Evidence-Based Use of Cardiac Procedures and Devices
http://www.100md.com
《新英格兰医药杂志》
The fundamental principle of evidence-based medicine is that clinical practice should rest on a sound scientific foundation established by clinical studies involving human subjects. The level of evidence needed for the adoption of new therapies should vary according to the clinical situation, and the higher the stakes, the better the evidence should be. The stakes grow higher in proportion to the size of the affected population, the severity of illness, and the cost of the therapy.
Several therapies in cardiovascular medicine are backed by substantial evidence of high quality. For example, the effects of statin therapy on mortality and the risk of myocardial infarction have been evaluated in 10 large randomized, clinical trials involving more than 70,000 patients. This extensive experience in multiple studies has demonstrated a consistent reduction in the rate of cardiac events. Even more important than its demonstration that statins "work" is the fact that this body of evidence is large enough to permit assessment of the benefits and risks of treatment in various segments of the target population. We are thus able to judge which groups of patients benefit sufficiently from statins to justify the use of this treatment.
The evidence regarding procedures and devices is much less extensive than the evidence regarding most drugs, despite the high stakes associated with these treatments. Randomly assigning patients to procedures is more difficult than randomly assigning them to drugs, and double-blind trials are practically impossible. Only a few specialists are capable of performing procedures and implanting devices, and they may be proponents of the therapy under consideration and therefore unwilling to support randomized trials. Moreover, the regulatory requirements for devices are not as strict as those for drugs, and there is no specific regulation at all for new surgical procedures. Because there are greater difficulties and fewer incentives to conduct clinical trials, the evidence base with regard to devices and procedures is only a fraction of the size of that available for many drugs.
Coronary bypass surgery is the quintessential major cardiovascular procedure; in 2001, more than 300,000 patients underwent such surgery in the United States alone, at a cost of at least $6 billion. Because the stakes are so high for bypass surgery, we should have solid evidence from randomized trials to guide practice. Yet only 2649 patients worldwide have been randomly assigned to bypass surgery or medical therapy in the comparative trials whose results have been published to date1; the largest of these trials included 780 subjects and completed enrollment 25 years ago. Clinical trials of bypass surgery were large enough to document that the treatment reduces mortality among selected patients, but not large enough to show clearly how the degree of benefit from surgery varies according to common clinical indicators. Many patients who are considered as candidates for bypass surgery today do not match the profiles of patients who were enrolled in older trials, so the generalizability of the evidence is uncertain. The relatively thin base of evidence from randomized trials has contributed to persistent concern about the excessive and inappropriate use of bypass surgery in some groups of patients.
Despite uncertainty about who benefits from coronary bypass surgery, the procedure has been embraced widely. The growth of surgery has been fueled by anxieties about sudden death from cardiac causes and patients' wishes for a seemingly definitive procedure. Coronary bypass is a complex operation that requires hospitals to develop large and sophisticated surgical programs to deliver care, and there are substantial economic incentives to increasing the volume of procedures performed. A 1977 editorial in the Journal2 suggested that bypass surgery was "at the crossroads" and might well develop into an industry, given its perceived improvement of survival, the large population of potentially eligible patients, and the strong economic incentives. These comments were prescient: between 1977 and 1997, the number of patients undergoing coronary bypass surgery annually in the United States increased from 82,000 to 366,000.
The implantable cardioverter–defibrillator is now at a similar crossroads. These devices are expensive (more than $20,000 apiece), and in 2000, only 34,000 implantations were performed in the United States. Early clinical trials showed that the implantation of defibrillators reduced mortality in the small population of patients who had survived a spontaneous episode of ventricular fibrillation or sustained ventricular tachycardia.3 The stakes have been raised by proposals that the use of these devices be expanded to the treatment of the much larger population of lower-risk patients with underlying heart disease who have never had a cardiac arrest. The implantable defibrillator has enormous intuitive appeal as a lifesaving device for this population; a cadre of well-trained physicians is capable of implanting it; and a sophisticated medical-device industry is poised to promote its widespread use.
