Eur J Nucl Med (1986) 12:219-225 European M I i t ' , l t~Jr Journal of I ~lUK./l~Ir., IL/

Medicine © Springer-Verlag 1986

The scintigraphic evaluation of myocardial infarction and regional ventricular performance using technetium-99m hexakis (t-butylisonitrile) technetium (I) (TBI): A new myocardial imaging agent* Stephen Campbell, B. Leonard Holman, James M. Kirshenbaum, Elliott M. Antman, John Lister-James, Alan Davison, Joseph Kozlowski, Robert J. English, and Alun G. Jones Departments of Medicine (Samuel A. Levine Cardiac Unit, Cardiovascular Divison) and Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA, Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA

Abstract. Technetium-99m hexakis (t-butylisonitrile) tech- netium (I) (99mTc-TBI) is a new myocardial perfusion imag- ing agent. To determine its potential in the evaluation of myocardial infarction, 15 patients with suspected or con- finned acute infarction were studied by bedside imaging in the coronary care unit. Good-quality planar scintigrams in multiple projections were obtained in 13 patients. Gated perfusion studies were performed in 14 patients, and for comparison 13 of these were restudied 24-72 h later by standard gated equilibrium blood pool radionuclide ventri- culography. Conventional and planar scintigraphic criteria for myocardial infarction (acute or old) agreed in 12 (92%) patients (k= 0.81, p < 0.05). All the infarctions detected by scintigraphy were associated with electrocardiographic Q- waves. Localization of infarction by the electrocardiogram and scintigraphy exhibited moderate agreement (k=0.49, p < 0.1). Regional wall motion analysis by standard radio- nuclide ventriculography and gated 99mTc-TBI scintigraphy were in complete agreement for 25 (64%) of 39 left ventric- ular segments (k = 0.35, p < 0.05). However, in 7 other seg- ments, associated with areas of infarction, regional wall motion abnormalities were noted only on gated 99mTc-TBI scintigraphy. Therefore, 99mTc-TBI scintigraphy can readily provide data on regional myocardial perfusion and wall motion, permitting detection and localization of areas of myocardial infarction. The superior imaging properties, ready availability and low cost of 99mTc point to the consid- erable potential value of 99rnTc-TBI in assessing patients with suspected or confirmed myocardial infarction.

Key words: 99mTc-TBI - Myocardial infarction

In the presence of typical clinical features the diagnosis of myocardial infarction is rarely in doubt (Alpert and Braunwald 1984). Many infarctions, however, are clinically unrecognized (Kannel and Abbot 1984) the prognosis in these cases appears similar to, or even slightly worse than,

* This work was done during the tenure of a British-American Research Fellowship of the American Heart Association and the British Heart Foundation, with Dr. S. Campbell the recipient

Offprint requests to: B. Leonard Holman, Department of Radiolo- gy, Brigham and Women's Hospital, Boston, MA 02115, USA

that after recognized infarctions (Geltman et al. 1979). Both prognosis and residual left ventricular function are deter- mined largely by the extent of myocardial damage (Geltman et al. 1979; Schelbert et al. 1976), which is related to the site of infarction (Geltman et al. 1979; Thanavaro et al. 1982). Therefore, investigations which can accurately detect and localize infarction are of great importance.

Planar myocardial perfusion scintigraphy can be used for this purpose (Henning etal. 1977; Iskandrian and Hakki 1985; Niess et al. 1979; Silverman et al. !980; Ta- maki et al. 1982; Wackers et al. 1976a, b, 1977, 1979) and can potentially evaluate residual left ventricular function if electrocardiographically gated methods of acquisition are used (Alderson et al. 1978). Most studies to date have uti- lized thallium-201 (2°lTh), which has provided much valu- able information, but has serious drawbacks as a myocar- dial imaging agent (Holman 1984a). Its low primary X-ray photon energies (69-83 keV) result in significant tissue at- tenuation and reduced resolution. In addition, imaging times are relatively long because the administered dose is limited by radiation dosimetry related to its long physical half-life (72 h). Finally, its cost remains an important limit- ing factor.

Consequently, there have been many attempts to devel- op a myocardial perfusion agent labeled with 99mTc, an inexpensive and readily available radionuclide with better physical decay characteristics for imaging. Unfortunately, despite promising results in animal models (Deutsch et al. 1981 a, b), success in man has been limited (Dudczak et al. 1983).

