

Hypothetical mechanisms linking bleeding to increased mortality.8 APT, antiplatelet therapy.

Bleeding definitions

Unlike ischaemic end points (eg, cardiovascular death, MI, stent thrombosis), for which there is a general consensus on definitions,15 numerous bleeding classification schemes have been proposed. Lack of standardisation has made it difficult to compare the results of different antithrombotic agents across trials. With a view to harmonising bleeding definitions among numerous stakeholder groups, the Bleeding Academic Research Consortium (BARC) was derived (table 1),16 which was modelled on the Academic Research Consortium's standardisation of ischaemic end point definitions such as stent thrombosis. While the BARC classification is a consensus based on expert opinion rather than on data, a close association between bleeding events defined according to BARC definitions and 1-year mortality in patients with ACS and after PCI has subsequently been shown in several reports.17 ,18

Table 1

Bleeding Academic Research Consortium definition for bleeding

Antiplatelet therapy


Timely initiation of antiplatelet therapy is a critical component of the successful management of ACS.1 ,3 Aspirin remains the cornerstone of antiplatelet therapy, and is administered as soon as an ACS is suspected at a loading dose (LD) of 300–325 mg oral (or 250–500 mg intravenous), followed by 75–100 mg daily. Higher daily doses have been associated with greater bleeding without evidence of improved efficacy.19 Aspirin has been shown to reduce 30-day mortality in ACS by ∼25%, and the risks of non-fatal reinfarction and stroke by ∼50%.20 ,21 Aspirin is typically continued indefinitely in patients with established coronary artery disease (CAD) (in the absence of significant bleeding complications).

Of note, aspirin reduces platelet activation by inhibiting the thromboxane A2-mediated pathway, and as such platelet aggregation is only partially inhibited by aspirin. Adenosine diphosphate (ADP) and thrombin activate platelets through P2Y12 receptor and protease-activated receptor-initiated pathways, respectively, even in the presence of aspirin.22 ,23 The adjunctive use of a second antiplatelet agent to inhibit ADP-mediated platelet aggregation is especially critical in patients with ACS undergoing PCI, and has been shown to provide more effective platelet inhibition and to reduce ischaemic events in NSTE-ACS24 ,25 and STEMI.26–28 Thus, the P2Y12 receptor antagonists clopidogrel, prasugrel and ticagrelor are routinely used and have been shown to complement aspirin in reducing ischaemic events in patients with ACS managed invasively.25 ,29 ,30 As the potency of P2Y12 has increased, the added efficacy of aspirin following PCI in ACS has been questioned. The Ticagrelor with Aspirin or Alone in High-risk Acute Coronary Intervention trial (n=9000) and the GLOBAL LEADERS (clinical study comparing two forms of antiplatelet therapy after stent implantation) trial (n=16 000) are both currently recruiting aiming to examine the efficacy of prolonged monotherapy with ticagrelor versus standard DAPT (aspiring and ticagrelor) in terms of bleeding and ischaemic end points. The results of which may provide important insights into the future rationale of antiplatelet therapy.


Clopidogrel, a thienopyridine derivative, inhibits ADP-mediated platelet activation by blocking the P2Y12 receptor, thereby reducing ischaemic events.24 The CURE trial31 confirmed the additive value of clopidogrel to aspirin in reducing the composite end point of death from cardiovascular causes, MI or stroke in patients with ACS treated conservatively or with PCI for refractory ischaemia. In this trial, clopidogrel principally reduced the rate of non-Q-wave MI, while increasing the risk of mild-to-moderate bleeding. However, clopidogrel did not significantly increase life-threatening or fatal bleeding, and the rates of all-cause mortality were similar in patients treated with aspirin alone or aspirin plus clopidogrel. On the basis of the CURE trial, DAPT with aspirin and clopidogrel became the standard of care in ACS. However, while clopidogrel remains the most widely used P2Y12 receptor inhibitor, substantial variability in the extent to which clopidogrel reduces platelet activation limits its utility in many patients. The active metabolites of clopidogrel are principally generated through hepatic metabolism, and genetic polymorphisms of the cytochrome P450 pathway may be present in up to 40% of patients that diminish clopidogrel conversion.32 ,33 Both loss and gain of function alleles have been described.34 Altered platelet reactivity on clopidogrel may result in either an increased risk of stent thrombosis and MI in patients who are hyporesponders35 or an increased risk of bleeding in patients in whose platelets are excessively inhibited.25 ,31 However, the data accumulated for the routine use of platelet function testing, to date, have not demonstrated improved clinical outcomes, with this currently reserved for use in selected high-risk cases only.36 Concerns about hyporesponders to clopidogrel led to pharmacodynamic assessment of clopidogrel dosing, which suggested that there was an advantage in loading with 600 mg rather than 300 mg.37 In the Antiplatelet Therapy for Reduction of Myocardial Damage During Angioplasty (ARMYDA-2) study, patients receiving a 600 mg LD of clopidogrel 4–8 hours before the procedure showed a 52% relative risk reduction (RRR) of the study composite end point (death, MI or target vessel revascularisation) at 30 days compared with patients receiving a 300 mg LD. The benefit of the ARMYDA-2 study was entirely due to a reduction of periprocedural MI, and no bleeding excess was reported, but this study was not powered to show differences in hard end points. The first trial to assess clinical outcomes, CURRENT OASIS-7, tested whether a double-dose regimen of clopidogrel (600 mg LD followed by 150 mg from days 2 to 7, then 75 mg maintenance dosing) was superior to a standard-dose regimen of clopidogrel (300 mg LD followed by 75 mg maintenance) in preventing cardiovascular death, MI or stroke at 30 days in patients with ACS who were treated with an early invasive strategy. The only statistically significant benefit difference observed was a 32% reduction in the risk of stent thrombosis (1.6% vs 2.3%; HR 0.68; 95% CI 0.55 to 0.85; p=0.001), but this was not associated with a difference in the major adverse cardiac events (MACE) between the two groups. Interestingly, although the success of clopidogrel at a standard-dose regimen against placebo in the CURE trial was driven by the benefit obtained in a population mostly medically treated, the medically treated patients did not benefit from higher doses of clopidogrel in CURRENT OASIS-7. These results suggest that the dose of clopidogrel should be tailored to whether the ACS is treated medically or with PCI. Such a hypothesis seems to be confirmed by the good clinical results obtained with prasugrel in the Trial to Assess Improvement in Therapeutic Outcomes by Optimising Platelet Inhibition With Prasugrel (TRITON-Thrombolysis In Myocardial Infarction (TIMI) 38)29 and ticagrelor in the Platelet Inhibition and Patient Outcomes (PLATO) trial,30 both of which showed that platelet inhibition should be more aggressive in patients with ACS undergoing PCI.


