ORIGINAL 
Annals of Nuclear Medicine Vol. 7, No. 2, 79-86, 1993 
Usefulness of resting thallium-201 delayed imaging for detecting myocardial viability in patients with previous myocardial infarction 
Tatsuo TSUKAMOTO, Kohshi GOTOH, Yasuo YAGI, Hisato TAKATSU, Yasushi TERASHIMA, Kenshi NAGASHIMA, Noritaka YAMAMOTO and Senri HIRAKAWA 
The Second Department of Internal Medicine, Gifu University School of Medicine 
To test the feasibility of resting thallium-201 (201Tl) initial and delayed scintigraphy for detecting the area of viable myocardium, we performed single photon emission computed tomography (SPECT) in 57 patients with previous myocardial infarction (MI). All had received coronary arteriography (CAG) and left ventriculography (LVG). Initial and delayed myocardial imagings were carried out 10 min and 2 hours, respectively, after the injection of 201Tl at rest. Redistribution was judged by visual interpretation and/or the circumferential profile curve, and found in the infarcted or its adjacent area in 40 of the 57 cases (70.2%)-A negative washout (net increase of 201Tl uptake in delayed image) was detected in 17 of these 40 cases. In 10 of the 57 patients, both exercise and rest-injected 201Tl myocardial images were obtained at exercise and rest, and compared visually. The areas of abnormal perfusion were smaller in the resting delayed images than those seen after exercise in 9 of the 10 cases, and were equal in one case. Thus, resting 201Tl delayed myocardial scintigraphy appears to reduce the underestimation of the size of the viable myocardium by the usual 201Tl images obtained after exercise or by single initial images obtained at rest in patients with previous MI. 
Key words: redistribution, negaive washout, myocardial viability, thallium-201,rest injection 

INTRODUCTION 
THALLIUM-201 (201Tl) single photon emission computed tomography (SPECT) is widely used to detect myocardial ischemia and identify myocardial viability.1 The area showing a persistent defect on the initial and delayed images after exercise is generally considered to be "infarcted", while the area with a redistribution of 201Tl on the delayed scan is considered to be consistent with viable myocardium. Since revascularization therapy, such as a coronary artery bypass graft (CABG) surgery or percutaneous transluminal coronary angioplasty (PTCA), are increasingly used, it becomes essential to detect the surviving myocardium accurately even in patients with previous myocardial infarction (MI).2-4 Our objective was to determine whether resting 201Tl delayed myocardial SPECT images would be useful in evaluating the viability of the myocardium in the "infarcted" and the adjacent area. Accordingly, we compared the size of perfusion abnormalities in the resting 201Tl images with those obtained after exercise. The correlation between the redistribution in resting scans and the wall motion on the left ventriculogram (LVG), or collaterals on the coronary arteriogram (CAG) and number of diseased coronary arteries, was also evaluated. 

MATERIALS AND METHODS 
Patients 
We evaluated 57 patients, 46 men and 11 women aged 26 to 76 years (average 56.2 years), who had experienced myocardial infarction at least 2 weeks before the performance of 201Tl scintigraphy. The diagnosis was confirmed by the increase in serum enzymes, ECG changes and clinical symptoms. None of these patients had received revascularization therapy during the acute phase. 
Resting 201Tl myocardial scintigraphy 
The patients were injected intravenously with 148 MBq (4 mCi) of 201Tl while seated. They were then placed supine on a scintigraphic imaging table. Single photon emission computed tomography (SPECT) imaging was carried out with a 180deg imaging arc (45deg RAO to 45deg LPO) and 19 acquisitions of 50 sec each. Data were acquired with a large field-of-view rotating gamma camera (ZLC-7500, Siemens Co. Ltd) equipped with a high resolution collimator, and stored in an on-1ine computer (Scintipac 2400, Shimazu, Japan). Initial and delayed images were acquired at 10 min and 2 hours after the administration of 201Tl, respectively. None of the patients experienced chest pain or discomfort during these studies. 