Recent reports from this institution have shown that the cationic complex 99mTc-hexakis (t-butylisonitrile) tech- netium (I) 99mTc-TBI) demonstrates prompt myocardial uptake in experimental animals after intravenous injection and that it is retained in the myocardium for several hours, permitting good-quality imaging (Jones et al. 1984). Its dis- tribution has also been shown to be directly and linearly related to regional myocardial blood flow in both normal and ischemic myocardium (Holman et al. 1985). Most im- portantly, preliminary studies in man have indicated similar properties, permitting planar, tomographic and gated imag- ing of excellent quality (Holman et al. 1984b). We have therefore investigated the potential of 99mTc-TBI myocar- dial scintigraphy in the detection and localization of myo- cardial infarction and the assessment of residual regional

220

left ventricular function in comparison with currently ac- cepted methods.

Patients and methods

Patients

This study was approved by the Human Subjects Commit- tee of the Brigham and Women's Hospital, Boston, Massa- chusetts and the use of the radioisotopes by the Radiation Safety Committee of the same institution. We studied 15 consecutive patients admitted to the coronary care unit with suspected acute myocardial infarction or continuing pain after a recent confirmed infarction, who were willing to give written informed consent and who did not develop major complications. Ten were male and five female; their mean age was 52.1 years (range 35-79 years).

Clinical assessment

On admission all patients were assessed by history, exami- nation and electrocardiographic findings. Blood was drawn for serum cardiac enzyme estimations. All of these investi- gations were repeated daily while the patients remained on the coronary care unit. The diagnosis of acute myocardial infarction required the presence of at least two of the fol- lowing three criteria: a history of typical chest pain persist- ing for at least 30 min, a characteristic rise in serum cardiac enzymes (creatine phosphokinase, total and MB fraction) and evolution of pathological Q-waves and/or ST-T wave changes on the electrocardiogram. Prior documentation of similar criteria indicated a previous infarction. For the pur- pose of comparison with scintigraphy, all the electrocardio- grams were read blindly and in random order by two experi- enced observers (JMK, EMA), to confirm the presence of infarction by electrocardiographic criteria and to determine its localization. The individual clinical details for each pa- tient are summarized in Table 1.

99mTe- TBI radionuclide studies

Preparation of 99mTc-TBL As described in detail previously (Holman et al. 1984b), the 99mTc-TBI was prepared by li- gand exchange from a sterile, pyrogen-free, solution of zinc bromide adduct of t-butylisonitrile and a freshly made stan- dard preparation of 99mTc-glucoheptonate. The radiophar- maceutical was tested to ensure its radiochemical purity and apyrogenicity, after which all subjects received 5-10 mCi by peripheral intravenous injection.

Imaging protocol. All 15 patients were injected with 99mTc- TBI while at rest in bed in the coronary care unit, a mean of 85.4 h (range 20-168 h) after the onset of acute symp- toms. Planar imaging in the anterior projection was per- formed immediately after injection, at the bedside, using a portable small-field-of-view Anger camera. One hour later, using the same camera, imaging was repeated in the anterior and the 30 ° and 70 ° left anterior oblique projec- tions. For each image 500,000 counts were acquired in a 128 x 128 matrix, which took from 3-5 rain. The unpro- cessed analogue images were recorded directly from the camera oscilloscope, on Polaroid film, for each projection. The data were also stored on hard disks and could be viewed in digital form, permitting enhancement and back- ground subtraction.

Immediately after acquisition of the planar series, elec- trocardiographically gated imaging (32 frames per R-R in- terval) was performed in the 30 ° left anterior oblique projec- tion; 3.5 million counts were acquired in a 64 x 64 matrix, which took approximately 20 rain. These data were stored on a hard disk and were displayed in a continuous loop cine mode, after nine-point smoothing, for analysis of re- gional wall motion.

Interpretation of 99mTe-TBIscintigrams. The series of planar analogue images were reviewed in random order by three experienced observers (BLH, SC and one idependent ob- server) blinded to the clinical and laboratory data. The pres- ence of perfusion abnormalities (either greatly reduced or absent perfusion, hereafter referred to as defects) was noted and scintigrams were rated as positive (defects present) or negative (no defect). Localization was determined as with conventional myocardial scintigraphy (Iskandrian and Hakki 1985). Disagreements were resolved by consensus. The gated studies were displayed in cine mode as previously described and reviewed similarly. The left ventricle was di- vided into three regions : anteroseptal, apical and inferopos- terior. The wall motion in each was qualitatively graded as normal, hypokinetic, akinetic or dyskinetic.

