|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 2007; 49:1352-1361, doi:10.1016/j.jacc.2006.12.035
(Published online 12 February 2007). © 2007 by the American College of Cardiology Foundation |
,*
* Department of Medicine, St. Luke’s Roosevelt Hospital, New York, New York
Departments of Imaging and Medicine and the Burns and Allen Research Institute, Cedars-Sinai Medical Center, and the Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, California
Heart Disease Prevention Program, University of California, Irvine, California.
Manuscript received August 22, 2006; revised manuscript received November 22, 2006, accepted December 8, 2006.
* Reprint requests and correspondence: Dr. Daniel S. Berman, Director of Cardiac Imaging, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Room 1258, Los Angeles, California 90048. (Email: bermand{at}cshs.org).
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Background: Whereas the prognostic effectiveness of MPS is well established, recent studies indicate that quantification of CAC also predicts cardiac outcomes. However, prognostic information is not yet available upon which to guide the management of patients who have had both tests.
Methods: We assessed the frequency of cardiac death and myocardial infarction over a mean follow-up of 32 ± 16 months in 1,153 patients undergoing both CAC scanning and MPS. Results were compared with those from a referent cohort of 9,308 patients who had earlier undergone MPS only.
Results: The frequency of myocardial ischemia rose with increasing CAC scores (p < 0.001), but ischemia was present in only 64 patients. Among the 1,089 nonischemic patients, of which only 3 (0.3%) underwent early revascularization, the annualized cardiac event rate was <1% in all CAC subgroups, including those with CAC scores >1,000. Kaplan-Meier analysis revealed similarly low cardiac event rates among nonischemic patients with CAC scores >1,000 and nonischemic patients with Bayesian coronary artery disease likelihood 
85%. Late myocardial revascularization rates were also similar in these 2 groups.
Conclusions: Among patients with nonischemic MPS studies, high CAC scores do not confer an increased risk for cardiac events. Thus, although patients with high CAC scores may be considered for intensive medical therapy to prevent future coronary artery disease events, a normal MPS study in such patients suggests no need for more aggressive interventions.
| |||||||||
Recently, coronary artery calcium (CAC) scanning has been advocated as a new method for risk stratifying patients at intermediate risk for ASCAD. Atherosclerotic lesions are frequently associated with the presence of calcium (3,4), which can be accurately detected using computed tomography and quantified according to calcium (5), volume (6), and percentile calcium scores (7). Measurements of CAC have been shown to predict the underlying magnitude of atherosclerotic burden (8–10), the frequency of stress-induced myocardial ischemia (11,12), and the likelihood of future cardiac events in longitudinal studies (13–18). To date, however, this newer method of risk stratification has not been compared directly to cardiac stress testing for prognostic purposes. Thus, there is as yet no information to guide physicians in their clinical assessment and subsequent management of patients who have undergone both forms of testing. Perhaps most puzzling may be the following question: do high CAC scores, an indicator of increased long-term risk, still confirm an increased likelihood of future cardiac events in patients with nonischemic MPS studies, an indicator of low risk? To address this issue, we undertook a study that compares 3-year outcomes among patients undergoing both exercise MPS and CAC scanning.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
1 year after undergoing dual isotope exercise MPS and CAC scanning within 6 months of each other (mean of 24 ± 38 days), between April 1998 and January 2005. This CAC group included a mixed sample of 702 patients (61%) who were physician-referred for CAC scanning, 92 (8%) who were self-referred, and 359 (31%) who underwent scanning as part of an ongoing research protocol (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research Study). Coronary artery calcium scanning was performed before MPS in 517 (55%) patients and on the same day or after MPS in 636 (55%) patients. The second group (hereafter referred to as the "MPS" group) consisted of 9,308 patients followed for 1 year after undergoing MPS on a clinical basis between March 1991 and September 1999. Patients with known coronary artery disease (CAD), valvular disease, or known cardiomyopathy were excluded. This MPS referent group allowed us to investigate whether there is a parallel between the prognostic relationship between CAD likelihood values and MPS results and CAC values and MPS results, in light of data indicating that CAC increases patients’ Bayesian likelihood for CAD (19). This research was approved by the CSMC Institutional Review Board, and all patients signed informed consent.
Stress testing and imaging protocol.
Patients were injected intravenously at rest with thallium-201 (3 mCi to 4 mCi). After rest MPS, patients underwent symptom-limited Bruce protocol treadmill exercise testing, with a 20 mCi to 30 mCi dose of technetium-99m sestamibi injected at near maximal exercise, and MPS imaging was then repeated. The MPS studies were acquired using elliptical 180° orbits and standard energy windows (19). Scintigrams were assessed semiquantitatively by visual interpretation of 20 myocardial segments according to a 5-point score, and summed scores were compared at stress and rest (20). A summed difference score 4, corresponding to 5% of the myocardial segments with reversible defect (21), defined the presence of ischemia.