(1) Visual interpretation of perfusion abnormalities and redistribution 
With a 64 x 64 matrix, the pixel size was adjusted so that the actual thickness of the slice was 6 mm/pixel. Six tomographic projection images (i.e. transverse, coronal, sagittal, vertical long axial, horizontal long axial and short axial) in each initial and delayed acquisition were reconstructed for visual interpretation. Thallium-201 SPECT images were interpreted jointly by three experienced physicians without knowledge of the patient's identity or catheterization findings. Redistribution was considered to have occurred when there was "filling in" of the defects or an improvement in the hypoperfusion in delayed SPECT images. The perfusion abnormalities were located with reference to the seven angiographic ventricular segments recommended by the American Heart Association (AHA).5 
(2) Analysis of redistribution with circumferential profile curves 
Redistribution was also evaluated from the circumferential profile curves generated from the resting initial and delayed SPECT images in the 57 patients. A vertical long axial image in the largest left ventricular (LV) lumen and a short axial image at the midway portion between the apex and the base was chosen, and a computer-generated circumference was constructed around the outer edge of the left ventricle. Radii were then constructed by the computer from the center to each point of the circumference with 10deg separation of each radius (36 radii).6,7 Peak activity per pixel along the radius was calculated, and a curve representing the 201Tl activity against the angular location was described. The highest count in each initial and delayed curve was adjusted to 
represent 100 ~ and the counts on all other radii were normalized to the highest count in each initial and delayed curve. Redistribution was judged to be positive when the fraction was improved by more than 10% on more than three consecutive radii, excluding the following areas : LV outflow tract (0-30deg, 340-360deg in the vertical long axial image) and the defect area with less than 30 ~ of the maximal counts on the initial image, No background was subtracted from the images. 
(3) Calculation of washout rate 
The washout rate along each radius was calculated from the original counts as follows :7 Washout rate (%) = (Initial counts -Delayed counts)/(Initial counts) x 100 
The negative washout area was defined as the area in which the calculated washout rate was negative (net increase of 201Tl uptake in delayed image) in more than three consecutive radii. 
201Tl myocardial scintigraphy after exercise 
In 10 of the 57 patients, 201Tl myocardial imaging was performed after graded exercise on a bicycle ergometer by means of standard protocols8 within one month before or after imaging at rest. Exercise was terminated when the patient developed chest symptoms (6/10) or lower limb fatigue (4/10). The same dose of 201Tl as used in the resting study (148 MBq) was injected intravenously 1 min before the end of the exercise. The patients were imaged at 10 min and again at 3 hours post exercise as described above. The initial and delayed SPECT images were analyzed visually, and the area showing re-distribution was compared with those of the rest images. 
Cardiac catheterization 
All patients received contrast LVG in the right and left oblique projections and CAG within two weeks of the performance of resting 201Tl imagings. A separate group of three experienced physicians scored regional wall motion by consensus in the seven myocardial segments of AHA5 with a four-point scale as normal, hypokinesis, akinesis or dyskinesis without knowledge of the scintigraphic data. Organic stenosis of more than 75% was considered to be significant. Patients were divided into three groups according to the number of diseased coronary arteries. The three observers scored the development of collaterals to the infarcted area on a three-point scale as good, poor or none. 
Statistical analysis 
Statistical analysis was performed by the chisquare test. A probability (p) value less than 0.05 was considered to be statistically significant. 