Like clopidogrel, prasugrel is a thienopyridine and a prodrug requiring conversion to an active metabolite before binding to and inhibiting the platelet P2Y12 receptor. However, prasugrel metabolism requires a single step (and bypasses several critical cytochrome P450 enzymes, most importantly the CYP2C19 allele), rather than the two-step conversion process required to metabolise clopidogrel. Therefore, prasugrel inhibits ADP-induced platelet aggregation more rapidly, more consistently and to a greater extent than even high doses of clopidogrel in patients32 ,33 and in those undergoing PCI.38 Additionally, prasugrel efficacy is less affected by external conditions and medications that interfere with clopidogrel metabolism, such as cigarette smoking and certain proton pump inhibitors.39

The TRITON-TIMI 38 trial compared prasugrel with clopidogrel in 13 608 patients presenting with ACS in whom PCI was planned and who were followed for 6–15 months.29 Compared with clopidogrel, prasugrel significantly reduced the rates of MI (9.7% vs 7.4%, p<0.001), urgent target-vessel revascularisation (3.7% vs 2.5%, p<0.001) and stent thrombosis (2.4% vs 1.1%, p<0.001). However, this anti-ischaemic advantage came at the cost of excess major bleeding (2.4% vs 1.8%, p=0.03), including life-threatening (1.4% vs 0.9%, p=0.01) and fatal (0.4% vs 0.1%, p=0.002) bleeding. There were no significant differences between the agents in cardiac mortality or all-cause mortality. On the basis of this trial, the European Society of Cardiology (ESC) recommended prasugrel in clopidogrel-naïve patients as an adjunct to primary PCI for STEMI, and in high-risk patients with NSTE-ACS undergoing PCI, particularly those with diabetes mellitus or previous stent thrombosis. Contraindications to its use include patients with previous stroke or transient ischaemic attacks (TIA) (in whom an increased risk of intracranial bleeding was evident), and caution is recommended in patients aged over 75 years or in those with low body weight (<60 kg), in whom a lower daily dose of 5 mg may be considered. Prasugrel should be discontinued 7 days prior to major surgery (vs 5 days for clopidogrel).40 Compared with patients receiving clopidogrel, prasugrel was associated with a marked increase in bleeding in patients requiring coronary artery bypass graft (CABG) surgery.19

In the ACCOAST trial, 4033 patients with troponin-positive NSTE-ACS were randomised in a double-blind fashion to prasugrel 30 mg versus placebo prior to cardiac catheterisation. The primary end point of cardiovascular death, MI, stroke, urgent revascularisation or need for glycoprotein inhibitor (GPI) IIb/IIIa bailout at 7 days occurred in 10.0% of prasugrel pretreated patients versus 9.8% of patients without pretreatment (p=0.81), and major bleeding was increased with pretreatment from 1.4% without pretreatment to 2.6% with pretreatment (p=0.006).41 Therefore, in patients presenting with STEMI, prasugrel may be administered before the coronary anatomy is known as primary PCI is performed in the vast majority of patients, and emergent CABG is uncommon. Conversely, in patients presenting with NSTE-ACS, CABG is more commonly required than with STEMI; as such prasugrel should be withheld until the coronary anatomy is known and PCI is planned.

Finally, in contrast to its benefits in patients with ACS undergoing PCI, prasugrel has not been demonstrated to provide incremental utility compared with clopidogrel in patients with NSTE-ACS managed medically (ie, without revascularisation). In the TRILOGY ACS trial, among patients presenting with NSTE-ACS who were managed medically, the use of either 5 or 10 mg of daily prasugrel showed no benefit in reducing the primary end point of cardiovascular death, MI or stroke compared with clopidogrel, with an excess in TIMI minor bleeding in the prasugrel-treated arm (3.3% vs 2.1%, p=0.02).42


Ticagrelor is a cyclo-pentyl-triazolo-pyrimidine and is a direct-acting reversible oral inhibitor of the P2Y12 receptor. Ticagrelor is absorbed quickly from the gastrointestinal (GI) tract, and reaches its peak concentration in ∼1.5 hours. Ticagrelor has been shown to provide more rapid and consistent inhibition of platelet function than clopidogrel.43 ,44

Among 18 624 patients with ACS (STEMI and NSTE-ACS) randomised in the PLATO trial,30 ticagrelor reduced ischaemic and vascular end points compared with clopidogrel in patients treated both invasively and medically. The primary composite end point of death from vascular causes, MI or stroke occurred in 9.8% of patients receiving ticagrelor versus 11.7% of those receiving clopidogrel (HR 0.84; 95% CI 0.77 to 0.92; p<0.001). Unlike prasugrel in the TRITON-TIMI 38 trial, ticagrelor use in PLATO resulted in a reduction in the rate of cardiac mortality (4.0% vs 5.1%, p=0.001) and all-cause mortality (4.5% vs 5.9%, p<0.001). While there were no significant differences in the rates of overall major bleeding between the ticagrelor and clopidogrel groups (11.6% and 11.2%, respectively, p=0.43), ticagrelor was associated with a higher rate of major bleeding not related to CABG (4.5% vs 3.8%, p=0.03). Compared with clopidogrel, ticagrelor resulted in more instances of fatal intracranial bleeding but fewer instances of fatal bleeding from other sources. Moreover, ticagrelor was effective in patients with prior stroke or TIA, the elderly and those of low body weight, and is thus not contraindicated in these subgroups. However, a significant interaction was present in the PLATO trial between ticagrelor and chronic aspirin dose, such that all patients treated with ticagrelor should be maintained on ≤100 mg of daily aspirin for optimal ischaemic efficacy.45 By blocking the cellular ENT1 receptor and increasing extracellular adenosine concentrations, ticagrelor may also cause the sensation of dyspnoea to a greater degree than clopidogrel. Nonetheless, in the PLATO trial, dyspnoea, severe or prolonged enough to require ticagrelor discontinuation, occurred in <1% of patients.46 Ticagrelor should be discontinued for 5 days before major surgery, although its pharmacodynamic effects are substantially diminished by 3 days. Based on the PLATO trial, the ESC guidelines recommend ticagrelor for patients with NSTE-ACS and STEMI treated either invasively or with a conservative medical approach, including those pretreated with clopidogrel.1

