Advances in quantitative perfusion SPECT imaging

Quantitative software for myocardial perfusion single photon emission computed tomography (SPECT) has advanced significantly over the last 25 years. The strength and availability of quantitative tools for perfusion SPECT have in many ways provided a competitive advantage to nuclear cardiology compared with other higher-resolution noninvasive imaging modalities for the detection of coronary artery disease. The purpose of this report is to review the advances in quantitative diagnostic software for cardiac SPECT over the past 25 years. The time period ending with the 1980s ("the past") saw the origins of nuclear cardiology with the development of planar thallium 201 imaging and perfusion SPECT imaging without electrocardiographic gating. The period from 1990 to the present saw the development of gated SPECT imaging providing both perfusion and functional information and attenuation correction SPECT with improved perfusion information. The report concludes with a look into the future, where hybrid multimodality imaging systems may provide a comprehensive noninvasive evaluation with previously unmatched accuracy in a single imaging session.     

Challenges for measurement of myocardial perfusion and perfusion reserve by SPECT imaging

Conclusions  The present study by Storto et al¹ provides additional evidence that quantitative estimates of myocardial perfusion and perfusion reserve can be derived from SPECT myocardial perfusion images by use of equipment, tracers, and techniques that are available in most nuclear cardiology laboratories. Additional clinical studies are needed to optimize the methods used to derive the quantitative estimates of perfusion and perfusion reserve from the SPECT imaging studies and, ultimately, to determine the applicability of these measurements to the daily practice of nuclear cardiology.     

The role of nuclear cardiology in clinical decision making

This review suggests that the field of nuclear cardiology is alive, well, and thriving, providing relevant information that aids in everyday clinical decision making for nuclear medicine and referring physicians alike. Despite the competition from other modalities, the clinically appropriate applications of nuclear cardiology techniques are likely to increase. The foundation of this optimism is based on the vast amount of data documenting cost-effective clinical applications for diagnosis, risk stratification, and assessing therapy in both chronic and acute coronary artery disease (CAD), the powerful objective quantitative analysis of perfusion and function provided by the technique, and the increasing general availability of the approach.     

Nuclear medicine studies of the heart

Nuclear imaging techniques are well established diagnostic tools in clinical cardiology, providing noninvasive information about myocardial perfusion, function and metabolism. The cost-effectiveness of radionuclide imaging in the diagnostic work-up of patients with coronary artery disease has been demonstrated. Additionally, the documented prognostic value of scintigraphic parameters is of clinical importance to guide decision making. Advances in technology, new radiotracers and new applications contribute to continuous growth in the field of nuclear cardiology. Multi-headed gamma camera systems lead to higher spatial resolution and sensitivity of cardiac single photon emission tomography (SPECT), and they also provide the opportunity for attenuation correction or electrocardiographic gating of SPECT images. Objective quantitative values of perfusion, function and metabolism are derived from scintigraphic data by use of improved software and hardware. With the latest developments in tracer technology, imaging of myocardial necrosis, receptor systems and autonomic innervation has become a reality and will lead to new clinical applications in the future. Received 12 December 1997; Revision received 16 February 1998; Accepted 18 February 1998     

Quantification of nuclear cardiac images: The Yale approach

Conclusion  Quantification of nuclear cardiac images provides a secondary support in reading myocardial perfusion images and improves the reproducibility in the diagnosis of cardiovascular diseases. The technology for the remote Web reading of nuclear cardiac images and the quantitative data allows for an easy and secured access to patient studies without the limitations of time and space. The recent increasing interests and applications in molecular targeted imaging have opened a new field in nuclear cardiology, and absolute image quantification of the focal tracer uptake in the myocardium is exceedingly critical for the quantitative analysis of molecular targeted images.     

