Valvular Heart Surgery
The field of cardiac surgery is urgently in need of new anticoagulant agents. Approximately 300,000 prosthetic valves are implanted yearly, after which lifelong anticoagulation therapy is mandatory as thromboprophylaxis. To date, vitamin K antagonists (VKAs), such as warfarin, remain the only licensed agents to prevent mechanical valve-induced thromboembolic events but have major shortcomings. The hazard of valve thromboembolic events is reduced with warfarin, but haemorrhagic complications are an inhibiting factor. Also, its use is impeded because of many drug and food interactions. As a result, patients selected for therapy often discontinue intake and are therefore insufficiently anticoagulated.1–3 Finally, routine monitoring of the international normalised ratio (INR) is necessary to ensure adequate anticoagulation; an INR outside the therapeutic range increases the risk of thrombus formation or haemorrhage. Ideally, new agents should be more effective in reducing thromboembolic events and lowering the risk of bleeding events, have less interaction with other intake and maintain the INR within therapeutic range without regular monitoring.
Coronary bypass surgery is an effective treatment for myocardial ischaemia, but 28% of saphenous vein grafts occlude within the first year after surgery.4 Early graft thrombosis is one of the reasons for this high occlusion rate and aspirin reduces the frequency of saphenous vein graft occlusion. Five to 10% of patients who take aspirin, however, do not have a complete antiplatelet effect from the usual doses prescribed and the effect of aspirin may vary over time. Intimal proliferation and smooth muscle cell hyperplasia also decrease vein graft patency and clopidogrel, but not aspirin, may inhibit intimal hyperplasia. Dual antiplatelet therapy with clopidogrel and aspirin for patients with cardiovascular disease has therefore been suggested to improve vein graft patency, but the Clopidogrel after surgery for coronary artery disease (CASCADE) trial could not confirm a better patency rate by adding clopidogrel to aspirin in patients undergoing coronary artery bypass grafting (CABG).5 Clopidogrel irreversibly inhibits platelet activation and increases intra- and postoperative bleeding as has been demonstrated in several studies of patients that underwent cardiac surgery after percutaneous coronary intervention (PCI), resulting in prolonged hospital stays and even mortality in some cases.6 A new platelet inhibitor agent that is as effective as clopidogrel but with a reversible effect is needed to optimise graft patency and decrease the risk of bleeding.
Stroke is one of the most serious complications of CABG surgery and intraoperative strategies can only reduce the rate to a certain extent. There is an unexplained peak risk of stroke after CABG around postoperative day two.7 The inflammatory process and hypercoagulability after surgery might be responsible for this peak and new anticoagulant treatment might be effective in reducing the incidence of postoperative stroke. New anticoagulant agents aiming at reducing both early graft thrombosis and postoperative stroke would mean a major evolution.
Several new anticoagulants have been or are currently evaluated to prevent adverse events in patients with atrial fibrillation (AF) and/or acute coronary syndrome (ACS) (see Table 1).8–18 Many of these newly introduced agents walk the same path. After trials have evaluated new agents for stroke prevention in AF or as treatment for deep-vein thrombosis, subsequent trials will likely evaluate its efficacy and safety in patients with ACS. Eventually, many of these patients will undergo CABG and the use of anticoagulant medication then becomes an issue regarding whether they will result in bleeding complications.
Agents AVE5026, idrabiotaparinux, otamixaban and RB006 are less suitable for long-term use in patients with mechanical heart valves because of their parental administration. Oral anticoagulants (see Figure 1 and Table 1) are easier to use in patients postoperative of cardiac surgery, but need to show their effectiveness and safety against existing therapies.
Dabigatran is taken orally as a prodrug called dabigatran etexilate (see Table 1). An oral bioavailability of approximately 6% is fully converted into dabigatran. In plasma, the peak level of the drug occurs after two hours. Eight per cent is excreted by the kidneys and the drug has a half-life of 14–17 hours.19–20 After initially being evaluated in hip and knee surgery, this factor IIa (fIIa) inhibitor was introduced as an anticoagulant in the field of cardiovascular diseases to prevent stroke in high-risk patients with AF. The Randomized evaluation of long-term anticoagulant therapy (RE-LY) trial21 compared two doses of dabigatran (110mg and 150mg twice daily) to the conventional adjusted dose (INR 2–3) of warfarin in >18,000 patients.
