ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 12, No. 4, 197-204, 1998 
Does myocardial thallium-201 SPECT combined with electron beam computed tomography improve the detectability of coronary artery disease? - Comparative study of diagnostic accuracy -
Keiko AOYAGI,* Tomio INOUE,* Yasuhiko YAMAUCHI,** Tsutomu IWASAKI** and Keigo ENDO* 
*Department of Nuclear Medicine, Gunma University School of Medicine **Second Department of Internal Medicine, Gunma University School of Medicine 
Objective. The aim of this study is to evaluate the diagnostic accuracy of myocardial 201Tl SPECT combined with EBT for detecting CAD. 
Methods. The study was based on 34 patients with suspected CAD, who had EBT and myocardial 201Tl SPECT. The CAD was diagnosed by the findings of coronary arteriography. Sensitivity, specificity and accuracy of EBT, myocardial 201Tl SPECT and the combined diagnosis on a per vessel basis and a per-patient basis were studied. 
Results. The sensitivity for detecting CAD of myocardial 201Tl SPECT, EBT and the combined diagnosis was 85%, 77%, and 62%, respectively. No significant difference in the accuracy of myocardial 201Tl SPECT, EBT and the combined diagnosis was observed on a patient basis and per vessel basis. In the over 70 yr age subgroup, the sensitivity and accuracy of EBT for detecting LAD lesion were significantly superior to those of myocardial 201Tl SPECT. Regardless of age-based subgroups and gender, the combined diagnosis did not contribute to an improvement in diagnostic accuracy. 
Conclusion. Although the sensitivity of EBT for detecting LAD lesion in patients over 70 yr of age was significantly higher than that of myocardial 201Tl SPECT, in the detectability of CAD, combined use of myocardial 201Tl SPECT and EBT offers no improvement. 
Key words: coronary artery disease, single photon emission computed tomography, thallium-201 chloride, electron beam computed tomography 
INTRODUCTION 
CORONARY ARTERY DISEASE is the leading cause of morbidity and mortality in industrialized nations. Not only invasive screening tests for detecting coronary artery disease (CAD) but also noninvasive screening tests have been performed. Even though exercise SPECT with 201Tl chloride (myocardial 201Tl SPECT) and exercise ECG with a treadmill are useful for screening tests in detect-
Received February 18, 1998, revision accepted July 6, 1998. For reprints contact: Keiko Aoyagi, M.D., Department of Nuclear Medicine, Gunma University School of Medicine, 3-39-22 Showa-machi, Maebashi, Gunma 371-8511. JAPAN. 
ing CAD, the results still included 10-25% false negatives.1-10 
 The recent development of electron beam computed tomography (EBT) enables us to detect CAD by assessing the degree of calcification in coronary arteries,1-19 but some pathologic studies and studies of EBT results showed that the severity of calcification did not always indicate the degree of lumen stenosis.15,20 Although only a few researchers compared the diagnostic accuracy of myocardial 201Tl SPECT with that of EBT,1,21 the clinical usefulness of myocardial 201Tl SPECT combined with EBT is still controversial. We had a chance to compare myocardial 201Tl SPECT and EBT. The aim of this study is to evaluate the clinical usefulness of myocardial 201Tl SPECT combined with EBT for detecting CAD. 
MATERIALS AND METHODS 
The study was based on 34 patients with suspected CAD, who received myocardial 201Tl SPECT, EBT and coronary arteriography (CAG) consecutively within a week. Catheterizations were performed in all patients because of the symptoms, abnormal ECG results and heart failure. 
 The study group comprised 20 men and 14 women whose age ranged from 37 to 82 yr with a mean age of 63 yr. The CAD was diagnosed by significant stenotic change in coronary arteriography, defined as having more than 75% narrowing in any major coronary artery. Seventeen out of 34 patients had a myocardial infarction. Patients with cardiomyopathy or old myocardial infarction with a normal coronary arteriogram were not included in this study. 
