Technetium-99m teboroxime scintigraphy

European Journal of Nuclear Medicine Original article Technetium-99m teboroxime scintigraphy Clinical experience in patients referred for myocardial perfusion evaluation Laurence Bontemps, X~nia Geronicola-Trapali, Yehia Sayegh, Odile Delmas, Roland Itti, and Xavier Andr6-Fou~t Department of Nuclear Medicine and Department of Cardiology, Cardiological Hospital, F-69003 Lyon, France Received 14 January and in revised form 21 April 1991 Abstract. In order to evaluate the clinical value of a new myocardial perfusion tracer, a series of 30 patients (25 male, 5 female, mean age 56 years) referred for thallium 201 stress/redistribution scintigraphy has been studied using stress/rest (n = 7) or rest/stress (n = 23) protocols with technetium 99m teboroxime (Cardiotec SQUIBB). In all cases coronary artery disease was known or highly probable, with a history of myocardial infarction in 18 cases. Medical treatment was not discontinued at the time of stress testing, and coronary angiography was available in 27 patients. Exercise tests for both tracers were carried out on a bicycle ergometer during the same day, and the levels of exercise achieved for the 2°1T1 study were very similar to those achieved for 99"Tc-teboroxime. Studies performed in three planar projections were evaluated using a model with four territories: septal and anterior assumed to correspond to the left anterior descending artery, lateral and lateroposterior (left circonflex), inferior and posterior (right coronary artery) and apex. Classification of results was: normal, ischaemic, infarcted and infarcted with ischaemia. On comparison with the 2°1T1 results, agreement was found in 86% (37/43) of normal regions and in 82% (63/77) of abnormal regions. Relative to documented coronary artery lesions (27 patients), sensitivity and specificity of 2°1T1 and 99~Tc-teboroxime for exact correspondence between arteries and territories were respectively: 2°1T1: sensitivity 64%, specificity 60%; 99mTcteboroxime: sensitivity 62%, specificity 77%. These results, obtained in a given clinical context, indicate the ability of Cardiotec to evaluate myocardial perfusion with a significant saving in time, since the complete study duration (stress and rest) was: 2°1T1, 4 h 35rain+ 21 rain; 99~Tc-teboroxime, 1 h 52 rain+_ 29 rain. Nevertheless, the high liver uptake was responsible for 68% of non-evaluable inferior segments and the limited acquisition time makes the applicability of SPET questionable. Offprint requests to: R. Itti, Centre de Mgdecine Nucl~aire, H6pital Cardiologique et Pneumologique Louis Pradel, 59 boulevard Pinel, F-69003 Lyon, France Key words: Myocardial perfusion scintigraphy - Thallium-201 - Technetium-99m teboroxime - Coronary artery disease - Stress testing Eur J Nucl Med (1991) 18:732-739 Introduction The quite ideal physical characteristics of technetium 99m for nuclear imaging, i.e. optimal gamma-emission energy and half-life, have promoted the development of new agents for myocardial perfusion scintigraphy (Deutsch etal. 1981). Compared with the reference tracer, thallium-201, these new molecules are potentially able to improve image quality and to reduce acquisition time, as it is possible to inject a higher activity without increased dosimetry. Two major classes of 99~Tc-labelled compounds are presently available: cationic agents with a supposed active myocardial uptake, such as sestamibi (Cardiolite, Du Pont, Billerica, Ma, USA) (Karcher et al. 1988; Taillefer et al. 1988; Wackers et al. 1989) and neutral lipophilic molecules (Johnson et al. 1987; Seldin et al. 1989; Zielonka et al. 1989; Johnson and Seldin 1990) that are passively diffusible through the myocardial cell membrane and behave similar to xenon, such as teboroxime (Cardiotec-Squibb Diagnostics, Princeton, NJ, USA). According to their particular biochemical properties, the time-course of uptake and clearance is dramatically different for the two categories: sestamibi has a medium uptake rate and a very slow myocardial clearance, allowing imaging for several hours after injection without any tracer redistribution, whereas teboroxime is highly extracted at the myocardial first pass and shows a rapid blood clearance (Narra et al. 1986; Gachon et al. 1988; Liu et al. 1988; Maublant et al. 1988 ; Okada et al. 1988). Therefore, images can be acquired as soon as 2 rain after i.v. injection, but after 10 15 rain hardly any cardiac activity remains. Imaging protocols have to be adapted to these characteristics. © Springer-Verlag 1991 733 The purpose of our study was to compare 2°iT1 and 99mTc-teboroxime planar myocardial imaging in a population of patients referred for myocardial perfusion evaluation. This means that the diagnosis of coronary artery disease was already established, a high proportion