RESULTS 
Incidence of redistribution in resting 201Tl scintigraphy 
All 57 patients exhibited abnormalities on the rest-ing initial 201Tl myocardial images based on the analysis of both their visual and circumferential profiles. In the delayed images, redistribution was found in 35 of 57 cases (61.4%) by visual interpretation, and in 34 of 57 cases (59.6 ~) by circumferential profile analysis, as shown in Fig. l. A total of 29 patients were judged to be redistribution-positive on both assessments, while 6 patients were positive only 
by visual interpretation, and 5 were positive only by circumferential profile analysis. The last 5 patients had relatively slight redistribution. Visual interpretation was useful in 6 patients since, in some cases, redistribution was small (less than 3 consecutive radii) but prominent. The other cases could be evaluated only from the projections other than the vertical long axial and short axial views. By either analysis, redistribution in resting 201Tl delayed images was observed in 40 of 57 patients (70.2%)-A region with negative washout, showing a net in-crease in counts in the delayed vs. the initial image, was found in 17 of the 40 redistribution-positive patients. Correlation of resting 201Tl redistribution with LV wall motion and CAG findings
The wall motion observed on contrast LVG and the redistribution on the resting 201Tl SPECT images in 399 left ventricular segments were compared in 57 patients (Fig. 2). Perfusion abnormalities (defect or hypoperfusion) in the resting initial images were detected visually in each grade of wall motion at the following rates : 6/175 (3.4%) in normokinesis, 96/129 (74.4%) in hypokinesis, 72175 (96.0~;) in akinesis, and 19/20 (95.0 ~) in dyskinesis. Redistribution as judged by visual interpretation and/or circumferential profile analysis of the abnormally per-fused segments on the initial image was detected in each grade of wall motion as follows: 32/96 (33.3%) in hypokinesis, 23172 (31.9%) in akinesis, and 2/19 (10.5%) in dyskinesis. In these redistributed segments, negative washout was seen in 4/32 (12.5%) of hypokinesis, 17/23 (73.9%) of akinesis, and 1/2 (50.0%) of dyskinesis. Thus, the incidence of redistribution was very close in the hypokinetic (33.3%) and akinetic segments (31.9%).
Table I shows the number of patients with or without 201T1 redistribution in each group classified by the number of diseased coronary arteries or by the degree of collateral circulation. There was no significant correlation between the appearance of redistribution on the resting delayed image and the number of diseased arteries or the development of collaterals (p>0.2, p>0.2, respectively, by chi-square analysis). The incidences of redistribution were also compared with the degree of stenosis in the infarct-related arteries. Of the 26 regions supplied by a totally occluded coronary artery, 19 regions were associated with the redistribution of 201Tl. However, the incidence of redistribution did not correlate significantly with the degree of stenosis statistically. 
Comparison of 201Tl myocardial scintigraphy at rest and exercise 
To evaluate the extent size of the perfusion visually in each initial 
of viable myocardium, the abnormality was compared and delayed image obtained 
at rest and after exercise in 10 patients with previous MI. As shown in Table 2, the size of the hypoperfused area was the largest in the initial image obtained immediately after exercise, and the smallest .in the resting delayed image. Comparison of the two delayed SPECT images showed that the abnormally perfused area was smaller in the resting delayed images in 9 of 10 patients, even though the interval between the injection of 201Tl and the delayed imaging was longer in the resting scans. In addition, in 7 of these 10 patients, the abnormally perfused area on the delayed image after exercise was larger than that on the resting initial image, whereas these two images were equivalent in the remaining 3 patients. These observations indicate that the delayed image obtained after exercise as well as the resting initial image tend to underestimate the viable myocardium. 
Case presentation 
Case 1: Figure 3 represents the initial and delayed SPECT images obtained at rest and following exercise in a patient with anteroseptal MI. Of the four images, the resting delayed image showed the smallest area of abnormal perfusion. This case demonstrates that the resting 201Tl delayed image is more likely to detect the largest amount of viable myocardium. Case 2: A patient with previous anterior MI and triple vessel disease is presented in Fig. 4. After CABG, the hypoperfused area became smaller than would be expected from the pre-CABG image. This case illustrates the underestimation of viable myocardium even on the resting delayed image. 

DISCUSSION 
The ability to differentiate the viable myocardium accurately from the irreversibly infarcted myocardium is essential in establishing the indications for revascularization therapy in patients with previous MI. Stress 201Tl myocardial scintigraphy with exercise or with the administration of dipyridamole9 has generally been performed to detect ischemia. Although these stress imagings are feasible for the detection of latent ischemic myocardium, one may underestimate the viable myocardium according to the studies on the 201Tl scintigraphy performed following revascularization by means of CABGIO or PTCA.2,3,11 Detecting the viable myocardium may be more essential than enhancing the ischemia in patients with an obvious MI history. In the present study, we carried out resting 201Tl myocardial scintigraphy to determine whether this procedure would be useful in reducing the underestimation of viable myocardium. Although resting redistribution has been reported in patients with acute MI12 or unstable angina,13,14 a study of resting delayed imaging is uncommon during the chronic phase of the patients with previous MI. In this study, we observed re-distribution in 70.2% of the patients with previous MI and no new episode of chest pain. It is suggested that a careful reading of the resting 201Tl images in combination with the circumferential profile analysis would be useful for detecting the viable myocardium of patients with previous MI. 
