May 11th, 2011
Optimal Medical Therapy and the Lack of COURAGE
William Borden, MD and John Spertus, MD, MPH
We welcome William Borden and John Spertus to answer questions from CardioExchange Editor-in-Chief Harlan Krumholz about their JAMA paper showing that optimal medical therapy (OMT) for patients with coronary artery disease (CAD) did not receive a meaningful boost from the publication of the COURAGE trial, despite the trial’s clear message showing the benefits of OMT.
Using data on 467,211 patients in the National Cardiovascular Data Registry to analyze the use of OMT in CAD patients prior to PCI and at the time of discharge, both before and after COURAGE was published, they found:
- Before PCI: OMT was used in 43.5% of patients before COURAGE and 44.7% after COURAGE.
- At discharge after PCI: OMT was used in 63.5% of patients before and 66.0% after COURAGE.
“Collectively, these findings suggest a significant opportunity for improvement and a limited effect of an expensive, highly publicized clinical trial on routine clinical practice,” the authors wrote.
Krumholz: Could the low rate of OMT be a result of patient preference?
Borden and Spertus: In our study, the low rate of OMT is not due to patient preference, since our database documents only those medications prescribed by the physician, not whether or not the patients were actually taking the medicines. Also, if a physician did not prescribe a medication because of patient preference, then that preference could be documented as a contraindication, which would not lower the OMT rate.
Krumholz: Are you sure about the quality of the medication data in the CathPCI registry?
Borden and Spertus: The medication data in the CathPCI registry are reported by the individual institutions according to rigorous data definitions and training. We cannot confidently exclude the possibility that these are not accurately abstracted, although registering patients’ medications on admission is a standard of care and should render these data easy to identify and record. Moreover, there is no reason to think that the quality of the medication data would have changed from the pre-COURAGE to the post-COURAGE period, and finding no meaningful increase in the use of OMT after COURAGE suggests that among the patients undergoing PCI, clinicians are not being substantially more aggressive in attempting OMT to see whether PCI could be avoided.
Krumholz: What do you recommend for someone with a high-risk study who does not want to try medications?
Borden and Spertus: Ultimately we believe that patients are autonomous and can request procedures that are medically reasonable. Based on the 2009 ACC/AHA Appropriateness Criteria for Coronary Revascularization, someone with a high-risk stress test on no medications would be considered appropriate for revascularization. However, we would want to be sure that the patient clearly understood that the procedure is not likely to make him or her live longer or prevent a heart attack. Furthermore, the patient would still be encouraged to take aspirin, thienopyridines, statins, and beta-blockers. The PCI does not obviate the benefit of these medications. The primary benefit of revascularization would be the alleviation of symptoms, which might also be accomplished with aggressive OMT alone.
Krumholz: Why do you think this is inadequate translation of COURAGE? Couldn’t this be a conscious decision by patients and their doctors after taking into account COURAGE?
Borden and Spertus: Our study does not preclude the possibility that some practitioners are following the COURAGE insights with attempting OMT in their stable angina patients and not referring them to PCI, as we only studied those patients who underwent PCI. In fact, a recent publication by Ahmed and colleagues noted a decrease in the proportion of PCI patients with stable angina in Northern New England. However, among those referred to PCI, we did not appreciate substantial increases in the use of OMT prior to treatment. While I doubt that this is a conscious decision of patients, there are a number of potential explanations, including possible clinicians’ disbelief of the COURAGE results, economic incentives to preferentially treat with revascularization, referral patterns, and concerns of interventionalists that if they don’t perform the PCI for a referring doctor that the doctor will not refer future patients. More work is clearly needed to illuminate the opportunities to further improve the use of OMT prior to, and after, PCI.
Krumholz: How would you improve current care?
Borden and Spertus: One key step to improving care may be through better collaboration among caregivers. Primary and interventional cardiologists could work together to ensure that patients, both before and after PCI, are on appropriate medical therapies. Supporting such collaboration and emphasizing medical management should be a priority of ongoing efforts toward improving our healthcare system.
6 Responses to “Optimal Medical Therapy and the Lack of COURAGE”
NEJM Journal Watch — Recent Cardiology Articles- Anticoagulation in Patients with Cancer-Associated Pulmonary Embolism: How Long is Long Enough?
