Comparison of post-stress ejection fraction and relative left ventricular volumes by automatic analysis of gated myocardial perfusion single-photon emission computed tomography acquired in the supine and prone positions

Danie l B e r m a n , MD, G u i d o G e r m a n o , PhD, H o w a r d Lewin , MD, Xing, p i n g K a n g , M D , Paul B. K a v a n a g h , M S , P o n c e Tapn io , CNMT, M i c h a e l Harr is , a n d J o h n F r i e d m a n , M D

Background. We have previously described an automatic method for measuring left ventricular ejection fraction (LVEF) for myocardial perfusion single-photon emission computed tomography (SPECT). The repeatabili ty of this method has not been previously described.

Methods and Results. This study compares LVEF and relative end-systolic and end-diastolic volumes assessed from myocardial perfusion SPECT by our automatic method in 180 consecutive patients undergoing gated myocardial perfusion SPECT with injection of 99mTc- labeled sestamibi in whom the acquisitions were performed sequentially in supine and prone positions. The algorithm operated completely automatically in the prone and supine positions in 178 Of the 180 patients. Very high correlations were observed for LVEF (r = 0.93), relative left ventricular end-systolic volume (r = 0 .98) , and relative left ventricular end-diastolic volume (r = 0.97). The mean paired absolute difference between LVEFs in the prone and supine position was 3.8 _4- 3.2, for left ventricular end-systolic volume was 4.9 _.+ 4.8 mi, and for left ventricular end-diastolic volume was 7.4 -+ 6.7 ml. When patients were classified by the extent and severity of stress perfusion defect, there was no significant difference in repeatability for the measurements in any category.

Conclusions. Our algorithm for automatic quantification of LVEF and relative end-systolic and end-diastolic volumes from gated 99mTc sestamibi myocardial perfusion SPECT is repeatable. When performed in the prone position, values of ejection fractions and ventricular volumes are essentially identical to those obtained in the supine position. (J Nucl Cardiol 1998; 5:40-7.)

Key words: 99"Tc-labeled sestamibi �9 myocardial perfusion �9 single-photon emission computed tomography �9 left ventricular function �9 left ventricular ejection fraction

For over three decades, left ventricular ejection fraction (LVEF) has been considered the single most representative index of global left ventricular function.'

From the Division of Nuclear Medicine, Department of Imaging, the Division of Cardiology, Department of Medicine, and the Division of Medical Physics and Imaging, CSMC Burns & Allen Research Institute, Cedars-Sinai Medical Center, Los Angeles, California and the Departments of Radiological Sciences and Medicine, UCLA School of Medicine, Los Angeles, California.

Received for publication Jan. 2, 1997; revision accepted Aug. 1, 1997. Reprint requests: Daniel S. Berman0 MD, Director, Nuclear Cardiol-

ogy, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Room A042, Los Angeles, CA 90048.

Copyright �9 1998 by American Society of Nuclear Cardiology. 1071-3581/98/$5.00+0 43/1/85157

40

LVEF has been found to be of prognostic importance in medically-treated chronic coronary artery disease, 2,3 acute myocardial infarction, 4.5 aortic valvular disease, 6 mitral valvular disease, 7 nonischemic cardiomyopathy, 8,9 and congestive heart failure. '~ It has also been found to be a strong predictor of outcome in various forms of cardiac surgical interventions. 6.'' In what remains perhaps the most common clinical application of the measurement, serial assessment of LVEF has been the standard approach to the assessment of potentially lethal adriamycin cardiotoxicity. '2 The measurement is also important in determining the appropriateness of a variety of medical therapies and is used to select patients for long-term angiotensin converting

Journal of Nuclear Cardiology Berman et al. 41 Volume 5, Number 1;40-7 Comparison of poststress EF and relative LV volumes

IJI. ILl

.-I G t - O t - -

I1.