Standard radionuclide ventriculography

Patients were restudied 24-72 h after the 99mTc-TBI scintig- raphy, using the same portable Anger camera to assess left ventricular function by routine gated equilibrium radionu- clide ventriculography (Maddox et al. 1978). Following in vivo red blood cell labeling with 20 mCi 99myc, imaging was performed in the 30 ° left anterior oblique position to permit a direct comparison to be made with the 99mTc-TBI gated studies. A total of 10 million counts were acquired in a 64 x 64 matrix, using 32 frames per R-R interval. The studies were processed using a previously described com- puter system (Maddox et al. 1978) to generate ejection frac- tion images. Regions comparable to those used in the analy- sis of the 99mTc-TBI gated studies were identified and graded independently of the 99mTc-TBI data as: normal, hypokinetic, akinetic or dyskinetic.

Contrast angiographic studies

Coronary arteriography in multiple views, with or without left ventriculography, was performed in 9 of the 15 subjects by the Judkins technique (Judkins 1967), either because of continuing chest pain or for prognostic reasons. All these tests were interpreted independently by the angiographers without knowledge of the 99mTc-TBI scintigraphic data.

Data analysis

Cohen's kappa test (Cohen 1968) was used to assess the level of agreement between 99mTc-TBI scintigraphy and conventional clinical criteria for detection of myocardial infarction, the electrocardiogram for infarct localization and standard radionuclide ventriculography for assessment of regional wall motion. Values for kappa (~c) are given with approximate 95% confidence limits in parentheses (K= 1 means complete agreement, while K=0 means that agreement is poor and can be accounted for by chance alone; intermediate values represent a scale of agreement).

22:1

Table 1. Clinical data and summary of investigations

Clinical data 99mTc-TBI Scintigraphy RVG

Case Age Sex Prior Acute Planar RWM RWM

no. (years) MI MI defects Ant- Ap In f- Ant- sep P sep

Angiography

(%DR)

Ap lnf-P LAD CX RCA

1 35 M 0 Ant-sep Technically Hk N N N N N 100 70 80 (Q) inadequate

2 61 M 0 0 0 N N N N N N 50 70 0 3 46 M Inf Ant-sep Ant + sep Ak Ak N Hk Hk N 90 0 0

(non Q) (Q) 4 55 F 0 0 0 N N N - - - (LMCA 70) 0 5 57 F 0 Ant Ant Hk Hk N Hk Ak N 100 70 90

(Q) (LMCA 90) 6 59 M Ant Ant Ant-lat + Hk Ak Hk Hk Dk Ak - -

(Q) extension Ap + inf-Ap

7 57 F 0 0 0 N N N N N N - - 8 56 F Non-Q 0 0 N N N N N N - - - 9 47 M Ant Ant + Ant + ap - - - N Ak Hk 100 90 100

(Q) Inf(Q) 10 46 M 0 Inf Inf, inf-P N Hk Hk N N N 0 0 70

(Q) + P-lat 11 71 F Ant-sep Ant-lat Ant + sep Hk N N N N N - - -

(Q) 12 39 M 0 Inf Inf-ap + P N Ak Ak N Ak Hk - - -

(Q) + sep 13 79 M 0 Inf Inf+ Inf-P N Hk Hk N N N -

(Q) 14 36 M 0 Ant + Ant-ap + Ak Ak N N Hk N 90 0 0

Inf(Q) Inf-ap 15 37 M 0 Inf(Q) Technically N Ak N N N N 0 0 100

inadequate

Ak = akinetic, Ant = anterior, Ap = apical, Cx = circumflex coronary artery, Dk = dyskinetic, DR = diameter reduction, F = female, Hk = hypokinetic, In f = inferior, LAD = left anterior descending, Lat = lateral, LMCA = Left main coronary artery, M = male, MI= myocardial infarction documented by conventional criteria, localized by electrocardiogram and designated Q or non Q-wave MI, N= normal, P = posterior, RCA =right coronary artery, R VG= standard equilibrium blood pool radionuclide ventriculography, R WM=regional wall motion, Sep ~ septal

Results

Technically adequate planar scintigrams were obtained in 13 of the 15 patients studied. Two studies were inadequate because of excessive liver and lung activity respectively. In all patients the immediate post-injection anterior image showed high lung activity. However, in most patients, much of this had cleared by the 1-h series and myocardial uptake was then prominent, although considerable liver activity remained present in most subjects. These findings are con- sistent with those reported in the preliminary studies in man (Holman et al. 1984b). Consequently, only the 1-h planar images were analyzed for myocardial perfusion de- fects. The results for individual patients are summarized in Table 1.