After a median follow-up period of two years, the primary end-point of stroke and systemic embolism was reached in 1.69% per year for warfarin, 1.53% for dabigatran 100mg (p<0.001 for non-inferiority) and 1.11% for the high dosage of dabigatran (p<0.001 for superiority). The primary safety end-point of major bleeding defined as “a reduction in the haemoglobin level of at least 20g per litre, transfusion of at least two units of blood, or symptomatic bleeding in a critical area or organ” was met in 3.36% per year in patients taking warfarin, 2.71 and 3.11% in patients taking the low and high dosage of dabigatran respectively. As a result of the superior effectiveness and comparable safety of the 150mg dose, dabigatran was approved by the US Food and Drug Administration (FDA) in the US in October 2010 for patients with AF. Also available is a 75mg twice daily dose for patients with severe renal failure, after which a debate has arisen. RE-LY did not investigate the use of dabigatran in such a low dose and no other data on the efficacy of this dose is available. Moreover, the dose of 110mg showed to be non-inferior to warfarin in both the efficacy and safety end-points, but was not approved.
Many discuss the possibility of dabigatran replacing VKAs in patients with mechanical heart valves. Although no studies have yet compared the two in such a setting, it is speculated that similar outcomes as the RE-LY trial would be the result. The international multicentre double-blind randomised RE-ALIGN study is currently in the start-up phase to compare the efficacy and safety in valvular surgery and start of enrolment is anticipated in June 2011.
Rivaroxaban targets factor Xa (fXa) and has a half-life of seven to 11 hours (see Table 1).22 It has shown to be effective in the prevention of thromboembolisms in patients undergoing orthopaedic surgery before it was evaluated in a phase II trial including patients with ACS.16 Results showed that rivaroxaban dose-dependently increased the rate of clinically significant bleeding for all doses. The primary efficacy end-point of death, myocardial infarction (MI), stroke, or severe ischaemia requiring revascularisation was 5.6% for rivaroxaban compared to 7% for placebo patients (p=0.10). The composite end-point without severe ischaemia was, however, significantly improved compared to aspirin alone (3.9 versus 5.5%).
The authors rightly stated that rivaroxaban increases bleeding and might reduce major adverse events. Recently presented results from the Randomized double blind study comparing once-daily oral rivaroxaban with adjusted dose oral warfarin for the prevention of stroke in patients with nonvalvular atrial fibrillation (ROCKET-AF) study at the 2010 American Heart Association (AHA) Scientific Sessions showed a reduction in the primary composite end-point of stroke or non-central nervous system embolisation by 21% in a per protocol analysis.23 Patients treated with rivaroxaban also had statistically significant fewer intracranial haemorrhages, critical organ bleeds and bleeding-related deaths compared to the warfarin group, but did, however, have an increased rate of haemoglobin/haematocrit drop and transfusions.
Rivaroxaban has a potential role in cardiac surgery as an adjunct to aspirin post-CABG, or as a replacement of VKAs in patients with mechanical heart valves.
Apixaban is administered in an active form, resulting in a rapid absorption with peaking plasma levels after three hours (see Table 1).20 Apixaban showed encouraging results in the Apixaban for prevention of acute ischemic and safety events (APPRAISE) trial as an additional therapy to aspirin and/or clopidogrel compared to a placebo in patients with ACS.18 There was a non-significant trend to a decrease of ischaemic events in both an apixaban 2.5mg twice daily and 10mg once daily dose. This did however come with a cost of increased bleeding in both doses. The authors concluded that the safety and efficacy may be dependent of background aspirin and clopidogrel therapy. A follow-up study, Examination of Apixaban (Factor Xa inhibitor) as adjunctive therapy in patients with recent acute coronary syndrome (APPRAISE-2), was halted after it became clear that the additional risk of bleeding would not be sufficient compared to the reduction of ischaemic events.24
Edoxaban (see Table 1) was compared to warfarin for the primary safety end-point of major and/or clinically relevant non-major bleeding in patients with AF.13 Of four edoxaban doses (30 and 60mg, both once and twice daily), the twice-daily doses significantly increased the incidence of bleeding, while the once daily 30mg (3.8%) and 60mg (3%) had similar rates as warfarin therapy (3.2%). As a result the agent is now in a phase III trial to evaluate the prevention of stroke in AF patients.12 Positive results could lead to a trial with ACS patients.