100'

8 0

60

40

20

y = 5 . 4 4 + O . 8 9 x r=0.93~

m B

t , - - _

i .,x=y: line

S B

0 r I ' I �9 I �9 I �9

o 20 40 60 eo lOO

Supine LVEF [%] Figure 1. Correlation between LVEF measured by gated perfusion SPECT in the supine and prone positions.

enzyme inhibitor administration after myocardial in- farction. ~3

Nuclear methods using planar gated equilibrium blood pool scintigraphy or, less commonly, first pass radionuclide angiography, have become widely accepted as accurate clinical methods for the noninvasive assess- ment of LVEF. Both the equilibrium and first pass methods for measuring LVEF from blood pool scintig- raphy are associated with a small error in repeated measurements, ~4~7 in part due to physiologic variation and also to subjective steps in implementing the com- puter algorithms used to make the measurements. Re- cently gated blood pool single-photon emission com- puted tomography (SPECT) has been proposed as a method for measuring LVEF, ~s.~9 but validated, auto- matic methods of assessment have not yet been reported. Automatic methods for determining LVEF from planar equilibrium blood pool scintigraphy have been reported previously. 2~

We have recently developed an automatic method for deriving LVEF from gated myocardial infusion SPECT using 99mTc-labeled sestamibi (99mTc sesta- mibi), 23 and have reported a high linear agreement (r = 0.91) between LVEF by this method and LVEF mea- sured from resting first pass radionuclide angiography. Since this perfusion SPECT LVEF measurement is completely automatic, it lends itself to the serial assess- ment of LVEF when accurate measurements are of critical importance. To date the repeatability of this method has not been described. The goal of this study

was to evluate the repeatability of automatic quantifica- tion of ejection and relative end-systolic and end- diastolic volumes by comparing measurements obtained from sequential supine and prone acquisitions.

METHODS

Patient Population. The study group consisted of 180 patients undergoing clinically indicated gated myocardial per- fusion SPECT with injection of 20 to 30 mCi of 99roTe

sestamibi after exercise (n = 94), adenosine or dipyridamole (n = 67) stress, or rest (n = 14). There were 110 men and 70 women with a mean age of 64 --. 12 years (range, 27 to 81). History of myocardial infarction was present in 35 (19%) patients, coronary artery bypass grafting in 28 (16%), and percutaneous transluminal coronary angioplasty in 37 (21%) patients. All patients were in sinus rhythm. The patient popu- lation represents a consecutive series of patients in whom gated myocardial perfusion SPECT was performed sequentially in supine and prone positions. Immediately following the standard supine data set the patient was placed in the prone position, and imaging was repeated. No modification to the standard imaging table was made; i.e., the prone emission data in part traversed the imaging table.

Quality Control. The raw projection data from the var- ious phases of the cardiac cycle were summed and reviewed in a rotating cine format for evidence of patient motion or gross gating error with the latter occasionally appearing in the form of dropped frames or flickering data. Gating error was also evaluated by inspecting the shape of the time-volume curve. Additionally, the endocardial edges determined by the auto-

42 Berman et al. Journal of Nuclear Cardiology Comparison of poststress EF and relative LV volumes January/February 1998

M. ILl > ,.J

0 m

4) 0 e- 4)

4) qlm

0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-0.4 0 .5

_*_..*_ _

n

| �9 �9 |

1 . 0 1 . 5

Mean ( log LVEF)

Figure 2. Biand-Altman analysis of ejection fraction data.