Since all of our patients were clinically stable and free of chest pain, it is hard to believe their ischemia could have deteriorated during these studies. Most of their redistributions were incomplete and occurred in the peri-infarcted area where blood flow is considered to be low, but adequate for survival. There-fore, this redistributed myocardium may be consistent with the "hibernating myocardium" pro-posed by Braunwald et al.15 In fact, following revascularization, some of our patients exhibited an improved 201Tl uptake in these areas as shown in Fig. 4. Rigo et al.16 stated that the presence of non-jeopardized collaterals contributed to the alleviation of ischemia during exercise. Imamura et al.17 and Nakatsuka et al.18 also found that the presence of viable myocardium in peri-infarcted areas, such as those detected by redistribution after exercise, was related to the degree of collaterals. However, the patients were evaluated after exercise in all of these studies. The initial perfusion abnormalities in those areas were caused by the enhanced myocardial ischemia. In our series, resting redistribution was not significantly correlated with either the number of diseased coronary arteries or the development of collaterals or the degree of stenosis in the infarctrelated arteries. A different mechanism may con-tribute to the resting redistribution such as damage to the microcirculation which can not be detected by CAG or a disorder of the thallium extraction,19 since myocardial blood flow was considered to be stable during resting studies. 
A negative washout was observed in 17 of the 40 redistribution-positive patients. This phenomenon was considered to be due to slow wash-in rather than to slow washout, because slow washout cannot account for this net increase in 201Tl accumulation on delayed images. The negative washout was observed in 73.9% of akinetic segments. It may be therefore related to an extremely slow uptake into the surviving myocytes in the severely damaged myocardium. 
We also compared the redistribution of 201Tl on resting images with that observed on the images obtained after exercise. It was demonstrated that resting 201Tl delayed images could detect a larger area of viable myocardium than the images after exercise in 9 of the 10 patients so tested. No patient showed smaller redistribution in resting images than those obtained after exercise. From these observations, we conclude that 3 hours is still insufficient for redistribution after exercise, and that the images obtained after exercise are not suitable for accurately detecting the size of the viable myocardium, even if they are useful in detecting the ischemic myocardium. Comparing the resting initial and the rest-ing delayed images, we found that the area of ab-normal perfusion became smaller in the resting delayed image in 7 of the 10 patients. The re-injection method20 has recently been developed. However, the image obtained by this method may perhaps underestimate the viable myocardium, since this image is basically similar to the resting initial images re resented as IR in Table 2. 
Recent studies on myocardial perfusion with 13N-ammonia and metabolism with 18F-fluorodeoxy-glucose by positron emission tomography (PET) have revealed that metabolic viability was observed in the segments with persistent defect on the 201Tl SPECT images after exercise.8,21,22 However, PET is not practical for studying every ischemic patient. Thus, it becomes important to accurately detect the viable myocardium by the commonly used 201Tl scintigraphy. As we have shown, the resting 201Tl initial and delayed myocardial imaging enables one to detect a larger amount of viable myocardium than that obtained after exercise. We conclude that resting initial and delayed 201Tl scintigraphy with both visual and circumferential profile analysis should be performed in patients with previous MI to evaluate myocardial viability. 

REFERENCES 
1. Bailey IK, Griffith LSC, Rouleau J, et al : Thallium-201 myocardial perfusion imaging at rest and during exercise. Circulation 55: 79-87, 1 977. 
