- Email: [email protected]
Clinical experience with technetium-99m teboroxime, a neutral, lipophilic myocardial perfusion imaging agent
Clinical Experience with Technetium-99m Teboroxime, a Neutral, Lipophilic Myocardial Perfusion Imaging Agent Lynne L. Johnson, MD, and David W. Seldin, MD
Technetium-99m (Tc-99m) teboroxime is a new technetium-based myocardial perfusion imaging agent (investigational code = SQ30217 [Cardiotec, Squibb Diagnostics]). A member of a class of neutral, lipophilic, technetium-containing complexes known as boronic acid adducts of technetium dioxime (BATO) complexes, this agent is chemically very different from the cationic tracer thallium-201 (D-201) and from the cationic technetium complex Tc-99m sestamibi (Cardiolite, Du Pont Imaging Agents). Tc-99m teboroxime has high myocardial extraction, rapid blood clearance, little lung uptake and rapid myocardial washout. A biexponential pattern of myocardial washout is demonstrated in animals and in man. Effective half-lives of the 2 washout components in man are 5.2 minutes and 3.8 hours and represent approximately 66 and 33% of the myocardial activity, respectively. The first half-life for the myocardium is approximately 11 minutes. As the agent washes out of the heart, hepatic uptake occurs, peaking at about 5 minutes afler injection. The liver is the major organ of excretion and receives, along with the large bowel, the largest radiation dose. Rapid imaging protocols using standard cameras have achieved good myocardial counts from 3 planar views acquired over a 4 to 5-minute period or for single photon emission (SPECT) images acquired computed tomography over a lo-minute period. An entire stress/rest procedure can be completed in 1 hour. Analysis of data from 155 patients from 4 centers using planar or SPECT imaging showed a sensitivity and specificity for blinded readings of 82 and 91%, respectively, when compared against overall clinical impression. There was a high agreement between blinded readings of Tc-99m teboroxime and TI-201 scans from the same patient (90%). Studies in progress include measurement of regional myocardial washout from dynamic SPECT acquisitions performed with a 3-headed SPECT camera, and combined function and perfusion tests using a portable igh count rate camera. (Am J Cardiol 1990;66:63E-67E)
From Columbia University, College of Physicians and Surgeons, Departments of Medicine and Radiology, New York, New York. Address for reprints: Lynne L. Johnson, MD, Department of Medicine, 630 West 168th Street, New York, New York 10032.
A
lthough myocardial perfusion imaging with thallium-201 (Tl-201) has become a mainstay in the diagnosis and management of coronary artery disease (CAD), Tl-201 is not an ideal imaging agent, especially when compared with technetium-99m (Tc99m). Two new technetium-based myocardial perfusion imaging agents, Tc-99m sestamibi and Tc-99m teboroxime, have recently been developed,and both have undergone clinical trialsm3 Tc-99m teboroxime is in a class of neutral, lipophilic, technetium-containing complexes known as boronic acid adducts of technetium dioxime (BATO) complexes. The pharmacokinetics of Tc-99m teboroxime are very different from those of the cationic tracer Tl-201 or the cationic technetium complex Tc-99m sestamibi. Tc-99m teboroxime has high myocardial extraction and rapid myocardial washout. It has shown promise as a myocardial perfusion imaging agent used with stresstesting. The rapid myocardial washout necessitates brief imaging protocols. PHARMACOKINETICS
AND
PREPARATION
Tc-99m teboroxime is produced by template synthesis when Tc-99m is added to a vial containing the appropriate vicinal dioxime, methyl boronic acid, and the vial is heated at 100°C for 15 minutes. The chemical structure is shown in Figure 1. Tc-99m teboroxime is supplied in kit form as a lyophilized powder in sterile evacuated vials Up to 100 mCi of Tc-99m in 1 ml of 0.9%sodium chloride is added to the contents of the vial. The vial is placed in a 100°C water bath for 15 minutes, and then allowed to cool. Two dosesof 15 mCi each (rest and stress) can be obtained from the samevial. The presenceof saline soluble contaminants and reduced technetium speciesis monitored with paper chromatography. One drop of the preparation is placed on each of two 1.3 cm X 11 cm Whatman 3 1 ET strips. One strip is developedin normal saline while the other is developedin a 50:50 (by volume) normal saline/acetone solution. The criterion for injection, which was met in all the phaseII and III clinical trials,& that the sum of saline soluble contaminants and reduced technetium speciesbe
THE AMERICAN JOURNAL OF CARDIOLOGY
OCTOBER 16,199O
A SYMPOSIUM:
lECHNETIUM49m
IMAGING
AGENTS
flGURE 1. Chendcal st~~twe of teclmeMum-99m teboroxime,amendwrofthedwsmkalgrarpcakdboro&acidadhcts of technetium diixime complexes. B = boron; Cl = cfkrhe; If = hydrogen; N = nitrogen; 0 = oxygen; Tc = tedmMium.
sion agents. These investigators found higher values for mean fractional extraction (measured over a range of flows) and for capillary permeability surface area product for Tc-99m teboroxime than for Tl-201 or Tc-99m sestamibi.’ Although Tc-99m teboroxime has high extraction and initial capillary permeation, it has a rapid myocardial washout. Thus, net extraction values, which combine the summation of tracer washin and washout, were not as different among the three. A biexponential pattern of myocardial washout is demonstrated in animals4 and in humans. Myocardial washout data in humans were collected from phase I studies performed at Duke University (unpublished data). Washout data from each individual patient were fitted to a biexponential curve, and the curves were combined using Systat software (Fig. 2). Effective half-lives of the 2 washout comoonents in humans are 5.2 minutes and 3.8 hours and representapproximately 66 and 33%c,f
flGURE 2. Myocardial washout of technetium-99m teboroxime. The ffrst half-fife of the myocantiun is approximately 11 minutes.
