Thallium-201 Single-Photon Emission Computed Tomography (SPECT) in the Assessment of Coronary Artery Disease Mario S. Verani,

Of all currently available techniques, thalliuml 201sin@e-photon em&&n computed tomqgraphy (SPECT)is the mosf tim&wtwl noninvasive methodforthe detwtlonofcoronaryartwydOsl ease (CAD). Recent pooled data show an overall sensltMtyof9o?~andaspedfkStyof70%for thallium-201 SPEC’L Of patlentswtth singlevessel coronary dbmase, 83% am identifhJ by SPECT.NearIy all patients with double- and tripleL vessel coronary d&wase (93% and 95%, respectidy) am also ldeWfhl. Ihallium~201 SPECT imagbg is also very effwtlve in diagnosing CAD uslngpharmacokgkstress testing. In certain patlent pqBulatlons (agm, in sedentary patients or thc#re using anti-kchemk medkatIuns), phamracologk stress testingwith dlpyridamole or atinosine may Ibg a l0gkal aRemaUve to exetesting. Moreover, many patients have physkal disabilitiesthatpwwl~apprapriateexen=ise testing. lntfavenous adenosMe k a very potent direct coronaryvasodilator, with tM advantage of an uttrashort haWife, which elhninates the need to administer an antagonkt In the majority of patients. In addltien, thwl0sageufadwwsine can be ad/us&d durling the infuslun, if necessary. Tke impwtance ofthalllum~201 SPECTduring 0xerdse of pharmacol0gk vasoMaWn transcends diagnosis, since It also plays an Important role in the pm waluatlon of patients withstabkangina or pwtmyocardial Infarction, R&k evaluation can be done with submaximal exeerciswlectwcardio@aphktestln&butthwe hmvldencethat th0additionofpwfiisknscintigraphy enhances the ability to predict future risk. In patients unable to exW or to rec&e dIpyr= idamoleoradenoslnwtress,dobutamlnehas

From the Section of Cardiology, Department of Medicine, Baylor College of Medicine, and Nuclear Cardiology, The Methodist Hospital, Houston, Texas. Address for reprints: Mario S. Verani, MD, Nuclear Cardiology, Baylor College of Medicine, 6535 Fannin, F-905, Houston, Texas 77030.

MD

receMlywneqgwlasyetanotheraltwnatlve.Alissmallwtth thoughthereportedexpwknce *butamine thallium p&uWn imagln& the test appears s&, well&Wrat~, and has good 88HsE tivttyandspecifkityforCADd&ecWn. (Am J Cardlol1992;70=3E-9E)

T

hallium-201 (*OlTl) myocardial perfusion imaging has been widely used in the United States and abroad for the past 15 years. Recently, there has been renewed interest in this technique, despite predictions that new compounds labeled with technetium-99m would supplant 201Tl There are several reasons1 for this renewed interest in 201T1m . (I) improved myocardial images have resulted from the advent of scanning with single-photon emission computed tomography (SPECT); (2) recent improvements have been reported in quantification of the SPECT images; (3) the popularization of perfusion imaging in conjunction with pharmacologic “stress” of the cardiovascular system; and (4) the use of reinjection techniques to assessmyocardial viability better. As a consequence, the number of myocardial perfusion imaging procedures continues to increase in many laboratories around the country. This review summarizes recent findings on the diagnostic value of 2*1T1SPECT during exercise and of thallium scintigraphy in general during pharmacologic stress. l

SENSIT’MW AND SPECIFICITY OF THAUJUM~201 SPECT Data on *OIT1SPECT in conjunction with exercise from > 1,000 patients have been pooled by Mahmarian et al (Table X).2-sThese data show an overall sensitivity of 90% for detection of coronary artery disease (CAD), with a range of 82-98%. Nearly all patients (99%) with prior myocardial infarction had abnormal SPECT images, but even in patients without prior myocardial infarction, the A SYMPOSIUM: CARDIAC IMAGING

3E

TABLE I Sensitivity and Specificity of Thallium-201 SPECT for Detecting Coronary Artery Disease Sensitivity* Study