Calcium scanning.
Scanning was performed using either electron beam (EBT) (Imatron C-150 or GE eSpeed, General Electric, Milwaukee, Wisconsin) (n = 934 patients) or multislice computed tomography (MSCT) (n = 219 patients), with acquisitions consisting of approximately 30 to 40 3-mm or 2.5-mm slices for the 2 tomographic systems, respectively (12). Foci of coronary artery calcification were identified and scored by an experienced technician, blinded to both patient characteristics and the MPS results, using semiautomatic commercial software on a NetraMD workstation (ScImage, Los Altos, California) by detection of at least 3 contiguous pixels (voxel size = 1.03 mm3) of peak density 130 HU within a coronary artery, with scoring verified by an experienced imaging cardiologist. Coronary artery calcium scores were calculated according to the method of Agatston et al. (5), and age- and gender-adjusted CAC percentile scores were determined according to the database of Raggi et al. (7).
Likelihood of CAD. The likelihood of ASCAD, exclusive of the MPS results, was calculated for each patient based on the Bayesian analysis of patients’ age, gender, risk factors, chest pain symptoms, and results of exercise electrocardiography according to a previously validated commercial software program (CADENZA, Advanced Heuistics, Bainbridge Island, Washington) (22).
Follow-up.
Patients were followed for the occurrence of hard cardiac events, which included either the occurrence of cardiac death, as noted and confirmed by a review of death certificate and hospital chart or physician’s record, or myocardial infarction, as evidenced by the appropriate combination of patients’ signs and symptoms and enzyme elevations (23). In addition, we identified if patients underwent cardiac catheterization or coronary revascularization procedures (coronary bypass or percutaneous coronary interventions), and defined these as "early" if they occurred 
60 days of noninvasive testing and "late" if they occurred >60 days after testing. The performance of late myocardial revascularization was a secondary end point in this study based on its use as a proxy variable for worsening clinical status (24). As previously described (25), initial patient follow-up was obtained by searching our hospital-based information system (WebVS) as well as the Social Security Death Index. Follow-up in the remaining patients was sought through a mailed questionnaire or a scripted telephone interview. To ascertain the cause of death, the information provided by WebVS and the death certificates obtained for all those who died in Los Angeles County were reviewed by 2 cardiologists blinded to subjects’ clinical information. Death from cardiac causes was defined as death from any cardiac cause (e.g., lethal arrhythmia, myocardial infarction, or pump failure).
Statistical analysis.
Continuous variables were compared using t tests or nonparametric Wilcoxon rank sum tests, depending on whether or not the assumptions for the test were met, whereas categorical variables were compared using the chi-square test. Chi-square test for trend was used when appropriate. Time to follow-up or events was calculated from the date of the second test to either the follow-up time or time of the first event. Early revascularizations <60 days were censored, except as noted. Kaplan-Meier survival curves were generated, and compared using the log-rank test. Cox proportional hazards survival models, risk adjusted for mean age, gender, dyspnea, and various coronary risk factors, were also generated. The Kaplan-Meier survival curves were compared using the log-rank test. The CAC group was divided for this purpose into 3 subgroups: those with CAC scores <400, 400 to 999, and 1,000. Similarly, the MPS group was also divided into 3 groups based on CAD likelihood values, according to a commonly used division: those with "low" (<15%), "intermediate" (15% to 84%), and "high" (>85%) CAD likelihood values (1). Because of clinical differences in the CAC cohort and MPS cohorts, subgroups were selected using patient matching techniques based on propensity scores that matched patients from the CAC cohort to patients from the MPS cohort on age, gender, absence of chest pain symptoms, shortness of breath, pre-MPS likelihood of ASCAD, and exercise duration (26). The propensity-matched groups were compared using either the paired t test in the case of exercise duration or the nonparametric Wilcoxon signed rank test for all continuous variables, whereas the McNemar test was used to compare the categorical variables.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
Frequencies of ischemia, revascularization rates, and cardiac events in the MPS referral cohort. As ASCAD likelihood increased, so did the frequency of inducible myocardial ischemia (Table 3). Among the MPS patients manifesting inducible ischemia, 27.7% underwent myocardial revascularization procedures, compared with only 0.4% among the patients with nonischemic MPS studies (p < 0.0001). Among the ischemic MPS patients, the frequency of revascularization procedures rose with increasing ASCAD likelihood, but the frequency of cardiac events did not. Among these patients, the percent ischemic myocardium was significantly greater in the 489 patients who underwent early myocardial revascularization compared with the 1,276 patients who did not (19.4 ± 10.0% vs. 11.6 ± 7.4%, p < 0.0001).
