ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol, 13, No.3, 135-140, 1999 
Functional evaluation of myocardial viability by 99mTc tetrofosmin gated SPECT - A quantitative comparison with 18F fluorodeoxyglucose positron emission CT (18F FDG PET)-
Yoichi KUWABARA, Satoshi WATANABE, Jiro NAKAYA, Masaki FUJIWARA, Rei HASEGAWA, Kouki MATSUNO, Toru KURODA, Yuji MIKAMI, Kiyotaka FUJII, Toshiharu HIMI and Yoshiaki MASUDA 
Third Department of Internal Medicine, Chiba University School of Medicine 
To validate functional analysis of gated SPECT in detecting myocardial viability, seventeen patients (male 15, female 2, mean age 58) with angiographically proven chronic ischemic heart disease (RCA 6, LAD 1O, LCX 1) and eight normal volunteers (all male) were studied. All patients underwent 18F FDG PET and 99mTc tetrofosmin (TF) gated SPECT within a week. After being displayed in a polar map, myocardial perfusion was regionally determined by the mean count in 9 segments at end diastole (ED) and end systole (ES) in gated SPECT. Systolic function was determined by the count increase ratio from ED to ES (WTI: ES - ED/ED). Glucose metabolism was assessed by 18F FDG PET in the segments correspondent to those defined for SPECT. TF %uptake of < 60% was defined as hypoperfusion, and FDG %uptake of < 50% was defined as reduced glucose metabolism. Results: The myocardial segments were classified into 3 categories: "normal" perfusion (n = 85), "mismatch" (reduced perfusion with reserved FDG uptake, n = 25) and "matched" reduced perfusion and metabolic reduction (n = 26). Mean WTI in "mismatch" segment was 0.38 +- 0.21 , and was significantly greater than that in "matched reduced" segments, 0.15 +- 0.20 (p < 0.001). It was also greater than that in "normal" segments, 0.27 +- 0.16. Regression analysis showed that association between WTI and FDG %uptake was significant (r = 0.57, p < 0.0005) for the ischemic segments ("mismatch" + "matched", n = 5 l), but the association was weak for the entire segments although it was statistically significant (r = 0.26, p = 0.02, n = 136). Conclusion: For the segments determined as infarct by perfusion image, systolic functional analysis by gated SPECT is helpful in differentiation of a viable myocardial region or artifact from a scar. Nevertheless, further clinical and technical assessment is required for ECG gating to eliminate overestimation of viability and to warrant clinical use. 
Key words: myocardial viability, gated SPECT, positron emission tomography (PET), 99mTc tetrofosmin 18F FDG (fluorodeoxyglucose) 
INTRODUCTION 
ALTHOUGH many clinical data indicate that imaging of myocardial perfusion provides clinically relevant information about the status of myocardial viability, underes-
Received August 10, 1998, revision accepted February 3, 1999. For reprint contact: Yoichi Kuwabara, M.D., Third Department of Internal Medicine, Chiba University School of Medicine, 1-8-1 Inohana, Chuo-ku, Chiba 260-8677, JAPAN. 
timation of myocardial viability by 99mTc SPECT imaging was observed.1,2 Systolic functional analysis is another important marker of myocardial viability. Myocardial segments with relatively well preserved wall thickening are expected to have viable tissue even if myocardial imaging indicates hypoperfusion, and myocardial segments with irreversible perfusion defect and with no wall thickening would be regarded as non-reversible. Electrocardiography (ECG)-gated acquisition of single photon emission computed tomography (gated SPECT) offers the potential for systolic function in addition to perfusion, since thickening parallels change in the intensity of counts throughout the cardiac cycle. This study is designed to compare wall thickening seen in gated SPECT In "mismatch" and "matched reduced" ischemic segments in terms of perfusion and metabolism, and to validate the usefulness of gated SPECT in detecting myocardial viability. 
MATERIALS AND METHODS 
Study design and study patients 
The study population consisted of 8 normal volunteers (all male, mean age 42 +- 10 years old) and 17 consecutive patients (male 15, female 2, mean age 58 +- 8 years old) admitted to our institute because of chronic ischemic heart disease. All patients had a previous angiographically proven myocardial infarction prior to acquisition of the nuclear studies (Median interval between the onset of myocardial infarction and acquisition of nuclear studies was 3 months, Table 1). The subjects who had myocardial infarction within a month or those who had diabetes mellitus were excluded from this study. All patients underwent 18F FDG PET with 75 g glucose loading and 99mTc tetrofosmin gated SPECT within 10 days. 
