Metabolic Imaging With -Methyl-p-[123I]-Iodophenyl-Pentadecanoic Acid Identifies Ischemic Memory After Demand Ischemia
http://www.100md.com
《循环学杂志》
the Division of Nuclear Medicine, University of Maryland Hospital and School of Medicine, Baltimore (V.D., M.Y.M.)
Cardiovascular Consultants, Kansas City, Mo (T.M.B.)
Cardiovascular Division, Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, NY (S.R.B.)
Oklahoma Heart Institute, Tulsa (R.D.P.)
Molecular Insight Pharmaceuticals, Inc, Cambridge, Mass (A.E.G., J.W.B.)
Division of Cardiology, Tufts-New England Medical Center and Tufts University School of Medicine, Boston, Mass (J.E.U.).
Abstract
Background— After myocardial ischemia, prolonged suppression of fatty acid metabolism may persist despite restoration of blood flow, which is called metabolic stunning. We hypothesized that a branched-chain fatty acid, -methyl-p-[123I]-iodophenyl-pentadecanoic acid (BMIPP), might identify the presence of myocardial ischemia late after demand ischemia at rest up to 30 hours later.
Methods and Results— In 32 patients with exercise-induced ischemia on thallium SPECT, BMIPP was injected at rest within 30 hours of ischemia. SPECT images were acquired beginning 10 minutes after injection (early) and again 30 minutes after injection (delayed). Thallium and BMIPP SPECT data were read separately by 3 observers blinded to other imaging and clinical data. Agreement between BMIPP and thallium data for the presence of an abnormality on the patient level was 91% (95% CI, 75 to 98) for the early BMIPP data and 94% (95% CI, 79 to 99) for the delayed BMIPP data. Agreement between delayed BMIPP and thallium was 95% among 21 patients studied on the same day, a mean of 6.2±1.4 hours after exercise-induced ischemia, and 91% among the 11 patients studied on the next calendar day, a mean of 24.9±2.6 hours after ischemia (P=NS). The magnitude of resting BMIPP metabolic defect by semiquantitative visual analysis was correlated to the magnitude of exercise-induced thallium perfusion defect (r=0.6, P<0.001 for early BMIPP; r=0.5, P=0.005 for delayed BMIPP).
Conclusions— Metabolic imaging with BMIPP identifies patients with recent exercise-induced myocardial ischemia. These findings support the concept that BMIPP imaging can successfully demonstrate the metabolic imprint of a stress-induced ischemic episode, also known as ischemic memory.
Key Words: imaging ischemia metabolism nuclear medicine radioisotopes
Introduction
After transient myocardial ischemia, recovery of regional abnormalities in perfusion, function, and metabolism may be temporally disparate.1 Recovery of regional perfusion is a prerequisite for recovery of regional function and metabolism. Although delayed recovery of regional function after reperfusion is well documented in the clinical setting as stunned myocardium,2 delayed recovery of regional metabolism, called metabolic stunning, is not well established.
Utilization of free fatty acids is the preferred metabolic pathway for high-energy ATP production in the normal myocardium.3 In the setting of myocardial ischemia and locally reduced oxygen supply, the myocardium shifts high-energy ATP production from fatty acid metabolism (mitochondria, aerobic) to glucose utilization (cytoplasm, anaerobic). Accelerated glucose metabolism in this setting is thought to provide compensatory energy production.4
-Methyl-p-[123I]-iodophenyl-pentadecanoic acid (BMIPP) is a methyl branched-chained fatty acid that is trapped in myocardial cells with limited catabolism.5 Because of its high myocardial uptake and long retention, BMIPP, which in clinical studies has been labeled with 123I, provides excellent images of the left ventricular myocardium. Uptake of BMIPP in the myocardium most likely reflects activation of fatty acids by coenzyme A and indirectly reflects cellular ATP production resulting from fatty acid metabolism.6 Thus, in the setting of myocardial ischemia, a reduction in ATP production secondary to diminished fatty acid metabolism is mirrored by decreased myocardial BMIPP uptake.7 In the clinical setting, the finding of persistent and prolonged disturbances in BMIPP uptake, long after resolution of ischemic symptoms, may provide a direct scintigraphic imprint as to the underlying cause of the patient’s symptoms. In patients presenting to the emergency department with acute chest pain and no myocardial infarction, sustained alterations of myocardial fatty acid metabolism as assessed by BMIPP have been described in the absence of abnormalities in regional myocardial blood flow.8 The potential importance of this observation is that BMIPP imaging may extend the time window for identifying myocardial ischemia long after resolution of chest pain and restoration of resting myocardial blood flow.