Switching between oral antiplatelet drugs

The availability of three P2Y12 inhibitors with different benefits, risks and costs has resulted in frequent relatively empiric switching between different regimens, for example, from the less potent clopidogrel to the more potent prasugrel or ticagrelor and vice versa after the acute event. Pharmacodynamic studies have suggested that the former practice is safe, with respect to both prasugrel47–49 and ticagrelor.50 ,51 In the PLATO trial,30 approximately half of the study population (46%) randomised to ticagrelor had received clopidogrel at presentation and then switched to their assigned treatment, with preserved efficacy and no safety concerns evident. Similarly, in the prasugrel-treated cohort of the TRILOGY ACS trial,29 96% of patients switched treatment following initial clopidogrel administration with no difference in event rates compared with patients naïve to clopidogrel at baseline. Additionally, there is no difference in major bleeding but an increase in BARC type 1 and 2 bleeding (mild bleeding) in patients switched from clopidogrel to a more potent agent.52 The TRIPLET study reported similar degrees of platelet inhibition in patients loaded with prasugrel or clopidogrel, and subsequently changed to prasugrel.53 Prior to any switching, individual assessment of the potential benefits of the added potency that the newer agents may provide should be considered, in order to optimise the balance between risk of subsequent bleeding and the potential reduction in ischaemic events.54

Intravenous P2Y12 inhibitors

Clopidogrel, prasugrel and ticagrelor are oral agents, and as such have a number of limitations including delayed absorption and onset of action in ACS, due to gastric hypoperfusion and reduced motility, an effect which may be exacerbated by narcotic analgesics such as morphine.55–57 In contrast, cangrelor is an intravenous, fast-acting (3–5 min for full effect), potent (inhibits >95% of ADP-induced platelet aggregation) and direct-acting platelet P2Y12 inhibitor that has a short plasma half-life (<10 min), allowing for complete reversal of platelet function within 1 hour of infusion discontinuation. The CHAMPION-PHOENIX trial58 assessed the efficacy of cangrelor (bolus plus 2 hours infusion) in 11 145 patients undergoing elective or urgent PCI, compared with a clopidogrel LD (300 or 600 mg) in the periprocedural period. The primary 48-hour end point of death, MI, ischaemia-driven repeat revascularisation or stent thrombosis was significantly reduced by cangrelor, and stent thrombosis was reduced by 38% (0.8% vs 1.4%, respectively, OR 0.62; 95% CI 0.43 to 0.90; p=0.01). There were several trial limitations. There was no mandate to treat with oral antiplatelet therapy upstream of the angiogram, which is considered routine practice in the ACS setting. Furthermore, clopidogrel was used as the oral antiplatelet of choice, often at a dose of 300 mg rather than 600 mg, in place of a more pragmatic comparison against prasugrel or ticagrelor. A meta-analysis of three PCI trials of cangrelor versus clopidogrel suggest a 19% RRR in periprocedural death, MI, ischaemia-driven revascularisation or stent thrombosis.59 Cangrelor use was observed to increase minor, but not major bleeding, and no increase in thrombocytopenia.44 ,45 In contrast to the increased rates of bleeding reported with overdosing of GPIs, unintentional cangrelor overdosing (defined as an excess of 20% of the bolus dose) does not lead to an increased rate of major bleeding complications.60 Cangrelor has recently received the US Food and Drug Administration and European Commission authorisation for the compound to be commercialised.61

Intravenous GPIs

GPI, including abciximab, eptifibatide and tirofiban, act at the final common pathway of platelet aggregation and have been shown to reduce ischaemic event rates when administered with heparin in patients undergoing PCI for NSTE-ACS or STEMI. A previous systematic review of early trials suggested that abciximab in STEMI may reduce 30-day mortality by approximately 30% without increasing haemorrhagic stroke or life-threatening bleeding62 (although major bleeding, thrombocytopenia and transfusions were increased with GPI use). However, most trials of GPI were performed a decade or more ago during the clopidogrel era, and before the introduction of the more potent P2Y12 receptor inhibitors.

In NSTE-ACS, several randomised trials have shown increased bleeding with routine upstream treatment with GPI in patients undergoing angiography and PCI, with no additive benefit demonstrated.11 ,63–65 GPI agents are therefore not recommended for upstream use. Nonetheless, in both NSTE-ACS and STEMI, the use of GPI as bailout therapy in cases of refractory large thrombus or slow or no reflow after PCI is considered reasonable (class IIa level of evidence C classification by the ESC), although bleeding is increased in this setting, especially when femoral access has been used. The efficacy and utility of bailout GPI therapy has never been formally assessed in a randomised trial setting (a difficult undertaking), although anecdotal experiences and expert opinion supports its use in this setting.66


Anticoagulation with unfractionated heparin (UFH), which inhibits both factor IIa and Xa, has historically been recommended for all patients with ACS, even those managed medically.67 In those treated with PCI, it may be discontinued immediately after successful PCI, or after up to 5 days in medically treated patients. In the contemporary era, there are four choices of anticoagulation to consider in the acute setting for NSTE-ACS prior to invasive management: UFH, low-molecular weight heparins (LMWH), fondaparinux and bivalirudin.