Gated SPECT 
Twenty mCi (740 MBq) of 99mTc tetrofosmin was intravenously administered at rest in a fasting period, and subjects were encouraged to drink 250 ml of milk shortly thereafter in order to accelerate the hepatobiliary clearance. Forty minutes after the injection, 16 frames per cardiac cycle were acquired on a three headed gamma camera and a computer system (Prism 3000XP, Odyssey super computer, Picker co., UTA, USA). Using a 120 deg. x 3 imaging arc, high resolution parallel-hole collimator, SPECT imaging was acquired in a 64 x 64 matrix size on 72 steps with 70 cardiac cycles per step, for a total imaging time of about 30 minutes. Gated frames were processed as 5.3 mm tomographic slices using a Butterworth filter with a 0.2 cut-off frequency and power of 8.0. The first frame was defined as the end diastolic phase. The end systolic phase was determined for each subject as a frame with the peak count on the count/time curve of mid slice of short axis imaging (Figure 1). For all patients, a polar map display was figured for the end diastolic phase and for the end systolic phase by using short axis slices after correcting cardiac rotation. Thereafter, wall thickening was quantitatively evaluated by the percent increase in counts from diastole to systole; a ratio defined as the wall thickening index (WTI; ES - ED/ED) on all pixels, and a polar map was also figured for WTI (Figure 2). Each polar map was divided into 9 segments: anterior, septal, inferior, lateral portions (they were further divided into basal and middle portions), and the apex. Regional myocardial perfusion was quantitatively assessed by %uptake of these segments on the polar map of the end diastolic 
phase. Regional wall thickening was assessed by the mean WTI of each segment on the polar map. 
18F FDG PET 
Transmission scans were obtained in all patients with 68Ge ring source of to correct for photon attenuation followed by administration of 370 MBq (10 mCi) 18F fluorodeoxyglucose (18F FDG). After a 60 minute distribution phase, 18F FDG PET imaging was performed with a HEADTOME II (Shimadzu, Kyoto, Japan) under 75 g oral glucose loading, producing 6 transaxial slices, each 8 mm thick.3 Nine myocardial segments were also defined on the six transaxial slices of PET images corresponding to the SPECT images. Since HEADTOME II could produce only transaxial slices, the middle portion of inferior segment was excluded from the analysis. 
Statistical analysis 
Analysis of variance (ANOVA) was used for comparing mean values, and Scheffe criteria was applied for multiple comparison. A linear regression analysis was used for evaluating the association between continuous variables. A two tailed p-value of less than 0.05 was defined as statistically significant. 
RESULTS 
Perfusion and systolic function normal subjects 
Demographic statistics of regional 99mTc tetrofosmin %uptake at end diastole and WTI values in normal volunteers are shown in Table 2. Among the normal subjects, mean regional %uptake ranged from 0.62 to 0.83. Regional mean WTI values ranged from 0.47 to 0.56, and no single WTI value among them was no less than 0.30 in all segments. 
Regional classification of myocardial viability 
Myocardial segments were regionally classified by perfusion and glucose metabolism. Myocardial perfusion was evaluated by mean %uptake of 99mTc tetrofosmin on the polar map in the end diastolic phase for each segment. The segment with mean %uptake of < 60% was defined as hypoperfusion based on the results for normal subjects. Glucose metabolism was evaluated by mean %uptake of FDG PET for each correspondent segment. A segment with a mean FDG uptake of < 50% was defined as reduced glucose metabolism. Of the 136 segments, 85 (63 %) segments were determined as normal perfusion. The remaining 51 (37%) segments demonstrated hypoperfusion. Of these, 25 (18%) segments demonstrated preserved (> 50%) 18F FDG PET uptake and were classified as "mismatch." Twenty six (19%) segments demonstrated abnormal perfusion and abnormal metabolism, and were classified as "matched reduced" (Figure 4). 
Wall thickening index (WTI) 
Figure 5 shows the regional mean WTI value in the segments with "mismatch" and "matched reduced." The mean WTI was 0.38 +- 0.21 in the segment with "mismatch" (i.e., preserved FDG uptake and reduced tetrofosmin uptake), and was significantly greater than mean WTI in the "matched reduced" segment, 0.15 +- 0.20 (p < 0.001). If the regional WTI was dichotomously defined as reduced (< 0.30) and normal (> 0.30) according to the data for normal subjects. Seventy percent of the "mismatch" segments had WTI above the threshold value. In contrast, only 30% had normal WTI value in "matched reduced" segments (p < 0.01). Figure 6-b shows the association between WTI and FDG %uptake in the segments with hypoperfusion. WTI had a strong positive linear correlation with %uptake of FDG (correlation coefficient (r) = 0.57, p < 0.0005). Their association is also evaluated for entire myocardial segments in Figure 6-a. Although their correlation was statistically significant (r = 0.26, p = 0.02), they did not seem to have a simple linear correlation. Sub-analysis was performed regarding the duration after the onset of myocardial infarction and the location of infarction. Percent uptake of 99mTc tetrofosmin in the segments with resent MI (1 month after to 3 months) was 0.49 +- 0.1 , and was similar to those in the segments with old MI (4 months and later), 0.45 +- 0.1 , but, FDG uptake was significantly greater for the segment with resent MI (0.61 +- 0.1 vs. 0.43 +- 0.2, p < 0.005). WTI was also greater in the segment with resent MI (0.32 +- 0.1 vs. 0.22 +- 0.2), although short of statistically significant (p = 0.1 ). As for segments with perfusion and glucose metabolism mismatch, 2 were located in the apex, 6 in the inferior, and 17 in the septal segment. 
