To bleed or not to bleed. Is that the question?
Aspirin was recognized as the first antiplatelet agent some 70 years after its discovery , but there was little perceived need for drugs with this action. Interest in inhibiting platelet aggregation was driven by an evolving recognition of the role of platelets in the development and progression of atherothrombosis . Even so, some 20 years after aspirin was shown to have an antiplatelet action the British National Formulary (BNF) in 1988  stated that ‘Clinical trials of aspirin in the prevention of myocardial infarction and of transient brain ischaemic attacks have been reported. The results are equivocal.’ The only alternative to aspirin was dipyridamole, which had even fewer uses. Sulphinpyrazone had been removed as an antiplatelet drug in 1986.
The revolution in the clinical use of antiplatelet agents would not begin until the publication of the ISIS-2 study in 1989 which showed impressive reductions in mortality with aspirin after acute myocardial infarction . The BNF in 1989 acknowledged this with the statement ‘Encouraging results have been obtained using aspirin 300 mg daily for the secondary prevention of cerebrovascular and cardiovascular disease; studies are still needed to determine whether lower doses (such as 75 mg daily or 300 mg on alternate days) might not be equally effective. Aspirin has also been shown to reduce mortality when given in a dose of 150 mg daily for 1 month after myocardial infarction’.
Progress in the subsequent 20 years has been dramatic. In the 1990s, the use of aspirin was extended to other presentations of vascular disease, while a modest benefit was shown in reducing thromboembolism in atrial fibrillation. The major disadvantage of aspirin is dose-related gastrointestinal irritation and bleeding. The meta-analyses of trials with a wide range of doses of aspirin was important in showing that small doses are as effective as large doses for the prevention of thromboembolic events [6, 7]. The use of low dose aspirin has reduced the risk of gastrointestinal irritation, but it remains at least double that with placebo.
Aspirin reduces but does not eliminate platelet activation, and the multiple pathways of platelet activation provide obvious targets for new drug therapies. Recognition of glycoprotein (GP)IIb/IIIa receptors on the surface of the platelet as the final common pathway of platelet activation then paved the way for drugs that could eliminate platelet aggregation altogether.
Exploiting our growing understanding of the mechanisms that underpin platelet activation and aggregation initially spawned ADP receptor antagonists and GPIIb/IIIa receptor antagonists. This precipitated an explosion in studies that have looked at their use as an adjunct to aspirin, in part driven by more intensive and interventional strategies for the management of coronary artery disease. As ‘plain old balloon angioplasty’ was superseded by the introduction of bare metal and then drug-eluting stents, attention turned to reducing the risk of instent thrombosis. The clinical consequences of acute stent occlusion were so serious, that combinations of antiplatelet drugs have been embraced enthusiastically as the best way of minimizing the risk.
An important, and as yet unresolved, issue is ‘resistance’ to antiplatelet drugs in in vitro tests of platelet aggregation. The incidence of resistance to aspirin may be 30%, and even higher in some disease states. There is evidence that failure to inhibit platelet activation, measured by in vitro platelet aggregation, with aspirin or clopidogrel may translate into a less favourable clinical outcome. However, interpretation of the studies is complicated by the lack of a standardized test for measuring platelet activity. Some newer antiplatelet drugs appear to be effective in a higher proportion of patients and may avoid this problem.
Until the 1990s, inhibition of the coagulation system could be achieved with parenteral unfractionated heparin (introduced into medical practice in the 1930s) or the oral vitamin K antagonists (4-hydroxycoumarin anticoagulants and phenindione) that were first used as medicines in the 1940s. Both classes of drug have actions at multiple points in the coagulation cascade and are inconvenient to use.
Twenty years ago, anticoagulation was largely confined to treatment of established venous thromboembolic disease, with lesser indications for prevention of thrombus formation in extracorporeal circulations. Interest slowly developed in the use of subcutaneous unfractionated heparin for prevention of venous thrombosis in the peri-operative period, although the evidence was initially confined to a limited number of surgical procedures and adoption into clinical practice was slow. Heparin was also given after thrombolysis with tissue plasminogen activator for acute myocardial infarction to reduce re-occlusion of the culprit coronary artery. Then, in the 1990s, warfarin was shown to be three times more effective than aspirin at preventing stroke in people with atrial fibrillation . This raised the prospect of large numbers of elderly people taking warfarin for prevention rather than treatment of arterial thromboembolism.