In this study, we hypothesized that, analogous to patients presenting to the emergency room with "supply-type" ischemia (with an acute coronary syndrome), metabolic stunning as assessed by BMIPP may also occur after demand ischemia (exercise on a treadmill). We determined whether the location and extent of myocardial ischemia after demand ischemia on a treadmill could be assessed with BMIPP imaging during the subacute phase at rest up to 30 hours after the ischemic event.
Methods
Patient Population
This investigation was an open-label, within-patient, comparative study conducted at 4 study sites. The target patients were those who had experienced documented exercise stress-induced ischemia during a clinically indicated stress/redistribution9 or reinjection10 thallium myocardial perfusion imaging study. The stress-induced ischemia event must have occurred no more than 30 hours before the planned administration of BMIPP. Each patient was to serve as his or her own control. Eligible patients fulfilled the following requirements: They were >18 years of age, had undergone a clinically indicated thallium stress/redistribution or reinjection perfusion test within 30 hours before administration of BMIPP, and had a stress myocardial perfusion defect in at least 1 vascular territory with a normal or near-normal finding for that same territory on the redistribution or reinjection thallium image. If female and of childbearing potential, the patient had to be using an acceptable form of birth control. All female subjects of childbearing potential had a negative serum pregnancy result on the day of the test before receiving BMIPP. Patients were excluded if there was a history of cardiomyopathy with a left ventricular ejection fraction <35%, if there was class IV congestive heart failure, if there was evidence of myocardial infarction in the same segment of reversible ischemia, if the patient was an insulin-requiring diabetic, or if the patient had a blood glucose level >200 mg/dL at screening. All patients signed informed consent, and all institutional review boards at the participating centers approved the study.
Study Design
After the clinically indicated thallium stress study demonstrating ischemia and after informed consent was obtained, baseline assessments were performed. Within 30 hours of the treadmill stress-induced ischemia, 3 to 6 mCi [123I]-BMIPP containing up to 0.5 mg BMIPP was injected at rest. The timing of the BMIPP injection relative to the time of stress-induced ischemia was not dictated per protocol. Rather, sites were asked to schedule the research BMIPP SPECT study to result in a wide range of postischemia BMIPP SPECT imaging times.
SPECT Imaging
SPECT imaging was performed with double-headed or triple-headed gamma cameras equipped with low-energy high-resolution collimators. A 10% to 20% window was centered on the 159-keV photopeak of 123I. Ten minutes after administration of [123I]-BMIPP, SPECT image acquisition began for the "early" BMIPP image. The total acquisition time was 10 minutes. At 35 to 60 minutes after injection, a second BMIPP SPECT acquisition (the "delayed" image) was started, with a total acquisition time of 20 minutes. Images were acquired over 360° (20 projections per head in a 64x64 digital matrix) with a triple-head camera or 180° (32 projections per head in a 64x64 digital matrix) with a dual-head camera in the step-and-shoot mode. Reconstruction was performed with a conventional filtered back-projection algorithm to an in-plane resolution of 10-mm full-width half-maximum. Corrections for attenuation or scatter were not performed. Background medications were not altered between the imaging sessions.