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Prior epidemiologic studies have largely assessed the role of CAC scanning in screening populations but not the typical diagnostic patients that are referred for cardiac stress testing. However, because high CAC scores have now become a potential reason for referring patients for cardiac stress testing (11,12), there is an increasing need to understand the prognostic interrelationship between CAC scores and measurement of inducible myocardial ischemia. Accordingly, in this study, we evaluated cardiovascular outcomes among patients who underwent both CAC testing and exercise MPS in close temporal proximity. In our population, revascularization was driven by the presence of myocardial ischemia on MPS imaging, but very few patients with normal MPS underwent revascularization, even when the CAC score was very high. This practice, which probably reflected our physicians’ confidence in the prognostic significance of a normal MPS study, resulted in a unique opportunity for us to examine the interrelationship between "anatomic" evidence of CAD and physiology, namely how clinical outcomes vary according to a wide spectrum of CAC scores among patients without evidence of inducible ischemia, as measured during MPS. Our findings indicate that the magnitude of CAC exerted little influence upon outcomes among the patients who did not have inducible ischemia during MPS. Most notably, among the nonischemic patients with CAC scores >1,000, the frequency of cardiac death or myocardial infarction over 
3 years of mean follow-up was <2%, thus yielding an annualized cardiac event rate of <1%.
Among patients with inducible myocardial ischemia, subsequent referral to cardiac catheterization is commonplace. In our study, 42% of CAC patients with inducible myocardial ischemia underwent early myocardial revascularization, with larger perfusion defects in these patients than those observed among ischemic patients who did not undergo myocardial revascularization. This "interventional referral bias" is the likely explanation for why cardiac event rates did not increase among our ischemic MPS subgroup as the likelihood of ASCAD increased. Similarly, because of this referral bias, we could not assess whether CAC scores added incremental information for predicting patient outcomes among patients with inducible myocardial ischemia.
Previous work by Budoff et al. (19) has suggested that progressively higher CAC scores increase the Bayesian likelihood of ASCAD, but complementary observations indicate that the likelihood of cardiac events still remains <1% for nonischemic MPS patients with a high Bayesian likelihood of CAD (1,23). Accordingly, we reasoned that there may be a parallel between outcomes observed in nonischemic patients who are stratified on the basis of ASCAD likelihood and those who are stratified on the basis of CAC scores. To assess this possibility, we also evaluated cardiac outcomes in a large series of patients who underwent exercise MPS studies in our laboratory, but from an earlier nonoverlapping period relative to the current recruitment of MPS patients for CAC scanning. Among the MPS patients without inducible ischemia, only a small increase in cardiac events occurred as the likelihood of ASCAD increased from small to high likelihood estimates, paralleling the pattern noted for CAC scores. Moreover, the frequency of cardiac events remained similarly low in the propensity-matched CAC and MPS subgroups that did not have inducible ischemia during exercise testing, including the subgroup of patients with CAC scores >1,000.
Pathophysiological considerations. The induction of myocardial ischemia is the end result of not only angiographic stenoses but also pathophysiological factors, such as the degree of endothelial dysfunction, which can influence the adequacy of myocardial perfusion during physiological stimulation (27,28) and contribute to the magnitude of perfusion defects observed during exercise MPS (29,30). For this reason, the assessment of myocardial perfusion during stress MPS complements both angiographic measurements (31,32) and Bayesian estimations of ASCAD likelihood in the prediction of cardiac events (1,23). The presence of higher levels of CAC is a marker of greater atherosclerotic burden (8,9), but whether CAC scores also reflect greater functional abnormalities known to be associated with adverse cardiac events is not presently clear. A recent study (33) suggests that higher CAC scores signify a greater magnitude of endothelial dysfunction, but CAC measurements correlate relatively poorly, for example, with C-reactive protein levels, a measure of systemic inflammation (18,34). Although we observed a progressively higher frequency of inducible perfusion defects as CAC scores increased, our study did not assess pathophysiological abnormalities that may govern this association. However, our follow-up results suggest that the inherently greater risk that may be associated with higher CAC scores may be identified by evaluating such patients for inducible myocardial ischemia.
Various cohorts of patients have been identified who appear to have a heightened risk of cardiac events in the presence of a normal MPS study. These include diabetic patients (35), patients with atrial fibrillation (36), and patients complaining of dyspnea (24). Our data, however, suggests that a high CAC score is not an additional factor that heightens short-term risk in the presence of a normal MPS study.