Electron beam computed tomography (EBT) 
From the level of the carina to the cardiac apex, forty contiguous 3-mm thick slices were taken at a 100-ms scan time in volume mode with Imatron C-150 (Imatron company, South San Francisco, CA). Transaxial images were reconstructed by the filtered back projection method, having a 512 x 512 matrix on each tomogram. Regions of interest (an area >= 0.51 mm2) were drawn over the area corresponding to three vessels (Left anterior descending artery (LAD), Left circumflex artery (LCx) and Right coronary artery (RCA)) on each slice to obtain the calcification score. The calcified lesion on the EBT image was defined as the area including a pixel with more than 130 Hounsfield units (HU) of CT density. The maximal density in each ROI was ranked from 1 to 4 as follows; 1: 130-199 HU, 2: 200-299 HU, 3: 300-399 HU and 4: 400 HU<=. A lesion score was obtained by multiplying the value for maximal density by the number of pixels of the calcified lesion in mm2. The total calcium score for each branch and all branches was the sum of lesion scores for slices involving coronary arteries. The calcium index was the natural log (1 + calcium score) for analyzing the diagnostic accuracy of EBT. 1,14 
Single photon emission computed tomography (SPECT) 
All patients fasted overnight before the myocardial 201Tl SPECT. During patients' exercise with the ergometer, 111 MBq of 201Tl chloride was injected intravenously when at least one of the following end points was reached; 1) ECG leads showing 0.1 mV < flat or downsloping ST segment depression; 2) Patients complaining of angina; 3) Achievement of 85% of the predicted maximal heart rate.22,23 After patients' heart rate returned to the base line condition, early images were taken with a triple head gamma camera (Prism 3000, Picker, Ohio) attached to a low-energy general purpose collimator (spatial resolution = 14.9 mm FWHM, on the center of the reconstructed image). A 20% symmetrical energy window centered on the 80 keV photopeak was used. These gamma cameras continuously rotated at 5deg. intervals and took 40 sec to obtain the data for the early images and 45 sec for the delayed images. The projection data were used to reconstruct the SPECT images with filtered back projection by means of a low-pass filter. Image matrices were 64 x 64. Short axial, horizontal long axial and vertical long axial images were displayed on the film. In addition 74 MBq of 201Tl was administered 10 min before the delayed imaging, 3 hours after the first 201Tl injection, which was conducted in the same way as the early stress imaging. 
 Tomograms were divided into 20 myocardial segments (Fig. 1). Each segment was assigned to one of the three major vascular territories; LAD territory included the anterior wall (segs. l, 2, 3, 5, 6 and 7), septum (segs. 8, 9 and 10) and apical (seg. 4), LCx territory included lateral wall (segs. 11, 12, 13, 14, 15 and 16, RCA territory included the inferior wall (segs. 17, 18 and 19) and apex (seg. 20). 
 For the myocardial 201Tl SPECT, regional 201Tl uptake was visually analyzed and scored by means of a four-point scoring system (0 = normal tracer uptake, 1 = faint tracer uptake, 2 = moderate reduction of tracer uptake, 3 = severe reduction of tracer uptake). The maximum 201Tl uptake score was used to evaluate the diagnostic accuracy of myocardial 201Tl SPECT in each territory or the entire myocardium. 
 SPECT images were interpreted independently by two experienced nuclear physicians who had no prior information about the results of CAG, calcification scores by EBT or the patient's history. The examination was followed by a discussion until a consensus was reached. 
Coronary arteriography (CAG) 
Coronary arteriography was performed in multiple projections by means of standard techniques. And coronary angiograms were analyzed at weekly conferences by several experienced cardiologists who had no information about calcification scores. The lesions in each coronary vessel were visually evaluated. The degree of coronary artery stenosis was decided by multiple projections to demonstrate the maximal coronary artery luminal narrowing. Angiographic abnormalities were considered as significant stenosis when >= 75% luminal diameter stenosis was found in LAD, LCx and RCA vessels. 