I

2 . 0

matic algorithm were evaluated to verify appropriate cardiac segmentation and edge assignment. In one patient study our algorithm tracked the heart with only moderate effectiveness because of abnormally high extra cardiac uptake; nonetheless, this data point was not excluded from analysis. One patient study was eliminated because of marked gating artifact. No other studies were eliminated.

tVEF Measurement . All studies were performed with a triple-detector camera (Prism 3000, Picker, Ohio) using low-energy high-resolution collimation, step-and-shoot de- tector rotation, 3-degree projections, and 25 seconds of data collection per projection, distributed over eight gating inter- vals and collected in 64 • 64 pixels 2 matrices. There was no "bad beat rejection," i.e., the cardiac beat length acceptance window was set at 100%. All projection data sets were then prefiltered with a two-dimensional Butterworth filter (or- der = 5 and critical frequency = 0.22 cycles/pixel, pixel size = 0.53 cm), and reconstructed over 180 degrees (45-degree fight anterior oblique to left posterior oblique) with filtered backprojection (ramp filter) and no attenuation correction, in 64 • 64 pixels 2 matrices. The reconstructed gated transaxial image sets were reoriented into gated short axis sets to which our automatic LVEF quantification software was applied.

Our automatic algorithm for the quantitative measurement of LVEF has been previously described and validated. 23 In short, the algorithm operates in three-dimensional space, and uses the gated short axis image sets. It segments the left ventricle, estimates and displays the endocardial surface, the epicardial surface and the valve plane for every gating interval, calculates from them the endocardial and epicardial volumes,

isolates the intervals corresponding to end-diastole and end- systole, and derives the related ejection fraction. The volumes are considered relative rather than absolute because the mea- surement of absolute volumes has yet to be validated. The software is implemented using the X-Windows and OSF-Motif interface standards and executes in about 15 seconds on the processing workstation (Odissey VP, Picker, Ohio) used by our triple-detector camera.

Petfusion Defect Analysis. A semi-quantitative visual interpretation was performed by means of short-axis and vertical long-axis myocardial tomograms and a 20-segment model as previously described. 24 By using a five-point scoring system for each segment (0 = normal, 4 = absent uptake), this model allows derivation of a summed stress score incorporating the severity and extent of stress perfusion defects. 25 Summed stress scores (SSS) of less than 4 were considered normal, 4 to 8 mildly abnormal, 8 to 12 moderately abnormal, and greater than 13 severely abnormal, zs

Statistical Analysis. All continuous values were sum- marized as mean --+ standard deviation. Relationships of LVEF and end-systolic volume by gated perfusion SPECT between the supine and prone positions were assessed by simple linear regression analysis. Analysis of agreement was also performed by means of the Bland-Altman method. 26

RESULTS

Interstudy Left Ventricular Ejection Fraction Repeatability. Mean LVEFs assessed from gated per- fusion SPECT were 58.2 __+ 14.2 (range 10% to 89%) in

Journal of Nuclear Cardiology Berman et al. 43 Volume 5, Number 1 ;40-7 Comparison of poststress EF and relative LV volumes

300

~___=200

2

100

LLI

0

y=-0.39+1.01:( r=0.978

f j j

y=x line

0 100 2 0 0 3 0 0

ESV supine [ml]

Figure 3. Correlation between end-systolic volume measured by gated perfusion SPECT in supine and prone positions. ESV, End-systolic volume.

300

" E 2oo,

Q.

100'

U.I

O' o

y=-1.63+0.99x r=0.971 �9 , , j

�9 . J �9

_ ~ [ ~ w ' O 0 t

f " " t '--*"~ " 1 i

100 200 3 0 0

EDV supine [ml]

Figure 4, Correlation between end-diastolic volume measured by gated perfusion SPECT in supine and prone positions. EDV, End-diastolic volume.

the supine position and were 57.2 ___ 13.2 (range 12% to 90%) in the prone position. The mean paired absolute difference between LVEFs was 3.8% _+ 3.2% (range 0% to 13%). The correlation between the prone and supine LVEF is illustrated in Figure 1. High correlation was noted (r = 0.93), with a slope near unity. The close relationship between ejection fractions in the supine

and prone positions was found across the spectrum of degrees of abnormality of stress myocardial perfusion. Bland-Altman analysis of repeatability of LVEF is shown in Figure 2. To assess whether ejection fraction measurements and their repeatability were affected by the extent and severity of perfusion defect, the corre- lations between prone and supine LVEFs were deter-

