Evaluation of chest pain in patients with low to intermediate pretest probability of coronary artery disease by electron beam computed tomography

Evaluation of Chest Pain in Patients With Low to Intermediate Pretest Probability of Coronary Artery Disease by Electron Beam Computed Tomography Paolo Raggi, MD, Tracy Q. Callister, MD, Bruce Cooil, PhD, Donald J. Russo, Nicholas J. Lippolis, MD, and Randolph E. Patterson, MD

MD,

Despite its limited sensitivity and specificity in patients with low to intermediate probability of coronary artery disease (CAD), exercise treadmill testing (ETT) is frequently used as the initial test for investigation of chest pain. Although myocardial perfusion imaging is a significantly more accurate test, its added cost to ETT is considerable. The cost of a non-contrast electron beam computed tomography (EBCT) scan is comparable to that of ETT and the calcium score (CS) correlates closely with the volume of atherosclerotic plaque. Therefore, we tested the hypothesis that EBCT might be an effective and cost-beneficial technique for the identification of angiographically obstructive CAD (>50% stenosis) in patients with low to intermediate pretest probability of disease. We calculated the theoretic cost of attaining a diagnosis of CAD based on a Bayesian model that utilizes published sensitivity and specificity levels for ETT, EBCT, and stress myocardial perfusion imaging. We then submitted a cohort of 207 patients with low to intermediate probability of disease both to EBCT and ETT in random order,

and estimated the cost of achieving a correct diagnosis by either route based on the number of expected further tests. An EBCT calcium score of 150 was chosen as a cut-point with a sensitivity of 74% and a specificity of 89% for the presence of obstructive CAD. The theoretic Bayesian model predicted substantial cost savings when EBCT was used as the initial test instead of ETT, with decreasing benefit as the prevalence of disease increased (44% saving at 0% prevalence; 15% saving at 100% prevalence). In the patient cohort, the diagnostic pathway starting with EBCT provided a 45% to 65% cost saving over the ETT pathway. We conclude that in patients with low to intermediate pretest probability of disease, a pathway based on EBCT as the initial test to investigate presence of obstructive CAD provides a substantial cost benefit over a pathway based on ETT. Such cost advantages decrease as the prevalence of disease increases. 䊚2000 by Excerpta Medica, Inc. (Am J Cardiol 2000;85:283–288)

hysicians are frequently faced with the task of excluding obstructive coronary artery disease P (CAD) in patients with new onset chest pain and a low

(CS) correlates closely with total burden of underlying atherosclerotic plaque,9 –12 EBCT may potentially be applied to the initial study of patients with chest pain symptoms to provide rapid and accurate information regarding the presence and extent of atherosclerotic disease while allowing substantial cost savings.13 We studied this hypothesis in a Bayesian cost model and a prospective group of 207 patients with new onset chest discomfort at low to intermediate pretest probability of CAD. We were thus able to compare the results of a hypothetical cost model with real patient data.

to intermediate pretest probability of disease. These patients are frequently submitted to exercise treadmill testing (ETT) as the initial diagnostic examination, although it is known for its limited sensitivity and specificity.1–3 Due to the low sensitivity of ETT, especially in patients with single- or double-vessel disease, some patients with CAD may remain undiagnosed.1–3 Further, given the low specificity of ETT, the cost of this diagnostic pathway often becomes considerable. Although myocardial perfusion imaging adds significantly to the diagnostic and prognostic accuracy of ETT,4,5 its cost is greater than that of ETT and electron beam computed tomography (EBCT) imaging. EBCT allows the identification and quantification of coronary artery calcification—a marker of CAD.6 – 8 Because the coronary artery calcium score From the EBT Research Foundation, Nashville, Tennessee; Vanderbilt University, Nashville, Tennessee; and Emory University School of Medicine, Atlanta, Georgia. Manuscript received May 5, 1999; revised manuscript received and accepted June 23, 1999. Address for reprints: Paolo Raggi, MD, EBT Research Foundation, 64 Valleybrook Drive, Hendersonville, Tennessee 37075. ©2000 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 85 February 1, 2000