Data Analysis 
On a patient-based diagnosis of the EBT image, a patient who had any individual coronary artery branch with a transformed calcification score of 4 or higher was diagnosed as CAD (1). And for per-vessel-based diagnosis, the territory with a transformed calcification score of 4 or higher was diagnosed as CAD.1 The patient or the territory with a 201Tl uptake score higher than 2 on the myocardial 201Tl SPECT image was defined as CAD. The final diagnosis for CAD was based on the results of CAG. Sensitivity, specificity and accuracy for detecting CAD in patients and each vessel were compared for EBT, myocardial 201Tl SPECT diagnosis, and the combined diagnosis by means of myocardial 201Tl SPECT and EBT. The diagnostic accuracy was also compared in subgroups based on gender and age (less than 50 yr, 50-69 yr and over 70 yr). 
 Regarding the combined diagnosis of EBT and myocardial 201Tl SPECT, patients or segments with positive results on both modalities were diagnosed as CAD. In assessment of each territory, evaluated segments with infarction defined as akinesis with echocardiograms or left ventricular graphy were excluded to assess the effect of exercise on myocardial 201Tl SPECT sensitivity in detecting ischemia. 
Statistical Analysis 
All values showing diagnostic accuracy were reported as percentages. Data were analyzed by the McNemar-test with 1 degree of freedom for comparing categorical values.24-26 All tests of statistical significance were defined at a probability of less than 0.05. 
RESULTS 
Eight of thirty-four patients (24%) showed normal coronary arteries on CAG. The CAG revealed single vessel disease in 12 patients (35%), two vessel disease in 8 patients (24%), and three vessel disease in 6 patients (18%) (Table 1). Regarding individual coronary arteries, 20 arteries of LAD (59%), 13 of LCx (38%) and 13 of RCA (38%) showed significant stenosis. Regarding the patients-basis diagnosis, sensitivity of myocardial 201Tl SPECT and EBT for detecting CAD was 85 and 77%, respectively (Fig. 2a). The combined diagnosis of myocardial 201Tl SPECT and EBT had a sensitivity of 62% which was significantly lower than that of myocardial 201Tl SPECT. In sensitivity on the per-vessel basis, there was no significant difference among myocardial 201Tl SPECT, EBT and their combined diagnosis, except in LAD-65%, 71% and 35% (p < 0.05) (Fig. 2a). The specificity of EBT for LAD and RCA lesions was significantly lower than that of myocardial 201Tl SPECT (43% vs. 93% in LAD lesion and 52% vs. 90% in RCA lesion). There was no significant difference among myocardial 201Tl SPECT and the combined diagnosis (63% vs. 75% on a patient-basis, 93% vs. 100% in LAD lesion, 95% vs. 100% in LCx lesion and 90% vs. 90% in RCA lesion, respectively) (Fig. 2b). No significant difference in the accuracy of myocardial 201Tl SPECT, EBT and the combined diagnosis was observed on a patients-basis and a per-vessel basis (79%, 68%, 65% on a patient basis, 77%, 58%, 65% in LAD lesion, 76%, 74%, 71% in LCx lesion and 77%, 55%, 74% in RCA lesion, respectively) (Fig. 2c). 
 In the 50-69 yr age subgroup, the specificity and accuracy of EBT for detecting LAD lesions were significantly lower than those of myocardial 201Tl SPECT, but in the over 70 yr age subgroup, the sensitivity and accuracy of EBT for detecting LAD lesion were significantly superior to those of myocardial 201Tl SPECT (Table 2a). For detecting LCx lesions, there was no significant difference in the diagnostic accuracy of myocardial 201Tl SPECT, EBT and the combined diagnosis, regardless of age (Table 2b). In the 50-69 yr and over 70 yr age subgroup, the specificity of EBT for detecting RCA lesion was significantly lower than that of myocardial 201Tl SPECT (Table 2c). Regardless of the subgroup, the combined diagnosis did not contribute to improving diagnostic accuracy. 
 In a comparison of the diagnostic accuracy of myocardial 201Tl SPECT and EBT on a patient-basis grouped by gender, only sensitivity of the combined diagnosis in the male patients was significantly lower than that of myocardial 201Tl SPECT (Table 3). Regardless of gender, combined diagnosis did not contribute to the improvement of diagnostic accuracy. 
