From interventional cardiologist Lloyd W. Klein, a 'balanced critique'
Percutaneous coronary intervention (PCI) is a highly effective therapy for symptomatic patients with coronary artery disease (CAD) and stable angina, especially when a single or few culprit lesions can be identified and treated with low risk. Additionally, PCI has been proven to reduce short and long-term mortality in patients with acute ischaemic syndromes, including myocardial infarction(MI) and cardiogenic shock.
In the Clinical Outcomes Utilizing Revascularization and Aggressive drug Evaluation (COURAGE) trial, the long-term outcomes of patients with stable angina treated initially with PCI combined with optimal medical care were no different from optimal medical care alone. The study showed, in my view definitively, that an initial strategy of PCI does not reduce the risk of death, MI or other major cardiovascular events and may not even significantly reduce angina when compared to an initial strategy of intensive medication in selected patients.
These findings have been greeted by the interventional community with a combination of shock and surprise, followed by anger directed at some cardiology societies and their leaders for interpreting the study results beyond the very strict and scientifically accurate conclusions which Drs William Boden and William Weintraub, the study’s designers and principal investigators, were careful to draw. A highly politically charged environment now exists in which many interventionists believe our field is under attack.
I think it is unfair to lay all of this at the feet of the COURAGE trial, which asked an important question and attempted to answer it accurately in a fair and reasonable manner.
The study outcome was not particularly surprising; cardiologists should have recognised, based on many previous studies, that PCI does not prevent MI or death in patients with stable angina. Questions of clinical judgment concerning when to proceed with PCI when angina continues at an unacceptable frequency or workload, or in the setting of prior MI when another infarct may be life-threatening or changing, remain open and were not directly addressed by COURAGE.
I believe some of my colleagues were performing PCI as an initial strategy when medical therapy could have been just as successful. COURAGE may give all of us the courage to counsel some patients, and teach many impressionable cardiology trainees, that modern medical therapy is a highly effective strategy in such patients and ought to be advised in a positive manner, not as a message of sub-optimal, non-aggressive, noncaring, cheap medical care. Herein lies the essential point of the trial, which, in a cooler, less politically charged atmosphere, would and should be accepted by all physicians.
That said, COURAGE has some important faults, which must be considered in applying its findings to individual patients. The trial was conducted primarily in US Veterans Hospitals and Canadian centres, where patients might be more likely to accept angina at lower workloads. The level of medical therapy given, and the intensity of the follow-up provided, was well beyond that provided by usual medical care in the community. The trial included just a small subgroup of highly selected CAD patients, excluding the majority of patients who are treated with PCI in contemporary practice. We do not know whether especially high-risk patients were enrolled in enough numbers to alter our advice to them based on this study, nor how many patients were approached and rejected enrolment, or how their physicians reacted, any of which could have led to serious bias.
The right message
My main concern is that in the rush to grab media, professional and public attention, it has been made to appear that having PCI performed under any circumstance is unnecessary. What should have been the message is the importance of careful patient selection for any strategic decision and the need to involve patients in deciding among treatment options.
PCI remains a proven treatment option for those whose quality of life (angina or activity level) is not improved by medical therapy, or for those patients who do not tolerate the side effects of medical therapy; the scores for physical limitation, angina frequency, and quality of life were significantly better for patients in the PCI group: one-third of the patients in the medications-only group switched over to PCI during the course of the trial because their angina was not sufficiently relieved by medication alone.
A better message for the public and medical community would have been that, although some low-risk patients who could reduce their pain with medication and lifestyle changes are treated with PCI, this is a decision that may still be appropriate after a candid discussion with their physician.
From COURAGE study principal investigator William S. Weintraub: 'conservative management'
THE CLINICAL Outcomes Utilizing Revascularization and Aggressive drug Evaluation (COURAGE) Trial showed that, as an initial management strategy in patients with stable coronary artery disease (CAD), percutaneous coronary intervention (PCI) did not reduce deaths, myocardial infarctions (MI), or other major cardiovascular events when added to optimal medical therapy.
