National health expenditures, which now total about $1.1 trillion dollars annually, represent almost 14% of the gross domestic product in the U.S. Healthcare costs threaten to spiral higher with the aging of the baby-boomers, a rise in per-beneficiary healthcare costs to nearly twice the rate of inflation, and a strong societal desire to further expand medical benefits. While healthcare spending may indeed need moderating, the heavy-handed techniques used by some managed-care organizations to control the volume and/or intensity of services are wearing thin.
An alternative approach-cost-effectiveness analysis-integrates economic and outcomes measures into a ratio that can be used as a standard for comparing different diagnostic or treatment strategies. While this type of analysis is in its infancy, it is useful for providers to begin thinking within this conceptual framework.
Because about $300 billion of annual healthcare expenditures goes to the assessment and care of ischemic heart disease, there has been great interest in the efficacy of nuclear techniques for stratifying patients into two therapeutic categories: those who will be treated medically and those who will undergo coronary angiography in anticipation of revascularization. From the standpoint of patient outcomes, both myocardial perfusion scintigraphy and coronary angiography attempt to determine which patients need revascularization. Cost-effectiveness analysis considers the relative costs of performing the tests and the subsequent costs incurred as a result.
These downstream costs need to be all-inclusive, regardless of who pays. Costs associated with both appropriate and inappropriate outcomes in relation to national standards must be part of the analysis. For example, consider the case of a female with nonanginal chest pain and a mildly positive treadmill test who is referred to angiography and subsequent angioplasty for a 70% right coronary artery stenosis. Cost-effectiveness analysis would include the costs of these procedures and follow-up but would also need to consider an additional factor to account for the fact that neither the angiogram nor the subsequent angioplasty were medically indicated. The 1999 American College of Cardiology/American Heart Association Guidelines on Coronary Angiography designate only three class I indications-reflecting general agreement that the procedure is useful and effective-for diagnostic coronary angiography: high-risk criteria on noninvasive testing, class III and IV angina on medical treatment, and resuscitation from sudden cardiac death.1
This article will consider the knowledge base and limitations of regular utilization of stress myocardial perfusion imaging as a "gatekeeper" to catheterization, and how this might translate into a more cost-effective management strategy for defined patient populations.
DEFINING RISK
From an economic standpoint, definitive recognition of the patient at very low risk for cardiac events is an important objective. Downstream costs in relation to the suspected disease process should then be minimal. The challenge is that the margin between low and high risk is narrow, according to current guidelines.2 For example, low risk is characterized by an annual cardiac event rate of less than 1%, whereas high risk is characterized by an event rate exceeding 5%.
Stress myocardial perfusion imaging is the single best strategy for identifying patients who have a less than 1% risk for cardiac death or myocardial infarction within the next 12 months. A normal perfusion study establishes this under a broad array of circumstances: using a variety of radionuclides, with planar or single-photon emission computed tomographic (SPECT) imaging, in patients with known or suspected coronary artery disease (CAD), in the elderly, in both men and women, after myocardial infarction, and after revascularization procedures (see table).
A multicenter study of some 4000 patients has shown that even in patients with an intermediate-risk Duke treadmill score, which on its own connotes a 3% to 5% one-year mortality risk, a normal perfusion study is associated with an aggregate rate of myocardial infarction and cardiac death of ischemic origin of less than 1% per year over the next seven years.3 The documentation is so extensive that the American Society of Nuclear Cardiology in 1997 published a statement in the Journal of Nuclear Cardiology that only rarely should a patient with a normal scan be referred to angiography within the next 12 months.4
Figure 1 shows a normal exercise thallium-201 study, including a normal left ventricular ejection fraction (LVEF) both poststress and at rest, using electrocardiographic (ECG) gating. The studies that led to our understanding of the role of myocardial perfusion imaging in risk stratification were performed before routine use of ECG-gating. Knowledge of poststress and rest LVEF should add considerably to the precision of risk-stratification.
Healthcare providers and clinicians know that myocardial perfusion imaging is not infallible, however, and need to be observant for objective signs of clinical instability. False-negative scans, although rare, can occur in patients who have ingested caffeine within 12 to 24 hours before pharmacologic stress (caffeine is a competitive antagonist of adenosine and dipyridamole) and in patients with small hearts, such as elderly females, in whom the relatively poor spatial resolution of this technique may occasionally underdetect disease. In addition, the databases used to establish the risk of subsequent events all presumed that testing was performed in accordance with American Society of Nuclear Cardiology procedural guidelines.5 A substantial number of exams are done in a way that does not comply with the ASNC guidelines, however, thereby potentially undermining the data.