% with MT

Tamaki4 (n = 104)

39

DePasquale5 (n = 210)

25

lskandrian et aI6 (n = 461)

18

Madda hi7 (n = 138)

47

Mahmarian et al2 (n = 360)

33

Van Train8 (n = 318)

40

TOTAL

31

Overal I 98 (80/82)

95 (17011791 82 (224/272) 95 (87/W

87 (1921221) 94 (1851196)

Mt

100 100 (47/47) 98 (49150~ 100 (43/43) 99 (73/74) 100

92 (123/134) 78 (174/222) 90 (44/49) 79

91 (85/W 64 (45/70) 83

ggt

(93811,042)

(322/3241

TVD

(15118)

68/m 90 (106/l 18) 85 (5631659)

disease; MI = myocardial

infarction;

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

Specificity

NormaIcy Rate

91 -

W/78)

DVD

96 (48/50)

sensitivity of zolTl SPECT for detecting CAD was quite high (average 85%). The more extensive the CAD, the more likely it will be detected by SPECT. In the pooled data, 83% of patients with single-vessel disease and nearly all patients with double- (93%) and triple(95%) vessel disease were identified by SPECT. The detection of all individual vessels with coronary stenosis is good, but not quite as high. Approximately 80% of all stenotic vessels are detected with SPECT, although detection of circumflex artery stenosis is lower. Another important observation is that the sensitivity increases as the severity of coronary stenoses progresses from mild ( < 50% luminal narrowing) to moderate (5&70% luminal narrowing) to severe stenosis (> 70% luminal narrowing). SPECT allows detection of nearly 90% of all stenoses with > 70% luminal narrowing.2 Analysis of pooled data* shows an overall specificity of 70% for detection of CAD with *OITl SPECT during exercise, with a range on the order of 50-90%. Thus, in some laboratories, a poor specificity for SPECT may be the main limitation of the technique. The number of false-positive interpretations may be substantially decreased, however, by paying attention to pertinent history, such as the patient’s gender and body habitus and the presence of anatomic factors leading to increased extracardiac attenuation. In our laboratory, quantification of myocardial perfusion images using an appropriate bank of normal perfusion data has enabled us to obtain a specificity of 90%.3 It has been previously emphasized that in many laborato4E

SVD

W/32)

90

*Actual numbers are shown in parentheses. tp = 0.0001 vs no MI; $p = 0.0001 vs SVD. CAD = coronary artery disease; DVD = double-vessel lVD = triple-vessel disease. Reprinted with permission from Am J Cardiol.*

No MI

(%>

99

U’2/73 87 (931107) 97

(32133)

84 (119/142) 88

W/66)

(56/W

(69/721

83 (3201387)

91 96 9% (326/351)

SPECT = single-photon

100

(13/13) 91 W/95) 98 (40/41) 100 (13/13) 100

(20122)

-

74 (23/31) 60

94

(35/58)

(1231131)

56

86

W/W

WI/281

m/60)

87 (65/75) 43 (15135)

82 (62/76)

9% (2 12/222)

70 (168/239)

89 (2091235)

emission computed

tomograhy;

SVD = single-vessel

disease;

ries an abnormal perfusion scan is often used as a reason to perform coronary angiography. This results in a significant post-test selection bias, by which patients with normal perfusion scans typically do not undergo coronary angiograms, whereas those with an abnormal scan often go on to have an angiogram. Such a selection bias decreases the apparent specificity of the test. To overcome this problem, some investigators prefer to report “normalcy rate,” i.e., the percentage of low-risk normal individuals with a normal scan. The normalcy rate is approximately 90% by 201T1SPECT.