DISCUSSION 
Accurate detection of myocardial viability is important for clinical decision making, since these segments are expected to be functionally reversible after revasucularization, and patients with such segments have a high risk of cardiac events. The radionuclide technique is frequently applied for the detection of myocardial viability and perfusion SPECT imaging using 201Tl or 99mTc agents. If the segment with asynergy detected by echocardiography or contrast left ventriculography has relatively preserved perfusion, it can be regarded as viable, but, underestimation is not rare in the viability assessment by perfusion imaging, especially when artifacts of the SPECT agents such as attenuation in the inferior or septal segments are involved. Currently, metabolic activity by 18F FDG PET has been regarded as a noninvasive gold standard for this assessment. A pattern of enhanced FDG uptake in the segments with reduced perfusion (termed the FDG blood flow "mismatch") indicates ischemic or hibernating myocardium that has shifted its metabolic substrate toward glucose utilization. Several studies have shown that these mismatch segments will manifest improved function after revascularization,4-7 and have a high mortality rate during 7 medical therapy. Myocardial wall thickening is also one of the important marker of the existence of vital tissue. This phenomenon may conceivably exist even if apparent inward movement of the myocardial wall is absent. In gated 99mTc tetrofosmin SPECT, regional wall thickening is sensitively measured by the increase in the count from diastole to systole due to a partial volume effect of the accumulated radioisotope. Regional myocardial perfusion with a less functional effect is also evaluated in gated SPECT at diastole. In this study, myocardial segments with hypoperfusion (TF %uptake < 60%) was classified in terms of viability by incorporating FDG uptake, then wall thickening in them was quantitatively compared by WTI, i.e., percent increase in counts from diastole to systole. The result demonstrated that WTI in the segment which was ischemic but viable was significantly greater than that in the non-viable segment, and good linear correlation was observed between WTI and FDG %uptake among hypo-perfused segments. These results suggested that WTI could distinguish the viable segment from the scar. Nevertheless. WTI demonstrated only weak correlation with FDG %uptake when the segments with normal perfusion were added to the regression analysis, and the expected slope began to lean as FDG uptake increasing. This phenomenon may partly be explained by the influence of limited spatial resolutions of 99mTc, in which the relationship between the measured maximum count in a region and the characteristic dimension of the object has a relatively steep slope of the curve over the range of dimensions 8-10 Figure 5 also showed that WTI in the segment which was ischemic but viable was even greater than that in the normal segments. Possible reasons for lower WTI values in the normal segments are; (1) The WTI values in the segments with infarction might be exaggerated because of the small denominator, i,e., low uptake of 99mTc tetrofosmin at end diastole. (2) Significant movement in normal segment can lead to image degradation. Although they are serious limitation in precise evaluation of wall thickening, they may facilitate the distinction of the viable region from the scar. In the sub-analysis, perfusion metabolism mismatch and greater WTI was more frequently observed in the segments with recent myocardial infarction, in which stunned myocardium is more likely to be expected. Many segments with perfusion metabolism mismatch was located in the inferior and septal segments. This phenomenon may partly be explained by the attenuation artifacts of SPECT agents in these segments. It was previously reported that the count density change from diastole to systole can supplement the attenuation artifacts in the diagnosis of SPECT studies I l Besides the wall thickening analysis, wall motion can also be evaluated by gated SPECT. Global functional analyses such as ejection fraction and regional motion has been assessed by edge detection of SPECT imaging through diastole to systole, and they have reportedly a good correlation with contrast ventriculography12 and blood pool studies 13,14 Nevertheless, quantitative wall motional studies have currently been limited to global function, and only visual analysis was performed for regional evaluation 15 Considering that motional analysis is readily available by means of contrast ventriculography, echocardiography or gated pool study for candidates for revascularization, assessment of wall thickening by gated SPECT can be another sensitive option for viability assessment in a region with severe functional damage. 