44 Berqaan et al. Journal of Nuclear Cardiology Comparison of poststress EF and relative LV volumes January/February 1998

UI

0 i

o

(It L _

q , - -

0 . 4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-0.4 0 .5

i . ' - " 1 �9 �9 � 9 1 4 9 � 9 1 4 9 � 9 �9 �9 ]

�9 . t ' . . . . / �9 l m � 9 1 4 9 v �9 U � 9 ~ " �9

. �9 ". . , I 0 f " A � 9 1 4 9 .1~ � 9 1 4 9 , , , A � 9 v �9 �9 00 v �9 �9

�9 . .** �9 - - -

I �9 �9 �9 �9 | �9 �9 �9 �9 I

1.0 1.5 2 .0 2 .5 Mean (log ESV)

Figure 5. Bland-Altman analysis of ESV data.

mined in the SSS categories. The mean SSS was 4.8 ___ 7.6 (range, 0 to 36). The correlation coefficient for SSS less than 4 (n = 115) was 0.91, for SSS 4 to 7 (n = 20) was 0.89, for SSS 8 to 12 (n = 17) was 0.96, and for SSS greater than 12 (n = 31) was 0.86. Thus the repeatability of the LVEF measurement did not vary appreciably throughout a wide spectrum of per- fusion defect sizes.

Interstudy Repeatability of Relative Left Ven- tricular End-Systolic and End-Diastolic Volumes. The mean ___ standard deviation of end-systolic vol- ume in the supine and the prone positions were 45.1 _ 34.0 ml (range, 3 to 180 ml) and 44.4 + 34.0 ml (range, 4 to 207 ml), respectively. The mean paired absolute difference between the left ventricular end- systolic volumes was 4.9 ___ 4.8 ml (range, 0 to 36 ml). The mean ___ standard deviation of end-diastolic vol- ume in the supine and prone position were 95.7 ___ 41 ml (range, 31 to 259 ml) and 92.0 _ 41.2 ml (range, 17 to 255 ml), respectively. The mean paired absolute difference between the left ventricular end-diastolic volumes was 7.4 _ 6.2 ml (range, 0 to 44 ml). Figure 3 illustrates a high correlation between prone and supine measurements of relative left ventricular end- systolic volumes (r = 0.98). Figure 4 illustrates the high correlation between prone and supine measure- ments of relative left ventricular end-diastolic vol- umes (r = 0.97). Bland-Altman analyses of repeatabil-

ity of end-systolic volumes and end-diastolic volumes are shown in Figures 5 and 6, respectively.

DISCUSSION

This study examined the repeatability of automati- cally quantified LVEF and relative end-diastolic and end-systolic volumes from myocardial perfusion SPECT. This test of repeatability was performed under conditions which severely challenged the automatic quantifying algorithm, because patients were examined in entirely different positions. Compared to supine imaging, the prone position is associated with a change in the location of the heart in the chest 27.28 as well as attenuation of a significant fraction of the emission data because of the imaging table. Nonetheless, excellent agreement be- tween ejection fractions and relative end-systolic and end-diastolic volumes obtained sequentially in the two positions was observed. It is likely that an even greater degree of reproducibility would be observed if patients were imaged in the same position for sequential acqui- sitions.

Prone vs. Supine Acquisitions. Prone imaging was initially described as a method which improves specificity of myocardial perfusion SPECT in the assessment of the inferior wall. 27.28 We then confirmed this improved specificity and also demonstrated that there was less patient motion with prone myocardial

Journal of Nuclear Cardiology Berman et al. 45 Volume 5, Number 1;40-7 Comparison of poststress EF and relative LV volumes

0 .2

A

�9 0.1

M,I

0 i

"-~ O0

o t-.

i t _

r -0.1 q , , .