 Illustrative cases are shown in Figs. 3 to 5. 
DISCUSSION 
Electrocardiographic exercise testing (ECG), exercise 201Tl SPECT and stress echocardiograms are useful techniques for evaluating CAD. Overall sensitivity of ECG, myocardial 201Tl SPECT and echocardiograms was reported to be about 70%, 90% and 80%, and overall specificity was about 80%, 70% and 90%, respectively.1,4,7,8,21 When a patient with positive results on both "SPECT and exercise ECG" was defined as a patient with CAD, the sensitivity and specificity of complemental diagnosis by means of exercise 201Tl SPECT and exercise ECG were reported to be 68% and 83%, respectively.l 
 Our results for sensitivity and specificity on a per-patient basis of myocardial 201Tl SPECT were almost the same as those previously reported,1,4,7,8,21 and the sensitivity of EBT in this study for the diagnosis of CAD was also almost the same as that previously reported,1,12,18.21 but the specificity of EBT was 38%, which was lower than that previously reported, due to low specificity in detecting LAD lesion. EBT can detect fine calcification more easily in LAD (Fig. 5b) than in LCx, which may lead to a high percentage of false positive results for LAD lesion. 
 With respect to the per-vessel basis, the sensitivity of exercise myocardial 201Tl SPECT for detecting LAD, LCx, and RCA lesions was reported as 83%, 63% and 87%, and regarding EBT, 94%, 88% and 100%, respectively.3,22,27 The specificity of myocardial 201Tl SPECT for detecting LAD, LCx and RCA lesions was reported as 95%, 96% and 87%, and of EBT, 54%, 80% and 76%, respectively.3,22,27 Sensitivity on a per-vessel basis of both exercise myocardial 201Tl SPECT and EBT and specificity on a per-vessel basis of EBT was inferior to those previously reported, which may be due to the different age population and diagnostic criteria as the cutoff point for calcification. 
 The cutoff points for the calcium score determined by EBT varied among institutions.1,28 Our cutoff point for the log transformed calcification score for predicting significant stenosis was 4.0 which was interpreted as 53.6 when using the Agaston method. In the report by Kajinami et al.1 the cutoff point for the log transformed calcification score was 3.0 in all patients. Rumberger described the optimal cutoff point for the calcium score as 123 when using the Agaston method,28 which was based on the receiver operating characteristic curve analysis. When we discuss the optimal cutoff points for the calcium score, we must pay attention to the age distribution of the patients described in previous studies.1 In fact, since a constant cutoff point was employed regardless of age in the present study, the specificity of EBT was sacrificed along with age (Table 2). 
 Regarding the combined diagnosis of "myocardial 201Tl SPECT and EBT," there was no improvement in diagnostic accuracy in sensitivity, specificity or accuracy, irrespective of age or gender. In our study, noticeably low sensitivity for detecting LCx lesions was observed in myocardial 201Tl SPECT and EBT (Fig. 2a). With respect to myocardial 201Tl SPECT, small area perfusion supplied through LCx might yield to sensitivity, but it might not be related to the sensitivity of EBT because of the evaluation of proximal coronary arteries. We anticipated that the detectability of LCx lesion by EBT may improve the low sensitivity of myocardial 201Tl SPECT, but for LCx lesion, there was no significant difference of the sensitivity between myocardial 201Tl SPECT and EBT. 
 Although sensitivity for detecting LAD lesions was reduced by combining the results of myocardial 201Tl SPECT and EBT (Fig. 2a), the sensitivity of EBT in patients over 70 yr was significantly higher than that of myocardial 201Tl SPECT. This may indicate the clinical usefulness of EBT examination. And the specificity of myocardial 201Tl SPECT in detecting LAD and RCA lesions was superior to that of EBT. A previous study also reported that the specificity and accuracy of EBT were lower than those of myocardial SPECT with 99mTc-MIBI (specificity 77% vs. 95%; accuracy 71% vs. 85%).29 From the viewpoint of screening tests for CAD, there was no reason to combine the results of SPECT and EBT. 