of patients having a history of prior myocardial infarction, and that Z°aT1 scintigraphy was indicated as an adjunct to coronary angiography in order to assess the severity and the extent of hypoperfusion related to the coronary lesions. The two methods were compared with each other and both with coronary angiography to assess the ability of 99mTc-teboroxime to detect the presence of reversible or fixed defects and to identify abnormal vessels. In addition, the timing of the complete examinations, i.e. stress/redistribution for Z°~T1 and stress/rest or rest/stress protocols for 99mTc-teboroxime was checked in order to demonstrate the time saved with the new molecule. Patients and methods steps of 30 W every 3 rain. Exercise was limited by the onset of typical angina or ST segment changes of more than 1 mm in positive cases, or by fatigue or achievement of at least 80% of the maximum predicted heart rate (MPHR) in negative cases. When peak exercise was reached, tracer was injected (74-111 MBq of Z°lT1, or 740-1110 MBq of 99mTc) and exercise was continued for I rain at reduced workload. Of the 30 patients, 28 underwent both stress tests on the same day, with at least 4 h 2°1T1 redistribution time before the injection of 99mTc-teboroxime. 2°lTl imaging. Between 5 and 10 rain after completion of the stress test the patient was placed supine under the gamma-camera (Siemens ZLC 75 fitted with an allpurpose low-energy collimator) and 128 x 128 pixel images were acquired (zoom x 2 300 K counts for the first view and same time for the two subsequent views, Elscint Apex 110 computer) in three projections: anterior (Ant), 45 ° left anterior oblique (LAO) and left lateral (LL). Four hours later, redistribution images were obtained in the same views. Patients A population of 30 patients was selected for this study on the basis of a clinical indication of 2°1T1 stress/redistribution scintigraphy for myocardial perfusion assessment in thecontext of known or strongly suspected coronary artery disease. Using a consent form approved by the ethical committee of the institution, all patients gave written informed consent to participate in the study. The patient population includes 25 men and 5 women. Mean age was 56 (range 28-73) years. The sample included 18 patients who had a history of prior myocardial infarction, subacute in 6 cases (12 days to 3 months) and remote in 12 cases (older than 3 months). Localization of the infarction was: anterior in 5 cases, inferior in 9 cases, lateral in 2 cases, inferior and lateral in 1 case and anterior and lateral in 1 case. In 5 cases there was a history of coronary by-pass surgery or percutaneous transluminal coronary angioplasty (PTCA). Biplane coronary angiography was available in 27/ 30 patients. In 23 cases angiography was performed within 15 days of the scintigraphic study (less than 3 days in 17 cases). In 4 other patients the time lapse was more than 15 days, but it could be assumed that there was no significant clinical change during this period. Methods 99mTc-teboroxime imaging. The drug was prepared from a lyophilized kit by addition of 80-100 mCi of 99mTcpertechnetate followed by a 15-rain period of heating at 100 ° C. The labelling efficiency was checked with paper chromatography. In our series the bound activity (mean + SD) was 92.3 _+1.44%. The stress tests were performed very close to the gamma-camera so that the patient could be transferred as quickly as possible to the imaging device and the acquisition could be started exactly 2 rain after tracer injection. In order to obtain comparable pictures with 2°1T1 images the patient was installed supine under the gamma-camera. Dynamic 128 x 128 acquisition at 10 s/frame was started at this time and continued for 5 rain. During the first minute the patient was positioned in the anterior projection, after which 10-20 s were taken to change the camera position from anterior to 45 ° LAO, and the acquisition was carried out for an additional minute without resetting the computer program. In the same way the last projection (LL) was recorded for 80 s. After 1-1.5 h the patient was again placed under the gamma-camera for the resting study. A second 20- to 30-mCi injection of 99mycteboroxime was given and imaging was then performed using the same pattern as described above (Fig. 1). This protocol (stress/rest) was followed for 7 patients, whereas an alternative protocol (rest/stress) was used in the remaining 23 patients. The procedure was basically the same, starting with a resting injection followed by the stress test and imaging. Exercise protocol. Exercise tests were performed using the same protocol for the 2°1T1 scintigraphy and the 99mTc-teboroxime scintigraphy. The patient was seated upright on a bicycle ergometer. Graded exercise was performed using a protocol equivalent to the Bruce protocol, starting at a workload of 30 W and increasing in Image processing and analysis. Unprocessed 2°1T1 images were used for the subsequent evaluation. Concerning the 99mTc-teboroxime the dynamic images were summed in order to obtain l-rain images in the anterior and 45 ° LAO views (excluding the dynamic images acquired dur- t. rtl ..