Study limitation 
Myocardial viability is determined only by nuclear study, and it does not promise improvement of ventricular function after coronary revascularization. Among the subjects studied, 4 patients have undergone coronary bypass operation. Of those, ail asynergic segments with WTI value of > 30% showed signs of wall motion recovery in contrast ventriculography after angioplasty. For simplicity of analysis and contrast with visual assessment in a clinical situation, definition of hypoperfusion with %uptake of < 60% was applied for all segments. Although middle of inferior segment was excluded from analysis, this may compromise the results. Since nonviable myocardium is encompassed by the minimum resolvable volume, normally contracting myocardium will contribute to an increase in the intensity of the scarred segment during systole, and this will overestimate the viability of the scar by the rotation and translation of the heart during the cardiac cycle. Further study involving gated SPECT reconstruction algorithms which compensate for cardiac motion is needed. In conclusion, for those segments determined as infarct by perfusion images, systolic functional analysis with gated SPECT is helpful in the differentiation of a viable myocardial region or artifact from scar, but, further clini-cal and technical assessment is required to eliminate overestimation of viability and to deserve clinical use. 
REFERENCES 
1. Rocco TP, Dilsizian V. Strauss HW, Broucher CA. Techne-tium-99m isonitrile myocardial uptake at rest. Relation to clinical markers of potential viability. J Am Coll Cardiol 14: 1678-1684, 1989.
2. Sawada SG, Allman KC, Muzik O, Beanlands RSB, Wolfe ER, Gross M, et al. Positron emission tomography detects evidence of viability in rest Technetium-99m sestamibi defects. J Am Coll Cardiol 23: 92-98, 1994.
3. Yamanouchi M. Yoshida K. Niwayama H, Nakagawa K, Aioi S, Shikama N, et al. Effect of the duration of fasting on myocardial fluorine- 18-fluorodeoxyglucose positron emission tomography images in normal males. Jpn Circ J 60 (6): 319-327, 1996.
4. Tillisch J, Brunken R, Marshall R. Schwaiger M, Mandelkem M, Phelps M, et al. Reversibility of cardiac wall-motion abnormalities predicted by positron tomography. NEJM 314 (14): 884-888, 1986.
5. Tamaki N, Yonekura Y, Yamashita K, Saji H, Magata Y, Senda M, et al. Positron emission tomography using fluorine-18 deoxyglucose in evaluation of coronary artery bypass grafting. Am J Cardiol 64 (14): 860-865, 1989.
6. Nienaber CA, Brunken RC, Sherman CT. Yeatman LA, Gambhir SS, Krivokapich J, et al. Metabolic and functional recovery of ischemic human myocardium after coronary angioplasty. J Am Coll Cardiol 18 (4): 966-978, 1991.
7. Eitzman D, al-Aouar Z. Kanter HL, vom Dahl J, Kirsh M, Deeb GM, et al. Clinical outcome of patients with advanced coronary artery disease after viability studies with positron emission tomography. J Am Coll Cardiol 20 (3): 559-565, 1992.
8. Galt JR, Garcia EV. Robbins WL. Effect of myocardial wall thickness on SPECT quantification. IEEE Trans Med Imag 9 (2): 144-150, 1990.
9. Ziffer JA, Cook CD, Folks RD, et al. Quantitative myocardial thickening assessed with sestamibi: clinical evaluation of a count-based method. J Nucl Med 30 (5): 1991.
10. Hoffmans EJ, Huang SC, Phelps ME. Quantitation in positron emission computed tomography: 1. Effect of object size. J ComputAssist Tomogr 3: 299-308, 1979.
11. DePuey EG, Nichols K. Salensky H, et al. Using gated technetium-99m-sestamibi SPECT to characterize fixed myocardial defects as infarct or artifact. J Nucl Med 36: 952-955, 1995.
12. Kouris K, Abdel-Dayem HM, Taha B, Ballani N. Hassan IM. Constantinides C. Left ventricular ejection fraction and volumes calculated from dual gated SPECT myocardial imaging with 99mTc-MIBI. Nucl Med Comm 13 (9): 648-655, 1992.
13. DePuey EG, Nichols K, Dobrinsky C. Left ventricular ejection fraction assessed from gated technetium-99m-sestamibi SPECT. J Nucl Med 34 (11): 1871-1876, 1993.
14. 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 36 (11): 2138-2147, 1995.
15. Chua T, Kiat H, Germano G. Maurer G, van Train K, Friedman J, et al. Gated technetium-99m sestamibi for simultaneous assessment of stress myocardial perfusion, post-exercise regional ventricular function and myocardial viability. Correlation with echocardiography and rest thallium-201 scintigraphy. J Am Coll Cardiol 23 (5): 1107-1114, 1994.