Detection o f myocardial infarction

Of the 13 patients with adequate scintigrams, 10 had docu- mented recent and/or old myocardial infarction, by conven- tional criteria; scintigraphic defects were present in 9 of the 10 patients. Only one patient (no. 8) had a false-negative scintigram; she had an old non-Q-wave infarction but no evidence of recent acute infarction. Three patients had no evidence of infarction and their final diagnoses were: hyper-

trophic cardiomyopathy and coronary artery disease (no. 2), unstable angina (no. 4) and atrial flutter (no. 7); all 3 had normal scintigrams. An example of a normal study (patient no. 7) showing uniform myocardial uptake of 99mTc-TBI is shown in Fig. 1. Therefore, the scintigraphic and conventional criteria were in agreement in 12 (92%) of the 13 patients, tc=0.81 (0.44, 1.1), p<0 .05 , and these data are summarized in Table 2. Scintigraphy detected both recent and old infarctions with similar accuracy, and did not appear able to distinguish one from the other, especially where they were in the same or similar territory (patients 6, 11), All the infarctions detected were associated with electrocardiographic Q-waves and 2 old non-Q-wave infarc- tions were not detected (patients 3, 8). 99mTc-TBI scintigra- phy accurately diagnosed all 8 patients with no prior history of infarction, 5 of whom had an acute Q-wave infarction.

Localization o f Q-wave infarction by scintigraphy

In the nine patients with Q-wave infarctions and scinti- graphic defects, localization of the infarction by the electro- cardiogram was as follows: anterior (including antero-sep- tal, antero-apical, and antero-lateral) (four patients), inferi- or (including infero-lateral and infero-posterior) (three pa-

222

Fig. 1. Digitized planar scintigrams in the anterior and 30 ° left anterior oblique projections in patient no. 7, with 30% background subtraction and an upper threshold set at 80% of the maximum count. Note the uniform uptake of the radionuclide in this patient without infarction. Persistent high activity is also seen in the liver

Table 2. Planar scintigraphic defects versus conventional criteria for myocardial infarction (M1) (recent and old)

MI No MI

Scintigraphic Present 9 0 tc = 0.81 (0.44,1.2) defects Absent 1 3 p < 0.05

K = Cohen's kappa with approximate 95% confidence limits in pa- rentheses

tients) and anterior plus inferior two (two patients). In the same group, four patients had anterior scintigraphic defects and a representative study (patient 3) is seen in Fig. 2, showing reduced tracer uptake in the septum and anterior wall of the left ventricle. Two patients had inferior defects and Fig. 3 is a representative study (patient 13), showing marked reduction in uptake of 99mTc-TBI in the inferior and inferoposterior walls of the left ventricle; finally, three patients had both anterior and inferior scintigraphic defects. These data are summarized in Table 3. Thus there was com- plete agreement between the electrocardiogram and scintig- raphy in six (67%) of the nine patients, to= 0.49 (0.03, 0.95), p < 0.1, which represents moderate agreement. In two cases (nos. 6,12) the scintigraphic defects were more extensive than anticipated from the electrocardiogram. Therefore, electrocardiography may have underestimated the extent and site of infarction, al though resting ischemia as well as infarction may have contributed to some of the scinti- graphic defects. In one of these patients (no. 12) the scinti- gram demonstrated anteroseptal and inferior defects while the electrocardiogram had only inferior Q-waves, but had transiently demonstrated anterior "reciprocal" ST depres- sion during the acute phase. In one other patient (no. 9)

there was electrocardiographic (and angiographic) evidence of anterior and inferior infarction, but scintigraphy only identified the former. Overlap of the liver on the inferior wall may be the reason that scintigraphy failed to identify the full extent of infarction in this subject.

Left ventricular regional wall motion analysis

Gated 99mTc-TB[ scintigraphy was performed in 14 of the 15 patients and 14 also had standard radionuclide ventricu- lography. However, only 13 had both. These yielded 39 left ventricular segments for comparative analysis of regional left ventricular wall motion.