SPECT Image Analysis
All SPECT images were interpreted by 3 expert nuclear cardiology readers after images were transferred to a core nuclear imaging laboratory. The readers interpreted the studies by consensus, using a 17-segment model of the myocardium and a semiquantitative visual analysis score on a 5-point scale, according to recommended guidelines.11 The stress/reinjection or redistribution thallium studies were read by readers blinded to patient clinical data and to the SPECT BMIPP data from the same patients. The early and delayed SPECT BMIPP studies were read at a distinct sitting from the thallium SPECT interpretation, separated by 4 weeks, and were interpreted by readers blinded to patient clinical data and to the thallium SPECT data. For the thallium studies, summed stress score (SSS) and summed rest score (SRS) were calculated, as was a summed difference score (SDS), which represents the extent and severity of reversible perfusion defects (SDS=SSS–SRS). For the BMIPP early and delayed images, SSS were calculated for each image set on the basis of the sum of the 17 segment scores for each image set.
Data and Statistical Analyses
Level of agreement between the early or delayed BMIPP data and the thallium stress data was calculated initially for the dichotomous classification of studies as normal or abnormal on a patient level (abnormal as a prespecified summed score >3). Agreement was also assessed on a vascular territory level, assigning individual segments to vascular territories as per guideline-recommended segmentation schema.11 Analyses were performed for all patients studied at all time points. A stratified analysis was also performed on the basis of whether the patients were studied with BMIPP on the same calendar day as the thallium study (n=21) or on the following day (n=11). Summed scores were compared by use of paired t tests. Correlations were also performed between the thallium stress data and the BMIPP early and delayed imaging using the Pearson correlation coefficients.
Results
Study Population
Most patients (24 of 32, 75%) were male. Patients ranged in age from 38 to 80 years; the mean age was 60.9 years. Most patients (24 of 32, 75%) were white. All patients had blood glucose levels <200 mg/dL at screening. A patient-reported history of myocardial infarction was present in 7 of 31 patients (23%). ECG evidence of myocardial infarction by the presence of pathological Q waves was present in 2 of 31 patients (6%). During stress testing, evidence of ECG ischemia was present in 11 of 31 patients (35%), and chest pain during exercise occurred in 9 of 31 patients (29%).
Distribution of Thallium Defects
Among the 32 patients studied, 31 (97%) were graded as having abnormal perfusion defect on the thallium stress images and 1 (3%) was graded as normal by the blinded reading panel. After thallium redistribution or reinjection at rest, 30 of the 31 patients (97%) with abnormal studies on stress demonstrated evidence of reversibility, and 1 patient (3%) demonstrated only an irreversible defect. The vascular distribution of thallium defects was 72% in the left anterior descending, 59% in the left circumflex, and 87% in the right coronary artery vascular territory.
Relation Between BMIPP Uptake and Thallium Ischemia
The thallium studies were rated as optimal/diagnostic quality in 84% of the stress studies and 75% of the redistribution or reinjection studies. Similarly, BMIPP studies were rated as optimal/diagnostic quality in 75% of the early studies and 97% of the delayed studies. The remainder were rated as suboptimal but diagnostic/interpretable.
Including all vascular territories and analyzing overall patient-level concordance, there was 91% agreement between the early BMIPP and the thallium data for the presence or absence of a scintigraphic abnormality (95% CI, 75 to 98). The data were similar for the delayed BMIPP, with 94% agreement (95% CI, 79 to 99). Agreement data on a patient level, by gender, and by vascular territory are summarized in Table 1.
Sensitivity and positive predictive values for detection of patients and vascular territories with reversible thallium defects are shown in Table 2.
Relation Between Same-Day Versus Next-Day BMIPP Studies and Thallium Ischemia
The protocol called for BMIPP injection and imaging to be performed within 30 hours of the stress-induced thallium ischemia. As the patients were recruited for the protocol, some were injected with BMIPP within the same calendar day as the thallium study, whereas others were injected on the next calendar day. Thus, there was a bimodal distribution with regard to the timing of the BMIPP injections and imaging: 21 patients were studied the same day (mean, 6.2±1.4 hours after exercise-induced ischemia; range, 4.4 to 9.2), and 11 patients were studied on the next calendar day (mean, 24.9±2.6 hours after exercise-induced ischemia; range, 21.7 to 30.5). The patient-level agreement of BMIPP with the thallium studies based on day of imaging is shown in Table 1. There appeared to be no significant difference between same-day and next-day BMIPP studies and thallium ischemia, with the caveats of relatively smaller number of patients studied the next day and thus relatively wide CIs.
