ORIGINAL 
Annals of Nuclear Medicine Vol. 5, No. 2, 47-51, 1991 
ECG-gated thallium-201 myocardial SPECT in patients with old myocardial infarction compared with ECG-gated blood pool SPECT 
Teruhito MOCHZUKl,*,**** Kenya MURASE Yasushi FUJIWARA ** Taketoshi ITOH *** Masao MIYAGAWA,* Shuji TANADA,* Atsushi IIO,* Ken HAMAMOTO* and W. Newlon TAUXE**** 
*Ehime University School of Medicine, Department of Radiology, **Second Department of Internal Medicine * * * Ehime-kenritsu Imabari Hospital and * * * * University of Pittsburgh School of Medicine, Department of Radiology, Division of Nuclear Medicine 
We evaluated one of the merits of ECG-gated thallium-201 single photon emission computed tomography (g-Tl SPECT), i.e., the ability to appreciate left ventricular (LV) wall motion. LV wall motion assessed by g-TI SPECT and by ECG-gated Blood Pool SPECT (g-BP SPECT) was classified into three grades and compared segment by segment. Tl-201 uptake by g-Tl SPECT was also classified into three grades and compared with those of wall motion in g-BP SPECT. Fifty patients with prior myocardial infarction were injected intravenously at rest with 111 to 185 MBq (3 to 5 mCi) of Tl-201 . The left ventricular regions were divided into anterior, septal, inferior and lateral segments (50 patients x 4 segments=200 segments in total). The grades of wall motion and Tl-201 uptake detected by g-Tl SPECT correlated well with those of wall motion in g-BP SPECT (94.5 % and 85 %, respectively). With g-Tl SPECT it was possible to evaluate left ventricular wall motion, providing clear perfusion images. 
Key words: thallium-201, cardial infarction, technetium-99m red blood cells, ECG-gated, SPECT, myocardial infarction

INTRODUCTION 
ECG-GATED THALLIUM-201 single photon emission computed tomography (g-Tl SPECT) has not been widely used as a routine examination procedure be-cause of its long data acquisition time. However, g-Tl SPECT has three major merits : (1) clearer per-fusion images [end-diastolic (ED) and end-systolic (ES) images], (2) left ventricular (LV) wall motion 1 and (3) systolic thickening of the LV wall as increased pixel counts in ES images compared with pixel 
Received December 14, 1990, revision accepted February 26, 1991. For reprints contact : Ken Hamamoto. Department of Radiology, Ehime University School of Medicine, 454 Shitsukawa, Shigenobu-cho. Onsen-gun, Ehime-ken, 791-02 JAPAN. 
counts in ED images.2 How well can g-TI SPECT appreciate the LV wall motion? Is the wall motion in the infarcted region always abnormal? This study was performed mainly to evaluate one of the merits of g-Tl SPECT, namely assessment of LV wall motion, comparing the grades of wall motion and Tl-201 uptake assessed by g-TI SPECT with those of wall motion in ECG-gated Blood Pool SPECT (g-BP SPECT).3,4 
MATERIALS AND METHODS 
Patients 
Materials were 50 patients with old myocardial infarction whose strokes occurred at least two months prior to the SPECT studies. A diagnosis of myocardial infarction was made on the basis of ECG changes in the acute phase and significantly increased myocardial enzymes (CPK and LDH). Of these, 14 cases were also confirmed by positive 99~Tc-pyrophosphate scintigraphy (hot scan).5,6 The g-Tl SPECT and g-BP SPECT were performed within a week. There were 41 males and 9 females. Ages ranged from 35 to 85 years with a mean of 62. I years. 
ECG-gated Tl-201 myocardial SPECT Ten minutes after intravenous injection of 111 to 185 MBq (3 to 5 mCi) of Tl-201 chloride, patients lay on the imaging table in the supine position with their left hands and forearms raised over their heads. The projection data were acquired through 180 degrees from 45 degrees RAO to 45 degrees LPO in 24 steps (80 to 100 beats per step) with a rotating gamma camera (Hitachi Gamma-View-T, Japan) equipped with a low energy high resolution collimator and interfaced to a computer (Hitachi HARP system). The acquisition required 30 to 40 minutes. An R-R interval was divided into about 20 fractions. The first three fractions were added for the reconstruction of ED images and three ES fractions as defined by planar Tc-99m MUGA studies were added for the reconstruction of ES images. After 9-point spstial smoothing, both ED and ES contiguous transverse images were reconstructed by filtered backprojection with Chesler filter. These were reformatted to short and long axial images perpendicular to the cardiac axes. The image size was a 64 x 64 matrix and the slice thickness was 6 mm. Attenuation correction was not performed. 
ECG-gated Blood Pool SPECT The patients were injected with 740 to 1,llO MBq (20 to 30 mCi) of Tc-99m pertechnetate to label red blood cells in vivo in the presence of stannous pyrophosphate. After the planar blood pool studies (first-pass and MUGA), projection data were acquired in the same way as with the g-Tl SPECT except that fifty beats per step were used. The data acquisition required 20 to 25 minutes. The ED and ES images were also processed in the same way as with the g-Tl SPECT. 