Subcutaneous enoxaparin (the most widely used LMWH) has theoretical advantages over UFH, although its longer half-life and lack of a readily available test to measure its effect on coagulation may complicate management in patients undergoing PCI. In the initial medical management of NSTE-ACS, the use of enoxaparin was shown in the SYNERGY trial to have superior efficacy compared with intravenous UFH, with reduced rates of MI (RR=0.83; 95% CI 0.70 to 0.99) and repeat revascularisation (RR=0.88; 95% CI 0.82 to 0.95), with similar bleeding (RR=1.00; 95% CI 0.80 to 1.24), although switching between these agents should be avoided.68–70 In comparison to subcutaneous enoxaparin, subcutaneous fondaparinux (a factor Xa inhibitor) has been shown to result in less major bleeding and lower long-term mortality in patients with NSTE-ACS and STEMI managed conservatively.71 ,72 However, in the OASIS 5 trial the dose of the comparator LMWH was considered by some to be unreasonably high, perhaps artificially favouring fondaparinux.38 Moreover, because of a risk of catheter-related thrombus during PCI, full-dose UFH or bivalirudin is required in addition to fondaparinux if PCI is performed.73 ,74 Fondaparinux is thus not recommended in patients with ACS undergoing PCI. Fondaparinux is still used for initial medical stabilisation of NSTE-ACS in the UK. Fondaparinux is not used for ACS in the USA.

Bivalirudin has emerged as an alternative antithrombotic agent in PCI for NSTE-ACS and STEMI, particularly in patients at high ischaemic risk and/or high risk of bleeding. UFH and LMWH are indirectly acting antithrombins which work by binding to antithrombin III, causing a conformational change that results in its activation through an increase in the flexibility of its reactive site loop.75 The activated antithrombin III then inactivates thrombin and other proteases involved in blood clotting. However, there are a number of pharmacological limitations to UFH and LMWH use including variable pharmacokinetics and pharmacodynamics, leading to frequent underdosing or overdosing. Heparins also secondarily activate platelets, and bind non-specifically to other proteins including platelet factor 4, which may result in heparin-induced thrombocytopenia. In contrast, bivalirudin is a small (20 amino acid) molecule direct thrombin (factor II) inhibitor, which has a more predictable pharmacokinetic and pharmacodynamic profile, a shorter half-life (25 vs 60–90 min with UFH and 4 hours or more for LMWH), has intrinsic antiplatelet activity and does not cause thrombocytopenia, providing an alternative with a number of theoretical advantages over UFH.76

Early observational data reported reduced bleeding and ischaemic complications with bivalirudin compared with UFH.77 ,78 Subsequent randomised trials in patients with ACS, in both NSTE-ACS (ACUITY and ISAR-REACT 4) and STEMI (HORIZONS-AMI), in which bivalirudin was compared with UFH plus GPI, demonstrated reduced bleeding and thrombocytopenia with comparable MACE with bivalirudin, with lower all-cause mortality.79 However, with the advent of potent oral P2Y12 receptor inhibitors, GPIs are less frequently used in Europe, and the bleeding advantage with bivalirudin versus UFH alone is likely of lesser magnitude than with bivalirudin versus UFH plus GPI. Moreover, because of its short half-life, stopping bivalirudin at the end of the PCI procedure was associated with an approximate 1% increase in acute (<24 hours) stent thrombosis in patients with STEMI in HORIZONS-AMI, although mortality was reduced at 30 days with bivalirudin (2.1% vs 3.1%, p=0.04), a difference which persisted for 3 years.80 An increase in acute stent thrombosis after discontinuation of bivalirudin has not been observed in NSTE-ACS.

Several additional studies have ignited a debate around bivalirudin use. Unfractionated Heparin versus Bivalirudin in Primary Percutaneous Coronary Intervention (HEAT-PPCI)81 was a single-centre trial comparing heparin only versus bivalirudin in consecutive STEMI patients undergoing PPCI. Unlike all other bivalirudin trials, bleeding was not decreased with bivalirudin in this trial, and the overall efficacy outcome favoured UFH because of more reinfarctions (2.7% vs 0.9%, p=0.004) and stent thromboses (3.4% vs 0.9%, p=0.001) with bivalirudin. However, the duration of bivalirudin use was very short and the activated clotting time with bivalirudin was low compared with other studies, although the HEAT-PPCI authors have highlighted that the bivalirudin infusion was stopped at the end of the procedure, in accordance with the licensed recommendations at the time of the trial initiation and that the optimum dose of the drug infusion remains subject of debate, putting in to question the generalisability of the findings from this study. However, bivalrudin use has subsequently dropped considerably in the UK as a result of this trial.

In contrast, the European Ambulance Acute Syndrome Angiography (EUROMAX) trial82 performed in 65 centres demonstrated a reduced rate of non-CABG major bleeding with prehospital bivalirudin compared with prehospital heparin and optional GPI (5.1% vs 8.5%, p<0.001). This reduction in bleeding with bivalirudin persisted regardless of GPI use, and regardless of arterial access route. However, there was again a significantly increased rate of acute stent thrombosis (<24 hours) (1.6% vs 0.5%, p=0.02), although this increased risk was eliminated if the PCI dose of the infusion was continued for several hours postprocedure.

The large-scale multicentre Bivalirudin versus Heparin with or without Tirofiban during Primary Percutaneous Coronary Intervention in Acute Myocardial Infarction (BRIGHT) trial randomised patients to either bivalirudin with a 3-hour PCI dose postprocedure infusion versus UFH alone versus UFH plus tirofiban (1:1:1 randomisation).83 Bivalirudin resulted in lower bleeding compared with UFH alone and UFH+GPI, and had no excess risk of acute stent thrombosis, presumably due to the 3-hour post-PCI high-dose bivalirudin infusion. It should be noted that clopidogrel was the only P2Y12 inhibitor used in this study.