Relation Between the Extent and Severity of BMIPP Metabolic Defect and Exercise-Induced Thallium Defect
The 32 studies graded by the blinded readers had a mean thallium SSS of 11.9±7.4, with no significant difference between early (12.3±7.3) and delayed (10.8±4.5) BMIPP SSS (P=0.76 and 0.41 for comparison of the early and delayed BMIPP with the thallium SSS, respectively). When the imaging data were analyzed by vascular territory, no significant difference was observed between the thallium SSS for each vascular territory and the BMIPP early or delayed SSS from that territory (Table 3). Early BMIPP SSS correlated with thallium SSS among all patients with abnormal thallium studies (r=0.61, P<0.001) (Figure 1). Similarly, there was good agreement between and delayed BMIPP SSS and thallium SSS (r=0.49, P=0.005). When a single patient with a large infarction in 1 territory was excluded, significant modest correlation remained (r=0.43, P<0.02).
These data suggest that the extent and severity of metabolic abnormality (as a reflection of preceding ischemia) increased in general proportion to the extent and severity of exercise-induced perfusion abnormality identified by the thallium imaging. An example of the similar extent and severity of rest metabolic BMIPP defect compared with exercise-induced thallium perfusion defects is shown in Figure 2.
Discussion
Depending on fuel availability and the physiological environment, the heart oxidizes the most efficient metabolic substrate for high-energy ATP production. In the setting of myocardial ischemia, suppression of fatty acid metabolism and a switch to glucose utilization represent metabolic adaptation of the stressed heart. Such metabolic adaptation has been well characterized in patients with hibernating myocardium. In this study, we demonstrate that suppression of fatty acid metabolism persists for up to 30 hours after exercise-induced myocardial ischemia, long after restoration of blood flow at rest, called metabolic stunning.
Alterations in myocardial fatty acid metabolism were first evaluated noninvasively in humans with the positron-emitting radiotracer C-11 palmitate, which requires an onsite cyclotron and a PET camera.12 Because most nuclear cardiology laboratories are equipped with SPECT cameras, investigators subsequently focused their attention on developing gamma-emitting fatty acid tracers.13 In contrast to palmitate, which is a straight-chain fatty acid that is rapidly catabolized, BMIPP is an iodine-labeled, methyl branched-chain fatty acid that is trapped predominantly in myocardial cells with limited catabolism.5 Uptake of BMIPP from the plasma into myocytes occurs via the CD36 transporter protein present on the sarcolemmal membrane.14 Once in the cell, BMIPP will back-diffuse to the plasma, accumulate in the lipid pool, or undergo limited - and -oxidation. Enzymatic conversion of BMIPP to BMIPP-CoA or triacylglycerol in the myocyte is ATP dependent and is an irreversible step. Such conversion prevents back-diffusion of BMIPP to the plasma and facilitates its cellular retention.6 The prolonged retention of BMIPP in the myocardium, combined with rapid clearance from the blood and diminished uptake in the liver and lung, results in excellent visualization and imaging of the myocardium by SPECT techniques. Thus, BMIPP provides a means of measuring myocardial fatty acid utilization in vivo.
The data from this study suggest that BMIPP can successfully image the suppression of fatty acid metabolism after exercise-induced ischemia at least up to 30 hours after an ischemic episode. There was excellent patient-level agreement (> 90%) between BMIPP and thallium data for the presence or absence of a scintigraphic abnormality on both the early and delayed BMIPP imaging. Furthermore, there was also good correlation between the extent and severity of the stress thallium perfusion abnormality (reflected by the summed segmental scores) and the extent and severity of the BMIPP metabolic abnormality on both the early and delayed BMIPP SPECT images. This is important because the extent and severity of the stress SPECT perfusion abnormality (often evaluated by the SSS) is the most powerful predictor of natural history outcome and is a more complete descriptor of both the extent of coronary artery disease and prognosis than simply classifying the scan as positive or negative.