64E
THE AMERICAN JOURNAL OF CARDIOLOGY
VOLUME 66
the myocardial activity, respectively.Thus, the first halflife for the myocardium is approximately 11 minutes. At 1 hour after injection, about 30%of the initial myocardial activity remains in the myocardium. However,with standard planar imaging, no detectabled activity is present over the heart (Leppo et al, unpublished data). Single photon emission computed tomography (SPECT) imaging at 1 hour doesshowsomemyocardial activity. As the agent washes out of the heart, hepatic uptake occurs. Hepatic uptake peaksbetween4.5 and 7.0 minutes after injection.(’ The liver is the major organ of excretion and is, along with the large bowel, the target organ for dosimetry. The liver receives3.6 rads and the upper tract of the large intestine 3.3 rads from 2 (rest and stress) 15mCi dosesof Tc-99m teboroxime. The total body doseis 5 10 mrads. This compares with a total body dose of 630 mrads from a 3-mCi dose of Tl-201. IMAGING TRIALS
PROTOCOLS
AND CLINICAL
Tc-99m teboroxime’s rapid myocardial washout necessitatesrapid imaging protocols. There are theoretical limitations to performing standard SPECT imaging when tracer activity within the field of view variessignificantly during the SPECT acquisition. Under these circumstances, projection images would be unequally weighted in the reconstruction procedure,potentially producing artifacts. Bok et al7 performed a phantom experiment addressingthis question. By simulating time dependence of the tracer concentration, these investigators found that when the activity imaged follows a decaying exponential function, a distortion factor may be defined as the ratio of the total imaging time divided by the tracer half-life in the object imaged. They found that a change in activity of less than a factor of 2 results in little image distortion. Since the first half-life of Tc-99m teboroxime in the myocardium is approximately 11 minutes, and standard SPECT acquisitions, beginning within 2 minutes of Tc-99m teboroxime injection, can becompletedin 10 minutes, there may be little image distortion from varying tracer activity. The first study using Tc-99m teboroxime and planar imaging in control subjectsand coronary diseasepatients was performed by Seldin et al’ (Table I), who used up right bicycle exercisewith the patient positionedin front of an Anger camera (Picker Dynamo, Picker International) to allow imaging as soonas blood pool activity cleared (about 2 minutes). Three standard planar views were acquired for 3, 5, and 6 minutes, respectively (Fig. 3), beginning with the anterior projection to allow visualization of the inferior wall before hepatic uptake occurred. Stressimaging wasperformedfirst, followed 2 hours later with a rest injection. A 1S-mCi doseof Tc-99m teboroxime was used for eachinjection. Thirty subjects- 10 normal volunteers and 20 patients with CAD documentedby recent angiography-all had stress perfusion imaging performed with both Tl-201 and Tc-99m teboroxime. There was no significant difference betweenthe 2 techniques for identifying abnormal vessels(Tc-99m teboroxime, 19 of 45; Tl-201, 21 of 45) or detecting CAD (Tc-
TABLE I Studies Using Technetium-99m Investigator
No.
Seldin and Johnson Lewo
Bellinger
Drane
Seldin and
20 15 15 5
and Planar imaging Camera/
Time
Total Imaging
Type
Window
(First Image)
Time
Counts/FOV (First Image)
Technical Factors
Planar 20% Planar 15% Planar SPECT
SFOV
20 minutes
700K
Liver third view
300-4OOK
Liver third view
350K
Liver third view Liver seen
Planar SPECT SPECT Planar
3 minutes
LFOV
60 seconds
4-5 minutes
LFOV Single head 20% LFOV Single head Triad 20% Scinticor SIM-400 1” collimator
60 seconds
45 minutes 10 minutes
3 minutes
60 seconds
All imaging was begun within 2 minutes of injectIon. Dose of 15 mCi used in all stress injections. EF = ejection fraction; FOV = field of view; L = large; S = small; SPECT = single photon emission
99m teboroxime, 16 of 20; Tl-201, 17 of 20). However, significantly more defectswere identified as fixed on resting Tc-99m teboroxime scansthan on Tl-201 redistribution scans. This might be due to residual background myocardial activity from the first (exercise) injection. Similar observationshave beenmade with Tc-99m sestamibi on same-daystress-reststudies when second-injection rest scans were compared to Tl-201 redistribution scans8Hepatic uptake of Tc-99m teboroxime, which was apparent in the second image (shallow left anterior oblique) in all subjects, obscured inferoapical segment analysis in 14 of 20 patients but did not interfere with abnormal vesselidentification. Seldin et al3were restricted to an imaging protocol lasting about 20 minutes, which included camerarepositioning time. Subsequently,it was shown independently by Leppo and by Bellinger that adequatemyocardial countscan be acquired by an Anger camera in 40 to 60 seconds/view after Tc-99m teboroxime injection (personalcommunication). Theseinvestigators showedthat using large field-of-view (LFOV) detectors, counts per image ranged from 300,000 to 400,000 (Fig. 4). Both of theseinvestigators used planar imaging protocols that were completed within 4 to 5 minutes. Standard SPECT imaging using single-head cameras and lo-minute acquisitions over 180” using standard “stop and shoot” software was performed by several groups of investigators (Table I). Continuous acquisition of counts over 180” shortensthe total acquisition time to under 6 minutes, minimizing the effectsof hepatic uptake (Fig. 5). The planar and SPECT data from 4 groups of investigators participating in the phaseIII trial were combined for analysis.9The “gold standard” in this trial was identified as“overall clinical impression,” which wasdefined as the results of all clinical data except for the Tc-99m teboroxime data and included angiography plus Tl-201 in 6 1 patients, angiography alone in 40 and Tl-201 alone in 54. For the 101 patients who had coronary angiography plus Tc-99m teboroxime study, the agreementfor detect-
Liver third view Liver uptake No liver
20 minutes 10 minutes 2 minutes 4-5 minutes
computed
tomography;
800K
Also EF data ? resolution
? = questionable.