= C ,,O o~ ¢,1 | .Z ¢'t m O Fig. 1. Example of 2°iT1 stress/redistribution and 99mTc-teboroxime stress/rest images in three planar projections (anterior, 45 ° left anterior oblique and left lateral), Case of inferior myocardial infarction with left ventricular failure and limited inferoseptal ischaemia (coronary angiography: 90% stenosis of dominant RCA; 50% stenosis of LAD) 735 ing the period necessary for patient repositioning) and 80-s images in the left lateral projection. Images were interpreted by two independent observers with consensus at the end of the reading, following a segmental model with three segments for each view. Tracer uptake was evaluated as normal or abnormal. Segments which could not be analysed, for instance because of liver superimposition in the left lateral projection, were systematically assessed as normal. The comparison of stress/redistribution or stress/rest results was interpreted in terms of myocardial perfusion with a fourlevel classification: normal, ischaemic, necrotic and necrotic plus ischaemic. The segmental results were summarized using a model with four vascular territories: septal and anterior, assumed to correspond to the LAD (left anterior descending) artery, lateral and latero-posterior (LCX=left circonflex), inferior and posterior (RCA = right coronary artery) and apex whose vascular involvement may vary according to the coronary artery distribution. Concerning the comparison with coronary artery angiography, stenosis of at least 50% lumen reduction (visual analysis by two "blinded" independent observers with consensus) were considered as significant. Results Comparison of stress tests The levels of exercise achieved for the 2°aT1 study and the 99mTc-teboroxime study were, respectively (mean_+ SD): - Workload: 105_+38 W (2°1T1) vs 109+34 W (99~Tcteboroxime) -% MPHR: 76.7_+12.4% (2°1T1) vs 77.3_+12.6% (99mTc-teboroxime) There was no statistically significant difference between these values. Comparison o f 201 Tl and 99mTc-teboroxime for myocardial perfusion assessment A total of 540 segments (3 segments for each projection) was analysed. Overall agreement between 2°~T1 and 99mTc-teboroxime for normal or abnormal assessment was 73% (393/540). This agreement was about the same for stress images (74% =200/270) and for rest/redistribution images (7/% = 193/270). Agreement for the anterior projection (82% =147/ 180) and for the left anterior oblique projection (74% = 133/180) was better than for the left lateral projection (63%=113/180). In this latter projection 68% (41/60) of the inferior segments were not evaluable for the 99mTc-teboroxime images and therefore assessed arbitrarily as normal. Table 1. Comparison of the final diagnosis, including stress and redistribution/restimaging,in terms of normal, ischaemic,necrotic or necrotic plus ischaemic segments, for Z°lT1and 99mTc-teboroxime Teboroxime Tc99m Thallium201 Normal Ischemia Necrosis Necrosisand Ischemia Normal Ischemia Necrosis Necrosis and Ischemia 37 (86%) 3 0 3 12 23 (61%) 1 2 0 0 13 (100%) 0 2 4 3 17 (65%) Total 43 38 13 26 For the segments with readings disagreement (147/ 540 segments) the majority (108/147 segments, i.e. 73%) showed abnormal 2°1T1 uptake with normal 99mTc-teboroxime uptake, whereas only 39/147 segments, i.e. 27%, showed normal /°~T1 uptake with abnormal 99mTc-teboroxime uptake. For the 137 discordant segments with available angiographic data, the diagnostic accuracy was 66% (91/137 correct segments) for 2°1T1 and 33% (46/137 correct segments) for 99mTc-teboroxime. According to the classification of territories as described above with reference to the 2°1T1 findings, the 99mTc-teboroxime results are given in Table 1. In total, 90 out of 120 territories (75%) were precisely evaluated for 99mTc-teboroxime when the 2°aT1 classification was taken as a reference. When the analysis was limited to two levels (normal/ abnormal) instead of the four levels previously mentioned, total agreement between 2°aT1 and 99mTc-teboroxime was found in 37 out of 43 normal territories (86%) and in 63 out of 77 abnormal territories (82%). Detailed analyses for anterior, lateral, inferior and apical regions are given in Fig. 2. The most relevant fact from these results is the low detection rate of inferior wall abnormalities (8 out of 18 =44%) with 99mTc-teboroxime scintigraphy. Comparison with coronary angiography Among the 27 patients for whom coronary angiography was available, the majority (12 patients) had