Similar to dabigatran, AZD0837 inhibits fIIa (see Table 1). Unlike its competitor, the initial results were less optimistic. A phase II study demonstrated a sufficient safety profile with a low incidence of bleeding, but the phase III randomised AZD0837 compared to warfarin for the prevention of stroke and systemic embolic events in atrial fibrillation (ASSURE) trial designed to evaluate AZD0837 in patients with AF was halted in 2009.19
Bivalirudin, a parental fIIa inhibitor, was evaluated in a randomised trial including over 13,000 patients with ACS. It demonstrated non-inferiority compared to heparin in a composite end-point of ischaemia and a reduction of major bleeding events.25 In off-pump coronary artery bypass patients, bivalirudin was associated with a similar rate of blood loss within 12 hours of surgery, but showed a better graft flow by angiography compared to patients treated with heparin and protamine.26 Even in bypass surgery with cardiopulmonary bypass, results with bivalirudin were similar to heparin with protamine reversal on all accounts of treatment success, 24-hour blood loss, transfusions, duration of surgery and mortality.27
The development of anticoagulants is an ever-expanding field and the number of agents currently evaluated is not limited to the agents mentioned previously. Oral fXa inhibitors betrixaban, eribaxaban, YM150, TAK442 and LY 517717 and fIIa inhibitor argatroban have not yet been compared to standard therapy in large randomised trials, but have had animal and small human studies for safety assessment.20 In the near future, more clinical evaluations will contribute to whether these agents can be used in patients with ACS or AF, eventually resulting in an assessment in cardiac surgery.
Aspirin and clopidogrel are the most used platelet aggregation inhibitors. Aspirin has demonstrated significant efficacy and a relatively safe profile in patients in need of cardiac surgery. Clopidogrel was introduced more recently and showed promising results in large randomised trials including patients with ACS,28–29 after which it is now indicated in patients treated with PCI. In patients with AF, clopidogrel in addition to aspirin, has been inferior to oral anticoagulation.30 Also, patients in whom warfarin was contraindicated showed a reduced rate of major vascular events, but also increased bleeding events.31
There are, however, some shortcomings of clopidogrel. Several genetic predispositions are linked to a nonresponsive effect of clopidogrel, which as a result leads to insufficient platelet aggregation inhibition. The drug has also been associated with adverse events in patients taking proton-pump inhibitors,32 statins,33–34 or with CYP2C19-carrier status.35 Furthermore, clopidogrel preoperative of coronary artery bypass surgery significantly increases the risk of bleeding complications.36 More recently, clopidogrel has even been linked with a more frequent occurrence of sepsis after CABG surgery.37 These arguments contribute to the debate as to whether clopidogrel is a useful adjunct and therefore other antiplatelets are considered as alternatives (see Table 2).38
Ticagrelor (AZD6140) was initially developed for patients with atherosclerosis to prevent thromboembolic events and might serve as an alternative to clopidogrel. Both bind to the P2Y12 receptor on platelets and thereby inhibit the aggregation response to adenosine diphosphate (ADP), but ticagrelor reversibly binds to the receptor (see Table 2). Furthermore, ticagrelor is the first chemical antiplatelet agent in which metabolic activation after oral intake is not required. The drug needs to be taken twice daily, because of an approximately 12-hour half-life time.39–41 The effect of tricagrelor has been deeply investigated over the last few years. Several randomised trials compared the new drug to clopidogrel to obtain efficacy, safety and tolerability data in patients with cardiovascular disease.39–44 However, data on its perioperative effect in cardiac surgery is scarce. The large Platelet inhibition and patient outcomes (PLATO) trial included over 18,000 patients to compare ticagrelor (180mg loading dose and 90mg twice daily thereafter) to clopidogrel (300–600mg loading dose and 75mg daily thereafter) in patients with ACS.44 Ticagrelor was shown to be significantly better in preventing the primary efficacy composite end-point of cardiovascular death, MI, or stroke at 12 months. A subgroup analysis in patients that underwent CABG was recently published in which 1,261 patients had received anticoagulation within seven days of surgery.45 In the trial it was recommended that clopidogrel was discontinued five days preoperatively and ticagrelor between 72 and 24 hours before surgery. Baseline characteristics were comparable between groups. The rate of bleeding, transfusions, chest-tube output and reoperations for bleeding was statistically comparable between ticagrelor and clopidogrel. At 12 months, the composite end-point was similar. However, both all-cause mortality and cardiovascular death were statistically significant in favour of ticagrelor. The authors concluded that tricagrelor within seven days of CABG was associated with reduced all-cause and cardiovascular mortality, but not with an increased risk of CABG-related bleeding. It has been suggested that there is either a positive effect of ticagrelor or a dismal effect of clopidogrel, as earlier studies suggested. Future studies have to confirm the positive effect of ticagrelor in patients undergoing cardiac surgery in the setting of ACS.