FIGURE 3. A, Thallium-201 exercise (fop) and redistribution (bottom) images in the anterior, 30” left anterior oblique and 60” left anterior oblique views in a patient with 99% stenosis of the left anterior descending coronary artery. There is decreased perfusion of the anteroseptum and apex with partial redistribution at 4 hours. R, Technetium-99m teboroxbne exercise (fop) and rest (bottom) images of the same patient showing a comparable anteroapical defect, which partially fills in on reinjection. Hepatic activity on the steep oblique view obscures the inferoapical segment on the exercise image and the anteroapical and inferoapical segments on the rest image. (Reproduced with permission from J NUC/ ~ed.3)
Tl-iE AMERICAN JOURNAL OF CARDIOLOGY
OCTOBER 16,199O
A SYMPOSIUM:
TECHNETIUM-5Sm
IMAGING
AGENTS
myocardium in the distribution of normal arteries. Their initial results suggestthat abnormal washoutin the distribution of stenosedvesselscan be detectedafter a single Tc-99m teboroxime injection with dynamic SPECT imaging using a 3-headedSPECT camera.’1 The rapid bolus injection of Tc-99m pertechnetate and a multicrystal scintillation camera have been the major components of the first-pass technique for measurement of left and right ventricular ejection fractions. There is an extensivecollection of published studiesthat apply this technique to rest and bicycle exercisestudiesin patients with valvular diseaseand CAD. The best data supporting the clinical usefulnessof exercise first-pass scintigraphy comesfrom the Duke databaseand the work of Robert Jones.Theseinvestigators found that the exerFIGURE 4. Technetium-SSm teboroxime exercise (fop] and cise left ventricular ejection fraction wasthe most imporrest (bottom) planar images acquired with rapid imaging setant singie independentpredictor of mortality in coronary quence (40 seconds/image) on large field-of-view camera showing prominent apieal cleft. Coronary angiography showed disease.12A newer model of a high count rate, portable insignificant coronary artery disease. Study performed by Ray multicrystal scintillation camera is now available (SIMRellinger, MD, Travis Air Force Base, California. ANT = an400, Scinticor). This camera,which hasa very small field terior; LAO = left anterior oblique. of view, can be used with treadmill exercise. Johnson, Seldin and co-workers developed a protocol using Tcing CAD was 76% and was not different from the agree- 99m teboroxime and the portable high count rate camera ment between coronary angiography and Tl-201 scintig- equippedwith a l-inch collimator to maximize resolution raphy (80%). For the 115 patients who underwent Tc- and a motion correction program developedby Port and 99m teboroxime and Tl-201 scintigraphy, the agreement colleagues.l 3 Before exercise, a low-energy radioactive between the 2 imaging modalities was 90%. The overall source is fastened on the skin over the sternum. The sensitivity and specificity for blinded reading were 82 and patient runs on the treadmill with the chest positioned 9 1%, respectively, when compared against overall clinical several inches in front of the camera detector and is injected with 15 mCi of Tc-99m teboroxime as a bolus at impression. peak exercise.Dynamic dual isotopeimaging is acquired STUDIES IN PROGRESS for 50 seconds,the treadmill is slowedand the patient is Becauseof the unique pharmacokinetic properties of seatednext to the treadmill. Three planar images using Tc-99m teboroxime, several clinical applications are be- only the technetium window are acquired within 4 mining pursued that take the aforementioned standard imag- utes. The low-energy sourceis usedto correct the dynaming protocols 1 step further. Tc-99m teboroxime is a ic first-pass data for patient motion. The potential drawneutral, lipophilic agent with rapid myocardial washout. The removal rate of a neutral lipophilic agent that freely enters and leaves a myocardial cell should be related to Vertical Horizontal Short Axis capillary blood flow. Tc-99m teboroxime thus bearssimilarities to xenon-133, the inert gas that was injected into the coronary arteries to quantitate regional myocardial blood flow from initial washout using the Schmidt-Kety Stress formula and a multicrystal scintillation camera.‘ODifferences in regional Tc-99m teboroxime washout may be usedto estimate regional differences in myocardial blood flow. To overcomecurve contamination by residual blood pool activity and overlap of coronary vascular bedsin any single planar projection, SPECT acquisition is preferable Rest for this application. Drane and co-workers** are performing dynamic SPECT imaging with Tc-99m teboroxime and a 3-headed SPECT camera. These investigators acquire ten 2-minute sequential SPECT acquisitions. In addition to measuring regional washout, they compare FIGURE 5. Technetium-SSm teboroxime stress (top) and rest the initial scansto normalized scansacquired at 18 to 20 (frottomJ single photon emission computed tomography reconminutes after injection to determine redistribution. This sfructions acquired as a 6-mirmte 160’ rotation on standard rotating gamma camera. Anteroseptal and inkroredistribution is actually differential washout due to the single-head basal detects show almost complete fill-in on res6ng scans. slower clearing of activity from myocardium in the distri- Study performed by David Yuille, St Luke’s Medical Center, bution of stenosedarteries compared to clearance from Milwaukee, Wisconsin.