triple vessel disease. For the other patients, the distribution of coronary lesions was as follows: 4 had non-significant stenosis, 7 had single vessel disease (3 LAD, 2 LCX and 2 RCA) and 4 had double vessel disease (1 LAD + LCX, 2 LAD + RCA, and 1 LCX + RCA). Sensitivity and specificity of 2°~TI and 99mTc-teboroxime for detection of a coronary artery lesion are given 736 i~T% 100 90 1 80 70 60 50 40 9O 80 70 60 50 40 30 30 20 20 10 10 0 0 Total Anterior Lateral Inferior TI Apex Fig. 2. Classification (normal/abnormal) of 99mTc-teboroxime results with reference to the 2°*T1 results. Values are expressed as percentage of agreement between both studies for all segments taken together (total) and for each individual territory (anterior, lateral, inferior and apex). [] Normal; [] abnormal on Fig. 3. To compute these values close correlation between a given coronary artery and the corresponding vascular territory was mandatory. Owing to coronary artery anatomic variabilities, the apical region has not been included. Globally, for all three coronary arteries, the sensitivity of 2°iT1 was 64% (32 abnormal territories for 50 coronary artery lesions) and the specificity was 60% (18 normal territories for 30 territories supplied by normal coronary arteries). For 99mTc-teboroxime the corresponding results were sensitivity, 62% (31/50) and specificity, 77% (23/30). Detailed results for individual coronary arteries are given on Fig. 3. Comparison of imaging time The total study time, including stress testing, delay between stress imaging and rest/redistribution and rest imaging, or the opposite in the case of a rest/stress protocol was compared individually for the 30 patients between 2°iT1 and 99mTc-teboroxime. Graphic results are given in Fig. 4, which shows that in a high proportion of cases the total study time was less than 1 h 30 min. For patient 16 a temporary breakdown of the electronics system of the bicycle ergometer was responsible for an abnormally long time lapse before the 99mTc-teboroxime study. Excluding this special case, the average study times were: 2°~Tl: 4 h 35 min+_21 min. 99~Tc-teboroxime: 1 h 52 min + 29 rain. There was no difference between the stress/rest protocol (1 h 55 rain+ 35 rain, n=7) and the rest/stress protocol (1 h 50 rain±28 rain, n=22). Tc Global 3"1 Tc L.A.D Tc L.C.X TI I1 Tc R.C.A. Fig. 3. Sensitivity and specificity for diagnosis of significant coronary artery stenosis with matching of coronary arteries and scintigraphic territories (LAD = anterior, LCX = lateral, R C A = inferior). Comparison of E°~T1 and 99~Tc-teboroxime for all territories taken together (global) and for each artery separately. [] Sensitivity; [] specificity Patient # 30 ................... f...................................................................................~..... ..................................................................................................................... t 20 ....................................................................................................... 10 ............................................................................................................................................ ............................................................................................................................................... 0 50 100 150 200 250 j 300 350 Total stress/rest study time (minutes) Fig. 4. Total study time, including stress testing, imaging and delay between stress and redistribution or rest imaging for 2°~T1 and 99~Tc-teboroxime scintigraphy. Individual values for patients I to 30 are given in graph form. [] Thallium-201 ; • teboroxime-Tc 99m Discussion Patient selection Besides the diagnosis of coronary artery disease in patients presenting with a more or less high pre-test probability (prevalence) of the disease, another important application of stress myocardial perfusion scintigraphy is the evaluation of perfusion defects in patients with known coronary artery lesions. In the first case it is important, in order to avoid any drug interference that could lead to false-negative results, that the patients have 737 stopped all medications for an adequate period of time prior to the study. In the second case, not only for ethical reasons but also for practical reasons, it is acceptable that the treatments are not discontinued, so that the corresponding perfusion evaluation reflects the real status of the patient, including the effects of his treatment. The radioisotope study is then a complement to the coronary angiography. As our aim was the comparison of 99mTc-teboroxime scintigraphy with the 2°1T1 reference data, and as much as possible with coronary angiography, patients included in this study were selected on the basis of known lesions and the evaluations have therefore been