- Spironolactone Use After Percutaneous Intervention for Acute Myocardial Infarction
- CRISPR/Cas9 Gene Editing for Transthyretin Cardiomyopathy
- Does Colchicine Reduce Cardiac Events After Myocardial Infarction?
- First-Line Therapy for Ventricular Tachycardia After MI: What's the Best Approach?
I believe primary care physicians, many, if not most cardiologists, and the public still believe that PCI is better than OMT. We need a large education campaign, and probably a shift in the physicians are reimbursed.
Competing interests pertaining specifically to this post, comment, or both:
I’m Canadian, and I believe this influences my thought processes.
A quantitative/qualitative survey of patients and practitioners beliefs is in order here- to elucidate the (erroneous) concept that a more patent vessel is beneficial or protective of adverse events(death, ACS/MI, hospitalization). And, how concordant or discordant that belief is between the specific patient and assigned practitioner or interventionalist.
But we have to allow for patient personal preferences. Most patients have an internal belief system where medicine is concerned. For instance some of my patients absolutely refuse to ever take a sleeping pill, even Benadryl. So if there are two alternative treatments, even if one has been shown to be superior (and that was only one study at that time) patients many have internal beliefs/wishes that lead them to prefer the other one.
COURAGE is not the only study to show the lack of prognostic benefit of PCI over OMT in stable coronary artery disease. Every trial ever conducted into this subject has had the same result but it doew not seem to alter practice. There are probably multiple reasons for this. People have already mentioned reimbursement issues and fears of losing referrals. I suspect that fear of litigation is another factor as Cardiologists are worried that if PCI is not offered and the patient has an MI it would be easy for a lawyer to claim that adequate treatment was witheld. Also, like most people who have made a big investment of time and effort in learning a skill, some interventionists seem to have a need to feel that that skill must be more beneficial than it really is (NB we are NOT talking about ACS here). There is also the need we have as doctors to “do something” and doing a PCI feels more like doing something than writing a presciption.
The interesting study by Lin et al. (Arch Intern Med. 2007; 167: 1604-1609) suggested a series of reasons behind excessive PCIs, despite the fact that no clear evidence supports such procedures in stable patients with CAD. Beyond the undeniable financial and emotional ties of interventionists and the industry to elective PCI procedures that might play a role in some unnecessary PCIs, one other part of the puzzle could be lack of comprehensive insight of the extent of atherosclerotic disease (which is far beyond what could be checked by coronary luminography) and the role of vulnerable plaques (which are not in general located in severely stenotic segments) by patients and practitioners, alike. In this sense, findings of the COURAGE study were fully concordant with the pathophysiologic nature of the disease; i.e. a local intervention would not bring dramatic results for a disease of systemic nature. Providing thorough educational initiatives for physicians and the public might help to improve the current practice trends.
Competing interests pertaining specifically to this post, comment, or both:
None
A really terrific discussion, and one that makes the issues brought up in Dr Krumholz’s blog regarding POBA vs PCI all the more compelling. The emphasis, in best cardiology practice and the training thereof, is on intervention – usually a hospital-based strategy, where rapid response to symptoms and signs is critical and financially rewarding to both interventionalist and facility. In this setting, institution of “optimal medical management” for patients who have undergone such intervention is mandated by hospital accreditation requirements – hence the higher OMT rate in PCI-d pts pre-COURAGE. . Busy cardiology, internal medicine and family practice offices, where evaluation of OMT-eligible patients is most frequently performed, have no such accreditation requirements, and often lack the support staff and the reimbursed time to educate patients adequately on the safe, effective use of these medicines (and the dietary and exercise issues that attend systemic vasculopathy). Obtaining the recommended medications is not simple for either the prescribing physician or the patient, and the concern about side effects in older eligible patients significant. Linking (advantageous) reimbursement for best practice by insurers to the thoughtful evaluation of outpatients with vascular disease, and required education of trainees that the use of mechanical intervention in vascular disease is a patch that requires long-term pharmacologic and lifestyle retooling of the vascular system might help in addressing these issues. The current patchwork of practice styles, requirements, and incentives does not seem to bode well for widespread implementation of OMT.
Competing interests pertaining specifically to this post, comment, or both:
None relevant to this subject matter.