CONCLUSION 
The clinical usefulness of combined diagnosis with myocardial 201Tl SPECT and EBT for CAD was investigated by statistical analysis for diagnostic accuracy. Although sensitivity in detecting LAD lesion in patients over 70 yr of age by means of EBT was significantly higher than that by means of myocardial 201Tl SPECT, in the detectability of CAD, combined use of myocardial 201Tl SPECT and EBT offers no improvement. 
ACKNOWLEDGMENTS 
The authors gratefully acknowledge the support and assistance of Hidenori Ootake, Masami Yamada, Hiroshi Kubota, Yoshihiko Hoshino, Noboru Hatori, Akira Hasegawa, MD, Takuji Toyama, MD, and Kumiko Tajima, MD of Gunma University School of Medicine. 
REFERENCES 
l. Kajinami K, Hamada S, Takekoshi N, Mabuchi H. Noninvasive prediction of coronary atherosclerosis by quantification of coronary artery calcification using electron beam computed tomography: Comparison with electrocardiographic and thallium exercise stress test results. J Am Coll Cardiol 26: 1209-1221, 1995. 
2. Massie BM, Botvinick EH, Brundage BH. Correlation of thallium-201 scintigrams with coronary anatomy: factors affecting region by region sensitivity. Am J Cardiol 44; 616-622, 1979. 
3. Tamaki N, Yonekura Y, Mukai T, Kodama S, Kadota K, Kanbara H, et al. Stress thallium-201 transaxial emission computed tomography: Quantitative versus qualitative analysis for evaluation of coronary artery disease. J Am Coll Cardiol 4: 1213-1221, 1984.  4. Taillefer R, DePuey EG, Udelson JE, Beller GA, Latour Y, Reeves F. Comparative diagnostic accuracy of Tl-201 and Tc-99m sestamibi SPECT imaging (perfusion and ECG-gated SPECT) in detecting coronary artery disease in women. J Am Coll Cardiol 29: 69-77, 1997. 
5. Verzijlbergen JF, Zwinderman AH, Ascoop CAPL, van der Wall EE, Niemeyer MG, Pauwels EKJ. Comparison of technetium-99m sestamibi left ventricular wall motion and perfusion studies with thallium-201 perfusion imaging: in search of the combination of variables with the highest accuracy in predicting coronary artery disease. Eur J Nucl Med 23: 550-559, 1996. 
6. Kiat H, Maddahi J, Roy LT, Train KV, Friedman J, Resser K, et al. Comparison of technetium-99m-methoxy isobutyl-isonitrile and thallium-201 for evaluation of coronary artery disease by planar and tomographic methods. Am Heart J 117: 1-11, 1989. 
7. Aldrich RF, Brensike JF, Battaglini JW, Richardson JM, Loh IK, Stone NJ, et al. Coronary calcifications in the detection of coronary artery disease and comparison with electrocardiographic exercise testing. Results from the National Heart, Lung and Blood Institute's type II coronary intervention study. Circulation 59: 1113-1124, 1979. 
8. Verani MS, Marcus ML, Razzak MA, Ehrhardt LC. Sensitivity and specificity of thallium-201 perfusion scintigrams under exercise in the diagnosis of coronary artery disease. J Nucl Med 19: 773-782, 1978. 
9. Iskandrian AS, Heo J, Kong B, Lyons E, Marsch S. Use of technetium-99m-isonitrile (RP30) in assessing left ventricular perfusion and function at rest and during exercise in coronary artery disease, and comparison with coronary arteriography and exercise thallium-201 SPECT imaging. Am J Cardiol 64: 270-275, 1989. 
10. Villanueva-Meyer J, Mena I, Narahara K. Simultaneous assessment of left ventricular wall motion and myocardial perfusion with technetium-99m-methoxy isobutyl-isonitrile at stress and rest in patients with angina: comparison with thallium-201 SPECT. J Nucl Med 31: 457-463, 1990. 
11. Eggen DA, Strong JP, McGill HC. Coronary calcification: relationship to clinically significant coronary lesions and race, sex, and topographic distribution. Circulation 32: 948-955, 1965. 