METHODS

Patients: The study was reviewed and approved by the Internal Review Board at our institution and all patients signed informed consent before enrollment. All patients were referred by primary care physicians for ETT or EBCT imaging due to the development of chest discomfort. The pretest probability of CAD for each study participant was estimated on the basis of age, sex, and the quality of reported symptoms as outlined in the tables of Diamond and Forrester.14 Patients were excluded if they were unable to sign informed consent, were affected by severe valvular 0002-9149/00/$–see front matter PII S0002-9149(99)00733-X

283

ETT results. Although none of the cardiologists interviewed suggested referring patients with a positive ETT for myocardial perfusion imaging before proceeding with coronary angiography, this alternative was included for completeness in the development of our model as previously done by other investigators (Figure 1).19 Electron beam computed tomography imaging protocol: All patients

underwent imaging with a C-100 scanner (Imatron, South San Francisco, California). Imaging was performed using 100-ms scan time and 3-mm single slice thickness, with a total of 36 slices obtained during 2 FIGURE 1. Flowchart of the algorithms. Angiography ⴝ cardiac catheterization with breath holding periods. Scanning selective coronary angiography; MPI ⴝ myocardial perfusion imaging. was electrocardiographically triggered at 80% of the RR interval. All heart disease, were unable to perform a treadmill areas of calcification within the borders of a coronary stress test, had suffered a previous episode of unstable artery with a minimal density of 130 Hounsfield units angina or myocardial infarction, or had undergone (HU) were computed for analysis and an atheroscleevaluation for coronary artery disease before our as- rotic plaque was defined as a minimum of 4 consecsessment. All patients were asked to undergo EBCT utive pixels with density of ⱖ130 HU (area ⬎1.37 and ETT in random order. The results of the tests were mm2). A CS was calculated according to the method reviewed by 2 independent investigators who were described by Agatston et al.20 blinded to the data. Electron beam computed tomography diagnostic Stress testing protocol: All patients underwent pathway: According to our EBCT diagnostic pathway, graded ETT according to a standard Bruce protocol all patients with coronary artery calcification and CSs aimed at reaching 85% of the maximal predicted heart ⱕ150 should not undergo further testing. These parate for age. To render the study more applicable to the tients should be followed clinically and given advice experience of daily practice, the criteria for grading on control of risk factors for CAD (Figure 1). Patients the results of ETT testing were developed from the with scores ⬎150 should be referred for ETT. If ETT recommendations of 12 community cardiologists. A is positive, coronary angiography is recommended. positive ETT response consisted of either flat or Patients with negative ETT results are treated medidownsloping ST-segment depression ⱖ2.0 mm mea- cally and are given advice on CAD risk factor modisured at 0.08 second after the J point, occurring in a fication. Patients with equivocal ETT results did not minimum of 2 contiguous leads with or without an- undergo further nuclear testing because the presence gina, or ST depression ⱖ1.0 mm in a minimum of 2 of CAD is already known. In this case, ETT is not contiguous leads accompanied by anginal chest pain. used to diagnose CAD, but to assess the physiologic Tests were interpreted as equivocal if they showed 1 significance of a previously discovered pathologic lemm ST depression at 0.08 second from the J point not sion. However, for completion of exposure, cost data accompanied by subjective symptoms. A test was are shown with and without the addition of the cost of considered negative if there were no inducible elec- myocardial perfusion imaging and that of the estitrocardiographic changes or any symptoms of chest mated number of ensuing coronary angiography prodiscomfort. A Duke treadmill score, which has been cedures for patients with equivocal ETT results (Figshown to provide relevant diagnostic and prognostic ure 1). A CS threshold of 150 was chosen because it information in patients undergoing ETT, was calcu- was close to values previously shown to have good lated according to previously published criteria.15–18 sensitivity and specificity for obstructive CAD13,21,22 Exercise treadmill testing diagnostic pathway: The and for identification of patients in need for revascuETT diagnostic pathway was devised based on the larization.23 Simultaneously, it was easier to memoanswers obtained from 12 consulting community car- rize, and therefore, utilize than values such as 147,22 diologists (Figure 1). The physicians indicated that 160,23 or 168.13,21 Finally, our study was completed by patients with equivocal ETT results should be referred performing a sensitivity analysis of our model with for nuclear myocardial perfusion imaging. Patients several lower scores. with either ETT or nuclear images suggestive of inCost analysis: The cost of EBCT was estimated ducible myocardial ischemia should be referred for from the average reimbursement paid for this test by 3 coronary angiography, whereas no further testing Tennessee commercial insurance companies (Table I). should be recommended for patients with negative To conform to the current literature, the cost of all 284 THE AMERICAN JOURNAL OF CARDIOLOGY姞