During gated 99mTc-TBI scintigraphy all the segments were discernible, despite evidence o f reduced perfusion in some; 21 segments were normal, 10were hypokinetic, 8 were akinetic and none were dyskinetic. During standard radionuclide ventriculography 28 segments were considered normal, 7 were hypokinetic, 3 akinetic and I dyskinetic. These data are compared in Table 4. There was complete agreement for 25 (64%) of the 39 segments, ~c= 0.35 (0.08, 0.62), p < 0.05. Of the 14 segments in which the two meth- ods disagreed, 7 had wall motion abnormalities in both stu- dies. Five were considered akinetic on gated 99mTc-TBI per- fusion scintigraphy and either hypokinetic or dyskinetic on standard radionuclide ventriculography and 2 hypokinetic segments on perfusion scintigraphy were considered akin- eric during ventriculography. All were in patients with defi- nite infarction. Of the remaining 7 segments, 5 were classi- fied as hypokinetic and 2 as akinetic during gated 99mTc- TBI perfusion scintigraphy, whereas all were classified as normal during standard radionuclide ventriculography. All of these segments were also in regions o f infarction, local- ized both by the electrocardiogram and planar scintigraphy.

Fig. 2. Digitized planar scintigrams in the 30 ° and 70 ° left anterior oblique projections in patient no. 3 with 15% background subtraction and an upper threshold set at 50% of the maximum. Reduction of tracer uptake is seen in the septum and anterior wall of the left ventricle (arrowed) consistent with anterior infarction. Some residual lung and persistent liver and spleen activity is also present

223

Fig, 3. Digitized planar scintigrams in the anterior, 30 ° and 70 ° left anterior oblique projections in patient no. 13, with 20% background subtraction and an upper threshold set at 80% of maximum counts. There is marked reduction in uptake of 99mTc-TBI in the inferior and inferoposterior walls of the left ventricle (arrowed) consistent with infarction. Again some residual lung and persistent liver activity is also seen

Table 3. Localization of Q-wave infarction : comparison of electro- cardiogram and 99mTc- TBI scintigraphy

ECG localization of MI

Ant Inf Ant and Inf

Site of Ant 3 0 t ~c = 0.49 scintigraphic Inf 0 2 0 (0.03,0.95) defects Ant and t 1 1 p < 0.1

Inf

Ant=anterior, Ant and Inf=both sites, Inf=inferior, ~c=kappa coefficient, with approximate 95% confidence limits in parentheses

Table 4. Regional wall motion analysis: comparison of gated 99mTc-TBI scintigraphy and standard radionuclide ventriculogra- phy (RVG)

RVG - Wall motion analysis

Norm Hypok Ak Dysk

Gated TBI Norm 21 0 0 0 ~=0.35 scintigraphic Hypok 5 3 2 0 (0.08,0.62) wall motion Ak 2 4 1 1 p < 0.05 analysis Dysk 0 0 0 0

Ak = akinesis, Dysk = dyskinesis, Hypok = hypokinesis, • = kappa coefficient, with approximate 95% confidence limits in parentheses, Norm = normal wall motion

Scintigraphy and angiographic findings

Of the 15 patients studied, 9 underwent coronary arteriog- raphy and 7 of these had technically satisfactory planar scintigrams. All 4 patients with > 90% stenoses of the left anterior descending coronary artery and electrocardio- graphic evidence of anterior infarction had anterior wall perfusion defects on planar scintigraphy; septal wall motion abnormalities were present in the 3 who had gated 99mTC- TBI studies, In addition, 2 patients with significant but non- occlusive coronary artery disease and no evidence of infarc- tion had normal perfusion and wall motion studies, (see Table 1).

Discussion

Detection and localization of myocardial infarction

This study has demonstrated that there is a high level of agreement between planar scintigraphic criteria using the

new myocardial perfusion imaging agent 99mTc-TBI and currently accepted clinical criteria for the diagnosis of myo- cardial infarction. Only one patient, with a prior non-Q- wave infarction, was not detected. In another, scintigraphy identified a new Q-wave infarction but did not identify an old non-Q-wave infarction. Thus, all the infarctions de- tected by scintigraphy were associated with electrocardio- graphic Q-waves. However, as this study evaluated only a small number of non-Q-wave infarctions, a definitive statement concerning the accuracy of this method is non-Q- wave infarction is not possible. Both old and recent infarc- tions were associated with scintigraphic defects and 99mTc- TBI scintigraphy did not appear able to distinguish one from the other.