Study limitations. A number of limitations should be considered. First, because our CAC group had a low event rate with only 11 hard cardiac events, the small but statistically significant increase in hard cardiac events among nonischemic patients with high CAC scores would need additional study in larger samples to further assess the robustness of our findings. Second, a recent study (37) has suggested that diabetes and metabolic syndrome may modify the relationship between CAC scores and myocardial ischemia, but these and other potential modifying factors were not assessed in this study. One recent study (38) examined the prognostic relationship between CAC scores and MPS in type II diabetics, but the smaller sample size and event rate in that study, as well as its lack of a nondiabetic control group, precluded comparisons to our study. Third, comparison of our CAC to MPS cohorts was limited by differences in the clinical characteristics of these 2 groups, and also by differences in temporal recruitment. We overcome this issue, in part, by limiting our comparisons to propensity-matched cohorts, but this matching did not take into account important advances in the therapeutic management of nonischemic patients over the last decade. Nevertheless, this limitation does not detract from the general principle that we demonstrated in this study: neither high ASCAD likelihood values nor high CAC scores appear to constitute a pattern of increased prognostic risk in the presence of a nonischemic exercise MPS study. Finally, our study does not exclude the possibility of higher CAC scores reducing the "warranty period" associated with a normal MPS as patients are followed over longer time intervals.
Clinical relevance. Although CAC scanning has shown promise as a means for risk stratifying patients with relatively low likelihood of ASCAD (13–18), the cost of such scanning is not widely covered by third-party insurers. A variety of factors may govern the current limited acceptance of this technology, including concern that it may lead to a substantial increase in further diagnostic tests and related healthcare costs. On the other hand, it is also possible that the judicious use of this technology could improve the risk assessment and management of patients in certain clinical subgroups in a cost-effective manner. For example, whereas a persistent trend to refer many patients with only a low likelihood of ASCAD for stress MPS studies has been documented (39), application of CAC scanning in such patients could potentially reduce the number of such patients who are then referred for MPS studies by preferentially limiting MPS referral to those patients with relatively higher CAC scores (12). However, support for such potential practice requires determination of the appropriate management for patients identified as having high CAC scores but nonischemic MPS studies. Our findings suggest that patients with this combination of findings are unlikely to benefit from myocardial revascularization procedures. Prospective study is required to determine whether a strategy of applying CAC scanning to patients with a low-to-intermediate likelihood of ASCAD, followed by the performance of MPS in those with a sufficiently elevated CAC score and reserving coronary angiography referrals for patients with inducible myocardial ischemia, represents a cost-effective strategy for managing patients.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Greenland and R. O. Bonow How Low-Risk Is a Coronary Calcium Score of Zero?: The Importance of Conditional Probability Circulation, April 1, 2008; 117(13): 1627 - 1629. [Full Text] [PDF]
|
|
|
|
|
|
M. P. Schenker, S. Dorbala, E. C. T. Hong, F. J. Rybicki, R. Hachamovitch, R. Y. Kwong, and M. F. Di Carli Interrelation of Coronary Calcification, Myocardial Ischemia, and Outcomes in Patients With Intermediate Likelihood of Coronary Artery Disease: A Combined Positron Emission Tomography/Computed Tomography Study Circulation, April 1, 2008; 117(13): 1693 - 1700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sanz, P. R. Moreno, and V. Fuster The year in atherothrombosis. J. Am. Coll. Cardiol., March 4, 2008; 51(9): 944 - 955. [Full Text] [PDF]
|
|
|
|
|
|
A. N. DeMaria, J. J. Bax, O. Ben-Yehuda, P. Clopton, G. K. Feld, G. S. Ginsburg, B. H. Greenberg, J. D. Knoke, W. Y.W. Lew, J. A.C. Lima, et al. Highlights of the year in JACC 2007. J. Am. Coll. Cardiol., January 29, 2008; 51(4): 490 - 512. [Full Text] [PDF]
|
|
|
|
|
|
J. J. Mahmarian Computed Tomography Coronary Angiography as an Anatomic Basis for Risk Stratification: Deja Vu or Something New? J. Am. Coll. Cardiol., September 18, 2007; 50(12): 1171 - 1173. [Full Text] [PDF]
|
|
|
|
|
|
R. J. Gibbons, P. A. Araoz, and E. E. Williamson The Year in Cardiac Imaging J. Am. Coll. Cardiol., September 4, 2007; 50(10): 988 - 1003. [Full Text] [PDF]
|
|
|
|
 |
|
 Is Coronary Calcium Scanning Helpful in Patients Without Inducible Ischemia? Journal Watch Cardiology, March 28, 2007; 2007(328): 3 - 3. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