Finally, in the Bivalirudin or Unfractionated Heparin in Acute Coronary Syndromes (MATRIX) trial, 7213 patients at 105 centres with NSTE-ACS (n=3203) or STEMI (n=4010) were randomised to heparin with optional GPI versus bivalirudin alone, with the latter group randomised again to either a several hour infusion of bivalirudin (low dose or high dose at operator discretion) or no infusion.84 Bivalirudin reduced all bleeding, major bleeding and fatal bleeding (BARC 5) (see table 1 for BARC classification of bleeding), and reduced all-cause mortality (2.3% vs 3.7%, p=0.04) in patients undergoing PCI via both radial or femoral access routes.70–72 As in EUROMAX and BRIGHT, acute stent thrombosis was slightly increased with bivalirudin, although not in patients who received a several hour PCI dose postprocedural infusion. Thus, these data collectively support the use of bivalirudin in STEMI with a 3–4 hours PCI dose postprocedural infusion, especially when GPI is being considered as adjunctive therapy.

In the USA, bivalirudin is the most widely used anticoagulant in the catheterisation laboratory. In Europe, its use has been more selective. The ESC guidelines currently suggest bivalirudin in patients with NSTE-ACS undergoing PCI, with a class I level of evidence A, while UFH is recommended for PCI if patients cannot receive bivalirudin (class I level of evidence C). However, in STEMI the ESC guidelines favour UFH with or without a GPI (class I level of evidence C), while bivalirudin with up to a 4-hour post-PCI infusion is given a class IIa level A recommendation in this setting.66

Factor Xa and thrombin inhibitors in ACS

Since factor Xa plays a central role in thrombosis, chronic inhibition of factor Xa may have a role in secondary prevention after ACS. However, studies examining the utility of factor Xa inhibitors in ACS have reported conflicting results. In the Apixaban for Prevention of Acute Ischaemic Events 2 (APPRAISE-2) trial of 7932 patients with recent ACS and at least two additional risk factors for recurrent ischaemic events, apixaban did not provide any ischaemic, thrombotic or mortality benefits, but did increase bleeding.85 In the 15 526 patient ATLAS ACS 2-TIMI 51 trial, the addition of two doses of rivaroxaban (2.5 and 5 mg twice daily) were evaluated in patients with ACS treated with aspirin therapy alone or DAPT.63 The doses when pooled together resulted in a reduced rate of the primary efficacy end point of death from cardiovascular causes, MI or stroke as compared with placebo (8.9% vs 10.7%, p=0.008). This benefit was realised at the expense of a significantly increased rate of non-fatal bleeding (2.1% vs 0.6%, p<0.001), including cerebral haemorrhage (0.6% vs 0.2%, p=0.009). Treatment with the chronic oral factor IIa inhibitor dabigatran in the Randomised Dabigatran Etexilate Dose-Finding Study in Patients With Acute Coronary Syndromes trial,86 apixaban in APPRAISE-285 and with the thrombin receptor antagonist vorapaxar in the Thrombin Receptor Antagonist for Clinical Event Reduction in Acute Coronary Syndrome87 trial failed to demonstrate reductions in MACE. Moreover, the risk-benefit profile of the novel oral anticoagulants with the more potent P2Y12 inhibitors such as prasugrel and ticagrelor (which have become the preferred agents in appropriate patients with ACS undergoing PCI) is unknown, as the vast majority of trial participants to date have received clopidogrel. Further study is warranted before the use of these agents may be routinely recommended in patients with ACS, given the propensity to increased bleeding events.

Duration of DAPT

Duration of DAPT has a major impact on both ischaemic and bleeding complications. The Patterns of Non-Adherence to Antiplatelet Regimens in Stented Patients (PARIS) registry reported that premature DAPT discontinuation following PCI, or even unplanned DAPT interruption for brief periods of time, were associated with increased rate of MACE.88 This observational study concluded that the increased risk of ischaemic events after DAPT cessation depend on the clinical circumstances and reasons for DAPT cessation, and is attenuated over time. PARIS highlighted that while most events after PCI occur in patients on DAPT, early risk for events due to disruption is substantial, irrespective of stent type. There have been a number of RCTs assessing the safety of shortened DAPT regimens; the ISAR-SAFE (6 vs 12 months),89 ITALIC (6 vs 24 months)90 and OPTIMIZE (3 vs 12 months)91 all showing no significant difference in death, MI or stent thrombosis. However, these studies only included approximately a third of patients with ACS within the studied cohort, which limit there generalisation to the higher-risk ACS population overall. The DAPT trial (Benefits and Risks of Extended Duration Dual Antiplatelet Therapy after PCI in Patients With and Without Acute Myocardial Infarction) provided evidence that prolonged DAPT after stent implantation reduces ischaemic events, although at the cost of increased bleeding, in patients with and without ACS (1.9% vs 0.8%, p=0.005 in the MI group, and 2.6% vs 1.7%, p<0.001 in those without MI).92 The benefit-risk ratio of prolonged DAPT was greater in the 31% of patients who presented with MI, in whom the absolute reduction in MACE rates were greater and the effect on mortality was neutral (figure 2). In contrast, mortality was significantly increased with prolonged DAPT in patients not presenting with a MI. These data suggest that the decision whether to use prolonged DAPT should be individualised. The authors of the DAPT trial have subsequently developed a risk score, aiming to assess the relative risk versus benefit of DAPT (see below).93 These observations are also consistent with the findings from the PEGASUS TIMI-54 trial,94 which assessed the value of prolonged DAPT (median follow-up of 33 months) with two different doses of ticagrelor versus aspirin only in n=21 162 patients who had an MI 1–3 years prior and had at least one other high-risk feature. A recent meta-analysis assessing the utility of prolonged DAPT with any of the P2Y12 inhibitors combined with aspirin following PCI in MI showed that there was a reduction in ischaemic events, including in the individual end points of cardiovascular death, recurrent MI and stroke. DAPT beyond 1 year increases major bleeding, but not fatal bleeding or non-cardiovascular death.95 In contrast, Palmerini et al96 performed a systematic network meta-analysis of 10 randomised trials (31 666 patients) demonstrating, in an all-comers population, that prolonged DAPT was effective in reducing MI and stent thrombosis after drug-eluting stent (DES) implantation, in the context that the studies were generally underpowered to detect differences in stent thrombosis and MI; additionally, its use also resulted in increased major bleeding, and non-cardiovascular and all-cause mortality (table 2).90 ,97–101