Our findings suggest that disturbances in cellular fatty acid metabolism may persists for up to 30 hours after demand ischemia, as evidenced by the presence of decreased BMIPP uptake despite restoration of regional blood flow at rest after thallium reinjection (3 to 4 hours after exercise). Long after restoration of blood flow at rest, BMIPP images (injected and acquired at rest) showed regional defects that reflected the extent and severity of exercise-induced thallium defects rather than thallium uptake at redistribution or reinjection. These data support the concept that BMIPP imaging can successfully demonstrate the metabolic imprint of a stress-induced ischemic episode, also known as ischemic memory. These data lay the groundwork for subsequent studies to assess more directly the clinical scenarios in which BMIPP data may provide important diagnostic and prognostic information to assist in clinical decision making. The findings of metabolic stunning after demand ischemia (exercise on a treadmill) extend the previous observation by Kawai and associates,8 who studied patients with "supply-type" ischemia (acute coronary syndrome) in the emergency room. All patients underwent myocardial perfusion SPECT at rest, BMIPP metabolic images within 2 days after the perfusion study, and coronary angiography within 1 to 4 days. Among patients with documented coronary artery stenosis or spasm (on ergonovine provocation), metabolic defects were present in 74% of patients on BMIPP imaging, whereas only 38% of patients showed perfusion defects on rest SPECT imaging (P<0.001). Both BMIPP and perfusion studies were normal in >90% of patients without coronary stenosis or spasm. Thus, BMIPP may have an important role in detecting myocardial ischemia remote from symptoms and in the absence of resting myocardial perfusion abnormalities. Although a myocardial perfusion study at rest can also identify patients with acute chest pain resulting from myocardial ischemia, myocardial perfusion may appear normal at rest in the subacute phase, particularly if the chest pain was >2 to 3 hours earlier.15,16
In this study, the distribution of timing of the rest BMIPP injection fell into a bimodal distribution. Twenty-one patients were studied with BMIPP on the same calendar day as the thallium study, and 11 patients had BMIPP imaging on the subsequent calendar day. When these data were analyzed separately, there did not seem to be any influence of the "same day/next day" injection time on the overall patient-level agreement between the BMIPP and the stress thallium studies. In a separate subgroup analysis, there also did not seem to be any influence of gender on the overall patient-level agreement. Exactly how long the "window" extends after an exercise-induced ischemic insult for BMIPP to identify suppression of fatty acid metabolism cannot be determined from the data in this study. Whether the window extends beyond 30 hours requires future studies with longer time intervals between the different isotopes.
This analysis has several limitations. Patients were eligible for this study on the basis of having a stress-induced ischemic defect on a clinically indicated thallium study. Thus, bias may have been introduced into the image analysis by the readers’ knowledge of the protocol design. However, the BMIPP data were read at a temporally distinct sitting, and the correlation between the summed scores of the BMIPP and the thallium data, as well as the similarity of the summed scores in paired comparisons (Table 3), suggests that bias could not fully explain the findings. The sequence of the imaging studies could not be randomized because, by design, patients were required to have demonstrated a reversible thallium defect on SPECT imaging to be eligible for enrollment. Although this may have engendered a sequence bias, it should be pointed out that the territories of interest, the ischemic areas with reversible thallium defects, already had essentially normal uptake of thallium on redistribution or reinjection imaging by definition; thus, the BMIPP defects are unlikely to have been even in part "caused" by the stress thallium defects. The relatively small number of patients renders preliminary any conclusions with regard to the similarity of same day/next day or early versus delayed data pending larger data sets in which a type 2 error can be more clearly ruled out. Moreover, patients in this study did not consistently undergo cardiac catheterization. Thus, data correlating the metabolic abnormalities and the site and severity of coronary stenoses are not available from this data set but will be an important focus for future studies. The lower agreement between the BMIPP and the thallium studies in the per-vascular-territory analysis relative to patient-level agreement is similar to data reported in other studies comparing 2 tracers for myocardial uptake.17 Because BMIPP assesses metabolic consequence of myocardial ischemia and thallium assesses myocardial perfusion, the differences observed in summed score in specific vascular territories could be secondary to actual differences in the underlying pathophysiology of perfusion versus metabolism rather than simple misregistration from the blinded readings. Although fully quantitative analysis of the thallium and BMIPP images would have been desirable to confirm our findings, quantitative normal databases are not yet available for BMIPP. Hence, quantification with comparison to normal limits was not possible for the BMIPP images.