VOL. 85

FEBRUARY 1, 2000

after the completion of either EBCT or ETT, divided by the number of patients tested. Bayesian model and statistical analysis: The Bayesian cost model compared all approaches described in Reimbursement Sensitivity Specificity the ETT and EBCT algorithms, as well as the perforETT $ 330.00 68%* 77%* mance of myocardial perfusion imaging and coronary EBCT $ 450.00 74%† 89%† angiography as the initial test across all pretest prob‡ ‡ Myocardial perfusion $1,200.00 90% 80% ability ranges of CAD. The model was devised using imaging values of sensitivities and specificities for these tests Angiography $4,800.00 100% 100% reported in the current literature (Table I).3,13,21,25 *Detrano et al3; †Rumberger et al21; ‡Berman et al.25 Sensitivity and specificity for coroAngiography ⫽ cardiac catheterization and coronary angiography. nary angiography were considered to be 100%. The sensitivity and specificity of the CS cut point of 150 were calculated by interpolation of published parameters13,21 and assessed to be 74% and 89%, respectively. The Bayesian analysis provided theoretic estimates of the direct cost of diagnosis of CAD according to hypothetical levels of pretest probability of disease. It also allowed us to project the direct cost of the diagnostic pathway followed by our prospectively collected patients for each of the diagnostic algorithms proposed. Finally, we performed a sensitivity analysis of our model by calculating the cost of the FIGURE 2. Relation between cost and prevalence (pretest probability) of CAD (>50% proposed EBCT and ETT pathways stenosis) for the 6 diagnostic strategies described in the text. Angiography ⴝ direct using different CS cut points and coronary angiography; Electron Beam-CT ⴝ primary EBCT algorithms (see text and Figure 1); Electron Beam-CT (alt II): initial testing by EBCT followed by ETT if CS is their related sensitivities and speci>150 and referral to nuclear imaging of patients with equivocal ETT results; Exercise ficities. A low pretest probability of Treadmill Test ⴝ primary ETT algorithm (see text and Figure 1); Exercise Treadmill Test disease was defined as ⬍15% and (alt I) ⴝ initial testing by ETT and referral of positive cases to nuclear imaging; Myoan intermediate probability was decardial Perfusion Imaging ⴝ initial testing by nuclear imaging and referral of positive fined as ⱖ15% and ⬍85%.4,5 Paired cases to coronary angiography. t tests were used to test the significance of differences in direct costs for the ETT and EBCT algorithms. All values are expressed as mean ⫾ SD. Mean and 95% confidence intervals for cost savings were calculated. TABLE I Reimbursement and Sensitivity and Specificity for Detection of Obstructive Coronary Artery Disease for Methodologies

RESULTS

Bayesian analysis: Figure 2 compares the direct cost of each of the diagnostic pathways based on the pretest probability of CAD and the sensitivities and specificities of each modality shown in Table I. As shown, the primary EBCT algorithm was less expensive than the other pathways at all probability levels. FIGURE 3. Cost savings provided by the primary and alternative EBCT algorithm relaHowever, the alternative EBCT aptive to the least expensive ETT algorithm at each prevalence level. proach requiring nuclear imaging in patients with equivocal ETT results other standard CAD diagnostic tests was derived from was slightly more expensive than the basic EBCT the publication by Patterson et al.24 The direct per- pathway. Figure 3 shows the cost saving of each of the patient cost to achieve the diagnosis of CAD was EBCT algorithms relative to the least expensive ETT calculated by adding the cost of all initial tests to the option at any level of pretest probability of CAD. The expected cost of all additional tests deemed necessary savings decrease as the pretest probability of CAD CORONARY ARTERY DISEASE/EBCT IN INTERMEDIATE PROBABILITY PATIENTS

