COURAGE or FAME…? Who should have percutaneous coronary intervention in stable coronary artery disease?
- Correspondence to Professor Keith A A Fox, Department of Cardiology, Centre for Cardiovascular Science, University of Edinburgh, Chancellor's Building, 49 Little France Crescent, Edinburgh EH16 4SB, UK;
A patient has just been diagnosed with stable coronary artery disease (CAD) and turns to the doctor and asks ‘Do I need a stent for this doctor?’ The answer, based on current evidence, is not simple.
First, there is poor appreciation among clinicians and among patients about the relatively benign prognosis in stable CAD. For example, in the 3825 patients with stable symptomatic CAD in the A Coronary Disease Trial Investigating Outcome with Nifedipine (ACTION) trial there were 1.5 deaths per 100 patient-years at risk and 1.4 myocardial infarctions (MI).1 Furthermore, the use of statins (62%), ACE inhibitors or angiotensin II receptor blockers (ARB) and other secondary prevention measures did not achieve the levels attained in the subsequent COURAGE trial (95% on statins at 1 year, 95% on aspirin, 89% on beta blockers, 72% on ACE or ARB).2 However, the overall figure masks a 10-fold range in risk, determined by concomitant risk factors including the extent of coronary disease3 (figure 1). So, the first response to the patient is reassuring; unless there are additional risk factors, the risks of death are relatively low and are reduced by secondary prevention treatments and lifestyle measures including smoking cessation.
The next question from the patient is ‘Okay, I will take the treatments you prescribe for preventing complications and do my best to adopt the lifestyle changes, but would I be better off with stent therapy?’ If the patient is symptomatic and is not controlled or is inadequately controlled on anti-anginal medications (because of troublesome angina or side effects) then it is entirely appropriate to consider revascularisation for symptom relief.4 ,5
The tenacious patient asks a third question: ‘But will I live longer and without a heart attack if I have stent treatment?’ The answer to this is unknown and the only sufficiently large study to address this question is the National Institutes for Health-funded ISCHAEMIA trial.6 This has started but will take several years to complete. Nevertheless, perceptions by patients (and some doctors) are different, with some believing that the stent procedure was performed to prevent MI and death.7 ,8
The tenacious and exceptionally well informed patient has been searching on the internet and asks ‘the COURAGE2 and BARI 2D9 trials tell me there is no advantage for intervention therapy but FAME 210 says there is—what should I believe?’ Based on the COURAGE and BARI 2D trials there was no difference in the rates of death or other major cardiovascular events between patients undergoing initial revascularisation and those undergoing medical therapy.2 ,9 In the COURAGE trial coronary artery anatomy was defined and patients needed to have a stenosis of at least 70% in one or more artery to be included. There was no difference, overall, in the rates of cardiovascular events (eg, death or non-fatal MI) when added to optimal medical therapy.2 However, percutaneous coronary intervention (PCI) did reduce symptoms of angina. It is possible that some patients with more severe lesions were not randomly assigned in the trial.2 ,11
A small nuclear substudy raised the possibility that those with a larger ischaemic volume (>10%) may benefit from the revascularisation strategy.12
The SYNTAX trial compared coronary artery bypass grafting (CABG) with PCI in patients with three-vessel or left main CAD, and concluded that CABG remains the standard of care for patients with left main or three-vessel coronary disease because of lower rates of major adverse cardiac events at 1 year and beyond.13 ,14 The benefits of CABG were most clear in those with high syntax scores (>33).13
So how do we interpret the FAME 1 and FAME 2 studies? In the FAME 1 study 1005 patients with multivessel CAD were randomly assigned to undergo PCI with implantation of drug-eluting stents guided either by angiography alone or guided by fractional flow reserve (FFR) measurements in addition to angiography.15 FFR provides an objective measure of the haemodynamic impact of a stenosis. The study found that the strategy of routine measurement of FFR in patients with multivessel coronary disease undergoing PCI and stent implantation significantly reduced the rate of death, non-fatal MI and repeat revascularisation.15 The composite of death or MI was also reduced using the FFR-guided strategy (relative risk 0.66, 95% CI 0.44 to 0.98).15 There were also strong trends for greater freedom from angina. In addition to the clinical impact, health economic analyses indicate that this strategy could be cost saving by reducing the inappropriate use of PCI. Therefore, the development of FFR is an important advance in guiding revascularisation and an advance on approaches that rely on measuring or estimating the severity of angiographic stenosis.
