Direct thrombin inhibitors and factor Xa inhibitors for atrial fibrillation

Abstract

Background

Description of the condition

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia (Wyndham 2000). It has been estimated that there are more than 2.3 million cases in the United States, with an estimated increase of up to 15.9 million persons by 2050. More than 46,000 new cases are diagnosed each year in the United Kingdom (Agarwal 2005; Testai 2010). The prevalence of AF increases with age with an estimated prevalence of 0.5% in the age group 50 to 59 years rising to approximately 9% in individuals older than 70 years. The lifetime risk of developing AF is approximately one in four (Agarwal 2005; Brieger 2009).

The majority of cases of AF, be it paroxysmal or permanent, are ascribed to cardiovascular disorders such ischaemic heart disease, hypertension, cardiac failure and valvular heart abnormalities. Other non-cardiac causes include hyperthyroidism, and only a minority of cases (estimated at 11%) have no identifiable cause (lone AF) (Agarwal 2005). The resultant arrhythmia leads to an increase in blood stasis within the atria. This, in combination with other factors such as an ageing vessel wall and blood component changes, leads to an increased risk in venous thromboemboli formation (Watson 2009). As a result, the main morbidity and mortality associated with atrial fibrillation is in relation to the risk of ischaemic stroke, which is increased five-fold (Hart 2001). However, this risk is thought to vary from one individual to another with the leading risk factors being: previous history of stroke or transient ischaemic attack (TIA), increasing age, hypertension, and structural heart disease in the presence of AF (Hughes 2008). These have led to several clinical prediction rules to estimate the risk of stroke in paroxysmal and permanent AF along with the best option for pharmacological prophylaxis. Of these the CHADS2 risk stratification score was found to have the highest ability to correctly rank-order patients by risk (Hughes 2008).

The mainstay for venous thromboemboli prophylaxis and stroke prevention in AF has thus far been using either a vitamin K antagonist (VKA) such as warfarin or an anti-platelet agent such as aspirin. An earlier systematic review of long term anticoagulants (warfarin) compared with antiplatelet treatment (aspirin) suggested that the included trials (all pre-1989) were too weak to confer any value of long term anticoagulation (Taylor 2001). However a more recent meta-analysis of 28,044 participants showed stroke was reduced by 64% for those on dose-adjusted warfarin and 22% for those on antiplatelet agents. Warfarin in comparison to aspirin leads to a 39% relative risk reduction in stroke (Hart 2007). The decision as to whether a patient receives warfarin or aspirin depends on risk versus benefit. Those at low risk or where warfarin is contraindicated may well be managed on aspirin alone, whereas patients at higher risk may benefit from warfarin. Patients who fall into an intermediate risk category may benefit from either treatment and this decision is largely based on individual risk. Table 1 summarises the criteria for low, intermediate and high risk stratification (Lafuente-Lafuente 2009).

Description of the intervention

The benefits of warfarin therapy in stroke reduction for AF patients are well established. However, these benefits are offset by increased side effects and the need for regular monitoring. The most serious complication for warfarin use is increased haemorrhagic risk. Two meta-analyses have suggested that there is a greater than two-fold increase in the risk of serious major haemorrhagic bleed with warfarin use when compared to placebo or aspirin (Segal 2001; Hart 2007). This risk is increased when warfarin and aspirin are combined without any benefit in stroke prevention (Flaker 2006).

Another significant problem with warfarin use is its narrow therapeutic window. To prevent under and over anticoagulation, patients on warfarin require regular monitoring of their international normalised ratio (INR). Most guidelines suggest patients on warfarin for AF should have an INR of between 2 and 3 (Lip 2007). Sub-optimal levels are associated with a greater risk of complications. One study looked at mortality within 30 days of admission to hospital with stroke. Among patients taking warfarin at the time of the stroke, 16% of those with an INR <2 died within 30 days compared to 6% with INR >2 (Hylek 2003). The same study also showed that increased haemorraghic risk was associated with an INR >4. Tight INR control requires regular monitoring and is thought to be one of the contributing factors to poor adherence to warfarin. A prospective cohort study of patients presenting to secondary care with AF found 56% of patients on anticoagulation treatment did not adhere to international guidelines. Reasons for this were thought to be due to poor understanding of treatment, logistics of regular monitoring and reluctance of physicians to correctly prescribe warfarin for fear of potential drug interactions and complications (Mehta 2004).

Several alternatives to warfarin have emerged over the past ten years. These include direct thrombin inhibitors (DTIs) and factor Xa inhibitors. These newer drugs have the potential for several advantages over traditional VKAs. For example, they do not require regular monitoring, have faster onset of action, and potentially fewer adverse interactions. However, what is not clear yet is how efficacious they are or the associated risks of adverse events (Verheugt 2010).