Analysis
In the 50 patients, each LV SPECT image was divided into four segments, i.e., anterior, septal. inferior and lateral walls, which made a total of 200 segments (50 pts x 4 seg). The anterior and inferior walls were evaluated with a vertical long axis and a short axis. The septal and lateral walls were evaluated with a horizontal long axis and a short axis. In the short axis, three slices (near the apex, in the middle and near the base) were evaluated. The ED and ES images were superimposed with a cut off level of 50% (30% in OMI regions with very low counts) of the maximum pixel counts for the edge detection of the LV wall on the screen. (1) One radiologist and one cardiologist classified the wall motion visually into three grades : normal. hypokinesis and akinesis (including dyskinesis). Then the grades of wall motion obtained by g-Tl SPECT and g-BP SPECT were compared segment by segment. (2) The intensity in the ED images in g-Tl SPECT was classified into three groups: normal (>70 % of maximum pixel counts), low (50 to 70%) and deficient (<50%)¥ The grades of Tl-201 uptake were then compared with those of the wall motion derived from the g-BP SPECT, analyzing all three of the cardiac axes. (3) The grades of Tl-201 uptake in the 91 segments with infarct defined by low or deficient Tl-201 uptake in ED images in the g-Tl SPECT were compared with those of wall motion in the g-BP SPECT. 

RESULTS 
(1) Table 1 summarizes the relationship between the grades of wall motion derived from the g-Tl SPECT and those from the g-BP SPECT. In the anterior and lateral walls, they were completely concordant. The overall concordance rate in the 200 segments was 94.5 %
(2) Table 2 summarizes the relationship between the grades of Tl-201 uptake assessed by g-Tl SPECT images and those of wall motion from the g-BP SPECT images. The concordance rate was 85 %
(3) In the 91 infarcted segments, wall motion was observed in 35 (38.5 ~) segments, i.e., in 20 (22.0%) segments wall motion was hypokinetic and in 15 (16.5 %) wall motion was normal. (see Table 3) 

CASE REPORT 
A sixty-two-year-old male with an inferior wall old myocardial infarction is represented in Fig. 1 . Al-though the vertical long axial ED image demonstrates uniformly decreased perfusion throughout the entire inferior wall, better wall motion and systolic thicken-ing near the apex than near the base can be seen in the ES image. The g-Tl SPECT provides : (1) clear ED and ES perfusion images, (2) information about wall motion, and (3) information about systolic thickening. 