2. Cloninger KG, DePuey EG, Garcia EV, et al : Incomplete redistribution in delayed thallium-201 single photon emission computed tomographic (SPECT) images : an overestimation of myocardial scarring. J Am Coll Cardiol 12: 955-963, 1988 
3. Verani MS, Tadros S, Raizner AE, et al : Quantitative analysis of thallium-201 uptake and washout before and after transluminal coronary angioplasty. Int J Cardiol 13 : 109-124, 1986 
4. Iskandrian AS, Hakii AH. Kane SA, et al : Rest and redistribution thallium-201 myocardial scintigraphy to predict improvement in left ventricular function after coronary arterial bypass grafting. Am J Cardiol 51: 1312-1316 1983 
5. Austeu WG, Edwards JE, Frye RL, et al : A reporting system on patients evaluated for coronary artery disease. Report of the Ad Hoc Comittee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 51 : News from the American Heart Association, 5-70, 1975 
6. Burow RD, Pond M, Schafer AW, et al : "Circumferential profiles" : a new method for computer analysis of thallium-201 myocardial perfusion images. J Nucl Med 20: 771-777, 1979 
7. 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 
8. 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 infarction. Am J Cardiol 62: 202-208, 1988 
9. Schmoliner R, Dudczak R, Kronik G, et al : Thallium-201 imaging after dipyridamole in patients with coronary multivessel disease. Cardiology 70 : 145-151, 1983 
10. Gibson RS, Watson DD, Taylor GJ, et al : Prospective assessment of regional myocardial perfusion before and after coronary revascularization surgery by quantitative thallium-201 scintigraphy. J Am Coll Cardiol 3 : 804-815, 1983 
11. Liu P, Kiess MC, Okada RD, et al : The persistent defect on exercise thallium imaging and its fate after myocardial revascularization : does it represent scar or ischemia? Am Heart J 110: 996-1001, 1985 
12. Gibson RS, Taylor GJ, Watson DD, et al : Prognostic significance of resting anterior thallium-201 defects in patients with inferior myocardial infarction. J Nuct Med 21 : 1015-1021, 1980 
13. Berger BC, Watson DD, Burwell LR, et al : Redistribution of thallium at rest in patients with stable and unstable angina and the effect of coronary artery bypass surgery. Circulation 60 : 1114-1125, 1979 
14. Brown KA, Okada RD, Boucher CA, et al : Serial thalliurn-201 imaging at rest in patients with unstable and stable angina pectoris : relationship of myocardial perfusion at rest to presenting clinical syndrome. Am Heart J 106: 70-77, 1983 
15. Braunwald E, Rutherford JD : Reversible ischemic left ventricular dysfunction : evidence for the "hibernating myocardium". J Am Coll Cardiol 8 : 1467-1470, 1986 
16. Rigo P, Becker LC, Griffith LSC, et al : Influence of coronary collateral vessels on the results of thallium-201 myocardial stress imaging. Am J Cardiol 44 : 452-458, 1979 
17. Imamura T, Araki H, Fukuyama T, et al : Significance of collateral circulation on peri-infarct zone : assessment with stress thallium-201 scintigraphy. Clin Cardiol 9: 137-144, 1986 
18. Nakatsuka M, Matsuda Y, Ozaki M, et al : Coronary collateral vessels in patients with previous myocardial infarction. Clin Cardiol 10: 791-795, 1987 
19. Iskandrian AS, Hakki AH, Amenta A, et al : Resting kinetics of thallium-201 in patients with coronary artery disease. Chest 86: 213-218, 1984 
20. Rocco TP, Dilsizian V, McKusick KA, et al : Companson of thallium redistribution wrth rest "reinjection" imaging for the detection of viable myocardium. Am J Cardiol 66: 158-163, 1990 
21. Brunken R. Schwaiger M, McKay MG, et al : Positron emission tomography detects tissue metabolic activity in myocardial segments with persistent thallium perfusion defects. J Am Coll Cardiol 10: 557-567, 1987 
22. Tamaki N, Yonekura Y, Senda M, et al : Value and limitation of stress thallium-201 single photon emission computed tomography : comparison with nitrogen-13 ammonia positron tomography. J Nucl Med 29: 1181-1188, 1988