Clinical trials of parenteral DTIs, such as hirudin, argatroban, and bivalirudin have been evaluated in acute settings such as percutaneous coronary intervention (PCI) and acute coronary syndrome (ACS) with mixed results (Hirsh 2005). The two oral DTIs furthest into clinical trials are ximelagatran and dabigatran. Trials of ximelagatran in preventing venous thromboemboli have shown superiority over placebo without increased risks of bleeding (Schulman 2003). The Stroke Prevention Using Oral Thrombin Inhibitor in Atrial Fibrillation (SPORTIF) III and SPORTIF V trials concluded ximelagatran was non-inferior to warfarin in preventing stroke with no increase in bleeding events (Olsson 2003; Albers 2005). However, serious concerns of hepatotoxicity have resulted in ximelagatran being withdrawn from the market (Kaul 2005).

Early trials of dabigatran have shown promise. In the Randomized Evaluation of Long Term Anticoagulation Therapy (RE-LY) trial, oral dabigatran, when given at a dose of 110 mg, was found to be associated with rates of stroke and systemic embolism that were similar to those associated with warfarin. At a dose of 150 mg, compared with warfarin, dabigatran was associated with lower rates of stroke and systemic embolism but similar rates of major haemorrhage (Connolly 2009).

Factor Xa inhibitors currently include idraparinux, apixaban, rivaroxaban, and edoxaban. The AMADEUS trial compared once weekly subcutaneous injections of idraparinux with oral warfarin (Amadeus Investigators 2008). While idraparinux was non-inferior to warfarin, there were significantly higher rates of bleeding. A biotinylated version entered phase III clinical trials but the trial was terminated early by the manufacturer. A number of clinical trials for oral factor Xa inhibitors are currently underway. The ROCKET AF study is a non-inferiority study comparing rivaroxaban to warfarin in atrial fibrillation patients (Investigators 2010). A recent randomised controlled trial looked at the effect of the oral factor Xa inhibitor apixaban against aspirin in those patients unsuitable for warfarin (Connolly 2011). This trial was stopped early as a result of the clear benefit of apixaban over aspirin in reducing stroke and adverse bleeding events. Apixaban also appeared superior when compared to warfarin (Granger 2011). Other direct thrombin and factor Xa inhibitors involved in clinical trials include AZD0837 (Lip 2009) and YM150 (Astellas 2007). Table 2 summarises the main direct thrombin and factor Xa inhibitors currently being investigated for thromboembolic prevention in atrial fibrillation.

How the intervention might work

Both the intrinsic and extrinsic coagulation pathways result in fibrin activation. Directly before this step is the conversion of prothrombin to thrombin which in turn is dependent upon activation of factor Xa. Warfarin interrupts this cascade indirectly through the inhibition of vitamin K dependent factors II, VII, IX, and X. In contrast, direct thrombin inhibitors (DTIs) bind to and inhibit thrombin, which is the most potent platelet agonist. This also has the advantage of preventing feedback activation of factors V, VIII, and XI. Inhibitors of factor Xa inhibit the formation of thrombin by binding directly to its precursor (Eriksson 2011).

Why it is important to do this review

Current management of anticoagulation for reducing stroke risk in AF patients involves a clinical decision of risk versus benefit in deciding who should receive and which type of anticoagulation to use. Current guidelines advocate the use of either aspirin or warfarin. While warfarin has shown clear superiority over aspirin in risk reduction, it may not be suitable for use in some patients. Long term use of warfarin requires regular monitoring through blood tests. This may be less suitable for patients with poor mobility, or in those who are housebound or have poor access to regular means for blood testing within the community. In addition, many of the patients being considered for warfarin are likely to be on other medications, and there are established risks of interactions with other drugs and subsequent effects on INR. Most importantly warfarin use is associated with higher rates of both minor and major bleeding events.

Newer anticoagulation drugs work along the coagulation pathways through, for example, direct inhibition of thrombin or other factors such as factor Xa. These newer drugs have the potential benefits over warfarin that they require no monitoring, have activity and inactivity through a shorter time frame than warfarin, and less potential of drug interactions. One study also found that dabigatran was as cost effective as warfarin although it is unclear if this is the case for other newer inhibitors (Patel 2010). However, there are concerns over the safety profile and risk of adverse events for some of these drugs. In addition, it remains unclear how these drugs could be monitored or rapidly reversed in the scenario of overdosing. Some drug trials have been stopped early and their results have not been published for unclear reasons that have further raised concerns over the safety profile of these newer drugs.

At least two other reviews (Aguilar 2007; Bruins 2011) appear in the Cochrane library that suggest some overlap with this one. However in the Aguilar 2007 review the authors compare only warfarin versus antiplatelet agents for stroke prevention in non-valvular AF, and the protocol from Bruins 2011 only evaluates Factor Xa inhibitors against VKA for AF. Therefore a systematic review is needed to assess the effect of both direct thrombin or factor Xa inhibitors in the prevention of stroke in AF.

Objectives

To assess the effectiveness of direct thrombin inhibitors and factor Xa inhibitors on clinical outcomes in patients with AF.