285

have required either no further testing, or myocardial perfusion imaging if the alternate route were followed. Positive Negative Equivocal In the remaining 20 patients with a CS of ⬎150 the ETT was negative, ETT Results 20 (10%)* 124 (60%) 63 (30%)† Median Duke treadmill score ⫺4 ⫹10 ⫺1 and they would have required no furEBCT results ther testing. The final mean per-capPatients with calcifications 13 (65%) 72 (58%) 40 (63%) ita cost of the primary EBCT pathPatients with calcium score ⬎150 4 (20%)‡ 20 (16%)‡ 11 (17%)‡ § § § way was $599 and that of the alterAverage calcium score 153 ⫾ 81 96 ⫾ 21 88 ⫾ 23 nate EBCT approach was $856. The *Patients in ETT algorithm potentially referred for coronary angiography. cost saving provided by the primary † Patients in ETT algorithm potentially referred for myocardial perfusion imaging. ‡ EBCT algorithm over the primary Patients in EBCT algorithm potentially referred for exercise treadmill testing. § p ⫽ NS. ETT algorithm amounted to 65% ($599 vs $1,701; average savings: $1,102 ⫾ 123; 95% CI, $860 to grows from 0% to 100% (from 48% to 33% saving for $1,344), whereas the average cost of the alternative the primary EBCT algorithm, and from 44% saving to EBCT approach was 45% lower than that of the alternative ETT protocol ($856 vs $1,556; average sav15% saving for the alternate EBCT pathway). Patient study: The mean age of the 207 patients ings: $700 ⫾ 91; 95% CI, $521 to $878). A comparincluded in this study was 50 ⫾ 9 (range 35 to 67, ison of savings provided by the primary and alterna57% men). The mean pretest probability for the entire tive EBCT versus ETT algorithms for several CS patient cohort was 29% (median 22%). Table II shows cut-points is provided in Table III as a sensitivity a comparison of ETT and EBCT results. ETT was analysis of our model. As expected, pathways utilizing positive in 20 patients (10%), equivocal in 63 (30%), lower CSs (higher sensitivity but lower specificity) are and negative in 124 patients (60%). The median Duke more expensive than pathways with higher scores and treadmill scores for these groups were ⫺4, ⫺1, and provide less savings. Nonetheless, substantial savings ⫹10 respectively. The original publication by Mark et can be attained via the EBCT pathways even with CS al15 indicates that patients with scores ⱖ⫹5 are at low cut points that are ⬍150. risk, and those with scores from ⫺10 to ⫹4 are at intermediate risk of CAD. Because myocardial perfu- DISCUSSION sion imaging is often recommended for patients at This study shows that a strategy based on the intermediate risk,17 and all patients with positive or application of EBCT for the initial investigation of equivocal ETT results in our study had intermediate chest discomfort in patients with low to intermediate risk scores, all of these patients would have potentially pretest probability of CAD, allows substantial cost required subsequent nuclear studies. According to the savings compared with strategies based on initial testprimary ETT pathway (Figure 1), our patients would ing by ETT and myocardial perfusion imaging. Simhave required 20 coronary angiographic studies and ilar conclusions were recently reached by Rumberger 63 myocardial perfusion imaging studies. The latter in et al,13 although these investigators did not use actual turn would have generated the need for several angio- patient data and reimbursement rates as done in this graphic procedures. Based on the cost of all predicted analysis. investigations, the projected average per-capita cost of ETT is widely employed as the initial test for the this protocol amounted to $1,701. When the option to diagnosis of CAD in patients with low or intermediate perform nuclear studies before proceeding with coro- pretest probability of disease, although it has known nary angiography was added to the group of patients limited sensitivity and specificity.1–3 Although the inwith positive ETT results, the average per-capita cost troduction of the Duke treadmill score improved both decreased to $1,556. In this option, after having un- the prognostic and the diagnostic accuracy of ETT, dergone a negative myocardial perfusion imaging Shaw et al18 recently demonstrated that 40% of the study, some of the patients may no longer need to patients with a low-risk score and 67% of the patients undergo coronary angiography. with a moderate-risk score have obstructive CAD. On EBCT, 125 of 207 patients (60%) showed Because of the uncertainty in the diagnosis, myocarcoronary calcifications on the initial scan (Table II). dial perfusion imaging is often recommended in these Coronary artery calcification was present in a high patients17 with a substantial financial cost. In our percentage in all ETT groups. Thirty-five patients study, all patients with either a positive or an equiv(17%) had a CS of ⬎150 and, according to our EBCT ocal ETT result reached a moderate Duke treadmill pathway, should have been referred for ETT testing. score level and would therefore have qualified for Because we performed simultaneous EBCT and ETT myocardial nuclear imaging. As shown, this approach we were able to predict what would have happened in was more expensive than the proposed EBCT pathan actual clinical situation for these 35 patients. ETT ways. However, it remains to be demonstrated was positive in 4 of the 35 subjects and, according to whether valuable prognostic information can be dethe protocol, these 4 patients would have required rived from the coupling of EBCT and ETT testing, as subsequent coronary angiography. Eleven patients un- has been shown in the association of plain ETT with derwent an ETT with equivocal results and would stress nuclear testing.4 TABLE II Patient Study: Comparison of Exercise Treadmill Testing (ETT) and Electron Beam Computed Tomography (EBCT) Results