DISCUSSION 
Long data acquisition time has been inherent in ECG-gated SPECT imaging. Before the development of SPECT, Budinger et al7 obtained three dimensional ECG-gated myocardial imaging by rotating the patient. That required two hours of acquisition with inadequate radionuclide counts and resolution. Even since more clinically oriented SPECT systems were developed with modest reducing time, no studies evaluating the usefulness of g-Tl SPECT in comparison with g-BP SPECT in wall motion have been reported, probably because of its still relatively long data acquisition time (about one hour). We have found that this acquisition time can be reduced to 30 to 40 minutes by increasing the T1-201 dose to (111 MBq to) 185 MBq [(3 mCi to) 5 mCi] and by acquiring fewer steps (24 steps). The patient raised his left hand and forearm over the head and maintained that position so that the camera could obtain the minimum radius of the orbit to get more counts and better resolution. Adding three frames out of 20 fractions of the R-R interval for ED and ES projection data allowed higher and enough counts to make g-Tl SPECT images clearer than the nongated images. Although a single head camera was used in this study, a multi-headed r-camera would shorten the acquisition time significantly. Moreover, the problem of the long data acquisition time is most likely to be overcome by the new technetium-99m labeled radiopharmaceuticals.8-10 The comparative study of wall motion by g-Tl SPECT and g-BP SPECT demonstrated that g-Tl SPECT had essentially the same ability to evaluate the local LV wall motion as the g-BP SPECT technique as shown by the excellent concordance (94.5%)-(see Table 1) Although the ED and ES images had enough pixel counts to evaluate the wall motion in most cases, in 8 out of 11 discordant segments, wall motion was underestimated compared with that of g-BP SPECT, and this might be attributed in part to the markedly reduced count rates in infarcted regions. The comparative study of the grades of Tl-201 uptake assessed by the g-Tl SPECT and those of wall motion assessed by the g-BP SPECT in 91 infarcted segments revealed some wall motion in 35 lesions (38.5%.)¥ Especially in 15 (16.5%) segments i,e., in 11 (12.1%) segments with low Tl-201 uptake and 4 (4.4%) with deficient, wall motion was considered as normal, where g-BP SPECT could not detect any abnormality resulting from the infarction. The wall motion approach in the blood pool study was therefore inadequate to detect myocardial infarction, but this may be acceptable for the evaluation or follow up of LV function in the region of known prior myocardial infarction. However, it is important to be aware of this incidence when interpreting wall motion and viability in the infarcted regions. For the evaluation of ischemia, Tl-201 stressredistribution myocardial scintigraphy11-15 was useful, while for the assessment of viability, the rest T-201 study was more suitable, because lesions with persistent defects in the delayed images 2 to 5 hours after the injection might be viable and reversible.16-20 Therefore, if the rest Tl-201 study is indicated for patients who have persistent defects or minimal redistribution in a delayed study, or who cannot undergo a stress study, the g-Tl SPECT would prove quite useful. Another advantage which is characteristic of g-Tl SPECT is that g-Tl SPECT can evaluate LV systolic thickening (not only wall motion) as increased pixel counts in ES images.2 For example a deficient perfusion area with no increased pixel counts indicates no function and probably a scar. On the other hand, a decreased perfusion area with increased pixel counts indicates functioning and viability. A decreased perfusion area but not an absolute defect with minimal (absent or inadequate) increased pixel counts may indicate a "hibernating myocardium.2l' We conclude that the Tl-201 ECG-gated myocardia SPECT is a useful diagnostic strategy that could evaluate wall motion, providing clear perfusion images. 

ACKNOWLEDGMENT 
The authors gratefully acknowledge the assistance of Ms. Judith K Holden in preparing the manuscript. 