Visual assessment of left ventricular perfusion and function with electrocardiography-gated SPECT has high intraobserver and interobserver reproducibility among experienced nuclear cardiologists and cardiology trainees

BACKGROUND: Stress electrocardiography (ECG)-gated single photon emission computed tomography (SPECT) for assessment of left ventricular perfusion and function improves the confidence of interpretation and enhances specificity for detection of coronary artery disease. The reproducibility of visual interpretation of ECG-gated SPECT images and the significance of training and experience have not been reported previously in a large series of consecutive patients. We evaluated both intraobserver and interobserver agreement of interpretation of ECG-gated SPECT images among 3 cardiology trainees and 3 experienced nuclear cardiologists from 3 institutions. METHODS AND RESULTS: Three nuclear cardiologists and 3 cardiology trainees who had fulfilled American College of Cardiology/American Society of Nuclear Cardiology Core Cardiology Training Symposium (ACC/ASNC COCATS) guidelines for level II training in nuclear cardiology independently evaluated 106 consecutive technetium 99m sestamibi SPECT images with ECG gating of either the stress or rest images. All cases were interpreted blindly, twice in random sequence, without clinical data. We assessed intraobserver and interobserver agreement for myocardial perfusion, left ventricular regional and global systolic function, and overall clinical impression, by means of percent agreement and Cohen's kappa statistic. Intraobserver agreement was good (82%-92%, kappa = 0.54-0.84) for assessment of myocardial perfusion, systolic function, and overall impression. Interobserver agreement was also good, ranging from 65% to 90% (kappa = 0.32-0.76), with better agreement found for assessment of function (77%-85%, kappa = 0.52-0.7) than for perfusion (65%-80%, kappa = 0.32-0.6). For all measures, there were no significant differences in reproducibility between nuclear cardiologists and cardiology trainees. CONCLUSIONS: Interpretation of ECG-gated SPECT images has high reproducibility and agreement among both nuclear cardiologists and cardiology trainees.     

Nuclear cardiology in a managed care environment

Health maintenance organizations (HMO) and nuclear cardiology represent mutual threats and mutual opportunities for each other. On the one hand, nuclear cardiology represents a cost center with HMOs exerting tremendous financial pressure on nuclear cardiology programs. On the other hand, nuclear cardiology can act as a sage gatekeeper to the cardiac catheterization laboratory and help HMOs effectively control the health care of an increasing percentage of the population. Through the process of negotiation, of determining each other's needs, an accommodation can take place between the two. The ability to correlate scan results with coronary angiography provides individual nuclear cardiology programs with the opportunity to demonstrate their accuracy. A Nuclear Cardiology Report Card based on these data can be developed for use, with HMOs creating the opportunity to compete not only on price but also on value. Carved out capitation rates for nuclear cardiology can be estimated on the basis of actual experience with an HMO population and by extrapolation from test frequency of the U.S. population. The financial disincentives of capitation and of managed care challenge the physician-patient relationship. Advocacy of the role of nuclear cardiology and an understanding of negotiation strategies can aid nuclear cardiologists in their attempts to provide quality care with commensurate compensation. Database software development, supported by a grant from DuPont Pharma, Billerica, Mass.     

Nonperfusion applications in nuclear cardiology: Report of a task force of the American society of nuclear cardiology

Conclusions  In conclusion, there seems to be a combination of factors that have contributed to the underutilization of nonperfusion applications of nuclear cardiology. For some procedures (i.e., infarct-avid imaging, probe radiocardiography, and functional evaluation with mental stress) there is a perceived lack of clinical application. For these, further clinical validation and education of referring physicians is necessary. Other nonperfusion studies (i.e., equilibrium RNV) have not kept pace technologically with competing modalities. For these, camera and computer manufacturers should be encouraged to develop and implement appropriate new hardware and software. For others (i.e., ¹²³I MIBG “nerve” imaging and fatty acid analog metabolic imaging) radiopharmaceutical availability, particularly in the United States, is severely limited. For these, radiopharmaceutical companies should be encouraged to sponsor clinical trials validating clinical efficacy and cost-effectiveness, and if justified, appropriate applications should be filed with federal agencies. Last, and perhaps most important, physicians practicing nuclear cardiology must be motivated to embrace these nonperfusion radionuclide applications when appropriate, target patient populations in their institutions who will most benefit, and convince referring physicians of the clinical efficacy and cost effectiveness. For these and so many other facets of nuclear cardiology, the old cliché applies: “Nothing ventured, nothing gained.” One of the concerns I had when I started my term as President of the ASNC was the lack of clinical or research activities in the other areas of nuclear cardiology outside of perfusion imaging. It seemed to me that we had put all of our eggs in one basket! Although perfusion imaging was and is still on the rise, other important technniques that we often used in the past have now been relegated to the group of techniques seldom used. To address this concern, we created a task force on the nonperfusion uses of nuclear cardiology to summarize the state of the art in those other areas, to try to understand the reasons for a deceleration in their use, and to propose ways to reverse this process. Gordon DePuey did a superb job with a group of top expers who met several times to discuss these less-used techniques. Their conclusions and suggestions are thought generating and I hope will serve as a stepping stone for nuclear cardiologists to reevaluate these other important components of nuclear cardiology. Mario S. Verani, MD Past President, ASNC     