Hence, rest imaging with [123I]-BMIPP, performed up to 30 hours after a stress-induced ischemic episode, can identify regions of suppressed fatty acid metabolism, also known as ischemic memory. Despite restoration of blood flow at rest (as evidenced by thallium reinjection), prolonged and persistent disturbances in fatty acid metabolism after an ischemic event represent a scintigraphic marker of metabolic stunning. Application of such a metabolic tracer, as opposed to a perfusion tracer, potentially extends the time window for noninvasive imaging an ischemic event beyond the resolution of symptoms.
Acknowledgments
Disclosure
Drs Dilsizian, Bateman, Bergmann, Des Prez, Magram, and Udelson and/or their institution received financial support for this study from Molecular Insight Pharmaceuticals, Inc. Drs Goodbody and Babich are employees of and shareholders in Molecular Insight Pharmaceuticals, Inc.
Footnotes
Guest Editor for this article was Robert O. Bonow, MD.
References
Taegtmeyer H, Woods A. Metabolic modulation as a principle for myocardial protection. Curr Hypertens Rep. 2003; 5: 443–444.
Braunwald E, Kloner RA. The stunned myocardium: prolonged, postischemic ventricular dysfunction. Circulation. 1982; 66: 1146–1149.
Taegtmeyer H. Metabolism: the lost child of cardiology. J Am Coll Cardiol. 2000; 36: 1386–1388.
Apstein CS, Eberli FR. Critical role of energy supply and glycolysis during short-term hibernation. Basic Res Cardiol. 1995; 90: 2–4.
Goodman MM, Kirsch G, Knapp FF Jr. Synthesis and evaluation of radioiodinated terminal p-iodophenyl substituted a- and b-methyl-branched fatty acids. J Med Chem. 1984; 27: 390–397.
Knapp FF Jr, Ambrose KR, Goodman MM. New radioiodinated methyl branched fatty acids for cardiac studies. Eur J Nucl Med. 1986; 12: S39–S44.
Hosokawa R, Nohara R, Fujibayashi Y, Okuda K, Ogino M, Hata T, Fujita M, Tamaki N, Konishi J, Sasayama S. Myocardial kinetics of iodine-123-BMIPP in canine myocardium after regional ischemia and reperfusion: implications for clinical SPECT. J Nucl Med. 1997; 38: 1857–1863.
Kawai Y, Tsukamoto E, Nozaki Y, Morita K, Sakurai M, Tamaki N. Significance of reduced uptake of iodinated fatty acid analogue for the evaluation of patients with acute chest pain. J Am Coll Cardiol. 2001; 38: 1888–1894.
Pohost GM, Zir LM, Moore RH, McKusick KA, Guiney TE, Beller GA. Differentiation of transiently ischemic from infarcted myocardium by serial imaging after a single dose of thallium-201. Circulation. 1977; 55: 294–302.
Dilsizian V, Rocco TP, Freedman NM, Leon MB, Bonow RO. Enhanced detection of ischemic but viable myocardium by the reinjection of thallium after stress-redistribution imaging. N Engl J Med. 1990; 323: 141–146.
Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan TJ, Verani MS. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: AHA Writing Group on Myocardial Segmentation and Registration for Cardiac Imaging. Circulation. 2002; 105: 539–542.