66E
THE AMERICAN JOURNAL OF CARDIOLOGY
VOLUME 66
back to this technique is the limited resolution of the high SPECT camera with rapid acquisition and the Scinticor count rate camera (equivalent to a 40 X 40 matrix on an SIM-400, a portable high count rate camera, are well SFOV Anger camera) (Fig. 6). Small defects may be suited to imaging Tc-99m teboroxime. The SIM-400 missed,but this deficit may be balanced by the additional gives ejection fraction data as well as perfusion imaging, diagnosticinformation added by the left ventricular ejec- but the image quality of the latter may be limited by the tion fraction numbers. The entire stressand rest proce- poorer resolution of a multicrystal camera. dure with 2 sets of planar images and ventricular funcAlthough both Tc-99m sestamibi and Tc-99m tebortional data can be obtained within 1 hour. oxime will probably competewith eachother and with Tl201 in stress/rest (redistribution) perfusion imaging, the SUMMARY very different pharmacokinetics of the 2 technetium perTc-99m teboroxime is a technetium-basedmyocardial fusion imaging agentspresent a definite choiceto nuclear perfusion imaging agent. It is one of the BAT0 com- imaging laboratories. A complete diagnostic study inpounds,which are neutral and lipophilic. Its pharmacoki- cluding stress and rest ejection fraction performed in 1 netics are very different from those of Tc-99m sestamibi, hour, compared to 5 hours for Tl-201 stressand redistria cationic technetium-basedmyocardial perfusion agent. bution, will probably find high patient approval. LaboraThe myocardial extraction for Tc-99m sestamibi is less tories must, however, be set up for rapid patient movethan that for Tl-201, whereasthe myocardial extraction ment betweentreadmill and camera. Alternatively, pharfor Tc-99m teboroxime is greater than that for Tl-201. macologic stress with intravenous dipyridamole or Tc-99m sestamibihasvery slow myocardial washout and adenosinecan be administered while the patient is lying negligible redistribution, whereas Tc-99m teboroxime under the camera, obviating the need to move the patient has rapid myocardial washout, with differential washout and potentially shortening pharmacologic stress/perfurates in the distribution of stenotic and normal vessels, sion imaging studies. Both Tc-99m teboroxime and Tcleading to apparent “redistribution” soonafter injection. 99m sestamibi offer the opportunity to acquire function The first half-life for the myocardium is 11 minutes. As and regional perfusion data in a single procedure, which Tc-99m teboroxime is washing out of the heart, hepatic will add important diagnostic and prognostic information uptake occurs,peaking at 4.5 to 7.0 minutes. The combi- to the study. nation of rapid myocardial washout and early hepatic REFERENCES uptake has necessitateddeveloping rapid imaging proto- 1. Wackers FJT, Berman DS, Maddahi J, Watson DD, Belier GA, Strauss H, Boucher CA, Picard M, Holman BL, Fridrich R, Inglese E, Delaloye B, Bischofcols. Protocolsusing standard Anger camerasand 40- to Delaloye A, Camin L, McKusick K. Technetium-99m hexakis 2-methoxyisobutyl 60-secondplanar acquisitions or 180’ rotational SPECT isonitrile: human biodistribution, dosimetry, safety, and preliminary comparison I for myocardial perfusion imaging J Nucl Med 1989;30:30 l-3 11. and lo-minute acquisitions beginning within 2 minutes of 2.to thallium-20 Kiat H, Maddahi J, Roy LT, Van Train K, Friedman J, Resser K, Berman DS. injection have minimized interference from hepatic up- Comparison of technetium 99m mcthoxy isobutyl isonitrile and thallium 201 for take, while yielding high counts over the myocardium. evaluation of coronary artery disease by planar and tomographic methods. Am J 1989;117:1-11. The sensitivity and specificity for detecting CAD or iden- Heart 3. Seldin DW, Johnson LL, Blood DK, Muschel MJ, Smith KF, Wall RM, tifying abnormal vesselsare comparable to those with Tl- Cannon PJ. Myocardial perfusion imaging with technetium-99m SQ30217: com201. Stress and rest studies can be performed within 1 parison with thallium-201 and coronary anatomy. J Nucl Med 1989;30:312-319. 4. Narra RK, Nunn AD, Kuczynski BL, Feld T, Wedeking P, Eckelmen WC. A hour. Although good imagescan be acquired using stan- neutral technetium-99m complex for myocardial imaging. J Nucl Med dard cameras,cameras with rapid counting or imaging 1989;30:1830-1837. 5. Leppo JA, Meerdink DJ. Comparative myocardial extraction of two technecapabilities are uniquely suited for this agent. A 3-headed tium-labeled BAT0 derivatives (SQ30217, SQ32014) and thallium. J Nucl Med
FIGURE 6. Exercise thallium-201 (Tl-201) planar images (fop) and technetium-99m teboroxime planar images (bottom) showing posterolateral defect in a patient with an old posterorateral infarct. Acquired with rapid imaging sequence (30 seconds/image) on SlM-409 multicrystal camera. ANT = anterior; LAO = left anterior oblique; ex = exercise.
1990;3 1:67-74. 6. Coleman RE, Maturi M, Nunn AD, Eckelmen WC, Juri PN, Cobb FR. Imaging of myocardial perfusion with Tc-99m SQ30217: dog and human studies b (abstr). .I Nucl Med 1986:27:893. 7. Bok BD, Bite AN, Clausen M, Wong DF, Wagner HN. Artifacts in camera based single photon emission tomography due to time activity variation, Eur J Nucl Med 1987;13:439-442. 6. Taillcfer R, Gagnon A, Laflamme L, Gregoire J, Leveille J, Phaneuf D-C, 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 1989;15:113-117. 9. Zielonka JS, Bellinger R, Coleman RE, Drane W, Johnson LL, Seldin DW, Leppo J, Reba R, Wasserman A. Multicenter clinical trial of 99m-Tc teboroxime (SQ30217, Cardiotec) as a myocardial perfusion agent (abstr). J Nucl Med 1989;30:1745. 10. Cannon PJ, Dell RB, Dwyer EM. Measurement of regional myocardial and a scintillation camera. J Clin Invest perfusion in man with ‘%mon 1972;51:964-977. 11. Dram WE, Decker M, Strickland P, Tineo A, Zmuda S. Measurement of regional myocardial perfusion using Tc-99m teboroximc (Cardiotec) and dynamic SPECT (abstr). J Nucl Med 1989;30:1744. 12. Pryor DB, Harrell FE, Lee KL, Rosati RA, Coleman E, Cobb FR, Califf RM, Jones RH. Prognostic indicators from radionuclide angiography in medically treated patients with coronary artery disease. Am J Cardiol 1984;53:18-22. P3. Port S, Gal R, Grenier R, Acharya K, Shen Y, Skrade B. First-pass radionuelide angiography during treadmill exercise: evaluation of patient motion and a method for motion correction (abstr). J Nucl Mrd 1989;30:770.