Many patients have mildly to moderately abnormal scans. Figure 2 shows a reversible defect inferiorly and inferoseptally, in the distribution of the right coronary artery. O'Keefe et al,6 Iskandrian et al,7 and Marwick et al8 have all shown that prognosis in patients with small defects is excellent with aggressive medical therapy and without surgical or percutaneous intervention. In our own experience, the cumulative hard event rate in patients with nonextensive perfusion defects was only 3% after 36 months of follow-up (Figure 3). As these patients clearly have CAD, they need follow-up by a cardiologist; about 3% annually will have worsening symptoms and/or disease progression despite aggressive medical therapy and attempts at risk factor modification.
GUARDING THE GATE
Several studies have examined whether using nuclear imaging as a gatekeeper to cardiac catheterization would result in significant cost savings. In our practice, about 40% of referred patients have normal scans, 30% have mild to moderate abnormalities, and 30% extensive defects.9 Although probably only the 30% with the most extensive abnormalities require angiography, a conservative stance would be to refer all but the 40% with normal scans.
Shaw et al evaluated this in a large multicenter study of patients with chest pain syndromes suggestive of CAD.10 About 5000 patients went directly to angiography, and a similar number had a perfusion study first. Of this second group, only patients with abnormalities on perfusion imaging were referred to angiography. Although patients in the angiography-first group had a higher rate of revascularization, clinical outcomes were the same. A strategy of performing nuclear imaging first, however, resulted in a cost savings of 35%, or about $3 million per 1000 such patients.
Most of the information on this topic thus far is derived from retrospective analyses of data from a single medical center or the combined experiences of several centers. The data are, however, impressive enough that practice patterns are beginning to change. It is increasingly common to use angiography with a view toward revascularization only when symptoms mandate it, or when risk profiling places a patient in a category associated with a 1% or greater risk of cardiac events for the next 12 months.
Large randomized prospective multicenter studies will help define any other criteria pertinent to making this critical decision for patients with known or suspected CAD. The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial, now under way, and the soon-to-start Bypass Angioplasty Revascularization Investigation (BARI)-2D trial, which focuses on diabetic patients, will compare medical therapy with revascularization. Both trials will take into account the incidence of cardiac events, cost-effectiveness, and quality of life in relation to ischemic burden. Until these data are available, angiography is indicated when a myocardial perfusion study shows a large perfusion defect, abnormalities in multiple vascular territories, reduced LV function along with a perfusion defect, ischemia in the distribution of the proximal LAD coronary artery, or other high-risk markers, such as abnormal lung uptake of thallium 201, transient ventricular dilation, or poststress stunning.
DR. BATEMAN is director of nuclear cardiology for Cardiovascular Consultants and Mid America Heart Institute in Kansas City, MO, and a professor of medicine at the University of Missouri-Kansas City.
---
CLINICAL SITUATIONS IN WHICH NORMAL MPI PREDICTS < 1% LIKELIHOOD OF CARDIAC EVENT WITHIN 12 MONTHS
--Planar T1-201
--Referred for possible CAD
--SPECT T1-201
--Known CAD
--SPECT MIBI
--Post-PTCA
--Exercise to 80% MPHR
--Post-CABG
--Adenosine
--Elderly
--Dipyridamole
--Strongly positive treadmill ECG response
Source: Bateman, Curr Opin Cardiol, 1996
---
References
1. ACC/AHA Guidelines for coronary angiography. J Am Coll Cardiol 1999;33:1756-1816.
2. ACC/AHA/ACP-ASIM Guidelines for the management of patients with chronic stable angina. J Am Coll Cardiol 1999;33:2092-2197.
3. Gibbons RJ, Hodge DO, Berman DS, et al. Long-term outcome of patients with intermediate-risk exercise electrocardiograms who do not have myocardial perfusion defects on radionuclide imaging. Circulation 1999;100:2140-2145.
4. ASNC position paper. Clinical relevance of a normal stress-rest myocardial perfusion scintigraphic study. J Nucl Cardiol 1997;4:172-173.
5. Garcia EV. Imaging guidelines for nuclear cardiology procedures. J Nucl Cardiol 1996;3:G1-G46.
6. O'Keefe JH, Bateman TM, Ligon RW, et al. Outcome of medical versus invasive treatment strategies for non-high risk ischemic heart disease. J Nucl Cardiol 1998;28-33.
7. Iskandrian A, Chae S, Heo J, et al. Independent and incremental prognostic value of exercise single photon emission computed tomography thallium imaging in coronary artery disease. J Am Coll Cardiol 1993;22:665-670.
8. Marwick TH, Shaw LJ, Lauer MS, et al. The noninvasive prediction of cardiac mortality in men and women with known or suspected coronary artery disease. Am J Med 1999;106:172-178.
9. Bateman TM, O'Keefe JH, Dong VM, et al. Coronary angiography rates following stress SPECT scintigraphy. J Nucl Cardiol 1995;2:217-223.
10. Shaw LJ, Hachamovitch R, Berman DS, et al. The economic consequences of available diagnostic and prognostic strategies for the evaluation of stable angina patients. J Am Coll Cardiol 1999;33:661-669.