performed during treatment. Image quality From the count density point of view, the image quality for 99mTc-teboroxime and 2°iT1 is comparable, but nonspecific activity (liver and lungs) may be higher for 99mTc-teboroxime. Liver uptake is a major problem only in the left lateral view, in which it may happen that the inferior wall is partly masked by the liver activity. In our experience, the inferior wall could not be properly analysed in 68% of cases, which could be a significant drawback for the method. Pulmonary activity is high during the first few minutes immediately after 99mTcteboroxime injection, but if acquisition starts at least 2 min after injection, significant lung activity persists in only very few cases. As demonstrated for 2°1T1, a persistent pulmonary activity may be clinically relevant and a consequence of some degree of left ventricular failure (Boucher et al. 1980). Therefore, such an observation should not be considered a degradation of image quality, but rather seen as a useful observation which may improve the diagnosis. Comparison 99mTc-teboroxime versus 2° ~Tl Sequential analysis of individual projections and territory analysis combining information from different projections showed a globally acceptable correlation between 99mTc-teboroxime and 2°~T1 for the normal/abnormal classification. The agreement was better for composite vascular territories (100 out of 120 territories=83%) than for individual segments (395 out of 540 segments = 73%). This difference can be explained by the fact that in anterior projection the segmental score (82%) is fairly identical to the territory score (83%), but in the other projections ( L A O = 7 4 % and lateral=63%) there is a clear lack of agreement. The worst results were obtained in the lateral view, and therefore the vascular territories, which are mostly dependent on this projection, like the inferior wall, showed the lowest scores (44% sensitivity for inferior wall abnormalities). In addition, the lateral view was also the last performed in the series of three views, and it is probable that there was some degradation in the image quality with time. One important fact is that in cases of disagreement between 99mTc-teboroxime and Z°IT1, a majority of segments (73%) showed abnormal 2°lT1 uptake with normal 99mTc-teboroxime uptake and only a minority (27%) normal 2°1T1 uptake with 99mTc-teboroxime uptake. In these cases of discrepancies, comparison with angiographic data showed better accuracy for z°aT1 (66% agreement) than for 99mTc-teboroxime (33%). These facts could indicate some loss of sensitivity for 99mTc-teboroxime versus 2°1T1. A major problem in the study protocol that could be the cause of the poor results observed for the inferior wall is the position of the patient for recording the lateral view (right lateral decubitus). This position was imposed by the time needed for the thallium study, which was extended for the 99mTc-teboroxime study in order to allow adequate comparison of the two. It is clear that owing to the short acquisition time needed to obtain good 99mTc-teboroxime images, it would be more efficient to image the patients in the upright (sitting) position, allowing the liver to drop downwards, which is not easily practicable for 2°iT1 because of the much longer acquisition time. The fact that the liver is superimposed on the inferior wall would therefore be less important and the diagnosis of inferior wall defects would improve (Johnson et al. 1987). When the diagnosis must be established in terms of ischaemia, necrosis or a combination of both, taking into account the stress as well as the rest or redistribution pictures, about 75% of the regions are correctly evaluated by 99mTc-teboroxime with reference to the 2°1T1 results, but this score is better for normal regions (86% correctly diagnosed) than for abnormal zones (70%). In necrotic areas (100%) the response is much better than when ischaemia is present, associated with necrosis or not, and the score drops down around 65%. This means that it is sometimes difficult to judge the reversibility of a 99mTc-teboroxime defect when there is evidence of 2°tT1 redistribution after a stress-induced defect. Therefore, it is suggested that this new tracer would in fact detect ischaemia less frequently than 2°lT1. The problem raised here may also be attributed to the fact that there is a fundamental difference between redistribution and rest reinjection, whether it is 2°aT1 or 99mTcteboroxime. In cases of resting ischaemia, a defect may appear to be persistent on resting images, even if there is a 4-h 2°1T1 redistribution, and this fact may be of even more importance if the redistribution delay is 24 h instead of 4 h. The usefulness of 24-h redistribution images for 2°~T1 is now widely accepted and the reinjection protocols