QUANTlFlCAIlON OF TMAWUM-201 SPECT IMAGES In our laboratory, we use a polar-plot display method, which is a variation of the method originally described by Garcia et al9with certain innovations that we have implemented.3 In summary, circumferential profiles are obtained on each of the short axis slices and the apical vertical long axis slices. Each of the slices is displayed concentrically on a “bull’s-eye” format, where the center of the bull’s-eye represents the apex of the left ventricle. The regions with normal myocardial activity are normalized to 100 and the abnormal regions scaled appropriately. Each patient’s polar map is compared to a normal data bank, and the regions with decreased thallium activity (those with activities > 2.5 standard deviations below the normal mean) are color-coded to represent perfusion defects. Such a technique allows for computer quantification of the total left ventricular perfusion defect

NOVEMBER 5, 1992

size (Figure l), as well as the amount of reversibility at 4 hours, 24 hours, or after thallium reinjection. EXERCISE THAUJUM~201 SPECT IN THE ASSESSMEW OF JEOPARDIZED IWOCARDIUM IN PATlEM’S WITH CORONARY ARTERY DISEASE Our group has recently evaluated the amount of myocardium jeopardized by stenoses of each of the 3 principal coronary arteries during exercise *OIT1 SPECT? In general, patients with left anterior descending artery stenosis have a larger myocardial perfusion defect size when compared with patients with right coronary artery or circumflex artery stenoses (Figure 2). In patients with left anterior descending artery stenosis, those in whom the stenosis is located in the proximal part of the artery usually have a larger myocardial perfusion defect than those with stenoses in the mid or distal location. Perhaps the most striking observation from this study, however, was the wide variation in myucardialpofusion defect sizescaused by stenosis uf similar severity and in similar locations. The implications of these data are that the coronary angiograms alone cannot predict the amount of myocardium in jeopardy. Thus, the knowledge of the functional expression of the coronary stenosis (i.e., the extent, severity, and reversibility of the perfusion defects) complements the information ob tained from the coronary angiograms. A further observation from our study’” is tha patients with more severe stenosis ( > 70% lumina diameter narrowing) in general have larger perfu sion defects than those with narrowings in simila r locations but of lesser severity ( < 70% stenosis).

FIGURE L Bu#‘eeye p&r map dthe thatttum-2Olmye cardtal actMty during exercise. The shmkd arm reprosents normattyprfused my0cardlum and the inner area In black the perluslon defect, which amouuted to 26% ofthe t&t ventricte In this patient with a praxlmal left anterior descmdtI?g(LpLD)stemk. t&x = teftctrcumflex; RCA = rtght coronary artery.

Here also, however, the coronary angiogram does not enable one to predict the extent or severity of the myocardial perfusion deficit. These observations are in keeping with the known difficulties in assessing the functional significance of angiographic coronary stenosisll and the variable contribution of coronary collaterals to the regional perfusion of different patients. One of the potential criticisms of using exercise 201Tl SPECT to assess the amount of jeopardized myocardium is that the perceived amount of jeopardized myocardium depends in large part on the intensity of the exercise test, ordinarily expressed by the maximal heart rate and double product

100% Stenosis 70199% Stenosis += 51-69% Stenosis O= Ml “p = .OoOl vs LAD PROX *p = .ows vs LAD PROX

60

I I

I

50 40 PDS (99 Lv)

30 00

I

IO or+

0 LAD PROX (N=32)

RCA PROX (N=29)

cx PROX (k12)

A SYMPOSIUM: CARDIAC IMAGING

5E

achieved. In other words, in patients with severe stenosis but with a poor exercise stress, the 201T1 images may not reflect the true extent of their jeopardized myocardium. In fact, it is to be expected that the intensity of the exercise stress plays a significant role in the sensitivity of the test? As a corollary of this observation, patients with significant coronary stenoses who perform an exercise test of low intensity may have an entirely normal 201T1scan.