Table 2

Summary of studies on duration of DAPT

Figure 2

Treatment effect by duration of antiplatelet therapy according to initial patient presentation with a myocardial infarction (from the dual antiplatelet therapy (DAPT) trial).92 Pint is the interaction between the relative treatment effect of 30-month versus 12-month DAPT on selected end points according to the presence or absence of baseline myocardial infarction for the original drug-eluting stent implantation.

Further analyses are required to determine which specific subsets of patients may benefit from prolonged DAPT. Patients at high risk for bleeding or low risk for ischaemic complications might be most optimally treated with <1 year of DAPT. In this regard, numerous randomised trials have randomised patients to <12 vs 12–24 months DAPT after DES (table 2). However, high-risk patients were excluded from these trials, and the results apply principally to patients undergoing PCI with non-acute ischaemic syndromes.

These data reinforce the need for an individualised approach when determining the duration of DAPT after PCI. Patients at low risk for bleeding but high risk for ischaemic complications may benefit from prolonged DAPT (eg, a patient aged 40 years undergoing primary PCI for STEMI), especially if first-generation DES were used, whereas those at low ischaemic risk but high bleeding risk (eg, a patient aged 85 years in whom a single stent was implanted for stable CAD) would likely benefit from a shortened DAPT regimen (3–6 months), especially if a second-generation DES (those with the lowest stent thrombosis potential) was used. The ESC guidelines suggest that P2Y12 inhibitor administration for a shorter duration of 3–6 months after DES implantation may be considered in patients with ACS who have a high bleeding risk, with a class IIb, level of evidence A recommendation.102 The recent US guideline update suggests that at least 1 year of DAPT is appropriate in all patients with ACS.103

Achieving balance: bleeding risk versus anti-ischaemic benefit

Approximately 40% of bleeding events in patients with ACS undergoing PCI via the femoral artery arises as a result of using this access site.104 ,105 These episodes are largely avoidable by radial artery access. Radial artery access also minimises vascular complications, an important consideration in high-risk patients with ACS. Thus, transradial access is now increasingly performed for ACS PCI, with a number of studies demonstrating its value in this setting. The Radial Versus Femoral Randomized Investigation in ST-Segment Elevation Acute Coronary Syndrome106 trial reported that in 1001 randomised patients with STEMI, radial compared with femoral access was associated with significantly lower rates of cardiac mortality (5.2% vs 9.2%, p=0.02) and bleeding (7.8% vs 12.2%, p=0.026). The larger Radial versus femoral access for coronary angiography and intervention in patients with acute coronary syndromes107 trial, in contrast, reported no significant difference in major bleeding between patients randomised to radial versus femoral access (0.7% vs 0.9%, respectively, p=0.23) among 7021 randomised patients with NSTE-ACS and STEMI. This lack of difference was attributed to the relative inexperience of the radial operators involved in this study, which may be particularly relevant in the context of primary PCI. However, the recently published large-scale Radial versus femoral access in patients with acute coronary syndromes undergoing invasive management: a randomised multicentre trial (MATRIX)108 trial (n=8404) observed a significant reduction in bleeding with radial artery access driven by a reduction in BARC major bleeding (1·6% vs 2·3%, p=0·01), a finding accompanied by a reduction in all-cause mortality (1·6% vs 2·2%, p=0·045).

Despite the potential benefits of radial artery access, non-access site bleeding is responsible for approximately 60% of all bleeding episodes after PCI in ACS, and has been associated with a greater increase in mortality than has access site bleeding.109 In the REPLACE-2 trial, independent predictors of bleeding after PCI included increased age, female gender, chronic kidney disease and anaemia.110 In the large-scale ADAPT-DES111 registry, major determinants of bleeding following PCI also included lower baseline haemoglobin, higher platelet reactivity on clopidogrel (greater inhibition) and use of chronic OAC therapy.112 This study also highlighted that bleeding from the GI tract is the most common source of major bleeding in patients on DAPT. GI bleeding can be ameliorated by prescribing a proton pump inhibitor, which is recommended in any patient with increased risk of GI bleeding. Non-steroidal anti-inflammatory drugs should be avoided if possible.1

Triple therapy

A substantial proportion (10%–20%) of patients undergoing PCI for treatment of ACS have coexistent conditions such as atrial fibrillation or metallic valve prostheses that require chronic OAC in addition to DAPT.113 There are limited randomised data to guide management in this setting, which is likely to become more common as the population ages. The What is the Optimal antiplatElet and anticoagulant therapy in patients with oral anticoagulation and coronary StenTing (WOEST) trial,114 in which 573 low-risk patients treated with warfarin undergoing a non-complex PCI were randomised to continued aspirin plus clopidogrel versus clopidogrel alone (plus placebo). Discontinuation of aspirin was associated with a marked reduction in bleeding complications, 19.4% compared with 44.4% (p<0.0001) (figure 3), without deleterious effects on the rates of cardiac death (1.1% vs 2.5%, p=0.21) or stent thrombosis (1.4% vs 3.2%, p=0.17). However, only approximately a quarter of the study population presented with an ACS, and the results are uncertain in patients at higher risk for stent thrombosis.