Localization of infarction by 99mTc-TBI scintigraphy and the electrocardiogram exhibited moderate agreement. Differences probably reflected the known limitations of the electrocardiogram (Wackers et al, 1977) and also those of planar scintigraphic imaging, owing to overlap between dif- ferent regions and from adjacent structures, especially the liver. Another factor may be the inability of perfusion scin- tigraphy to distinguish infarction from ischemia at rest al- though, interestingly, one patient with unstable angina did not have any scintigraphic defects. The anatomical con- straints could be largely overcome by tomographic imaging (Holman et al. 1979), which may be feasible in stable pa- tients.

Planar 99~"Tc-TBI perfusion scintigraphy is therefore able, like 2°1T1 scinfigraphy, to detect and localize myocar- dial infarction. The sensitivity of detection by 99mTc-TBI was similar to that reported for 2°1T1 (Henning et al. 197"7; Wackers et al. 1976b, 1979) even though many of our sub- jects were investigated more than 24 h after an acute event, at which time thallium scintigraphy is less sensitive (Wackers et al. 1976b). Our studies with 99mTc-TBI also indicate a similar relationship of infarct localization com- pared with the electrocardiogram to that seen with 2°1T1 (Henning et al. 1977; Niess et al. 1979; Wackers et aI. 1976 b). Neither technique can distinguish acute/recent from old infarction (Henning et al. 1977; Wackers et al. 1876 b); to do so requires imaging with infarct avid agents such as 99mTc-pyrophospate (Holman 1980). This represents a limitation of all perfusion based imaging techniques in de- tecting acute infarction unless there is conclusive evidence of no prior infarction, or a recurrent infarction occurs in a site anatomically distant f rom an old infarction. However, because the scintigraphic defects represent the total amount of left ventficular myocardial necrosis and rest ischemia, the prognostic value of perfusion imaging may be great, as has indeed been shown with 2°iT1 (Silverman et al. 1980). Similarly, the ability to identify old infarction in a stable

224

population may have important prognostic implications and scintigraphy may be a useful non-invasive complement to the electrocardiogram in this respect (Niess et al. 1979), especially as many infarctions are clinically silent (Kannel and Abbott 1984).

Assessment of regional wall motion

Gated 99mTc-TBI scintigraphy and standard blood pool ra- dionuclide ventriculography exhibited fair agreement in the assessment of left ventricular wall motion. Some minor dif- ferences would be expected, as the analysis of both studies was essentially qualitative. In addition, differences in pa- tient and camera position and the time delay between the two studies may have contributed in some cases to apparent disagreement. However, for the majority of the 39 segments analyzed the two investigations were either in complete agreement or only minor disagreement. Of interest is that 7 segments appeared normal with standard radionuclide ventriculography but were either hypokinetic or akinetic with 99mTc-TBI scintigraphy. As all of these segments repre- sented areas of infarction, detected by the electrocardio- gram and planar scintigraphy, it may be that gated perfu- sion wall motion analysis is more sensitive than equilibrium blood pool analysis. Alternatively it is possible that some recovery of ischemic dysfunction occurred between the two studies as the equilibrium blood poll ventriculography was invariably performed at least 24 h after the 99mTc-TBI study. However, if gated 99mTc-TBI scintigraphy is indeed more sensitive at identifying wall motion abnormalities, it would be a compelling reason to perform combined perfu- sion and functional imaging; this is eminently feasible, as demonstrated in this investigation.