286 THE AMERICAN JOURNAL OF CARDIOLOGY姞

VOL. 85

FEBRUARY 1, 2000

TABLE III Comparison of Cost Savings Provided by Primary and Alternative Electron Beam Computed Tomography (EBCT) Algorithms Over Respective Exercise Treadmill Testing (ETT) Algorithms at Various Calcium Score Cut-Points Score Cut Point

Sensitivity*

Specificity*

Mean Cost of Primary EBCT Pathway ($)

20 50 80 100 150

92% 88% 84% 81% 74%

63% 79% 84% 87% 89%

743 688 644 633 599

Relative Savings in Primary Pathways EBCT vs ETT 957 1,013 1,056 1,068 1,102

⫾ ⫾ ⫾ ⫾ ⫾

120 121 122 122 123

(56%) (60%) (62%) (63%) (65%)

Mean Cost of Alternative EBCT Pathway ($) 1,305 1,148 982 911 856

Relative Savings in Alternate Pathways EBCT vs ETT 251 408 573 644 700

⫾ ⫾ ⫾ ⫾ ⫾

92 95 95 93 91

(16%) (26%) (37%) (41%) (45%)

*Derived from Rumberger et al.13,21

The lower cost of the EBCT algorithms over the myocardial perfusion imaging pathway shown in this study is probably secondary to a lower sensitivity for obstructive CAD of EBCT compared with nuclear imaging. As a less sensitive tool, EBCT detects fewer patients with true or suspected obstructive CAD, therefore, mandating fewer subsequent angiographic studies. Because nuclear imaging is substantially more expensive than EBCT, the pathway including myocardial perfusion studies as the primary tool will be more expensive. Although more patients with obstructive CAD may be overlooked with EBCT, a greater proportion of patients with subclinical atherosclerotic disease can be identified with this tool than with other technologies. Notably, a large proportion of patients in our study had evidence of atherosclerotic disease as demonstrated by presence of coronary calcifications on EBCT testing while failing to show inducible myocardial ischemia on ETT. Therefore, the benefits of performing EBCT as the initial diagnostic test may extend beyond the mere identification of flow-limiting stenoses and include the possibility of early detection of disease and aggressive modification of the risk factors responsible for the initiation and progression of CAD.26 Because the majority of acute coronary syndromes occurs in patients with noncritical arterial disease,27–29 patients with a CS of ⬍150 should be considered to have true CAD and should receive treatment with aspirin, statins, and probably ␤ blockade. Close clinical follow-up should be recommended and myocardial perfusion imaging advised for those patients who continue to experience anginal symptoms despite medical therapy. The small sample size of the patient cohort may be seen as a partial limitation of this study. Nevertheless, the reported cost savings and corresponding confidence intervals still suggest a sizeable cost advantage of the EBCT pathways compared with other traditional noninvasive strategies in patients with low to intermediate pretest probability of disease. Our analysis did not address the cost effectiveness of this new diagnostic approach in terms of complex variables, such as the improved quality-adjusted life expectancy of patients who underwent testing.24 Future long-term studies are necessary to further address this question.