PHARM&COLtbGlC MYOCARDIAL PERFUSION IMAGING As discussed, the intensity of the exercise test is critical in order to obtain the highest possible yield from the 201T1SPECT images. We estimate that 20-30% of patients with suspected or documented CAD cannot perform a maximal exercise test. Of course, diverse physical disabilities may entirely preclude many other patients from performing exercise testing. For these patients, pharmacologic perfusion imaging is the logical approach. Dip-& won I-B@ Dipyridamole is a potent, albeit indirect, coronary vasodilator that acts by blocking the cellular reuptake of adenosine, thereby increasing the blood and tissue concentrations of adenosine, which in turn produces maximal or near-maximal coronary vasodilation I2 In normal coronary arteries, dipyridamole increases the flow to 34fold times the baseline values (the so-called ?oronary vascular reserve”). In arteries with stenosis of >50% of the lumen diameter, the increase in flow will be proportionately less, however. In arteries with moderate stenosis (50-70% luminal narrowing), the coronary flow may still increase 2-3-fold above the resting values. On the other hand, in arteries with more severe stenosis ( > 70%), the flow increases much less or may not increase at all. This differential flow reserve in normal versus stenotic arteries provides the physiologic rationale for injecting imaging agents such as 201Tlthat are distributed according td the coronary blood fl~w.~~y~~When 201Tl is injected under conditions of coronary hyperemia, a differential distribution of 201Tl is observed, with greater amounts of 201Tltaken up by the normally perfused areas and lesser amounts by the abnormally perfused areas. Leppo15 has pooled previously reported data on dipyridamole planar perfusion imaging and found an overall sensitivity of 90% and specificity of 70% for CAD detection. The sensitivity and specificity of 201Tl imaging were similar during exercise, or l

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

dipyridamole stress testing in the patients who received both tests on different days. In a study from our laboratory, Borges-Neto et all6 have demonstrated high sensitivity and specificity of 201TlSPECT imaging following high doses of oral dipyridamole. Thus far, no reports are available describing the corresponding values with intravenous dipyridamole; one would anticipate similarly high sensitivity and specificity. In addition to its prominent role in the diagnosis of CAD, dipyridamole 201Tl SPECT imaging has been found to be very useful in the cardiovascular risk stratification of patients undergoing peripheral vascular operations and in patients recovering from an acute myocardial infarction. Recent evidence also suggests that pharmacologic vasodilation may be the ideal stress in patients suspected of having CAD and who also have a left bundle branch block. Patients with this conduction abnormality frequently show perfusion defects, usually involving the interventricular septum, that may closely mimic CAD. Interestingly, this is not a problem with dipyridamole imaging, and thus pharmacologic stress appears to be more specific than exercise scintigraphy in patients with left bundle branch block. Although dipyridamole perfusion imaging is of proven value and has a good safety record, it has two potential drawbacks. First, dipyridamole may not elicit maximal coronary vasodilation in up to 20% of patients. I7 Second ?the elimination half-life of intravenous dipyridamole is several minutes, so that the effects of the drug as well as the side effects may persist for several minutes after 201Tl administration. These side effects occur in approximately 50% of patients receiving dipyridamolell and in some cases are severe enough to require administration of the adenosine antagonist, aminophylline. Adendne perfusion imagln@ Recently, adenosine perfusion imaging has been proposed as an alternative to dipyridamole perfusion imagingal There are some distinct advantages of using adenosine as a coronary vasodilator. First, the ultrashort plasma half-life of adenosine (2-10 seconds) ensures that the effects of the drug will abate rapidly following termination of the test. Second, administration of adenosine may be titrated up or down, and the full effect of the new dose will be apparent in a matter of l-2 minutes. Third, adenosine may elicit more consistent maximal coronary vasodilation than dipyridamole. l6 Our initial protocol with adenosine SPECT imaging used a stepwise titration of adenosine, beginning with a dose of 50

NOVEMBER 5, 1992

pg/kg/min and increasing the dose every minute to a maximum of 140 kg/kg/min! In this protocol, 201T1was injected after 1 minute at the highest dose, which was then continued for an additional 2 minutes and then discontinued. This protocol proved to be very safe and yielded a high sensitivity and specificity for CAD detection. Subsequently, the protocol was simplified to a continuous infusion of 140 pg/kg/min for a total of 6 minutes, with thallium administered 3 minutes after the infusion

started. This protocol has now been used in several thousand patients in the United States with an excellent safety record, high sensitivity and specificity, and excellent comparability with exercise 201T1 SPECT imaging. 19,2oA recent report from our laboratory by Nishimura et a121with computerized quantification of the SPECT images showed a sensitivity of 87% and specificity of 90% (Figure 3). Nguyen et a122and Iskandrian et a12” have also reported large series of patients receiving adeno-