Advantages and limitations of 99mTc-TBI

The main potential advantages of 99mTc labeled myocardial imaging agents, over the currently used 2°1T1, are related largely to the relative properties of the isotopes. 99mTc has a higher emission photopeak (140 keV versus 69-83 keV) resulting in less tissue attenuation and greater image resolu- tion. This may be especially valuable if tomographic imag- ing is to be employed in an effort to size infarctions. In addition, its shorter physical half-life (6 h vs 72 h) permits higher activity to be administered safely and would m a k e serial studies eminently feasible; it is also more readily and cheaply available from on site generator systems. Therefore, various attempts have been made to produce 99mTc-labeled myocardial agents (Deutsch et al. 1981 a, b; Dudczak et al. 1983), but 99mTc-TBI is the first to produce images in man of sufficient quality to be of clinical use. Because of its physical properties, planar studies were acquired in only 3-5 min and adequate gated studies in 20 min or less. Pre- vious studies have shown that excellent tomographic images may also be obtained (Holman et al. 1984b). The informa- tion provided by planar 99mTc-TBI scintigraphy was quali- tatively similar to that seen with a°lT1 in previous studies in patients with suspected or confirmed myocardial infarc- tion (Henning etal. 1977; Iskandrian and Hakki 1985; Niess et al. 1979; Silverman et al. 1980; Tamaki et al. 1982; Wackers et al. 1976a, b, 1977, 1979). However, no attempt has been made to obtain both perfusion and gated wall motion data in one study with a°IT1 in patients with myo- cardial infarction, possibly because of the time it would take to acquire an adequate study. However, these data

were readily obtained with 99mTc-TBI which, combined with the rapid on-site availability of 99mTc-TBI, makes it a potentially more useful agent for assessment of patients with myocardial infarction.

99mTc-TBI does suffer from several limitations. Like all perfusion agents, as discussed earlier, it cannot distinguish recent from old infarctions or infarction from ischemia at rest. The high initial lung uptake and subsequent slow clear- ance mean that satisfactory myocardial imaging cannot be performed until approximately I h after injection of the compound. Although the lung activity does decline, that in the liver remains high (Holman et al. 1984b) and this may interfere with imaging of the inferior wall of the left ventricle. This is mainly a problem in the 70 ° left anterior oblique position and may usually be partly overcome by cranial angulation of the camera. Furthermore, animal stu- dies suggest that analogues of TBI may have more favorable biodistributions with rapid lung and liver clearance and hence less interference from these organs.

Thus, 99mTc-TBI has shown considerable promise in the assessment of patients with suspected or proven myocardial infarction and can provide unique early information on the extent of infarction and ischemia as well as on regional function of the left ventricle. Such an agent may prove useful in the evaluation of methods of acute physical or pharmacological intervention in myocardial infarction.

Acknowledgement. We are grateful to Yolanda Gooden for her patience and expertise in typing the manuscript

References

Alderson PO, Wagner HN Jr, Gomez-Moeiras J J, Rehn TG, Becker LC, Douglas KH, Manspeaker HF, Schindledecker GR (1978) Simultaneous detection of myocardial perfusion and wall motion abnormalities by cinematic Z°IT1 imaging. Radiology 127:531-533

Alpert JS, Braunwald E (1984) Acute myocardial infarction: patho- logical, pathophysiological and clinical manifestations. In: Braunwald E (ed) Heart disease, a textbook of cardiovascular medicine. Philadelphia, WB Saunders, pp 1262-1300

Cohen J (1968) Weighted kappa: nominal scale agreement with provision for scaled disagreement or partial credit. Psychol Bull 70 : 213-220

Deutsch E, Bushong W, Glavan KA, Elder RC, Sodd VJ, Scholz KL, Fortman DL, Lukes SJ (1981 a) Heart imaging with ca- tionic complexes of technetium. Science 214:85-86

Deutsch E, Glavan KA, Sodd VJ, Nishiyama H, Ferguson DL, Lukes SJ (1981 b) Cationic Tc-99m complexes as potential myo- cardial imaging agents. J Nucl Med 22:897-907

Dudczak R, Angelberger P, Homan R, Kletter K, Schmoliner R, Frischauf H (1983) Evaluation of 99mTc-Dichloro bis (1,2-Di- methylphosphino) Ethane (99'~Tc-DMPE) for myocardial scin- tigraphy in man. Eur J Nucl Med 8:513-515

Geltman EM, Ehsani AA, Campbell MK, Schechtman K, Roberts R, Sobel BE (1979) The influence of lacation and extent of myocardial infarction on long-term ventricular dyshythmia and mortality. Circulation 60:805-814

Henning H, Schelbert HR, Righetti A, Ashburn WL, O'Rourke RA (1977) Dual myocardial imaging with technetium-99m pyrophosphate and thallium-201 for detecting, localizing and sizing acute myocardial infarction. Am J Cardiol 40:147 155

Holman BL, Hill TC, Wynne J, Lovett RD, Zimmerman RE, Smith EM (1979) Single-photon transaxial emission computed tomography of the. heart in normal subjects and in patients with infarction. J Nucl Med 20:736-740