Unfractionated heparin is usually given by continuous intravenous infusion and requires regular monitoring of the activated partial thromboplastin time with frequent dose adjustment to ensure adequate anticoagulation and to avoid excessive dosage. Therefore, low molecular weight (LMW) heparins were a major advance in parenteral anticoagulation when they were introduced into clinical practice at the start of the 1990s. They offer standardized doses, subcutaneous administration and do not need monitoring in most situations. In the UK, use of LMW heparin for prevention of deep vein thrombosis in hospitalized patients has been a major thrust of national health policy. The place of heparin in therapy seemed assured, but the arrival of oral anticoagulants with a rapid onset of action threatens to dislodge parenteral anticoagulants from their perch.
Warfarin (and the other vitamin K antagonists) is also an inconvenient drug to use, both for patients and their doctors. The wide interindividual variation of effective doses, slow onset of action, saturable metabolism, narrow therapeutic index, drug and food interactions and the need for regular monitoring (with the International Normalized Ratio or INR) combine to make it difficult to maintain therapeutic anticoagulation. Several studies have shown that in clinical practice patients have INRs outside the target range for at least 30% of the time. The INR is more often sub-therapeutic, but almost half of the INRs outside the therapeutic range are high and expose the patient to the risk of haemorrhage [9, 10].
Several new oral anticoagulants that act at specific points in the coagulation cascade are now available. They have a faster onset of action than warfarin, predictable anticoagulant effect from standardized doses, few drug interactions and, in common with LMW heparin, do not require monitoring. These drugs are clearly more convenient than warfarin, but the latter is a cheap medicine, with additional costs associated with monitoring that will be difficult to extract in the short term. The major barrier to replacement of vitamin K antagonists in clinical practice will probably be cost.
Combinations of drugs that inhibit complementary platelet activation pathways are increasingly used (such as aspirin with clopidogrel for acute coronary syndromes and vascular stenting or aspirin with dipyridamole for transient cerebral ischaemic attacks and sometimes following ischaemic stroke). In addition, regimens that involve inhibition of both platelets and coagulation are becoming more common and these have raised concerns about significant increases in bleeding complications. An anticoagulant may be prescribed for another indication, such as atrial fibrillation in someone with a drug-eluting coronary stent.
Is bleeding the Achilles heel of the new generation of drugs?
The major complication of all antiplatelet drugs and anticoagulants is bleeding. Of greatest concern is intracerebral haemorrhage, which carries a high morbidity and mortality. However, gastrointestinal blood loss and minor bleeding or bruising are all far more common. Many new drugs that affect haemostasis have greater efficacy than established drugs in preventing thromboembolic events, but does this come at the price of more bleeding? While unfractionated heparin and vitamin K antagonists have antidotes that can reverse their effect (protamine and vitamin K respectively), the newer drugs do not (such as Dabigatran). There is emerging evidence with certain new agents of non-inferiority to older drugs in preventing and treating thrombosis, but with a lower bleeding risk. Other new drugs (or Dabigatran at higher doses) confer greater efficacy but similar rates of bleeding to warfarin and heparin. Regulators are now debating the merits of greater efficacy against bleeding risk when choosing doses to license.
Combinations of drugs that affect haemostasis have highlighted the dangers of interfering with multiple steps in these processes, and the need for strategies to manage bleeding. ‘Triple’ antiplatelet and anticoagulant therapy carries a particularly high bleeding risk, with a three to five fold increase in bleeding compared with dual antiplatelet therapy. Up to 21% of patients taking ‘triple’ therapy require a blood transfusion, usually for gastrointestinal haemorrhage . The importance of optimizing risks and benefits in these situations cannot be over emphasized.
The BJCP themed issue
The articles in this themed issue explore the array of new and established drugs (and the potential roles of those under development) for inhibition of platelet function and anticoagulation. The important questions of antiplatelet drug resistance and dealing with bleeding during use of the new anticoagulant drugs are considered.
One thing is certain, that the options for both antiplatelet therapy and anticoagulation will be very different in the next few years as new drugs jostle for niches in both the prevention and treatment of thromboembolic disease.