Figure 3

One-year incidence of bleeding with double therapy (clopidogrel and warfarin) versus triple therapy (aspirin, clopidogrel and warfarin) following percutaneous coronary intervention.46

One other study has examined the issue of dual versus triple therapy in patients with atrial fibrillation after PCI. The ISAR-TRIPLE investigators tested the hypothesis that 6 weeks (n=307) versus 6 months (n=307) of clopidogrel after DES implantation would result in a net superior clinical outcome (ie, a composite of death, MI, stent thrombosis or TIMI major bleeding at 9 months) in patients treated with aspirin and OAC. In this trial, 6 weeks of clopidogrel was not superior to 6 months in terms of the primary outcome, and did not reduce the rate of TIMI major bleeding as expected (5.3% vs 4.0%, HR 1.35, 95% CI 0.64 to 2.84, p=0.44). Much like WOEST, the results of the ISAR-TRIPLE study were more hypothesis-generating than proscriptive.

The use of prasugrel or ticagrelor instead of clopidogrel in a triple therapy regimen is not recommended given the likelihood of substantially increased bleeding and the lack of published clinical data.8 In a modest-sized observational study, use of prasugrel with warfarin lead to a fourfold increase in TIMI minor and major bleeding, when compared with clopidogrel and warfarin.115

Conversely, the use of non-vitamin K antagonist (VKA) oral anticoagulants (rivaroxaban, apixaban, dabigatran, edoxaban) may prove safer in the management of patients requiring triple therapy, since these agents have been demonstrated to reduce major bleeding and intracranial haemorrhage in patients with non-valvular atrial fibrillation compared with warfarin.116–118 Several multicentre trials are now currently underway designed to determine whether triple therapy incorporating a non-VKA versus warfarin to standard DAPT can improve the bleeding profile, specifically with apixaban (AUGUSTUS (NCT02415400)); dabigatran (REDUAL-PCI (NCT021664864)) and rivaroxaban (PIONEER AF-PCI (NCT01830543)).

The most recent ESC guidelines on myocardial revascularisation note that dual therapy with a (N)OAC and clopidogrel 75 mg/day may be considered as an alternative to initial triple therapy in selected patients (class IIb, level of evidence B).66 Acknowledging the paucity of clinical data, a European consensus document has been published which recommends various combinations of therapy based according to stent type and patient presentation (figure 4).119

Figure 4

Choice of antiplatelet and anticoagulant in patients with acute coronary syndrome (ACS) and atrial fibrillation.119 CAD, coronary artery disease; PCI, percutaneous coronary intervention.

Specific groups at risk of bleeding

A number of risk factors have consistently been shown to predict bleeding in ACS and PCI. These include older age, female sex, lower body weight, anaemia, renal insufficiency, prior bleeding and chronic oral anticoagulant use. Increasing age is a strong risk factor for bleeding (figure 5).8 ,120 Age-related hyperfibrinolysis and thrombus instability may predispose to increased bleeding in the elderly.121 The Global Registry of Acute Coronary Events (GRACE) registry demonstrated that the adjusted odds of having a major haemorrhage prior to discharge increased by about 30% per decade of age.120 Conversely, hypercoagulability in older patients, caused by elevated levels of activated factors such as VII, IX and X, in addition to increased platelet reactivity, has been associated with an increased risk of acute stent thrombosis.122–124 Women with ACS have also been identified as being at increased risk of vascular and bleeding complications, with the GRACE registry observing a 43% excess of major in-hospital bleeding in women compared with men.125 Chronic kidney disease also contributes to bleeding risk, with registries of contemporary ACS demonstrating an approximate 50% increase in the risk of in-hospital major bleeds in patients with renal insufficiency.120 Moreover, patients with renal failure often inadvertently receive higher than recommended doses of antithrombotic drugs. Renal insufficiency is also a risk factor for more extensive atherosclerosis and stent thrombosis. Anaemia, which may be present in 15% of patients presenting with ACS and in over a third of elderly patients with acute MI, has consistently been reported to be predictive of subsequent bleeding complications.126 ,127 Anaemia also has the potential to worsen the myocardial insult during ischaemia and infarction, both by decreasing the oxygen content of the blood supplied to the jeopardised myocardium128 and by increasing myocardial oxygen demand through necessitating a higher cardiac output to compensate for inadequate systemic oxygen delivery. Anaemia has been identified as a powerful and independent predictor of MACE in patients across the spectrum of ACS.12 Moreover, a history of a prior bleeding diathesis, with or without persistent anaemia, independently increases the risk of further bleeding complications following initiation of DAPT.129–131

Figure 5

Predictors of bleeding in acute coronary syndromes.120 GPI, glycoprotein inhibitor; LMWH, low-molecular weight heparins.

Bleeding risk scores

Clinical scoring systems are increasingly used in the management of ACS, with the GRACE score now part of the ESC guideline recommendations for identifying patients at high risk for death and MI following presentation with an NSTE-ACS. The use of equivalent bleeding risk scores is not yet routine. There have been a number of clinical bleeding risk scoring tools developed over the last decade.120 ,132 ,133 Many of the risk factors that predict bleeding, such as age, renal dysfunction and vascular disease, also predict a higher incidence of ischaemic events and in-hospital mortality, reducing their ability to identify patients who might benefit from more versus less potent antithrombotic therapy. Some risk factors, however, have high discrimination to predict bleeding (but not ischaemia), specifically anaemia, prior bleeding and triple therapy. The CRUSADE bleeding score133 includes eight factors: female sex, history of diabetes, prior vascular disease, heart rate, systolic blood pressure, signs of congestive heart failure (CHF), baseline haematocrit <36% and creatinine clearance, and was tested in a cohort of 89 000 patients who presented with a NSTE-ACS, providing a method of quantification of bleeding risk for in-hospital major bleeding, across all modes of treatments for ACS.