Advances in our understanding of the pathophysiologic basis for acute coronary syndromes and the important role that plaque rupture or fissure plays in the genesis of MI clearly indicate that non-flow-limiting coronary stenoses are the most common progenitors of most “hard” clinical events. Furthermore, there are trial data demonstrating that urgent/emergent PCI in patients with STsegment elevation myocardial infarction (STEMI) or high-risk non-STEMI derives a prognostically-important reduction in death or subsequent MI. The COURAGE trial does not challenge this important therapeutic premise.
Distinction from ACS
PCI performed electively in patients with chronic angina and stable CAD is virtually identical technically to that performed in acute coronary syndrome (ACS) patients. This may lead to the belief that PCI would confer a more durable clinical benefit (ie, beyond mere angina relief or improved exercise performance) in this population of patients as well. However, prior to the COURAGE trial, this had not been tested with coronary stents in patients where optimum medical therapy was used in both arms. A critical design feature of COURAGE from its inception was the adoption of an aggressive LDL-C target of 60-85 mg/dl, as well as the establishment of secondary lipid targets for HDL-C >40 mg/dl and triglycerides <150 mg/dl, in the PCI group as well as the medical therapy group.
Death or MI as the primary outcome measure during long-term follow-up was chosen as the most rigorous test of efficacy in COURAGE, since such a composite endpoint was also the benchmark used in the earlier coronary artery bypass graft (CABG) surgery versus medical therapy trials of the 1970s and 1980s, as well as trials comparing PCI to CABG.
Another strength of the COURAGE trial relates to the MI definition as part of the composite primary endpoint. There was a fair and balanced three-component definition for MI with spontaneous MI requiring an MB-CK elevation 1.5 times the upper normal limit (UNL) or a troponin two times the UNL, whereas a peri-PCI MI required an MB-CK or troponin value three times the UNL or five times the UNL respectively, while for perioperative MI post-CABG, an MB-CK or troponin elevation ten times the UNL was required. Furthermore, in all cases of suspected MI, clinical evidence of an acute ischaemic syndrome presentation together with the above abnormal biomarkers was required to qualify as a “COURAGE MI”.
"The COURAGE results do not indicate
that PCI is inherently worthless...or inappropriate"
An additional design strength of COURAGE was that it was undertaken as a large US and Canadian multicentre clinical trial and it aimed to enrol heart disease patients from a diversity of geographic, ethnic, clinical, and health care system conditions so as to be broadly representative of contemporary clinical practice; however, a low percentage of women (15%) and non-Caucasian (14%) patients was a trial limitation.
A particularly important design feature of COURAGE was that prospective health status assessment (using the Seattle Angina Questionnaire [SAQ] and SF-36 instruments) was embedded in the trial and represents the most comprehensive approach to quantifying patient-centred outcomes ever undertaken in a large strategy trial of CHD management.
With a sample size of 2270, there was 85% power to detect a 22% relative difference in event rate from 21 to 16.4%. The COURAGE results do not indicate that PCI is inherently worthless or that PCI is ineffective or inappropriate as an initial management strategy in stable CAD patients; it means that PCI is not the only viable and clinically-defendable initial strategy and that optimum medical therapy alone may be suitable and appropriate for many patients (and physicians) who may decide to defer PCI until after they have made an assessment of the efficacy of optimum medical therapy.
CAD is a systemic problem that requires systemic treatment. Flowlimiting lesions cause angina and ischaemia but may not necessarily be the lesions predisposing to death, MI, and ACS. The COURAGE results support the ACC/AHA clinical practice guidelines that optimum medical therapy should be considered an appropriate and preferred first approach in stable CAD patients with mild and moderate symptoms, while PCI should be reserved for patients with severe symptoms or those who become unstable or remain symptomatic on optimum medical therapy.