Is the available evidence sufficient to support the expansion of the indications for these devices? By the end of 2003, five trials of prophylactic defibrillator implantation had been published, but their results varied significantly.3 Two new trials are reported in this issue of the Journal (by Bristow et al., pages 2140–2150, and Kadish et al., pages 2151–2158), and the results of two additional trials were announced at the meeting of the American College of Cardiology held in New Orleans in March 2004. The results from all nine trials and 6200 patients have yet to be fully presented and assessed, but the emerging evidence suggests that the prophylactic implantation of defibrillators reduces mortality among patients with a left ventricular ejection fraction of 30 percent or less. The release of these new data will increase the pressure on large payers (especially Medicare) to expand coverage for these devices, and it is very likely that the rate of implantation will increase dramatically, propelled by the same forces that fueled the expansion of bypass surgery.
As with bypass surgery, the key question about the prophylactic use of defibrillators is not just whether it "works" overall, but how much the risks, benefits, and costs vary among the different subgroups of patients with heart disease. Variation in outcomes among different groups of patients is typical of many therapies and is likely in this instance, given the heterogeneity of the results of the early trials.3 Some patients may not reap enough benefit to justify the risk and cost of prophylactic defibrillator implantation, but we do not yet know how best to identify them. Furthermore, in the face of uncertainty, many physicians would rather err on the side of caution by putting in a defibrillator. There is a need to bring together the emerging evidence to determine who obtains sufficient benefit from a prophylactically implanted defibrillator to justify its use. Ideally, the clinical-trial investigators will pool their data to determine how the effectiveness of defibrillator treatment varies according to key clinical factors such as ejection fraction, presence or absence of a history of coronary revascularization, and time since myocardial infarction. Pooling trial data offers the best opportunity to establish the efficacy and cost effectiveness of these devices in various segments of the large population of patients who would potentially be eligible for this expensive therapy.
The enormous growth of coronary bypass surgery occurred before the advent of evidence-based medicine. It remains to be seen whether there will now be sufficient demand for solid evidence of efficacy and cost effectiveness to promote the appropriate use of the implantable defibrillator and other new devices and procedures.
Source Information
From the Donald W. Reynolds Cardiovascular Clinical Research Center, Stanford University School of Medicine, Stanford, Calif.
References
Yusuf S, Zucker D, Peduzzi P, et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994;344:563-570.
Braunwald E. Coronary-artery surgery at the crossroads. N Engl J Med 1977;297:661-663.
Ezekowitz JA, Armstrong PW, McAlister FA. Implantable cardioverter defibrillators in primary and secondary prevention: a systematic review of randomized, controlled trials. Ann Intern Med 2003;138:445-452.(Mark A. Hlatky, M.D.)
Several therapies in cardiovascular medicine are backed by substantial evidence of high quality. For example, the effects of statin therapy on mortality and the risk of myocardial infarction have been evaluated in 10 large randomized, clinical trials involving more than 70,000 patients. This extensive experience in multiple studies has demonstrated a consistent reduction in the rate of cardiac events. Even more important than its demonstration that statins "work" is the fact that this body of evidence is large enough to permit assessment of the benefits and risks of treatment in various segments of the target population. We are thus able to judge which groups of patients benefit sufficiently from statins to justify the use of this treatment.
The evidence regarding procedures and devices is much less extensive than the evidence regarding most drugs, despite the high stakes associated with these treatments. Randomly assigning patients to procedures is more difficult than randomly assigning them to drugs, and double-blind trials are practically impossible. Only a few specialists are capable of performing procedures and implanting devices, and they may be proponents of the therapy under consideration and therefore unwilling to support randomized trials. Moreover, the regulatory requirements for devices are not as strict as those for drugs, and there is no specific regulation at all for new surgical procedures. Because there are greater difficulties and fewer incentives to conduct clinical trials, the evidence base with regard to devices and procedures is only a fraction of the size of that available for many drugs.
Coronary bypass surgery is the quintessential major cardiovascular procedure; in 2001, more than 300,000 patients underwent such surgery in the United States alone, at a cost of at least $6 billion. Because the stakes are so high for bypass surgery, we should have solid evidence from randomized trials to guide practice. Yet only 2649 patients worldwide have been randomly assigned to bypass surgery or medical therapy in the comparative trials whose results have been published to date1; the largest of these trials included 780 subjects and completed enrollment 25 years ago. Clinical trials of bypass surgery were large enough to document that the treatment reduces mortality among selected patients, but not large enough to show clearly how the degree of benefit from surgery varies according to common clinical indicators. Many patients who are considered as candidates for bypass surgery today do not match the profiles of patients who were enrolled in older trials, so the generalizability of the evidence is uncertain. The relatively thin base of evidence from randomized trials has contributed to persistent concern about the excessive and inappropriate use of bypass surgery in some groups of patients.