THE AMERICAN
JOURNAL
OF CARDIOLOGY
OCTOBER
16,199O
Technetium-99m (Tc-99m) teboroxime is a new technetium-based myocardial perfusion imaging agent (investigational code = SQ30217 [Cardiotec, Squibb Diagnostics]). A member of a class of neutral, lipophilic, technetium-containing complexes known as boronic acid adducts of technetium dioxime (BATO) complexes, this agent is chemically very different from the cationic tracer thallium-201 (D-201) and from the cationic technetium complex Tc-99m sestamibi (Cardiolite, Du Pont Imaging Agents). Tc-99m teboroxime has high myocardial extraction, rapid blood clearance, little lung uptake and rapid myocardial washout. A biexponential pattern of myocardial washout is demonstrated in animals and in man. Effective half-lives of the 2 washout components in man are 5.2 minutes and 3.8 hours and represent approximately 66 and 33% of the myocardial activity, respectively. The first half-life for the myocardium is approximately 11 minutes. As the agent washes out of the heart, hepatic uptake occurs, peaking at about 5 minutes afler injection. The liver is the major organ of excretion and receives, along with the large bowel, the largest radiation dose. Rapid imaging protocols using standard cameras have achieved good myocardial counts from 3 planar views acquired over a 4 to 5-minute period or for single photon emission (SPECT) images acquired computed tomography over a lo-minute period. An entire stress/rest procedure can be completed in 1 hour. Analysis of data from 155 patients from 4 centers using planar or SPECT imaging showed a sensitivity and specificity for blinded readings of 82 and 91%, respectively, when compared against overall clinical impression. There was a high agreement between blinded readings of Tc-99m teboroxime and TI-201 scans from the same patient (90%). Studies in progress include measurement of regional myocardial washout from dynamic SPECT acquisitions performed with a 3-headed SPECT camera, and combined function and perfusion tests using a portable igh count rate camera. (Am J Cardiol 1990;66:63E-67E)
From Columbia University, College of Physicians and Surgeons, Departments of Medicine and Radiology, New York, New York. Address for reprints: Lynne L. Johnson, MD, Department of Medicine, 630 West 168th Street, New York, New York 10032.
A
lthough myocardial perfusion imaging with thallium-201 (Tl-201) has become a mainstay in the diagnosis and management of coronary artery disease (CAD), Tl-201 is not an ideal imaging agent, especially when compared with technetium-99m (Tc99m). Two new technetium-based myocardial perfusion imaging agents, Tc-99m sestamibi and Tc-99m teboroxime, have recently been developed,and both have undergone clinical trialsm3 Tc-99m teboroxime is in a class of neutral, lipophilic, technetium-containing complexes known as boronic acid adducts of technetium dioxime (BATO) complexes. The pharmacokinetics of Tc-99m teboroxime are very different from those of the cationic tracer Tl-201 or the cationic technetium complex Tc-99m sestamibi. Tc-99m teboroxime has high myocardial extraction and rapid myocardial washout. It has shown promise as a myocardial perfusion imaging agent used with stresstesting. The rapid myocardial washout necessitates brief imaging protocols. PHARMACOKINETICS
AND
PREPARATION
Tc-99m teboroxime is produced by template synthesis when Tc-99m is added to a vial containing the appropriate vicinal dioxime, methyl boronic acid, and the vial is heated at 100°C for 15 minutes. The chemical structure is shown in Figure 1. Tc-99m teboroxime is supplied in kit form as a lyophilized powder in sterile evacuated vials Up to 100 mCi of Tc-99m in 1 ml of 0.9%sodium chloride is added to the contents of the vial. The vial is placed in a 100°C water bath for 15 minutes, and then allowed to cool. Two dosesof 15 mCi each (rest and stress) can be obtained from the samevial. The presenceof saline soluble contaminants and reduced technetium speciesis monitored with paper chromatography. One drop of the preparation is placed on each of two 1.3 cm X 11 cm Whatman 3 1 ET strips. One strip is developedin normal saline while the other is developedin a 50:50 (by volume) normal saline/acetone solution. The criterion for injection, which was met in all the phaseII and III clinical trials,& that the sum of saline soluble contaminants and reduced technetium speciesbe
THE AMERICAN JOURNAL OF CARDIOLOGY
OCTOBER 16,199O
A SYMPOSIUM:
lECHNETIUM49m
IMAGING
AGENTS
flGURE 1. Chendcal st~~twe of teclmeMum-99m teboroxime,amendwrofthedwsmkalgrarpcakdboro&acidadhcts of technetium diixime complexes. B = boron; Cl = cfkrhe; If = hydrogen; N = nitrogen; 0 = oxygen; Tc = tedmMium.
sion agents. These investigators found higher values for mean fractional extraction (measured over a range of flows) and for capillary permeability surface area product for Tc-99m teboroxime than for Tl-201 or Tc-99m sestamibi.’ Although Tc-99m teboroxime has high extraction and initial capillary permeation, it has a rapid myocardial washout. Thus, net extraction values, which combine the summation of tracer washin and washout, were not as different among the three. A biexponential pattern of myocardial washout is demonstrated in animals4 and in humans. Myocardial washout data in humans were collected from phase I studies performed at Duke University (unpublished data). Washout data from each individual patient were fitted to a biexponential curve, and the curves were combined using Systat software (Fig. 2). Effective half-lives of the 2 washout comoonents in humans are 5.2 minutes and 3.8 hours and representapproximately 66 and 33%c,f
flGURE 2. Myocardial washout of technetium-99m teboroxime. The ffrst half-fife of the myocantiun is approximately 11 minutes.