12. Breen JF, Sheedy PF, Schwartz RS, Stanson AW, Kauffman RB, Moll PP, et al. Coronary artery calcification detected with ultrafast CT as an indication of coronary artery disease. Radiology 185: 435-439, 1992. 
13. Baron MG. Significance of coronary artery calcification. Radiology 192: 613-614, 1994. 
14. Kajinami K, Hamada S, Takekoshi N, Mabuchi H. Quantification of coronary artery calcification using ultrafast computed tomography: reproducibility of measurements. Coron Artery Dis 4: 1103-1108, 1993. 
15. Janowitz WR, Agaston AS, Viamonte M Jr. Comparison of serial quantitative evaluation of calcified coronary artery plaque by ultrafast computed tomography in persons with and without obstructive coronary artery disease. Am J Cardiol 68: 1-6, 1991. 
16. Simons DB, Schwartz RS, Edwards WD, Sheedy PF, Breen JF, Rumberger JA. Noninvasive definition of anatomic coronary artery disease by ultrafast computed tomographic scanning: A quantitative pathologic comparison study. J Am Coll Cardiol 20: 1118-1126, 1992. 
17. Wong ND, Detrano RC, Abrahamson D, Tobis JM, Gardin JM. Coronary artery screening by electron beam computed tomography. Facts, controversy and future. Circulation 92: 632-636, 1995. 
18. Budoff MJ, Georgiou D, Brody A, Agatston AS, Kennedy J, Wolfkiel C, et al. Ultrafast computed tomography as a diagnostic modality in the detection of coronary artery disease. A multicenter study. Circulation 93: 898-904, 1996. 
19. Simon A, Giral P, Levenson J. Extracoronary atherosclerotic plaque at multiple sites and total coronary calcification deposit in asymptomatic men. Association with coronary risk profile. Circulation 92: 1411-1421, 1995. 
20. Mautner GC, Mautner SL, Froehlich J, Feuerstein IM, Proschan MA, Roberts WC, et al. Coronary artery calcification: Assessment with electron beam CT and histomorphometric correlation. Radiology 192: 619-623, 1994. 
21. Patterson RE, Horowitz SF, Eisner RL. Comparison of modalities to diagnose coronary artery disease. Semin Nucl Med XXIV: 286-310, 1994. 
22. DePasquale EE, Nody AC, Depuey EG, Garcia EV, Pilcher G, Bredlau C, et al. Quantitative rotational thallium-201 tomography for identifying and localizing coronary artery disease. Circulation 77: 316-327, 1988. 
23. Schlant RC, Froelicher VF Jr, Blomqvist CG, Hall RJ, Brandenburg RO, McCallister BD, et al. Guidelines for exercise testing a report of the American College of Cardiology/American Heart Association. Task force on assessment of cardiovascular procedures (Subcommittee on exercise testing). J Am Coll Cardiol 8: 725-738, 1986. 
24. Dwyer AJ. Matchmaking and McNemar in the comparison of diagnostic modalities. Radiology 178: 328-330, 1991. 
25. Lilienfeld DE, Stolley PD. Foundations of Epidemiology. New York Oxford; Oxford University Press, pp. 122-131, 1994. 
26. Armitage P, Berry G. Statistical Methods in Medical Research. Oxford; Blackwell Scientific Publications, pp. 402-468, 1994. 
27. Hamada S, Takamiya M, Haruo S. Evaluation of coronary artery calcification by ultrafast CT. Nippon Acta Radiologica 51: 1299-1305, 1991. 
28. Rumberger JA, Sheedy PF, Breen JF, Schwartz RS. Electron beam computed tomographic coronary calcium score cutpoints and severity of associated angiographic lumen stenosis. J Am Coll Cardiol 29: 1542-1548, 1997. 
29. Yao Z, Liu XJ, Shi R, Dai R, Zhang S, Liu Y, et al. A comparison of 99mTc-MIBI myocardial SPET with electron beam computed tomography in the assessment of coronary disease. Eur J Nucl Med 24: 1115-1120, 1997.