1. Froelicher VF, Lehmann KG, Thomas R, Goldman S, Morrison D, Edson R, Lavori P, Myers J, Dennis C, Shabetai R, Do D, Froning J. The electrocardiographic exercise test in a population with reduced workup bias: diagnostic performance, computerized interpretation, and multivariate prediction. Ann Intern Med 1998;128:965–974. 2. Froelicher VF Jr. The application of electrocardiographic screening and exercise testing to preventive cardiology. Prev Med 1973;2:592– 607. 3. Detrano R, Gianrossi R, Mulvihill D, Lehmann K,Dubach P, Colombo A, Froelicher V. Exercise-induced ST depression in the diagnosis of coronary artery disease: a meta-analysis. Circulation 1989;80:87–98. 4. Hachamovitch R, Berman DS, Kiat H, Cohen I, Cabico JA, Friedman J, Diamond GA. Exercise myocardial perfusion SPECT in patients without known coronary artery disease. Incremental prognostic value and use in risk stratification. Circulation 1996;93:905–994. 5. Berman D, Hachamovitch R, Lewin H, Friedman J, Shaw L, Germano G. Risk stratification in coronary artery disease: implications for stabilization and prevention. Am J Cardiol 1997;79:10 –16. 6. Wexler L, Brundage B, Crouse J, Detrano R, Fuster V, Maddahi J, Rumberger J, Stanford W, White R, Taubert K. Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications. A statement for health professionals from the American Heart Association. Circulation 1996;94: 1175–1192. 7. Breen JF, Sheedy PF II, Schwartz RS, Stanson AW, Kaufmann RB, Moll PP, Rumberger JA. Coronary artery calcification detected with ultrafast CT as an indication of coronary artery disease. Radiology 1992;185:435– 439. 8. Rumberger JA, Sheedy PF, Breen JF, Fitzpatrick LA, Schwartz RS. Electron beam computed tomography and coronary artery disease: scanning for coronary artery calcium. Mayo Clin Proc 1996;71:369 –377. 9. Simons DB, Schwartz RS, Edwards WD, Sheedy PF, Breen JF, Rumberger JA. Non invasive definition of anatomic coronary artery disease by ultrafast computed tomographic scanning: a quantitative pathologic comparison study. J Am Coll Cardiol 1992;20:1118 –1126. 10. Mautner GC, Mautner SL, Froehlich J, Feuerstein IM, Proschan MA, Roberts WC, Doppman JL. Coronary artery calcification: assessment with electron beam CT and histomorphometric correlation. Radiology 1994;192:619 – 623. 11. Rumberger JA, Simons DB, Fitzpatrick LA, Sheedy PF, Schwartz RS. Coronary artery calcium area by electron-beam computed tomography and coronary atherosclerotic plaque area. A histopathologic correlative study. Circulation 1995;15:2157–2162. 12. Sangiorgi G, Rumberger JA, Severson A, Edwards WD, Gregoire J, Fitzpatrick LA, Schwartz RS. Arterial calcification and not lumen stenosis is highly correlated with atherosclerotic plaque burden in humans: a histologic study of 723 coronary artery segments using non-decalcifying methodology. Electron beam computed tomography and coronary artery disease: scanning for coronary artery calcification. J Am Coll Cardiol 1998;31:126 –133. 13. Rumberger JA, Behrenbeck T, Breen JF, Sheedy PF. Coronary calcification by electron beam computed tomography and obstructive coronary artery disease: a model for cost and effectiveness of diagnosis as compared with conventional cardiac testing methods. J Am Coll Cardiol 1999;33:453– 462. 14. Diamond GA, Forrester JS. Analysis of probability as an aid in the clinical diagnosis of coronary artery disease. N Engl J Med 1979;300:1350 –1358. 15. Mark DB, Shaw L, Harrell FE Jr, Hlatky MA, Lee KL, Bengtson JR, McCants CB, Califf RM, Pryor DB. Prognostic value of a treadmill exercise score in outpatients with suspected coronary artery disease. N Engl J Med 1991;325: 849 – 853. 16. Mark DB. An overview of risk assessment in coronary artery disease. Am J Cardiol 1994;73:19B-25B. 17. Shaw LJ, Hachamovitch R, Iskandrian AE. Treadmill test scores: attributes and limitations. J Nucl Cardiol 1997;4:76 –78. 18. Shaw LJ, Peterson ED, Shaw LK, Kesler KL, DeLong ER, Harrell FE Jr, Muhlbaier LH, Mark DB. Use of a prognostic treadmill score in identifying diagnostic coronary disease subgroups. Circulation 1998;98:1622–1630. 19. Maddahi J, Gambhir SS. Cost-effective selection of patients for coronary angiography. J Nucl Cardiol 1997;4:S141–151. 20. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano

