Over the past few months, the question of whether to stent patients with stable CAD has received a great deal of attention following publications on the topic in both Circulation and the Archives of Internal Medicine, which showed conflicting results.
In the Circulation study, New York-based patients with stable CAD who underwent PCI had lower rates of mortality, MI, mortality/MI and revascularization compared with routine medical treatment (RMT), whereas in the Archives meta-analysis of randomized controlled trials (RCTs), initial stent implantation for stable CAD did not lead to reductions in death, nonfatal MI, unplanned revascularization or angina vs. optimal medical therapy (OMT) alone.
With the debate still raging, Cardiology Today Intervention asked Spencer B. King III, MD, MACC, co-author of the Circulation paper, and William E. Boden, MD, principal investigator of the COURAGE trial, to share their thoughts on the pros and cons of stenting patients with this indication.
Hannan EL. Circulation. 2012;125:1870-1879.
Stergiopoulos K. Arch Intern Med. 2012;172:312-319.
Stable CAD, as it has been investigated in trials, is not a condition that carries the high risk that unstable disease does. For that reason, we frequently confuse the benefit of intervention with survival statistics. The two studies referenced above show the usual divergence of randomized trial results of selected patients and registry results of a broader but likely confounded population of patients. Of course, there are patients with stable CAD and extensive ischemia who are at high risk of death, but those have not been extensively investigated. Fortunately, studies are now under way to evaluate that high-risk group with extensive ischemia.
Spencer B. King III
The major benefit of stenting in patients with the kind of stable CAD that has been evaluated in trials is relief of angina and limiting ischemia. To that end, stenting is dramatically successful when the problem is defined by lesions in epicardial coronary arteries which limit blood flow to the myocardium. Patients symptomatic from this condition can gain relief by limiting or reducing oxygen consumption or by restoring nutrient blood flow (ie, stenting). This need not be a hierarchal approach to the problem, first with anti-anginal therapy only and then stenting reserved for those in whom medical therapy fails. Proper informed consent presents the patient with options, and if conditions are right (ie, flow-limiting stenoses which are amenable to stenting with a high chance of eliminating the blood flow deficit), the patient may choose that approach. In situations in which there is no survival benefit to be gained by either approach, the patient should be aware of that fact and make the decision with clear knowledge of what benefit is to be expected. For the right patients, there is no more effective therapy than stenting to eliminate angina and allow enjoyment of unrestricted activities. Even in 14 comparative studies that included some asymptomatic patients, freedom from angina with PCI (angioplasty or stenting) was 70% more likely than with medical therapy alone.
Of course, all this presupposes that the information is available to even consider if stenting is an option. Some would contend that angiography should only be performed in patients without high clinical risk if medical therapy has failed. The assumption is that angiography is not a diagnostic test but a preamble to stenting. Driving this attitude is the widespread use of ad hoc PCI. Despite the convenience of such an approach, performing PCI in the same procedure as a diagnostic angiogram makes true and informed consent very difficult, if not impossible. We should remember that the trials that showed medical therapy alone to be comparable to medical therapy with stenting were all done on patients who had coronary angiograms, and therefore all the information needed for deciding if equipoise between stenting or not stenting was available. This is not the case for stable CAD patients who have not undergone angiography. In these patients, there is no stenting option.
Recently, I was in Rotterdam and saw CT angiography technology that provided a flow map of the coronary circulation in which noninvasive fractional flow reserve could be identified for any segment of the coronary tree. Perhaps, in the future, decisions about the utility of stenting can be obtained noninvasively. For now, however, the angiogram is necessary if stenting is to be considered among the therapeutic choices. To avoid angiography in patients not at high risk removes their freedom of choice. Once this information is obtained, medical therapy alone will be the best choice for some, but for others, restoration of normal perfusion to the myocardium will offer great benefit. Regardless of the course chosen, the patient can then concentrate on the most important goal — secondary prevention of cardiac events.
Wijeysundera HC. Ann Intern Med. 2010;152:370-379.
Disclosure: Dr. King has received honorarium for serving on the data safety monitoring board from Medtronic, Merck, nContact Surgical and Wyeth Pharmaceuticals, and has received equity for serving as a consultant from Celonova Biosciences.
Increasingly, clinicians are faced with discordant evidence from clinical trials and observational studies that complicate decision-making for patients with symptomatic stable CAD. Although we have very convincing data that total or subtotal coronary occlusion following plaque rupture or fissuring in ACS patients is a CV emergency that cannot be optimally managed medically, the clinical benefit and prognostic improvement in both incident MI and death rates with PCI is firmly established.
William E. Boden
By contrast, several RCTs, such as COURAGE, BARI-2D, Japan Stable Angina Pectoris study, as well as numerous meta-analyses of these RCTs, including a very recent one of eight stent trials in the era of modern medical therapy published in the Archives, have failed to demonstrate any incremental clinical benefit for PCI above and beyond OMT alone for the reduction of death or nonfatal MI, hospitalization for ACS, need for unplanned revascularization and a durable, sustained effect on angina relief — findings quite in contrast to those achieved with PCI in acute MI or ACS patients.