Assessment of cardiac function and myocardial metabolism by nuclear medicine]

Simultaneous assessment of myocardial perfusion and cardiac function came to be possible by 99mTc myocardial perfusion agents. We can use ECG-gated SPECT and first pass radionuclide angiocardiography for it. ECG-gated SPECT made it possible to assess wall motion using wall thickening and QGS (quantitative gated SPECT) analysis, which are useful in various clinical situations. First pass radionuclide angiocardiography gives good assessment of cardiac function during stress, and supports the diagnosis of myocardial ischemia. On the other hand, the assessment of myocardial metabolism is another specific feature of nuclear cardiology. 123I-BMIPP SPECT is applicable to various cardiac diseases such as ischemic heart disease, and 18F-FDG PET has been considered as the gold standard of myocardial viability. Recently, gamma camera for 18F-FDG imaging has been developed, which may make FDG imaging more popular.     

Significance of 99mTc-labeled perfusion agents in the simultaneous assessment of myocardial perfusion and cardiac function

Simultaneous assessment of left ventricular myocardial perfusion and systolic function was accomplished by utilizing ECG-gated myocardial perfusion SPECT. This development in nuclear cardiology will be attributed to the recent advances in new 99mTc-labeled perfusion agents, multi-detector SPECT system and software for automatic edge-detection of the left ventricle. In this article, we described about the clinical utilities of this method in detecting "hibernating myocardium," severe coronary artery disease patients with exercise-induced LV dysfunction, in predicting functional recovery after reperfusion therapy for acute myocardial infarction patients and in diagnosing patients with right heart diseases.     

Analysis of referral patterns,predictive accuracy,and impact on patient management of myocardial perfusion imaging in a new nuclear cardiology laboratory

BACKGROUND: Most of the published data on myocardial perfusion imaging (MPI) come from large tertiary-referral medical centers with extensive experience in cardiac imaging as well as a large volume of procedures. Whether the results of MPI remain as reliable in new nuclear cardiology laboratories with smaller volumes of procedures is unknown. The purpose of this study was to analyze the referral patterns, predictive accuracy, and impact of MPI on clinical practice in a newly opened nuclear cardiology laboratory. METHODS AND RESULTS: We performed a prospective study on all patients referred for MPI at our nuclear cardiology laboratory during its first year of operation. Patients were followed up for 3 months after the MPI study to determine whether they underwent coronary angiography. The study population consisted of 334 patients. Their mean age was 56 +/- 10 years, and 80% were men. Of the patients, 30% were asymptomatic, 29% had angina, and only 6% had recent acute myocardial infarction or unstable angina. Fifty-one patients (fifteen percent) were subsequently referred for coronary angiography. The positive and negative predictive values of MPI were 91% and 86%, respectively. The presence of reversible perfusion defects (P =.02) and the presence of multiple perfusion defects (P =.01) on MPI were the most important determinants of subsequent referral to coronary angiography. CONCLUSIONS: MPI stress testing retains its high diagnostic accuracy in a new nuclear cardiology laboratory with a relatively small volume of procedures. Furthermore, MPI findings in this population had a strong impact on the clinical practice of the referring physicians in terms of subsequent referral to coronary angiography.     

KSNM60 in Cardiology: Regrowth After a Long Pause

The Korean Society of Nuclear Medicine (KSNM) is celebrating its 60th anniversary in honor of the nuclear medicine professionals who have dedicated their efforts towards research, academics, and the more comprehensive clinical applications and uses of nuclear imaging modalities. Nuclear cardiology in Korea was at its prime time in the 1990s, but its growth was interrupted by a long pause. Despite the academic and practical challenges, nuclear cardiology in Korea now meets the second leap, attributed to the growth in molecular imaging tailored for many non-coronary diseases and the genuine values of nuclear myocardial perfusion imaging. In this review, we describe the trends, achievements, challenges, and perspectives of nuclear cardiology throughout the 60-year history of the KSNM.     