Holman BL, Wynne J (1980) Infarct avid (hot spot) myocardial scintigraphy. Radiol Clin North Am 18:487-499

225

Holman BL, Nuclear Cardiology (1984a) In: Braunwald E (ed) Heart disease, a textbook of cardiovascular medicine. Philadel- phia, WB Saunders, pp 351-399

Holman BL, Jones AG, Lister-James J, Davison A, Abrams M J, Kirshenbaum JM, Tumeh SS, English RJ (1984b) A new Tc- 99m-labelled myocardial imaging agent hexakis (t-butylisoni- trile)-technetium (I) [Tc-99m TBI] : Initial experience in the hu- man. J Nucl Med 25:1350-1355

Holman BL, Campbell CI, Lister-James J, Jones AG, Kloner RA (1985) The effect of reperfusion and hyperemia on the biodistri- bution of the myocardial imaging agent Tc-99m TBI. J Nucl Med 26:P13 (abstr)

Iskandrian AS, Hakki AH (1985) Thallium-201 myocardial scintig- raphy. Am Heart J 109:113-129

Jones AG, Abrams MJ, Davison A, Brodack JW, Toothaker AK, Adelstein S J, Kassis AI (1984) Biological studies of a new class of technetium complexes: The hexakis (alkylisonitrile) techne- tium (I) cations. Int J Nucl Med Biol 11 : 225-234

Judkins MP (1967) Selective coronary arteriography, Part 1 :A per- cutaneous transfemoral technic. Radiology 89 : 815-824

Kannel WB, Abbott RD (1984) Incidence and prognosis of unrec- ognized myocardial infarction. N Engl J Med 311 : 11~1147

Maddox DE, Holman BL, Wynne J, Idoine J, Parker JA, Uren R, Heill JM, Cohn PF (1978) The ejection fraction image: a non invasive index of regional left ventricular wall motion. Am J Cardiol 41 : 123(>1238

Niess GS, Logic JR, Russell RO Jr, Rackley CE, Rogers WJ (1979) Usefulness and limitations of thallium-201 myocardial scintig- raphy in delineating location and size of prior myocardial in- farction. Circulation 59:1010-1019

Schelbert HR, Henning H, Ashburn WL, Verba JW, Karliner JS, O'Rourke RA (1976) Serial measurements of left ventricular ejection fraction by radionuclide angiography early and late after myocardial infarction. Am J Cardiol 38:407-415

Silverman K J, Becker LC, Bulkley BH, Burow RD, Mellits ED, Kallman CH, Weisfeldt ML (1980) Value of early thallium-201 scintigraphy for predicting mortality in patients with acute myocardial infarction. Circulation 61:996-1003

Tamaki S, Nakajima H, Murakami T, Yui Y, Kambara H, Kadota K, Yoshida A, Kawai C, Tanakin N, Mukai T, Ishii Y, Tori- zuka K (1982) Estimation of infarct size by myocardial emission computed tomography with thallium-201 and its relation to creatine kinase-MB release after myocardial infarction in man. Circulation 66:994-1001

Thanavaro S, Kleiger RE, Province MA, Hubert JW, Miller JP, Krone RJ, Oliver GC (1982) Effect of infarct location on the in-hospital prognosis of patients with first transmural myocar- dial infarction. Circulation 66 : 742-747

Wackers FJ Th, Busemann Sokole E, Samson G, van der Schoot JB, Wellens HJJ (1976 a) Myocardial imaging in coronary heart disease with radionuclides, with emphasis on thallium-201. Eur J Cardiol 4:273-282

Wackers FJ Th, Buseman Sokole E, Samson G, van der Schoot JB, Lie KI, Wellens HJJ (1976b) Value and limitations of thal- lium-201 scintigraphy in the acute phase of myocardial infarc- tion. N Engl J Med 295 : 1-5

Wackers FJ Th, Becket AE, Samson G, Buseman Sokole E, van der Schoot JB, Vet AJJM, Lie KI, Durrer D, Wellens H (1977) Location and size of acute transmural myocardial infarction estimated from thallium-201 scintiscans. Circulation 56:72-78

Wackers FJ Th, Lie KI, Liem KL, Buseman Sokole E, Samson G, van der Schoot J, Durrer D (1979) Potential value of thal- lium-201 scintigraphy as a means of selecting patients for the coronary care unit. Br. Heart J 41 : 111-117

Received March 8, 1986 / June 7, 1986