Ter-Pogossian MM, Klein MS, Markham J, Roberts R, Sobel BE. Regional assessment of myocardial metabolic integrity in vivo by positron emission tomography with 11C-labeled palmitate. Circulation. 1980; 61: 242–255.
Feinendegen LE. Measurement of myocardial fatty acid metabolism by iodine-123 heptadecanoic acid kinetics. J Nucl Med. 1987; 28: 924–925.
Tanaka T, Kawamura K. Isolation of myocardial membrane long-chain fatty acid binding protein: homology with a rat membrane protein implicated in the binding or transport of long-chain fatty acids. J Mol Cell Cardiol. 1995; 27: 1613–1622.
Udelson JE, Beshansky JR, Ballin DS, Feldman JA, Griffith JL, Handler J, Heller GV, Hendel RC, Pope JH, Ruthazer R, Spiegler EJ, Woolard RH, Selker HP. Myocardial perfusion imaging for evaluation and triage of patients with suspected acute cardiac ischemia; a randomized controlled trial. JAMA. 2002; 288: 2693–2700.
Fram DB, Azar RR, Ahlberg AW, Gillam LD, Mitchel JF, Kiernan FJ, Hirst JA, Mather JF, Ficaro E, Cyr G, Waters D, Heller GV. Duration of abnormal SPECT myocardial perfusion imaging following resolution of acute ischemia: an angioplasty model. J Am Coll Cardiol. 2003; 41: 452–459.
Acampa W, Cuocolo A, Sullo P, Varrone A, Nicolai E, Pace L, Petretta M, Salvatore M. Direct comparison of technetium 99m-sestamibi and technetium 99m-tetrofosmin cardiac single photon emission computed tomography in patients with coronary artery disease. J Nucl Cardiol. 1998; 5: 265–274.(Vasken Dilsizian, MD; Tim)
Cardiovascular Consultants, Kansas City, Mo (T.M.B.)
Cardiovascular Division, Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, NY (S.R.B.)
Oklahoma Heart Institute, Tulsa (R.D.P.)
Molecular Insight Pharmaceuticals, Inc, Cambridge, Mass (A.E.G., J.W.B.)
Division of Cardiology, Tufts-New England Medical Center and Tufts University School of Medicine, Boston, Mass (J.E.U.).
Abstract
Background— After myocardial ischemia, prolonged suppression of fatty acid metabolism may persist despite restoration of blood flow, which is called metabolic stunning. We hypothesized that a branched-chain fatty acid, -methyl-p-[123I]-iodophenyl-pentadecanoic acid (BMIPP), might identify the presence of myocardial ischemia late after demand ischemia at rest up to 30 hours later.
Methods and Results— In 32 patients with exercise-induced ischemia on thallium SPECT, BMIPP was injected at rest within 30 hours of ischemia. SPECT images were acquired beginning 10 minutes after injection (early) and again 30 minutes after injection (delayed). Thallium and BMIPP SPECT data were read separately by 3 observers blinded to other imaging and clinical data. Agreement between BMIPP and thallium data for the presence of an abnormality on the patient level was 91% (95% CI, 75 to 98) for the early BMIPP data and 94% (95% CI, 79 to 99) for the delayed BMIPP data. Agreement between delayed BMIPP and thallium was 95% among 21 patients studied on the same day, a mean of 6.2±1.4 hours after exercise-induced ischemia, and 91% among the 11 patients studied on the next calendar day, a mean of 24.9±2.6 hours after ischemia (P=NS). The magnitude of resting BMIPP metabolic defect by semiquantitative visual analysis was correlated to the magnitude of exercise-induced thallium perfusion defect (r=0.6, P<0.001 for early BMIPP; r=0.5, P=0.005 for delayed BMIPP).
Conclusions— Metabolic imaging with BMIPP identifies patients with recent exercise-induced myocardial ischemia. These findings support the concept that BMIPP imaging can successfully demonstrate the metabolic imprint of a stress-induced ischemic episode, also known as ischemic memory.