C3

ee

_~*_.__ . _ = _

- . . J , . , . I 0_. �9 �9 �9

. : : , - .- .

-0 .2 . . . . , ~

0 . 5 1 . 0 1 . 5 2 . 0

Mean (log EDV)

Figure 6. Bland-Altman analysis of EDV data.

2 . 5

perfusion SPECT. 29 We and others, however, do not advocate the routine replacement of supine imaging with prone acquisitions because of a commonly ob- served artifactual defect in the anteroseptal wall on prone reconstructions in all patients. 29 This anterosep- tal defect would be a source of false-positive studies because it would be incorrectly attributed to left anterior ascending coronary artery disease. After ap- proximately 1 year of consecutive prone imaging studies, we returned to supine myocardial perfusion SPECT as a standard because of the frequency of the artifactual anteroseptal defects. In recent years, how- ever, when using 99mTc sestamibi as the stress perfu- sion agent, we have used prone imaging as an adjunct to supine SPECT. With the use of 99mTc sestamibi, because the redistribution process is minimal during the first 2 hours after injection, we have suggested that improved specificity and normalcy rates can be ob- tained by performing prone SPECT when supine SPECT raises the question of true inferior wall perfu- sion defect versus an artifactual defect caused by diaphragmatic attenuation or patient motion. Further- more, in some patients, supine SPECT is not well- tolerated and prone imaging alone might be used. In addition to demonstrating the repeatability of ejection fractions and relative ventricular volumes between supine and prone imaging, the present study validates the application of quantitation of ejection fraction or relative volumes from prone images alone by demon-

strating the essential equality between prone and supine measurements.

Limitations. In addition to differences in imaging characteristics inherent in the use of prone and supine positions, other factors may have been involved in the small differences noted between supine and prone ejection fractions. The sequence of positions was constant; i.e., supine acquisitions were performed first and prone acquisitions second. This may have intro- duced small errors on a physiologic basis. For exam- ple, slightly higher catecholamine concentrations on the early versus slightly later poststress images may have resulted in a slightly higher ejection fraction on the supine study. 3~ Conversely, since an element of postexercise stunning of the left ventricle may be present soon after stress, 3] this factor may have resulted in a reduced LVEF in the supine positions in some severely ischemic patients. The automatic tech- nique for evaluation of volumes has recently been preliminarily validated for absolute volume measure- ment by comparison with 2-dimensional echocardiog- raphy32; however, the present study examines the repeatability of only relative end-systolic and end- diastolic measurements.

Compadson With Previous Studies. Previous studies with equilibrium blood pool scintigraphy demon- strated excellent repeatability of resting LVEFs. 14-] 7.33 A study with rest first-pass radionuclide ventriculography has also demonstrated high repeatability of LVEF with a

46 Berman et al. Journal of Nuclear Cardiology Comparison of poststress EF and relative LV volumes January/February 1998

correlation coefficient of 0.98. 34 The present study dem- onstrates results which are comparable to these previous studies, despite the fact that the acquisitions were per- formed with the patients in two different positions.

CONCLUSION

Our algorithm for automatic quantification of LVEF and relative end-systolic and end-diastolic vol- umes from gated 99mTc sestamibi myocardial perfu- sion SPECT is repeatable. When performed in the prone position, values of ejection fractions and ven- tricular volumes are essentially identical to those obtained in the supine position. These studies suggest that gated myocardial perfusion SPECT might be an effective method for serial assessment of LVEF on a routine clinical basis.

References

1. Folse R, Braunwald E. Determination of fraction of left ven- tricular volume ejected per beat and of ventricular end-diastolic and residual volumes: experimental and clinical observations with precardial dilution technique. Circulation 1962;25:674.

2. Emond M, Mock MB, Davis KB, Fisher LD, Holmes DR Jr, Chaitman BR, et al. Long-term survival of medically treated patients in the coronary artery surgery study (CASS) registry. Circulation 1994;90:2645-57.