64E
THE AMERICAN JOURNAL OF CARDIOLOGY
VOLUME 66
the myocardial activity, respectively.Thus, the first halflife for the myocardium is approximately 11 minutes. At 1 hour after injection, about 30%of the initial myocardial activity remains in the myocardium. However,with standard planar imaging, no detectabled activity is present over the heart (Leppo et al, unpublished data). Single photon emission computed tomography (SPECT) imaging at 1 hour doesshowsomemyocardial activity. As the agent washes out of the heart, hepatic uptake occurs. Hepatic uptake peaksbetween4.5 and 7.0 minutes after injection.(’ The liver is the major organ of excretion and is, along with the large bowel, the target organ for dosimetry. The liver receives3.6 rads and the upper tract of the large intestine 3.3 rads from 2 (rest and stress) 15mCi dosesof Tc-99m teboroxime. The total body doseis 5 10 mrads. This compares with a total body dose of 630 mrads from a 3-mCi dose of Tl-201. IMAGING TRIALS
PROTOCOLS
AND CLINICAL
Tc-99m teboroxime’s rapid myocardial washout necessitatesrapid imaging protocols. There are theoretical limitations to performing standard SPECT imaging when tracer activity within the field of view variessignificantly during the SPECT acquisition. Under these circumstances, projection images would be unequally weighted in the reconstruction procedure,potentially producing artifacts. Bok et al7 performed a phantom experiment addressingthis question. By simulating time dependence of the tracer concentration, these investigators found that when the activity imaged follows a decaying exponential function, a distortion factor may be defined as the ratio of the total imaging time divided by the tracer half-life in the object imaged. They found that a change in activity of less than a factor of 2 results in little image distortion. Since the first half-life of Tc-99m teboroxime in the myocardium is approximately 11 minutes, and standard SPECT acquisitions, beginning within 2 minutes of Tc-99m teboroxime injection, can becompletedin 10 minutes, there may be little image distortion from varying tracer activity. The first study using Tc-99m teboroxime and planar imaging in control subjectsand coronary diseasepatients was performed by Seldin et al’ (Table I), who used up right bicycle exercisewith the patient positionedin front of an Anger camera (Picker Dynamo, Picker International) to allow imaging as soonas blood pool activity cleared (about 2 minutes). Three standard planar views were acquired for 3, 5, and 6 minutes, respectively (Fig. 3), beginning with the anterior projection to allow visualization of the inferior wall before hepatic uptake occurred. Stressimaging wasperformedfirst, followed 2 hours later with a rest injection. A 1S-mCi doseof Tc-99m teboroxime was used for eachinjection. Thirty subjects- 10 normal volunteers and 20 patients with CAD documentedby recent angiography-all had stress perfusion imaging performed with both Tl-201 and Tc-99m teboroxime. There was no significant difference betweenthe 2 techniques for identifying abnormal vessels(Tc-99m teboroxime, 19 of 45; Tl-201, 21 of 45) or detecting CAD (Tc-
TABLE I Studies Using Technetium-99m Investigator
No.
Seldin and Johnson Lewo
Bellinger
Drane
Seldin and
20 15 15 5
and Planar imaging Camera/
Time
Total Imaging
Type
Window
(First Image)
Time
Counts/FOV (First Image)
Technical Factors
Planar 20% Planar 15% Planar SPECT
SFOV
20 minutes
700K
Liver third view
300-4OOK
Liver third view
350K
Liver third view Liver seen
Planar SPECT SPECT Planar
3 minutes
LFOV
60 seconds
4-5 minutes
LFOV Single head 20% LFOV Single head Triad 20% Scinticor SIM-400 1” collimator
60 seconds
45 minutes 10 minutes
3 minutes
60 seconds
All imaging was begun within 2 minutes of injectIon. Dose of 15 mCi used in all stress injections. EF = ejection fraction; FOV = field of view; L = large; S = small; SPECT = single photon emission
99m teboroxime, 16 of 20; Tl-201, 17 of 20). However, significantly more defectswere identified as fixed on resting Tc-99m teboroxime scansthan on Tl-201 redistribution scans. This might be due to residual background myocardial activity from the first (exercise) injection. Similar observationshave beenmade with Tc-99m sestamibi on same-daystress-reststudies when second-injection rest scans were compared to Tl-201 redistribution scans8Hepatic uptake of Tc-99m teboroxime, which was apparent in the second image (shallow left anterior oblique) in all subjects, obscured inferoapical segment analysis in 14 of 20 patients but did not interfere with abnormal vesselidentification. Seldin et al3were restricted to an imaging protocol lasting about 20 minutes, which included camerarepositioning time. Subsequently,it was shown independently by Leppo and by Bellinger that adequatemyocardial countscan be acquired by an Anger camera in 40 to 60 seconds/view after Tc-99m teboroxime injection (personalcommunication). Theseinvestigators showedthat using large field-of-view (LFOV) detectors, counts per image ranged from 300,000 to 400,000 (Fig. 4). Both of theseinvestigators used planar imaging protocols that were completed within 4 to 5 minutes. Standard SPECT imaging using single-head cameras and lo-minute acquisitions over 180” using standard “stop and shoot” software was performed by several groups of investigators (Table I). Continuous acquisition of counts over 180” shortensthe total acquisition time to under 6 minutes, minimizing the effectsof hepatic uptake (Fig. 5). The planar and SPECT data from 4 groups of investigators participating in the phaseIII trial were combined for analysis.9The “gold standard” in this trial was identified as“overall clinical impression,” which wasdefined as the results of all clinical data except for the Tc-99m teboroxime data and included angiography plus Tl-201 in 6 1 patients, angiography alone in 40 and Tl-201 alone in 54. For the 101 patients who had coronary angiography plus Tc-99m teboroxime study, the agreementfor detect-
Liver third view Liver uptake No liver
20 minutes 10 minutes 2 minutes 4-5 minutes
computed
tomography;
800K
Also EF data ? resolution
? = questionable.