cannot replace the long delay images. Correlations with coronary angiography Since most patients already had coronary abnormalities, the correlation with coronary angiography was not lira- 738 ited to a single sensitivity assessment (without specificity), regardless of the regions where the coronary versus scintigraphic abnormalities were found, but an exact territorial correlation between the coronary lesions and the corresponding vascular beds was required. In this way regional sensitivities (uptake defect in a region supplied by a stenosed artery) and specificities (normal uptake in a region supplied by a normal artery) could be defined. Under these very strict conditions it is obvious that the values of regional sensitivities and specificities are relatively low for both Z°tT1 and 99mTc-teboroxime (between 60% and 77%), and many important factors are not being taken into account, such as the anatomic variability of the coronary network, the haemodynamic significance of a given degree of stenosis and the presence of collaterals. These restrictions apply for both tracers, and the important fact in this comparative study is that the results, including the limitations, are comparable for/°iT1 and 99mTc-teboroxime. The best value obtained in this approach was 77% global specificity (all territories taken together but with individual correspondence required) for 99mTc-teboroxime, which means that this tracer would allow a reduced proportion of falsepositive results compared to 2°1T1 (specificity=60%) without significant loss of sensitivity (64% for 2°1T1 vs 62% for 99mTc-teboroxime). Detailed analysis of the data reported in Fig. 3 nevertheless shows that this gain in specificity is marked for the LAD and for the RCA territory. In the last case the increased specificity can be attributed to the fact that the inferior wall has been arbitrarily assessed as normal in cases where the lateral image suffered from extensive superimposition of the liver. Timing of the studies During this protocol major attention was devoted to the time gain that the new molecule could provide. This benefit appears at different levels, together with the time needed by the camera for acquisition of each set of pictures and the complete study time, including the rest and stress images and the delay required between the two studies. Due to the high 99myc dose which can be injected, a complete set of three views can be acquired within less than 5 rain, which is at least three times less that required for Z°lT1 (and on many occasions much more than three times less). The rapid clearance of the tracer allows identical doses to be injected at rest and at stress, regardless of the order in which the two studies are performed, and therefore both types of pictures are acquired for the same time period, which is not usual for the "same-day" protocols proposed for other 99mTclabelled tracers, like sestamibi, where the second injection must be under more active conditions in order to cover the remaining activity from the first injection (Taillefer et al. 1989). This fact makes the comparison of successive pictures easier. Concerning the total study time, it should theoretically be shorter in a rest/stress protocol than for a stress/ rest protocol because in the first case the stress test may start immediately after the resting pictures during the washout phase, so that the exercise time takes place during an "active" phase. In the case of a stress/rest protocol, on the other hand, the exercise time must be added to the stress-rest delay and, in addition, this delay should be a little more prolonged in order to allow the patient to come back to a physiologically resting state. In our experience we have nevertheless not found any difference between the times for the two protocols, probably because the studies were included in a routine day program so that the camera was not always available at the optimal time for recordings. It is clear that compared to 2°1T1, 99mTc-teboroxime allows an important reduction in the total study time, which could be of major interest in the optimal management of a nuclear medicine department. On the other hand, there is an important drawback that must be taken into consideration for use of 99mycteboroxime in the clinical routine: due to the very short myocardial time, the performance of SPET is probably difficult with a single-headed camera. Nevertheless, the feasibility of SPET has been demonstrated in preliminary works by some authors (Drane et al. 1989; Zielonka et al. 1989). Their first results are promising even if they still have to be confirmed by further investigations. Conclusions Even if the more than 15 years' use of Z°lT1 for myocardial perfusion assessment had not strongly marked the routine in nuclear cardiology, the appearance on the market of 99mTc-teboroxime