FWURE 3. Tblllum-20% slmephoton 8mlsslon computed tol mography (SECT) lmag@s during adenosine Infusion (b@ andat rtdsH@h (bottom) in a patlentwlth left anterior degcendlng stm. In each part, three repr8s8ntatlvesllc8fsar0simwn In the shwt axls (upperrow), horlzuntal lung axls (tiddk mw), and vertical loa axis (bwer row). A defect (In orange color) lmrdved theantetlw (ANT)wall and apex In the adenoeine lmagesaml flil8d-In In the redktrlbution lma@~. INF = Interior; SEPT = septum. (R0prlnbd with pen&sslon from I Am CoN C3iW0l.~~)

A SYMPOSIUM:

CARDIAC

IMAGING

7E

sine 201TlSPECT in whom the sensitivity was 90% and specificity S&b. Side effects are common during administration of adenosine to patients?24 The most frequent side effects are facial flushing, chest pain, headaches, and dyspnea. Although frequent, these side effects disappear within l-2 minutes of discontinuing the infusion and almost never require administration of aminophylline. The known negative dromotropic effect of adenosine in the atrioventricular node accounts for an occurrence of firstdegree atrioventricular block in approximately 10% of patients and a 3.6% frequency of second-degree atrioventricular block.24Occasionally ( < 1%), thirddegree atrioventricular block occurs. Fortunately, aswith the other side effects, atrioventricular block is transient and often disappears even when the infusion is continued. When necessary, the infusion may be terminated and the atrioventricular block will typically abate within several seconds. Although most of the experience thus far reported with adenosine has been with 201TlSPECT perfusion imaging, adenosine may also be used in combination with technetium-99m sestamibi (Verani et al, unpublished observations) or technetium99m teboroxime imaging.25Gupta et a126have used adenosine stressin combination with positron emission tomography (PET) with excellent results. At the time of this writing, adenosine has yet to be approved by the Food and Drug Administration for use in conjunction with perfusion imaging, although adenosine is currently approved for emergency therapy of supraventricular tachycardia. Debutamine

myocardial

peffwhn

imagln@

tion with radionuclide angiography by Freeman et a1.2gRecently, Pennell et a130reported on the utility of dobutamine in conjunction with 201TlSPECT (maximal dobutamine dose 20 kg/kg/min). Again, high sensitivity (97%) and specificity (80%) were reported. Dobutamine stress imaging is performed in our laboratory using the following protocol. Dobutamine is infused intravenously at an increasing dose of 5, 10, 20, 30, and 40 pg/kg/min, with increases every 3 minutes. At the maximal tolerated dose, 201Tlis injected and dobutamine maintained for an additional 2 minutes. The initial SPECT images are obtained immediately after stress,and the redistribution images are obtained 4 hours later. In a series of 144 patients recently studied in our laboratory, dobutamine stress was well tolerated, despite a high frequency of side effects. Sensitivity and specificity were 87% and 90%, respectively.31 Thus, dobutamine represents a third alternative for pharmacologic perfusion imaging in patients who are unable or unwilling to exercise. Together, these 3 drugs have considerably broadened the application of myocardial perfusion imaging. REFERENCES L Verani MS. Tl-201 myocardial perfusion imaging Czar Opin h&Z 19913: 797m. 2. Mahmarian JJ, Verani MS. Exercise thallium-201 perfusion scintigraphy in the assessmentof ischemicheart disease.Am J Catdiol1991;67(suppt):2D-11D. 3. Mahmarian JJ, Boyce TM, Goldberg RK, Ckanougher MK, Roberts R, Verani MS. Quantitative exercise thallium-201 single-photon emission cornputed tomography for the enhanceddiagnosisof ischemicheart disease.J Am co11cardib1199&15:31&329. &Tan&i N, Yonekura Y, Mukai T, Kodama S, Kadota K, Kambara H, Kawai C, Torizuka K Stressthallium-201transaxial emissioncomputed tomography: quantitative versus qualitative analysis for evaluation of coronary dis-