Prasugrel is a thienopyridine similar to clopidogrel and ticagrelor in its binding to the P2Y12 receptor to prevent platelet aggregation as a result of ADP (see Table 2). It is comparable to clopidogrel in multiple aspects. Prasugrel also needs enzymatic metabolism,46 after which the peak plasma concentration is reached after approximately 30 minutes. The unbound half-life is estimated to be ≈7 hours.47 Some advantages regarding interactions with inducers or inhibitors of metabolism and renal or hepatic comorbidities are noted.48–50 Nevertheless, prasugrel irreversibly binds the platelet receptor and ticagrelor would, therefore, be preferred.
Cangrelor is another thienopyridine reversibly binding P2Y12 and therefore has a major advantage over prasugrel (see Table 2). Cangrelor furthermore has an effect that is worn off 60 minutes after parenteral injection, as a result of its less than five-minute half-life.51 The large Cangelor versus standard therapy to achieve optimal management of platelet inhibition (CHAMPION) randomised trial evaluated the use of cangrelor compared to a placebo in over 5,000 patients undergoing PCI.51 Treatment with cangrelor did not significantly decrease the rate of the composite end-point consisting of death, MI, or ischaemia-driven revascularisation at 48 hours. At 48 hours, secondary end-points stent thrombosis and death were significantly reduced. This advantage was, however, associated with an increased risk of major bleeding, driven by more groin haematomas. A second CHAMPION trial randomised almost 9,000 patients to cangrelor or clopidogrel. The primary end-point was similar in both groups, but patients treated with cangrelor showed an increased risk of major bleeding (p=0.06), although not according to the criteria by Thrombolysis in myocardial infarction (TIMI) or Global utilization of streptokinase and tissue plasminogen activator for occluded coronary arteries (GUSTO).52 Because of the parenteral administration is in unlikely that cangrelor will have a major contribution in the surgeons practice, but the short time before normalisation of platelet function has some benefits. No evaluation in surgical patients has yet been done and therefore, no evidence-based conclusions can be made.
The lower rate of stroke after PCI compared to CABG surgery suggests that the addition of clopidogrel to aspirin could decrease the incidence of cerebrovascular accidents. This effect of dual antiplatelet therapy on stroke rate however, has not been validated. The high occlusion rate of venous grafts after CABG surgery is another complication that could be lowered with adequate anticoagulation therapy. Although no studies have shown that a combination of aspirin with another agent increases graft patency, the higher rate of bleeding complications with dual antiplatelet therapy is clear. The addition of clopidogrel to aspirin for patients undergoing CABG surgery is therefore not justified.
The negative aspects of VKAs as thromboprophylaxis after mechanical valve replacement lead to a higher usage of bioprostheses. Until what age the use of bioprosthesis can be advocated is not established as it is unclear if the complications of VKAs outweigh the risk of reoperation as a result of the limited durability of bioprostheses.
The data on new anticoagulant therapy in surgical patients is limited. Before any advice can be given as to which drugs can be used as alternatives to VKAs, the efficacy and safety needs to be addressed in prospective trials. Multiple studies in patients with ACS or AF are currently in the start-up, inclusion, or follow-up phase. The RE-ALIGN randomised trial investigating the use of dabigatran in patients requiring mechanical valve replacement will start enrolment mid-2011. If results are similar as in the RE-LY trial, dabigatran will probably extend its indication to valvular heart disease, thereby ending the 60-year reign of VKAs. This could decrease the rate of bleeding and result in a renewed interest in mechanical valves. Ticagrelor could potentially have a role in CABG. The subgroup analysis of the PLATO trial suggests a better outcome compared to clopidogrel. A new study specifically focusing on bypass patients is needed. The new drugs that are currently under investigation fail to demonstrate a lower bleeding rate compared to VKAs. Hopefully, the next generation of drugs will not only show the same efficacy, but also demonstrate a better safety profile.
- Although preliminary data suggests that the use of new antiplatelets in patients with ACS undergoing CABG lowers mortality, this has to be established in a randomised controlled trial.
- New anticoagulants may eliminate the need for regular INR monitoring of patients with mechanical heart valves.
- Trials that specifically address the issue of antiplatelet therapy for the prevention of vein graft occlusion should be performed.