Effectiveness of nuclear cardiology training guidelines: A comparison of trainees with experienced readers

Background

To evaluate the effectiveness of published nuclear cardiology training guidelines, the diagnostic accuracy of image interpretation by nuclear cardiology trianees was compared with that of experienced nuclear cardiologists.

Methods and Results

The accuracy of three experienced nuclear cardiologists and three trainees with level II experience following Society of Nuclear Medicine/American College of Cardiology/American Society of Nuclear Cardiology guidelines in the interpretation of 114 exercise ^99mTc-labeled sestamibi single-photon emission computed tomographic imaging studies was evaluated. Studies were selected randomly and included patients with less than 5% likelihood of coronary artery disease, as well as patients with angiographically demonstrated single and multivessel disease. Studies were interpreted by each reader without knowledge of clinical or exercise data. Each reader classified perfusion as normal or abnormal. Accuracy was assessed according to sensitivity, normalcy rate, and predictive accuracy. In addition, the ability of experienced readers and trainees to identify abnormal perfusion in patients with multivessel disease was compared. Trainees had high accuracy, comparable to experienced readers for sensitivity, normalcy rate, and predictive accuracy, as well as the ability to identify abnormal perfusion in patients with multivessel disease. In all categories, experienced interpretors demonstrated a trend toward greater accuracy with less observer variability than did trainees.

Conclusion

Structured training in nuclear cardiology following Society of Nuclear Medicine/American College of Cardiology/American Society of Nuclear Cardiology guidelines during clinical cardiology fellowship is effective, and trainees possess the skills to interpret myocardial perfusion images accurately. Interpretive skills can be expected to improve further with experience.     

Myocardial perfusion imaging agents: SPECT and PET

The advent of myocardial perfusion imaging 30 years ago was a major landmark, which heralded the emergence of the field of nuclear cardiology into clinical practice. Over the years, the different tracers cited in this review have been used with SPECT or PET imaging technologies for the noninvasive evaluation of regional myocardial blood flow, which has enhanced our ability to diagnose CAD, assess prognosis, detect viable myocardium, and evaluate the efficacy of therapies aimed at improving myocardial blood flow. In the future, new SPECT perfusion agents should be developed and validated in the experimental laboratory for feasibility in the clinical setting. Hopefully, such new radiolabeled perfusion agents will have a high first-pass extraction, will be more linear with flow increases in the hyperemic range, and will be labeled with Tc-99m. The clearance rates from the myocardium after initial uptake should be slow enough, as with Tl-201, to acquire high-quality poststress gated SPECT images. Ideally, such perfusion agents should also be extracted intracellularly with quantitative uptake reflecting the degree of viability (eg, as with Tl-201). Absolute quantitation of myocardial blood flow in milliliters per minute per gram by use of SPECT technology would be highly desirable, particularly to increase the detection rate of multivessel disease in which flow reserve is uniformly diminished. This is often categorized as balanced ischemia. Absolute quantitation is a major strength of PET perfusion tracers, as is the ability to accurately correct for attenuation, thereby providing high sensitivity and specificity for CAD detection. The roll-off or plateau in myocardial uptake with hyperemia is also seen with the PET perfusion tracers such as N-13 ammonia and Rb-82. Despite the advent of molecular imaging and the introduction of new imaging agents by which to noninvasively evaluate biologic processes such as apoptosis and angiogenesis in vivo, myocardial perfusion imaging will remain the mainstay of nuclear cardiology in the near future. Continued research and development for this imaging technique are warranted for the reasons cited in this review.     

Corridor4DM: The Michigan method for quantitative nuclear cardiology

Quantitative software for the analysis and review of myocardial perfusion emission computed tomography images is an indispensable tool in the nuclear physician’s evaluation of patients with known or suspected heart disease. The Corridor4DM (4DM) application (formerly known as 4DM-SPECT), developed at the University of Michigan Medical Center, is a quantitative software application providing automated processing, analysis, and reporting of myocardial perfusion and function from cardiac emission computed tomography studies in a tightly integrated, user-centered environment. With health care placing increased emphasis on higher quality and efficiency in diagnostic imaging, quantitative analysis and review software applications need to provide a comprehensive environment supporting correlative review of multimodality images, integrated report generation, and remote review capabilities. The current and future design capabilities of the 4DM software application are discussed with respect to the changing landscape of imaging, where cardiac computed tomography, positron emission tomography, structured reporting, and remote review are expanding the base requirement specifications for quantitative software.     