Recently, the DAPT study investigators have developed a risk prediction tool to identify patients who may (or may not) benefit from prolonged (>12 months) DAPT after PCI. Nine clinical and angiographic variables were identified in multivariable analysis that differentiated the late risk of bleeding versus ischaemia on prolonged DAPT. The subsequently derived DAPT score ranges from −2 to 10, with a higher score conferring greater ischaemic risk and increased benefit of long-term DAPT therapy. The factors included in the risk model following regression analysis were age, smoking, diabetes, MI at presentation, prior PCI or prior MI, paclitaxel-eluting stent, stent diameter <3 mm, CHF or left ventricular ejection fraction <30% and vein graft stent. Patients with a DAPT score of ≥2 (∼50% of patients) had reduced rates of ischaemic MACE and stent thrombosis with prolonged DAPT, with only a small increase in bleeding, and a neutral effect on mortality. In contrast, patients with a DAPT score of <2 had a lesser reduction in ischaemic events, a marked increase in bleeding and greater all-cause mortality with prolonged DAPT. Unfortunately, the DAPT score did not incorporate the variables of prior bleeding, baseline anaemia or chronic OAC use, and outcomes on ticagrelor were not studied. Further prospective evaluation and external validation is required to assess the discriminative ability of this prediction tool, particularly in specific subsets of patients.93

Strategies to minimise bleeding complications

The balance between minimising ischaemic versus bleeding complications is complex and mutifactorial. While there are no guidelines that specifically encapsulate the current contemporary trial data on the various facets of minimising bleeding, particularly in the context of PCI in the ACS setting, there are certain principles that may guide management in this often complex conundrum. Radial access should be considered wherever possible, particularly in the elderly, to minimise access site complications. Antiplatelet therapy type and duration should be administered on a individualised basis, taking into account the complexity of the PCI procedure, along with anatomical considerations such as myocardial jeopardy if the stent were to become compromised, and the innate risk profile of the patient, with the use of risk scores to encapsulate the risk. Additionally the duration of therapy should be tailored, such that those with multivessel disease, particularly in cases of high disease burden, may require long-term therapy, while PCI to short, focal lesions with no other bystander disease in low-risk patients (eg, young, non-diabetic) may allow for a shortened duration of therapy. Further data regarding the optimal duration of DAPT therapy and specifically which subsets of patients would benefit from prolonged versus shortened therapy will be forthcoming in subsequent years, and may provide important insights into how DAPT therapy can be tailored in the future.

Management of bleeding complications

Sudden interruption of DAPT may result in adverse ischaemic events (eg, acute stent thrombosis and MI from untreated atherosclerotic disease), therefore the decision to discontinue DAPT in those with bleeding must be considered carefully. Interruption of antiplatelet therapy related to bleeding episodes has been independently associated with increased mortality, so cessation of DAPT should only be undertaken after consideration of the potential ramifications.134 Minor bleeding should be managed by simple means (eg, mechanical compression) without the interruption of antiplatelet drugs. For an acute GI bleed, consideration should be given to continuing DAPT along with endoscopic management with cautery or clipping of a visibly bleeding vessel and/or use of haemostatic gel.135 Most surgeries may be performed while the patient remains on DAPT for protection against stent thrombosis, accepting some increase in bleeding. When DAPT cessation is required, the need for treatment interruption should be weighed against the risks of stent thrombosis and MI. DAPT should usually be reintroduced as early as possible once bleeding has resolved. These recommendations have been summarised in an ESC position statement as follows:8

  • Minor bleeding should preferably be managed without interruption of active treatments (class I, level of evidence C).

  • Major bleeding requires interruption and/or neutralisation of both anticoagulant and antiplatelet therapy, unless bleeding can be adequately controlled by specific haemostatic interventions such as the use of reversal agents include protamine, fresh frozen plasma, vitamin K, and platelet transfusions (class I, level of evidence C).

  • Blood transfusions may have deleterious effects on outcome and should therefore be considered individually, but withheld in haemodynamically stable patients without overt bleeding with haematocrit >25% or haemoglobin level >8 g/dL (class I, level of evidence C).

  • Endoscopic procedures should be performed without cessation of DAPT, unless risk of acute GI haemorrhage outweighs the risk of stent thrombosis. Close liaison with interventional cardiologists and gastroenterologists is recommended to optimise patient care.136


Patients with ACS require potent antiplatelet agents and anticoagulants to reduce ischaemic and thromboembolic risk, all of which increase bleeding to some degree according to their mechanism of action, potency and duration of treatment, a risk which is further modified by specific patient characteristics. Understanding the risk factors for bleeding is an essential consideration in ACS, as bleeding has a significant impact on early and long-term morbidity and mortality. Thus, each patient should undergo an individualised assessment of the risks of bleeding and ischaemia. Reducing the frequency of bleeding complications while maintaining anti-ischaemic effectiveness is an important goal in the management of ACS, and has the potential to reduce morbidity, mortality and healthcare costs. Further studies are required to identify optimal risk-stratification protocols to maximise the risk-benefit profile of different therapies in the patients with ACS, an increasingly important consideration as newer drugs are introduced that may additively increase bleeding when combined with other agents.

Key messages

  • Current antiplatelet and antithrombotic therapy is complex and acts on a number of pathways. These therapies are underpinned by a large amount of randomised controlled data; further data are required to continue to improve understanding in the field, and therefore improve patient care and outcomes.

  • Patient therapy should be individualised based on their overall respective bleeding and ischaemic risks.

  • Adoption of risk scores such as the dual antiplatelet therapy score may prove a useful adjunct to guiding therapy if further prospective trials confirm its utility.

  • The use of radial access for percutaneous coronary intervention (PCI) should be adopted whenever possible to minimise access site bleeding complications. The transfemoral approach remains an important mode of access, which needs to be trained and adopted when necessary.

  • Management of bleeding following PCI in the ACS setting often requires a multidisciplinary approach.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to:

  • RCP credits: Log your activity in your CPD diary online (—pass mark is 80%.

  • EBAC credits: Print out and retain the BMJ Learning certificate once you have completed the MCQs—pass mark is 60%. EBAC/ EACCME Credits can now be converted to AMA PRA Category 1 CME Credits and are recognised by all National Accreditation Authorities in Europe (

Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal's username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.



  • Contributors KDS, AG and GWS wrote and edited the manuscript. AM, JC and SJ all helped edit the manuscript at various stages.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.