Despite uncertainty about who benefits from coronary bypass surgery, the procedure has been embraced widely. The growth of surgery has been fueled by anxieties about sudden death from cardiac causes and patients' wishes for a seemingly definitive procedure. Coronary bypass is a complex operation that requires hospitals to develop large and sophisticated surgical programs to deliver care, and there are substantial economic incentives to increasing the volume of procedures performed. A 1977 editorial in the Journal2 suggested that bypass surgery was "at the crossroads" and might well develop into an industry, given its perceived improvement of survival, the large population of potentially eligible patients, and the strong economic incentives. These comments were prescient: between 1977 and 1997, the number of patients undergoing coronary bypass surgery annually in the United States increased from 82,000 to 366,000.
The implantable cardioverter–defibrillator is now at a similar crossroads. These devices are expensive (more than $20,000 apiece), and in 2000, only 34,000 implantations were performed in the United States. Early clinical trials showed that the implantation of defibrillators reduced mortality in the small population of patients who had survived a spontaneous episode of ventricular fibrillation or sustained ventricular tachycardia.3 The stakes have been raised by proposals that the use of these devices be expanded to the treatment of the much larger population of lower-risk patients with underlying heart disease who have never had a cardiac arrest. The implantable defibrillator has enormous intuitive appeal as a lifesaving device for this population; a cadre of well-trained physicians is capable of implanting it; and a sophisticated medical-device industry is poised to promote its widespread use.
Is the available evidence sufficient to support the expansion of the indications for these devices? By the end of 2003, five trials of prophylactic defibrillator implantation had been published, but their results varied significantly.3 Two new trials are reported in this issue of the Journal (by Bristow et al., pages 2140–2150, and Kadish et al., pages 2151–2158), and the results of two additional trials were announced at the meeting of the American College of Cardiology held in New Orleans in March 2004. The results from all nine trials and 6200 patients have yet to be fully presented and assessed, but the emerging evidence suggests that the prophylactic implantation of defibrillators reduces mortality among patients with a left ventricular ejection fraction of 30 percent or less. The release of these new data will increase the pressure on large payers (especially Medicare) to expand coverage for these devices, and it is very likely that the rate of implantation will increase dramatically, propelled by the same forces that fueled the expansion of bypass surgery.
As with bypass surgery, the key question about the prophylactic use of defibrillators is not just whether it "works" overall, but how much the risks, benefits, and costs vary among the different subgroups of patients with heart disease. Variation in outcomes among different groups of patients is typical of many therapies and is likely in this instance, given the heterogeneity of the results of the early trials.3 Some patients may not reap enough benefit to justify the risk and cost of prophylactic defibrillator implantation, but we do not yet know how best to identify them. Furthermore, in the face of uncertainty, many physicians would rather err on the side of caution by putting in a defibrillator. There is a need to bring together the emerging evidence to determine who obtains sufficient benefit from a prophylactically implanted defibrillator to justify its use. Ideally, the clinical-trial investigators will pool their data to determine how the effectiveness of defibrillator treatment varies according to key clinical factors such as ejection fraction, presence or absence of a history of coronary revascularization, and time since myocardial infarction. Pooling trial data offers the best opportunity to establish the efficacy and cost effectiveness of these devices in various segments of the large population of patients who would potentially be eligible for this expensive therapy.
The enormous growth of coronary bypass surgery occurred before the advent of evidence-based medicine. It remains to be seen whether there will now be sufficient demand for solid evidence of efficacy and cost effectiveness to promote the appropriate use of the implantable defibrillator and other new devices and procedures.
Source Information
From the Donald W. Reynolds Cardiovascular Clinical Research Center, Stanford University School of Medicine, Stanford, Calif.
References
Yusuf S, Zucker D, Peduzzi P, et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994;344:563-570.
Braunwald E. Coronary-artery surgery at the crossroads. N Engl J Med 1977;297:661-663.
Ezekowitz JA, Armstrong PW, McAlister FA. Implantable cardioverter defibrillators in primary and secondary prevention: a systematic review of randomized, controlled trials. Ann Intern Med 2003;138:445-452.(Mark A. Hlatky, M.D.)