Assessing the need for nuclear cardiology and other advanced cardiac imaging modalities in the developing world

Background

In 2005, 80% of cardiovascular disease (CVD) deaths occurred in low- to middle-income countries (i.e., developing nations). Cardiovascular imaging, such as myocardial perfusion SPECT, is one method that may be applied to detect and foster improved detection of at-risk patients. This document will review the availability and utilization for nuclear cardiology procedures worldwide and propose strategies to devise regional centers of excellence to achieve quality imaging around the world.

Methods

As a means to establish the current state of nuclear cardiology, International Atomic Energy Agency member and non-member states were queried as to annual utilization of nuclear cardiology procedures. Other sources for imaging statistics included data from medical societies (American Society of Nuclear Cardiology, European Society of Cardiology, and the European Association of Nuclear Medicine) and nuclear cardiology working groups within several nations. Utilization was calculated by dividing annual procedural volume by 2007 population statistics (/100,000) and categorized as high (>1,000/100,000), moderate-high (250-999/100,000), moderate (100-249/100,000), low-moderate (50-99/100,000) and low (<50/100,000).

Results

High nuclear cardiology utilization was reported in the United States, Canada, and Israel. Most Western European countries, Australia, and Japan reported moderate-high utilization. With the exception of Argentina, Brazil, Colombia and Uruguay, South America had low usage. This was also noted across Eastern Europe, Russia, and Asia. Utilization patterns generally mirrored each country’s gross domestic product. However, nuclear cardiology utilization was higher for developing countries neighboring moderate-high “user” countries (e.g., Algeria and Egypt); perhaps the result of accessible high-quality training programs.

Conclusions

Worldwide utilization patterns for nuclear cardiology vary substantially and may be influenced by physician access to training and education programs. Development of regional training centers of excellence can guide utilization of nuclear cardiology through the application of guideline- and appropriateness-driven testing, training, continuing education, and quality assurance programs aiding developing nations to confront the epidemics of CVD.     

Practice and future aspects of nuclear cardiology

The field of nuclear cardiology has demonstrated sustained growth in recent years owing to its increasingly recognized value for clinical applications and patient management. Computer advances in this field have allowed the technology of ECG-gated SPECT to become a routine part of nuclear cardiology. In our laboratory, myocardial perfusion and left ventricular function during stress (bicycle exercise or dobutamine infusion) were analyzed in a single examination by means of gated SPECT. This procedure has the potential to provide comprehensive information with which to evaluate patients with ischemic heart disease. 123I-BMIPP is a branched-chain free fatty acid, and its distribution could provide useful information about metabolic function in patients with ischemic heart disease (including minor infarction). The solid-state gamma camera 2020 tc Imager is now commercially available and has been clinically applied. The lightweight and compact design of the camera allows mobility of the unit between departments and floors. This technique would be useful for assessing left ventricular function under emergency conditions. In this paper, the techniques of examinations are described, and potential assessments are addressed. We look forward to further advances in nuclear cardiology for the accurate diagnosis and management of patients with various cardiac diseases.     