3. Lee KL, Pryor DB, Pieper KS, Harrell FE Jr, Califf RM, Mark DBN, et al. Prognostic value of radionuclide angiography in medically treated patients with coronary artery disease: a compar- ison with clinical and catheterization variables. Circulation 1990; 82:1705-17.

4. The Multieenter Postinfarction Research Group. Risk stratification and survival after myocardial infarction. N Engl J Med 1983;309: 331-6.

5. Shah PK, Maddahi J, Staniloff HM, Ellrodt AG, Pichler M, Swan HJC, et al. Variable spectrum and prognostic implications of left and right ventricular ejection fractions in patients with and without clinical heart failure after acute myocardial infarction. Am J Cardiol 1986;58:387-93.

6. Bonow RO, Picone AL, Mclntosh CL, et al. Survival and func- tional results after valve replacement for aortic regurgitation from 1976 to 1983: impact of preoperative left ventricular function. Circulation 1985;72:1244:

7. Enriquez-Sarano M, Schaff HV, Orszulak TA, Bailey KR, Tajik AJ, Frye FL. Congestive heart failure after surgical correction of mitral regurgitation. A long-term study. Circulation 1995;92:2496- 503.

8. Fruhwald FM, Dusleag J, Eber B, Fruhwald S, Zweiker R, Klein W. Long-term outcome and prognostic factors in dilated cardio- myopathy. Angiology 1994;45:763-70.

9. Keogh M, Freund J, Baron DW, Hickie JB. Timing of cardiac transplantation in idiopathic dilated cardiomyopathy. Am J Cardiol 1988;61:418-22.

10. Gradman A, Deedwania P, Cody R, Massie B, Packer M, Pitt B, et al. Predictors of total mortality and sudden death in mild to moderate heart failure. J Am Coil Cardiol 1989;14:564-70.

I h Coronary artery surgery study (CASS): A randomized trial of coronary artery bypass surgery--survival data. Circulation 1983; 68:939-50.

12. Alexander J, Dainiak N, Berger HJ, Goldman L, Johnstone D, Reduto L, et al. Serial assessment of doxorubicin cardiotoxicity with quantitative radionuclide angiocardiography. N Engl J Med 1979;300:278-83.

13. Pfeffer MA, Braunwald E, Moye LA, Basta L, Brown EJ, Cuddy TE, et al. Effect of captopril on mortality and morbidity in patients with left ventricular of dysfunction after myocardial infarction: results of the survival and ventricular enlargement trial. N Engl J Med 1992;327:669-77.

14. Hecht HS, Josephson MA, Hopkins JM, Singh BN. Reproducibil- ity of equilibrium radionuclide ventriculography in patients with coronary artery disease: response of left ventricular ejection fraction and regional wall motion to supine bicycle exercise. Am Heart J 1982;104:567-74.

15. Maddahi J, Berman DS, Diamond G, Gray R, Shah PK, Forrester J. Evaluation of left ventricular ejection fraction and segmental wall motion by multiple gated equilibrium cardiac blood pool scintigraphy. In: Cady L, editor. Computer tech- niques in cardiology. New York (NY): Marcel Dekker Pub- lisher; 1978. p. 389-416.

16. Wackers FJ, Berger H J, Johnstone DE, Goldman L, Reduto LA, Langou RA, et al. Multiple gated cardiac blood pool imaging for left ventricular ejection fraction: validation of the technique assessment of variability. Am J Cardiol 1979;43:1159-66.

17. Marshall RC, Berger HJ, Reduto LA, et al. Variability in sequential measures of left ventricular performance assessed by radionuclide angiocardiography. Am J Cardiol 1978;41:53 I-6.

18. Gil JB, Moore RH, Tamaki N, Miller DD, Barlai-Kovach M, Yasuda T. et al. Multigated blood-pool tomography: new method for the assessment of left ventricular function. J Nucl Med 1986;27:1916-24.