FIGURE 3. A, Thallium-201 exercise (fop) and redistribution (bottom) images in the anterior, 30” left anterior oblique and 60” left anterior oblique views in a patient with 99% stenosis of the left anterior descending coronary artery. There is decreased perfusion of the anteroseptum and apex with partial redistribution at 4 hours. R, Technetium-99m teboroxbne exercise (fop) and rest (bottom) images of the same patient showing a comparable anteroapical defect, which partially fills in on reinjection. Hepatic activity on the steep oblique view obscures the inferoapical segment on the exercise image and the anteroapical and inferoapical segments on the rest image. (Reproduced with permission from J NUC/ ~ed.3)
Tl-iE AMERICAN JOURNAL OF CARDIOLOGY
OCTOBER 16,199O
A SYMPOSIUM:
TECHNETIUM-5Sm
IMAGING
AGENTS
myocardium in the distribution of normal arteries. Their initial results suggestthat abnormal washoutin the distribution of stenosedvesselscan be detectedafter a single Tc-99m teboroxime injection with dynamic SPECT imaging using a 3-headedSPECT camera.’1 The rapid bolus injection of Tc-99m pertechnetate and a multicrystal scintillation camera have been the major components of the first-pass technique for measurement of left and right ventricular ejection fractions. There is an extensivecollection of published studiesthat apply this technique to rest and bicycle exercisestudiesin patients with valvular diseaseand CAD. The best data supporting the clinical usefulnessof exercise first-pass scintigraphy comesfrom the Duke databaseand the work of Robert Jones.Theseinvestigators found that the exerFIGURE 4. Technetium-SSm teboroxime exercise (fop] and cise left ventricular ejection fraction wasthe most imporrest (bottom) planar images acquired with rapid imaging setant singie independentpredictor of mortality in coronary quence (40 seconds/image) on large field-of-view camera showing prominent apieal cleft. Coronary angiography showed disease.12A newer model of a high count rate, portable insignificant coronary artery disease. Study performed by Ray multicrystal scintillation camera is now available (SIMRellinger, MD, Travis Air Force Base, California. ANT = an400, Scinticor). This camera,which hasa very small field terior; LAO = left anterior oblique. of view, can be used with treadmill exercise. Johnson, Seldin and co-workers developed a protocol using Tcing CAD was 76% and was not different from the agree- 99m teboroxime and the portable high count rate camera ment between coronary angiography and Tl-201 scintig- equippedwith a l-inch collimator to maximize resolution raphy (80%). For the 115 patients who underwent Tc- and a motion correction program developedby Port and 99m teboroxime and Tl-201 scintigraphy, the agreement colleagues.l 3 Before exercise, a low-energy radioactive between the 2 imaging modalities was 90%. The overall source is fastened on the skin over the sternum. The sensitivity and specificity for blinded reading were 82 and patient runs on the treadmill with the chest positioned 9 1%, respectively, when compared against overall clinical several inches in front of the camera detector and is injected with 15 mCi of Tc-99m teboroxime as a bolus at impression. peak exercise.Dynamic dual isotopeimaging is acquired STUDIES IN PROGRESS for 50 seconds,the treadmill is slowedand the patient is Becauseof the unique pharmacokinetic properties of seatednext to the treadmill. Three planar images using Tc-99m teboroxime, several clinical applications are be- only the technetium window are acquired within 4 mining pursued that take the aforementioned standard imag- utes. The low-energy sourceis usedto correct the dynaming protocols 1 step further. Tc-99m teboroxime is a ic first-pass data for patient motion. The potential drawneutral, lipophilic agent with rapid myocardial washout. The removal rate of a neutral lipophilic agent that freely enters and leaves a myocardial cell should be related to Vertical Horizontal Short Axis capillary blood flow. Tc-99m teboroxime thus bearssimilarities to xenon-133, the inert gas that was injected into the coronary arteries to quantitate regional myocardial blood flow from initial washout using the Schmidt-Kety Stress formula and a multicrystal scintillation camera.‘ODifferences in regional Tc-99m teboroxime washout may be usedto estimate regional differences in myocardial blood flow. To overcomecurve contamination by residual blood pool activity and overlap of coronary vascular bedsin any single planar projection, SPECT acquisition is preferable Rest for this application. Drane and co-workers** are performing dynamic SPECT imaging with Tc-99m teboroxime and a 3-headed SPECT camera. These investigators acquire ten 2-minute sequential SPECT acquisitions. In addition to measuring regional washout, they compare FIGURE 5. Technetium-SSm teboroxime stress (top) and rest the initial scansto normalized scansacquired at 18 to 20 (frottomJ single photon emission computed tomography reconminutes after injection to determine redistribution. This sfructions acquired as a 6-mirmte 160’ rotation on standard rotating gamma camera. Anteroseptal and inkroredistribution is actually differential washout due to the single-head basal detects show almost complete fill-in on res6ng scans. slower clearing of activity from myocardium in the distri- Study performed by David Yuille, St Luke’s Medical Center, bution of stenosedarteries compared to clearance from Milwaukee, Wisconsin.