Activity and practice of nuclear cardiology in the Czech Republic, 2001

Radionuclide myocardial perfusion imaging (MPI) has been on the rise in Europe and the USA. Details on nuclear cardiology in the Czech Republic are not available as yet, as it is impossible to obtain comprehensive data from official registers owing to different methods of reporting and data evaluation. A questionnaire concerning nuclear cardiology activity and practice in 2001 was sent to all nuclear medicine departments in the Czech Republic. All 48 departments completed the questionnaire. In 2001, 50 planar and 54 tomographic (SPET) scintillation cameras were used. The average age of the SPET cameras was 5 years (13% of SPET cameras were >8 years old). Out of the 48 centres, 39 (81%) provided a nuclear cardiology service; the total number of cardiological studies was 15,740 in 2001 (1.5 studies/1,000 population/year). The most frequently employed method was MPI (81.7%), the frequency of which had increased by 10% compared with 2000; 26 of the 39 (67%) departments reported that MPI activity was increasing. Nevertheless, the Czech Republic nuclear cardiology activity remained below the European average (2.2/1,000 population in 1994) and, particularly, below activity in the USA (15/1,000 in 1997). The activity was rather unevenly spread. Whereas two centres with >1,000 studies/year accounted for 20% of the total MPI studies, 16 of 39 (41%) departments exhibited low activity (<200 studies/year) and accounted for only 15% of the total MPI studies. The use of SPET increased from 91% in 2000 to 94% in 2001 (only three institutes performed planar examinations). The most widely used tracer was (99m)Tc-MIBI (60% of total MPI), followed by (201)Tl (21%) and (99m)Tc-tetrofosmin (19%). ECG-gated SPET was employed by 20/39 (51%) centres, of which 11 (28%) performed it as a standard examination; 39% of the total MPI studies included this technique. Thirteen percent (5/39) of the departments used attenuation correction, and 69% (27/39) of the departments used a prone projection. Equilibrium radionuclide ventriculography, with 2,317 examinations (14.7%), ranked second among all nuclear cardiology methods, followed by first-pass angiocardiography (406 studies, 2.6%) and (18)F-FDG (163 studies, 1%).     

Northern exposure: Nuclear cardiology in the Canadian health care system

The Canadian health care system may provide valuable insights into the future practice of nuclear cardiology in the United States. Rationing of medical care is not legislated by the Canadian health care system, although resource allocation is required of Canadian physicians and hospital administrators. Canadian nuclear cardiologists and physicians are not restricted in the ordering of diagnostic studies, despite the decreased availability in imaging systems and the centralization of equipment and personnel in Canada. Canadian imaging equipment is, in general, used more with less average idle time per unit. Delays in the performance of nonemergent imaging studies are more common in Canadian imaging laboratories. The number of out-of-hospital nuclear medicine laboratories is not increasing, because of government constraints on licensing and the general requirement that only radiologists or certified nuclear medicine physicians can operate these laboratories. A survey of 71 nuclear cardiology laboratories in the United States and Canada reveal that 21% of all cardiac imaging studies are performed for post-myocardial infarction risk stratification in Canada, compared with only 11% in United States laboratories. Rest and reinjection thallium imaging studies are performed more than twice as often in the United States laboratories. Canadian laboratories perform a higher average number of myocardial perfusion (2123 vs 1789) and ventricular function (773 vs 554) studies as compared with their United States counterparts. No other significant differences in clinical usage patterns were identified. A total of 130,000 nuclear cardiologies were performed in Canada in 1993, with less than 5% growth in the number of Canadian studies projected for 1994. Forty-five percent of Canadian perfusion studies are performed with ^99mTc-labeled sestamibi frequently using a 2-day protocol (60%) with electrocardiogram gating (30%). Positron emission tomography (PET) can be performed in only six Canadian cities. Canadian PET centers are government funded, located in university teaching hospitals, and principally, used for the purpose of research. Stress echocardiography is not widely performed in Canada because of the heavy clinical volume of standard echocardiographic studies at most hospitals, which reduces the time available for stress echocardiography. No separate billing code is available for stress echocardiography studies in Canada. Canadian cardiologists have accepted the value of rest and stress nuclear studies for the management of their patients and have concluded that it is more time efficient to perform clinical duties in lieu of stress echocardiographic studies. In conclusion, the realities of the Canadian health care system are that universal health care is valuable as long as it is consistent high quality medical care, and that the cost of universal coverage must be borne by the taxpayer using the system. The fact that nuclear cardiology has continued to thrive in the Canadian health care system suggests that future health care modifications in the United States will not exert a significant impact on the practice of nuclear cardiology.     

Acute myocardial perfusion imaging for the technologist

Conclusion  Acute myocardial perfusion imaging involves careful planning from the nuclear medicine or nuclear cardiology laboratory to ensure optimal results are achieved. The role of the technologist is to ensure a high-quality study is performed on every patient who is referred to the laboratory. This is one of the most important roles because the decision for further evaluation can be based on the interpretation of the acute images. When acute myocardial perfusion imaging is used appropriately, in conjunction with standard methods of evaluation for patients presenting to the emergency department with chest pain syndromes perceived to be cardiac in origin, it can be of great benefit. It offers a more definitive diagnosis of chest pain syndromes and can be used to reduce the expense of otherwise costly hospital stays, even in patients with moderate risk of ischemic heart disease.