CORONARY ARTERY DISEASE/EBCT IN INTERMEDIATE PROBABILITY PATIENTS

287

R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15:827– 832. 21. Rumberger JA, Sheedy PF, Breen JF, Schwartz RS. Electron beam computed tomography calcium score cutpoints and severity of associated angiographic luminal stenosis. J Am Coll Cardiol 1997;29:1542–1548. 22. Guerci AD, Spadaro LA, Goodman KJ, Lledo-Perez A, Newstein D, Lerner G, Arad Y. Comparison of electron beam computed tomography scanning and risk factor assessment for the prediction of angiographic coronary artery disease. J Am Coll Cardiol 1998;32:673– 679. 23. Arad Y, Spadaro LA, Goodman K. Predictive value of electron beam computed tomography of the coronary arteries. 19 month follow-up of 1173 asymptomatic subjects. Circulation 1996;93:1951–1953. 24. Patterson RE, Eisner RL, Horowitz SF. Comparison of cost-effectiveness and utility of exercise ECG, single photon emission computed tomography, positron emission tomography, and coronary angiography for diagnosis of coronary artery disease. Circulation 1995;91:54 – 65. 25. Berman DS, Kiat H, Friedman JD, Diamond G. Clinical application of

288 THE AMERICAN JOURNAL OF CARDIOLOGY姞

VOL. 85

exercise nuclear cardiology studies in the era of health care reform. Am J Cardiol 1995;75:3D-13D. 26. Callister TQ, Raggi P, Cooil B, Lippolis NJ, Russo DJ. Effect of HMG-CoA reductase inhibitors on coronary artery disease as assessed by electron beam computed tomography. N Engl J Med 1998;339:1972–1978. 27. Davies MJ, Thomas AC. Plaque fissuring—the cause of acute myocardial infarction, sudden ischaemic death, and crescendo angina. Br Heart J 1985;53: 363–373. 28. Papanicolaou MN, Califf RM, Hlatky MA, McKinnis RA, Harrell FE Jr, Mark DB, McCants B, Rosati RA, Lee KL, Pryor DB. Prognostic implications of angiographically normal or insignificantly narrowed arteries. Am J Cardiol 1986; 58:1181–1187. 29. Ambrose JA, Tannenbaum MA, Alexopoulos D, Hjemdahl-Monsen CE, Leavy J, Weiss M, Borrico S, Gorlin R, Fuster V. Angiographic progression of coronary artery disease and the development of myocardial infarction. J Am Coll Cardiol 1988;1:56 – 62.

FEBRUARY 1, 2000