REFERENCES 
1. Mochizuki T, Murase K, Tanada S, et al: Clinical evaluation of the thallium-201 ECG-gated myocardial SPECT. J Nucl Med 30: 864, 1989 
2. Mochizuki T. Murase K, Miyagawa M, et al: Assessment of systolic thickening with thallium-201 ECG-gated SPECT : new parameter for local left ventricular function. Radiology 177 (p): 98-99, 1990 
3. Moore ML, Murphy PH, Burdine JA: ECG-gated emission computed tomography of the cardiac blood pool. Radiology 134: 233-235, 1980 
4. Tamaki N, Mukai T, Ishii Y, et al : Multiaxial tomography of heart chambers by gated blood-pool emission computed tomography using a rotating gamma camera. Radiology 147 : 547-554, 1983 
5. Parkey RW, Bonte FJ, Meyer SL, et al : A new method for radionuclide imaging of acute myocardial infarc-
tion in humans. Circulation 50: 540-546, 1974 
6. Fujiwara Y, Mochizuki T, Itoh T, et al: Quantitative analysis of acute myocardial infarction using single photon emission computed tomography using technetium-99m pyrophosphate. J Cardiogr 16: 555-562, 1986 
7. Budinger TF, Cahoon JL, Derenzo SE, et al: Three dimensional imaging of the myocardium with radionuclides. Radiology 125: 433-439, 1977 
8. Okada RD, Glover D, Gaffney T, et al: Myocardial kinetics of technetium-99m-methoxy-2-methylpropyl-isonitrile. Circulation 77: 491-498, 1988 
9. McGhie I, Feber TL, Akers MS, et al: Assessment of the relationship between ventricular function and perfusion in patients with myocardial infarction using gated tomographic imaging with Tc-99m MIBI and radionuclide ventriculography. J Nucl Med 30: 801 1989 
10. Gerundini P, Maffioli L: Cationic complexes of technetium for myocardial imaging. J Nucl Med 30: 1415-1419, 1989 
ll. Pohost GM, Zir LM, Moore RH, et al : Differentiation of transiently ischemic from infarcted myocardium by serial imaging after a single dose of thallium-201. Circulation 55 : 294-302, 1977 
12. Maddahi J, Garcia EV, Berman DS, et al: Improved noninvasive assessment of coronary artery disease by quantitative analysis of regional stress myocardial distribution and washout of thallium-201 . Circulation 64: 924-935, 1981 
13. Rozanski A, Berman DS, Gray R, et al: Use of thallium-201 redistribution scintigraphy in the preoperative differentiation of reversible and nonrevers-ible myocardial asynergy. Circulation 64 : 936-944, 1981 
14. Garcia EV, Train KV, Maddahi J, et al: Quantification of rotational thallium-201 myocardial tomography. J Nucl Med 26: 17-26, 1985 
15. Depasquale EE, Nody AC, DePuey EG, et al: Quantitative rotational thallium-201 tomography for identifying and localizing coronary artery disease. Circulation 77: 316-327, 1988 
16. Gutman J, Berman DS, Freeman M, et al: Time to complete redistribution of thallium-201 in exercise myocardial scintigraphy. Am Heart J 106: 989-995, 1983 
17. Ziessman HA. Sigler CJ, Wells TM et al: Utility of delayed 24 hour redistribution on SPECT thallium-201 myocardial perfusion studies. J Nucl Med 29: 769, 1988 
18. Hecht HS. Andreae G, Myler RK, et al : The role of 24 hour tomographic thallium-201 myocardial imaging in revascularization. J Nucl Med 29 : 769, 1988 
19. Rocco T, Dilsizian V, Maltais F, et al : Thallium reinjection after delayed imaging demonstrates fill-in to regions with "fixed" defects. J Nucl Med 29: 769, 1988 
20. Tamaki N, Yonekura Y, Yamashita K, et al: Relation of left ventricular perfusion and wall motion with metabolic activity in persistent defects on thallium-201 tomography in healed myocardial infraction. Am J Cardiol 62: 202-208, 1988 
21. Braunwald E, Rutherford JD. Reversible ischemic left ventricular dysfunction : evidence of the "hibernating myocardium". J Am Coll Cardiol 8 1467 1470, 1986