19. Corbett JR, Jansen DE, Lewis SE, Gabliani GI, Nicod P, Filipchuk NG, et al. Tomographic gated blood pool radionuclide ventricu- lography: analysis of wall motion and left ventricular volumes in patients with coronary artery disease. J Am Coil Cardiol 1985;6: 349-58.

20. Goris ML, McKillop JH, Briandet PA. A fully automated deter- mination of the left ventricular region of interest in nuclear angiography. Cardiovasc Intervent Radiol 1981;4:117-23.

21. Douglass KH, Tibbits P, Kasecamp W, Han ST, Koller D, Links JM, et al. Performance of a fully automated program for measure- ment of left ventricular ejection fraction. Ear J Nucl Med 1982;7: 564-6.

22. Lee FA, Fetterman R, Zaret BL, Wackers FJTh. Rapid radionu- clide derived systolic and diastolic cardiac function using cycle- dependent background correction and Fourier analysis. Proc Com- put Cardiol 1985:443-6.

23. Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, et al. Automatic quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med 1995;36:2138-47.

24. Berman DS, Kiat H, Friedman J, Wang FP, Van Train K, Matzer L, et al. Separate acquisition rest thallium-201/stress tc-99m sestamibi dual isotope myocardial perfusion SPECT: a clinical validation study. J Am Coil Cardiol 1993;22:1455-64.

25. Hachamovitch R, Berman DS, Kiat H, Cohen I, Cabico JA, Friedman J, et al. Exercise myocardial perfusion SPECT in patients without known coronary artery disease: incremental prognostic value and use. Circulation 1996;93:905-14.

26. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical assessment. Lancet 1986;1:307- 10.

Journal of Nuclear Cardiology Berman et al. 47 Volume 5, Number 1 ;40-7 Comparison of poststress EF and relative LV volumes

27. Esquerre JP, Coca FJ, Martinez S J, Guiraud RF. Prone decubitus: a solution to inferior wall attenuation in thallium-201 myocardial tomography. J Nucl Med 1989;30:398-401.

28. Segall GM, Davis MJ. Prone versus supine thallium myocardial SPECT: a method to decrease artifactual inferior defects. J Nucl Med 1989;30:548-55.

29. Kiat H, Van Train K, Friedman JD, Germano G, Silagan G, Wang FP, et al. Quantitative stress-redistribution thallium-201 SPECT using prone imaging: methodologic development and validation. 1992;33:1509-15.

30. Rozanski A, Berman DS, Gray R, Diamond G, Raymond M, Prause J, et al. Preoperative prediction of reversible myocardial asynergy by postexercise radionuclide ventriculography. N Engl J Med 1982;307:212-6.

31. Ambrosia G, Betocchi S, Pace L, Losi MA, Perrone-Filardi P, Socicelli A, et al. Prolonged impairment of regional contractile

function after resolution of exercise-induced angina: evidence of myocardial stunning in patients with coronary artery disease. Circulation 1996;94:2455-63.

32. Zanger D, Ghatnagar A, Hansner E, Botello M, Nuquist C, Weissman N, et al. Automated calculation of ejection fraction from gated Tc-99m sestamibi images--comparison to quanti- tative echocardiography [abstract]. J Nucl Cardiol 1997;411, Part 2]:$78.

33. Van Royen N, Jaffe CC, Krumholz HM, Johnson KM, Lynch PJ, Natale D, et al. Comparison and reproducibility of visual echoear- diographic and quantitative radionuclide left ventricular ejection fractions. Am J Cardiol 1996;77:843-50.

34. Friedman JD, Berman DS, Kiat H, Bietendorf J, Hyun M, Van Train KF, et al. Rest and treadmill exercise first-pass radionuclide ventriculography: validation of left ventricular ejection fraction measurements. J Nucl Cardiol 1994;1:382-8.