Although dipyridamole and adenosine represent ease. JAm Coil Cardbl1984;4:121>1221. important advancesthat have broadened the appli- 5. DePasqualeEE, Nody AC, DePuey EG, Garcia EV, Pilcher G, Brediau C, cation of myocardial perfusion imaging in patients Roubin G, G&er A, Gruentzig 4 D’amato P, Berger W. Quantitative rotathallium-201 tomography for identifying and localizing coronary artery who could not previously undergo exercise stress tional disease.circulation 1988;77:3&327. testing, both dipyridamole and adenosine are con- 8. Iskandrian AS, He0 J, Kong B, Lyons E. Effect of exercise level on the of thallium-201 tomographic imaging in detecting coronary artery distraindicated in patients with asthma or bronchos- ability ease:analysisof 461 patients.JAm Cd Car&i 1989;14:1477-1486. pasm and must be used carefully in patients with 7. Maddahi J, Van Train K, Prigent F, Garcia EV, Friedman J, Ostrzega E, low resting systolic blood pressures (< 100 mm Berman D. Quantitative sin@ephoton emissioncomputed thallium-201tomogfor detection and localization of coronary artery disease:optimization Hg). Further, neither dipyridamole nor adenosine raphy and prospectivevalidation of a new technique.J Am CulI Cardzbl1989;14:1689should be used in patients who have recently used 1699. Van Train KF, Maddahi J, Berman DS, Fiat H, Arreda J, Prigent F, methylxanthine compounds (theophylline or caf- 8.Friedman J, and the Participants of the Multicenter Trial. Quantitative analysis feine) because these drugs antagonize the adeno- of tomographicstressthallium-201myocardial scintigrams:a multicenter trial. J sine receptors in the cell membrane and thus may Nucl Med 1990,31:116&1179. 9. Garcia EV, VanTrain K, Maddahi J, Prigent F, Friedman J, Areeda J, preclude the adenosine-induced vasodilation. Waxman A, Berman DS. Quantification of rotational thallium-201 myocardial Mason et a127were the first to report dobu- tomography.J hruclMed 1985;26:17-26. 10. Mahmarian JJ, Pratt CM, Boyce TM, Verani MS. The variable extent of tamine stress in combination with planar 201T1 jeopardized myocardium in patients with single vessel coronary artery disease: perfusion scintigraphy. These authors reported a quantification by thallium-201 single-photonemissioncomputed tomography.J high sensitivity and specificity for this test. Subse- Am Coil Car&i 1991;17:355-362. il White CW, Wright CB, Doty DB, Hiratza LF, EasthamCL, Harrison DC, quently, dobutamine was used in combination with Marcus ML. Does visual interpretation of the coronary arteriogram predict the 201Tlscintigraphy by Zellner et a128and in combina- physiologic importance of a coronary stenosis?N Engi J Med 1984;310:81-24. 8E

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

NOVEMBER 5, 1992

l2. Albro P, Gould KC, Wescott RJ, Hamilton GW, Ritchie JL, Williams DL. Noninvasive assessmentof coronary stenosis by myocardial imaging during pharmamlogic coronary vasodilatation, III: clinical trial. Am J Cardiol 1978;42: 75l-760, 13. Gould KL. Noninvasive assessmentof coronary stenoses by myocardial perfusion imaging during pharmacologiccoronary vasodilatation. I. Physiologic basisand experimental validation. Am J Curdiol 1978;41:267-278. 14. Gould KL, Westcott RJ, Albro PC, Hamilton GW. Noninvasive assessment of coronary stenosisby myocardial imaging during pharmacologic coronary vasodilatation: II. Clinical methodology and feasibility. Am J Cardioi 1978;41:27%%7. 15. Leppo JA. Dipyridamole-thallium imaging: the lazy man’s stress test. J Nucl Med 1989;30:281-287. 16. Barges-Neto S, Mahmarian JJ, Jain A, Roberts R, Verani MS. Quantitative thallium-201single photon emissioncomputed tomographyafter oral dipyridamole for assessingthe presence,anatomic location and severity of coronary artery disease.J Am Coil Cardbll988; 11:962-969. 17. RossenJD, Quillen JE, Lopez JAG, StenbergRG, Talman CL, Winniford MD. Comparison of coronary vasodilation with intravenous dipyridamole and adenosine.J Am CoU Car&I 1991;18:485~91. 18. Verani MS, Mahmarian JJ, Hixson JB, Boyce TM, Staudacherl&4. Diagnosis of coronary artery diseaseby controlled coronary vasodilation with adenosine and thallium-201 scintigraphy in patients unable to exercise. Circulation 1990;82:8M7. 19. Gupta NC, Esterbrooks DJ, Hilleman DE, Mohiuddin SM. Comparison of adenosineand exercisethallium-201single-photonemissioncomputedtomography (SPECT) myocardial perfusion imaging.J Am CuZ! CardbZ 1992,19:24& 257. 20. Nishimura S, Mahmarian JJ, Boyce TM, Verani MS. Equivalencebetween adenosine and exercise thallium-201 myocardial tomography: A multicenter, prospective,crossovertrial. J Am CoU Car&Z 1992 (in press). 21, Nishimura S, Mahmarian JJ, Boyce TM, Verani MS. Quantitative thallium201 single-photon emission computed tomography during maximal pharmacologic coronary vasodilation with adenosinefor assessingcoronary artery disease. JAm CoEl Curdiol 1991;18:73&745.