would have been regarded enthusiastically as a major advance in cardiac imaging. In the present situation, any new drug must be compared to the reference tracer 2°1T1 and must be superior to it. Our results show comparable diagnostic capabilities for both Z°IT1 and 99mTc-teboroxime, and they correlate with coronary angiography, which suffers from the same limitations as when an anatomic examination is compared to a functionnal one. Therefore, the new tracer appears to be very useful for the clinical practice, and its specific features allow the optimization of study timing, which can be very important for busy nuclear medicine departments. A very short imaging time of about I rain for single projections may even suggest the development of new protocols, taking into account the rapid tracer clearance and the ability to obtain repeat images with short delays. For this reason, 99mTc-teboroxime can be considered for the future to approach a "real-time" myocardial imaging agent. 739 Acknowledgements. The authors would like to thank the technical team of the nuclear medicine department for their active and efficient participation in this study. References Boucher CA, Zir LM, Beller GA, Okada RD, McKusick KA, Strauss HW (1980) Increased lung uptake of thallium-201 during exercise myocardial imaging: clinicai, hemodynamic and angiographic implications in patients with coronary artery disease. Am J Cardiol 46:189-196 Deutsch E, Glavan KA, Sodd VJ, Nishiyama H, Ferguson DL, Lukes SJ (1981) Cationic 99myccomplexes as potential myocardial imaging agent. J Nucl Med 22:897-907 Drane WE, Decker M, Strickland P, Tineo A, Zmuda S (1989) Measurement of regionai myocardial perfusion using Tc-99m teboroxime (Cardiotec) and dynamic SPECT (abstract). J Nucl Med 30:1744 Gachon P, Moins N, Renoux M, Maublant J (1988) A comparison between thallium-201 and two new technetium-99m labeled myocardial blood flow imaging agents in cultures of cardiac cells (abstract). J Mol Cell Cardiol 20 [Suppl 5] :49 Johnson LL, Seldin DW (1990) Clinical experience with technetium-99m teboroxime, a neutrai, lipophilic myocardial perfusion imaging agent. Am J Cardiol 66 : 63E-67E Johnson LL, Seldin DW, Muschel M J, Smith KF, Blood DK, Cannon PJ (1987) Comparison of planar SQ30217 and T1-201 myocardial imaging with coronary anatomy (abstract). Circulation 76 [Suppl 4] :217 Karcher G, Bertrand A, Moretti JL, Maublant J, Mathieu E, Itti R, Bourguignon M, Baconnet B (1988) Comparative study by two independent observers of the uptake abnormalities of 201 thallimn and Tc-99m-MIBI in 81 patients with coronary artery disease (abstract). J Nucl Med 29:793 794 Liu P, Dawood F, Houle S, Riley R, McLaughlin PR (1988) Myocardial uptake and clearance of new myocardial perfusion tracers SQ32014 and SQ30217 in an isolated heart model: comparison with thallium (abstract). Circulation 78 [Suppl II] : 126 Maublant J, Gachon P, Moins N (1988) Hexakis (2-methoxy isobutylisonitrile) technetium-99m and thallium-201 chloride: uptake and release in cultured myocardial cells. J Nucl Med 29 : 48-54 Narra RK, Feld T, Wedeking P, Matyas J, Nunn AD (1986) SQ30217, a technetium 99m labelled myocardial imaging agent which shows no interspecies differences in uptake (abstract). Nuklearmedizin 25:489 491 Okada RD, Glover D, Gaffney T, Williams S (1988) Myocardial kinetics of technetium-99m-hexakis-2-methoxy-2-methylpropyl-isonitrile. Circulation 77:491-498 Seldin DW, Johnson LL, Blood DK, Muschel M J, Smith KF, Wall RM, Cannon PJ (1989) Myocardial perfusion imaging with technetium-99m SQ 30217: comparison with thallium-201 and coronary anatomy. J Nucl Med 30:312-219 Taillefer R, Dupras G, Sporn V, Rigo P, L6veille J, Boucher P, Perez-Balino N, Camin LL, McKusick KA (1988) Myocardial perfusion imaging with a new radiotracer, technetium-99m-hexamibi (methoxy-isobutyl isonitrile): comparison with thallium201 imaging. Clin Nucl Med 14:89-96 Taillefer R, Gagnon A, Laflamme L, Gr6goire J, L6veitle J, Phaneuf DC (1989) Same day injections of Tc-99m methoxy isobutyl isonitrile (hexamibi) for myocardial tomographic imaging: comparison between rest-stress and stress-rest injection sequences. Eur J Nucl Med 15:113-117 Wackers F, Berman D, Maddahi J, Watson D, Beller G, Srauss HW, Boucher C, Picard M, Holman BL, Fridrich R, Inglese E, Delaloye B, Bischof-Delaloye A, Camin L, McKusick K (1989) Technetium-99m hexakis-2-methoxyisobutyl isonitrile: human biodistribution, dosimetry, safety and preliminary comparison to thallium-201 for myocardial perfusion imaging. J Nuel Med 30:301-311 Zielonka JS, Bellinger R, Coleman RE, Drane W, Williams C, Goris M, Johnson L, Seldin D, Leppo J, Reba R, Wassermann A (1989) Multicenter clinical trial of 99m-Tc teboroxime (SQ 30,217; Cardiotec) as a myocardial perfusion agent (abstract). J Nucl Med 30:1745