The FAME 2 trial set out to test whether composite outcomes were superior using a PCI strategy in patients with CAD and severe stenosis (FFR <0.80) versus a medical strategy in those with similarly defined lesions.10 Those with less severe stenosis (FFR >0.80) were not part of the randomisation but were followed separately. The composite endpoint of the trial was lower with the PCI strategy. However, the composite endpoint of death, MI or revascularisation was driven almost entirely by revascularisation.10 Therefore, what was shown was that having defined high-grade stenosis, with both the cardiologist and the patient being aware that there was high-grade stenosis, was more likely to lead to further revascularisation than a strategy with initial revascularisation. There was no impact on death or MI.10 Importantly, if the design of the COURAGE and the BARI 2D trials had included revascularisation as part of a composite endpoint, then those trials would have become ‘positive’ because of more revascularisation in the conservatively managed patients. However, there was neither an impact on death or on MI in those studies. Therefore, unfortunately although FAME 2 is informative, it does not answer the question of the impact of PCI compared with medical therapy alone on death or on MI (it is insufficiently powered to do so). Another way of interpreting the data is that ‘96%’ of patients with stable angina, despite having high-grade stenoses, do well and are free of death or MI after 1 year, irrespective of the PCI or no initial PCI strategy.10
So how do we respond to our tenacious patient? To guide investigation and management of the patient with stable ischaemic heart disease three issues need to be considered (figure 2):
▸ Symptomatic status and response to anti-anginal therapy (including the patient's informed wishes);
▸ The extent and severity of myocardial ischaemia.
In all patients emphasis must be placed on reducing risk with evidence-based therapies including statins, ACE inhibitors/ARB, lifestyle changes and, critically, smoking cessation. Persistence with therapy is a challenge.16 It is important to remember that chest pain of new or recent onset is within the spectrum of acute coronary syndrome, rather than chronic stable angina, and should be managed differently.17
In the patient with severe or disabling symptoms of chronic angina, despite medical therapy, functional or other non-invasive imaging is not indicated, and the patient should proceed to angiography (figure 2). For those with three-vessel or left main disease evidence supports coronary bypass surgery and this should be discussed with the patient and the heart team.18 For those with severe stenoses (FFR <0.8) a PCI strategy is appropriate if technically feasible. For those with FFR greater than 0.8 perfusion imaging can establish whether reversible ischaemia is present. Some patients without haemodynamically significant proximal lesions nevertheless have an ischaemic burden due to more diffuse or distal CAD, and some may have spasm superimposed on minor disease or even arteries with no visible lesions.
For the symptomatic patient without disabling symptoms the first step is to optimise medical therapy. If there is a high probability of CAD based on risk factors, (high ‘prior probability’) then proceeding to angiography is an appropriate next step.18 In those with an intermediate probability of coronary disease, functional imaging can define the extent and distribution of ischaemia, and angiography is indicated in those with clearly defined ischaemia.18 In those without ischaemia, alternative diagnoses need to be considered.
Asymptomatic patients with known or suspected CAD can be categorised based on their risk factors and probability of underlying CAD. In those with a low probability, a CT calcium score can be used to rule out significant coronary atheroma.4 ,18 ,19 Those with a high probability of underlying CAD should proceed to functional imaging.4 Medical therapy, with review, is appropriate for the intermediate group.
Insufficient attention is paid to persistence with evidence-based secondary prevention measures, lifestyle modification and smoking cessation. These measures have the potential to impact substantially on long-term outcome. So for our tenacious patient we can explain that we now have better evidence that stent therapy will improve symptoms and reduce the likelihood of needing a subsequent interventional procedure, but we still do not know whether this therapy improves survival and the risk of MI.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.