66E
THE AMERICAN JOURNAL OF CARDIOLOGY
VOLUME 66
back to this technique is the limited resolution of the high SPECT camera with rapid acquisition and the Scinticor count rate camera (equivalent to a 40 X 40 matrix on an SIM-400, a portable high count rate camera, are well SFOV Anger camera) (Fig. 6). Small defects may be suited to imaging Tc-99m teboroxime. The SIM-400 missed,but this deficit may be balanced by the additional gives ejection fraction data as well as perfusion imaging, diagnosticinformation added by the left ventricular ejec- but the image quality of the latter may be limited by the tion fraction numbers. The entire stressand rest proce- poorer resolution of a multicrystal camera. dure with 2 sets of planar images and ventricular funcAlthough both Tc-99m sestamibi and Tc-99m tebortional data can be obtained within 1 hour. oxime will probably competewith eachother and with Tl201 in stress/rest (redistribution) perfusion imaging, the SUMMARY very different pharmacokinetics of the 2 technetium perTc-99m teboroxime is a technetium-basedmyocardial fusion imaging agentspresent a definite choiceto nuclear perfusion imaging agent. It is one of the BAT0 com- imaging laboratories. A complete diagnostic study inpounds,which are neutral and lipophilic. Its pharmacoki- cluding stress and rest ejection fraction performed in 1 netics are very different from those of Tc-99m sestamibi, hour, compared to 5 hours for Tl-201 stressand redistria cationic technetium-basedmyocardial perfusion agent. bution, will probably find high patient approval. LaboraThe myocardial extraction for Tc-99m sestamibi is less tories must, however, be set up for rapid patient movethan that for Tl-201, whereasthe myocardial extraction ment betweentreadmill and camera. Alternatively, pharfor Tc-99m teboroxime is greater than that for Tl-201. macologic stress with intravenous dipyridamole or Tc-99m sestamibihasvery slow myocardial washout and adenosinecan be administered while the patient is lying negligible redistribution, whereas Tc-99m teboroxime under the camera, obviating the need to move the patient has rapid myocardial washout, with differential washout and potentially shortening pharmacologic stress/perfurates in the distribution of stenotic and normal vessels, sion imaging studies. Both Tc-99m teboroxime and Tcleading to apparent “redistribution” soonafter injection. 99m sestamibi offer the opportunity to acquire function The first half-life for the myocardium is 11 minutes. As and regional perfusion data in a single procedure, which Tc-99m teboroxime is washing out of the heart, hepatic will add important diagnostic and prognostic information uptake occurs,peaking at 4.5 to 7.0 minutes. The combi- to the study. nation of rapid myocardial washout and early hepatic REFERENCES uptake has necessitateddeveloping rapid imaging proto- 1. Wackers FJT, Berman DS, Maddahi J, Watson DD, Belier GA, Strauss H, Boucher CA, Picard M, Holman BL, Fridrich R, Inglese E, Delaloye B, Bischofcols. Protocolsusing standard Anger camerasand 40- to Delaloye A, Camin L, McKusick K. Technetium-99m hexakis 2-methoxyisobutyl 60-secondplanar acquisitions or 180’ rotational SPECT isonitrile: human biodistribution, dosimetry, safety, and preliminary comparison I for myocardial perfusion imaging J Nucl Med 1989;30:30 l-3 11. and lo-minute acquisitions beginning within 2 minutes of 2.to thallium-20 Kiat H, Maddahi J, Roy LT, Van Train K, Friedman J, Resser K, Berman DS. injection have minimized interference from hepatic up- Comparison of technetium 99m mcthoxy isobutyl isonitrile and thallium 201 for take, while yielding high counts over the myocardium. evaluation of coronary artery disease by planar and tomographic methods. Am J 1989;117:1-11. The sensitivity and specificity for detecting CAD or iden- Heart 3. Seldin DW, Johnson LL, Blood DK, Muschel MJ, Smith KF, Wall RM, tifying abnormal vesselsare comparable to those with Tl- Cannon PJ. Myocardial perfusion imaging with technetium-99m SQ30217: com201. Stress and rest studies can be performed within 1 parison with thallium-201 and coronary anatomy. J Nucl Med 1989;30:312-319. 4. Narra RK, Nunn AD, Kuczynski BL, Feld T, Wedeking P, Eckelmen WC. A hour. Although good imagescan be acquired using stan- neutral technetium-99m complex for myocardial imaging. J Nucl Med dard cameras,cameras with rapid counting or imaging 1989;30:1830-1837. 5. Leppo JA, Meerdink DJ. Comparative myocardial extraction of two technecapabilities are uniquely suited for this agent. A 3-headed tium-labeled BAT0 derivatives (SQ30217, SQ32014) and thallium. J Nucl Med
FIGURE 6. Exercise thallium-201 (Tl-201) planar images (fop) and technetium-99m teboroxime planar images (bottom) showing posterolateral defect in a patient with an old posterorateral infarct. Acquired with rapid imaging sequence (30 seconds/image) on SlM-409 multicrystal camera. ANT = anterior; LAO = left anterior oblique; ex = exercise.
1990;3 1:67-74. 6. Coleman RE, Maturi M, Nunn AD, Eckelmen WC, Juri PN, Cobb FR. Imaging of myocardial perfusion with Tc-99m SQ30217: dog and human studies b (abstr). .I Nucl Med 1986:27:893. 7. Bok BD, Bite AN, Clausen M, Wong DF, Wagner HN. Artifacts in camera based single photon emission tomography due to time activity variation, Eur J Nucl Med 1987;13:439-442. 6. Taillcfer R, Gagnon A, Laflamme L, Gregoire J, Leveille J, Phaneuf D-C, 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 1989;15:113-117. 9. Zielonka JS, Bellinger R, Coleman RE, Drane W, Johnson LL, Seldin DW, Leppo J, Reba R, Wasserman A. Multicenter clinical trial of 99m-Tc teboroxime (SQ30217, Cardiotec) as a myocardial perfusion agent (abstr). J Nucl Med 1989;30:1745. 10. Cannon PJ, Dell RB, Dwyer EM. Measurement of regional myocardial and a scintillation camera. J Clin Invest perfusion in man with ‘%mon 1972;51:964-977. 11. Dram WE, Decker M, Strickland P, Tineo A, Zmuda S. Measurement of regional myocardial perfusion using Tc-99m teboroximc (Cardiotec) and dynamic SPECT (abstr). J Nucl Med 1989;30:1744. 12. Pryor DB, Harrell FE, Lee KL, Rosati RA, Coleman E, Cobb FR, Califf RM, Jones RH. Prognostic indicators from radionuclide angiography in medically treated patients with coronary artery disease. Am J Cardiol 1984;53:18-22. P3. Port S, Gal R, Grenier R, Acharya K, Shen Y, Skrade B. First-pass radionuelide angiography during treadmill exercise: evaluation of patient motion and a method for motion correction (abstr). J Nucl Mrd 1989;30:770.
THE AMERICAN
JOURNAL
OF CARDIOLOGY
OCTOBER
16,199O