More recently, Hannan and colleagues published an analysis of long-term (2-4 years) comparative outcomes in patients with stable CAD who did and did not undergo PCI from the large, prospective New York State PCI Registry. Overall, 9,586 patients were followed prospectively, of whom 89% received PCI, 2% underwent bypass surgery and only 11% received what the authors called “routine medical treatment.” The details of RMT are unfortunately not collected in the registry, so we have no way of knowing how intensive or multifaceted medical therapy was as the comparator in this analysis.
Because so few patients (n=1,100) received RMT alone, the propensity matching comprised only 933 matched pairs of patients (n=1,866), but the study did show that compared with those who received RMT alone, patients who received PCI plus RMT had a significantly lower rate of death or MI (P=.003), mortality (P=.02), MI (P=.007) and subsequent revascularization (P=.005) over the follow-up (median, 2.87 years). The rate of PCI for stable CAD pre-COURAGE (2003-2007) in this analysis was 88.4%, almost identical to the rate observed post-COURAGE (2008; 88.7%). Notably, no differences were reported in comparative outcomes for patients aged younger than 65 years or for patients who underwent PCI for single-vessel CAD between the PCI-plus-RMT and RMT-alone groups.
In all the RCTs comparing the initial management strategy of either revascularization or OMT, diagnostic coronary angiography was first performed to define anatomic coronary stenoses after which patients who exhibited one or more significant flow-limiting stenoses were assigned to revascularization (generally PCI) with OMT vs. OMT alone. Despite knowledge of the anatomy and the relief of angina and ischemia with PCI, every single trial has failed to demonstrate any reduction in short- or long-term “hard” clinical events; only angina relief has been shown — as expected — to be favorably altered and even then studies have failed to show that this is a durable effect beyond 1 to 2 years as compared with OMT alone.
Although it is frequently argued that angina relief is an important therapeutic target, the fact remains that medical therapy is often not utilized first as a viable treatment strategy or, if it is, such medical therapy is not utilized optimally or up-titrated as needed in an attempt to abolish or diminish angina and ischemia. In support of this contention is the large NCDR observational analysis in almost 500,000 stable CAD patients by Borden in which the use of OMT pre- and post-PCI was compared in the pre-COURAGE era with the post-publication of COURAGE. Disappointingly, only about 45% of stable CAD patients who were obviously candidates for OMT were receiving it pre-PCI in both comparative time periods, and only about 65% of patients in both time periods were receiving OMT after PCI. Thus, less than half of patients in whom coronary angiography was performed for symptomatic stable CAD were receiving OMT and only two-thirds were discharged on such treatment post-PCI. Even worse, in the Hannan paper from the NY State Registry, only 11% of stable CAD patients were receiving even routine (much less optimal) medical therapy. The data very convincingly show that we woefully underutilize medical therapy and do not follow our own professional society guidelines that state medical therapy in such patients should be first-line therapy.
Lastly, the controversy of whether or not to use PCI as an initial approach to stable CAD management will likely continue until more definitive data emerge from clinical trials. Earlier this year, the FAME II trial was terminated because, in patients randomized to PCI that was guided by FFR as compared with OMT alone without revascularization, there was a significantly lower rate of hospitalization for ACS and subsequent unplanned, urgent revascularization. At present, we have no information about whether death or MI was favorably altered by this approach vs. OMT alone. In mid-2012, an important, new NHLBI-funded RCT (ISCHEMIA) is scheduled to begin enrollment in 8,000 high-risk stable CAD patients to address whether patients with moderate-severe myocardial ischemia who receive optimal myocardial revascularization combined with OMT will have a lower rate of long-term CV death or MI as compared with OMT alone. Until then, we should base treatment decisions in stable CAD patients with chronic angina on the best available evidence, and should recognize that there are three equally effective management approaches — PCI, CABG and OMT — that can be applied to individualize the best treatment for stable CAD patients.
The BARI-2D Study Group. N Engl J Med. 2009; 360: 2503-2515.
Boden WE. N Engl J Med. 2007; 35: 1503-1516.
Boden WE. Arch Intern Med 2012; 172:319-321.
Borden WB. JAMA 2011; 305: 1882-1889.
Cannon CP. N Engl J Med 2001; 344:1879-1887.
Fox KA. Lancet . 2002; 360:743-751.
FRISC II Investigators. Lancet 1999; 354:708-715.
Katritsis DG. Circulation 2005; 111:2906-2912.
Jeremias A.Am J Med. 2009; 122:152-161.
Keeley EC. Lancet. 2003; 361:13-20.
Kushner FG. J Am Coll Cardiol. 2009; 54:2205-2241.
Mehta SR. JAMA. 2005; 293:2908-2917.
Nishigaki K. J Am Coll Cardiol Intv. 2008. 1:469-479.
Schomig A. J Am Coll Cardiol. 2008; 52:894-904.
Trikalinos TA. Lancet . 2009; 373:911-918.
Wijeysundera HC. Circ Cardiovasc Qual Outcomes 2009; 2:123-126.
Disclosure: Dr. Boden reports no relevant financial disclosures.