22. Nguyen T, Heo J, Ogilby JD, Iskandrian AS. Single-photon emission

computed tomography with thallium-201 during adenosine-inducedcoronary hyperemia-relation with coronary arteriography, exercise thallium imaging and two-dimensional echmardiography.J Am CoU Car&Z 199O;Kx 13751383. 23. Iskandrian AS, Heo J, Nguyen T, Lyons E, Paugh E. Left ventricular dilatation and pulmonary thallium uptake after single-photon emission computer tomographyusingthallium-201during adenosine-inducedcoronary hyperemia.Am J Car&Z 1990$6:807-811. 24. Abreu A, Mahmarian JJ, Nishimura S, Boyce TM, Verani MS. Tolerance and safety of pharmacologic coronary vasodilation with adenosine in association with thallium-201 scintigraphy in patients with suspectedcoronary artery disease.J Am Co11Cardiol 1991;18:730-735. 25. Iskandrian AS, Heo J, Nguyen T, Beer S, Cave V, Cassel D, Iskandrian BB. Tomographicmyocardialperfusion imagingwith technetium99mteboroxime during adenosine-inducedcoronary hyperemia: correlation with thallium-201 imaging.J Am Co11Car&b1 1992;19:307-312. 28. Gupta NC, Esterbrooks D, Mohiuddin S, Hilleman D, SunderlandJ, Shiue CY, Frick MP. Adenosine in mymardial perfusion imaging using positron emissiontomography.Am Heart J 2991;122:29%301. 27. Mason JR, Palac RT, Freeman ML, Virupannavar S, Loeb HS, Kaplan E, Gunnar RM. Thallium scintigraphy during dobutamine infusion: nonexercisedependentscreeningtest for coronary disease.Am Heart J 1984;107:481-485. 28. Zellner JL, Elliott BM, Robinson JG, Hendrix GH, Spiker KM. Preoperative evaluation of cardiac risk using dobutamine-thallium imaging in vascular surgery.Ann Vbsc Sq 1990;4:23%243. 29. Freeman ML, Palac R, Mason J, Barnes WE, Eastman G, Virupannavar S, Lneb HS, Kaplan E. A comparison of dobutamine infusion and supine bicycle exercisefor radionuclide cardiac stresstesting.C/in Null Med 1984;9:251255. 30. Pennell DJ, Underwood SR, Swanton RH, Walker JM, Eli PJ. Dobutamine thallium myocardial perfusion tomography.J Am CoU Cardbl 1991;18: 1471-1479. 3L Hays JT, Mahmarian JJ, Cochran AJ, Verani MS. Dobutamine thallium201 single-photon emission tomography in the evaluation of patients with suspectedcoronary artery diseaseunable to undergo exerciseor pharmacologic vasodilation. (In press) 1992.

A SYMPOSIUM: CARlIlAC IMAGING

9E