Key Words: imaging ischemia metabolism nuclear medicine radioisotopes
Introduction
After transient myocardial ischemia, recovery of regional abnormalities in perfusion, function, and metabolism may be temporally disparate.1 Recovery of regional perfusion is a prerequisite for recovery of regional function and metabolism. Although delayed recovery of regional function after reperfusion is well documented in the clinical setting as stunned myocardium,2 delayed recovery of regional metabolism, called metabolic stunning, is not well established.
Utilization of free fatty acids is the preferred metabolic pathway for high-energy ATP production in the normal myocardium.3 In the setting of myocardial ischemia and locally reduced oxygen supply, the myocardium shifts high-energy ATP production from fatty acid metabolism (mitochondria, aerobic) to glucose utilization (cytoplasm, anaerobic). Accelerated glucose metabolism in this setting is thought to provide compensatory energy production.4
-Methyl-p-[123I]-iodophenyl-pentadecanoic acid (BMIPP) is a methyl branched-chained fatty acid that is trapped in myocardial cells with limited catabolism.5 Because of its high myocardial uptake and long retention, BMIPP, which in clinical studies has been labeled with 123I, provides excellent images of the left ventricular myocardium. Uptake of BMIPP in the myocardium most likely reflects activation of fatty acids by coenzyme A and indirectly reflects cellular ATP production resulting from fatty acid metabolism.6 Thus, in the setting of myocardial ischemia, a reduction in ATP production secondary to diminished fatty acid metabolism is mirrored by decreased myocardial BMIPP uptake.7 In the clinical setting, the finding of persistent and prolonged disturbances in BMIPP uptake, long after resolution of ischemic symptoms, may provide a direct scintigraphic imprint as to the underlying cause of the patient’s symptoms. In patients presenting to the emergency department with acute chest pain and no myocardial infarction, sustained alterations of myocardial fatty acid metabolism as assessed by BMIPP have been described in the absence of abnormalities in regional myocardial blood flow.8 The potential importance of this observation is that BMIPP imaging may extend the time window for identifying myocardial ischemia long after resolution of chest pain and restoration of resting myocardial blood flow.
In this study, we hypothesized that, analogous to patients presenting to the emergency room with "supply-type" ischemia (with an acute coronary syndrome), metabolic stunning as assessed by BMIPP may also occur after demand ischemia (exercise on a treadmill). We determined whether the location and extent of myocardial ischemia after demand ischemia on a treadmill could be assessed with BMIPP imaging during the subacute phase at rest up to 30 hours after the ischemic event.
Methods
Patient Population
This investigation was an open-label, within-patient, comparative study conducted at 4 study sites. The target patients were those who had experienced documented exercise stress-induced ischemia during a clinically indicated stress/redistribution9 or reinjection10 thallium myocardial perfusion imaging study. The stress-induced ischemia event must have occurred no more than 30 hours before the planned administration of BMIPP. Each patient was to serve as his or her own control. Eligible patients fulfilled the following requirements: They were >18 years of age, had undergone a clinically indicated thallium stress/redistribution or reinjection perfusion test within 30 hours before administration of BMIPP, and had a stress myocardial perfusion defect in at least 1 vascular territory with a normal or near-normal finding for that same territory on the redistribution or reinjection thallium image. If female and of childbearing potential, the patient had to be using an acceptable form of birth control. All female subjects of childbearing potential had a negative serum pregnancy result on the day of the test before receiving BMIPP. Patients were excluded if there was a history of cardiomyopathy with a left ventricular ejection fraction <35%, if there was class IV congestive heart failure, if there was evidence of myocardial infarction in the same segment of reversible ischemia, if the patient was an insulin-requiring diabetic, or if the patient had a blood glucose level >200 mg/dL at screening. All patients signed informed consent, and all institutional review boards at the participating centers approved the study.
Study Design
After the clinically indicated thallium stress study demonstrating ischemia and after informed consent was obtained, baseline assessments were performed. Within 30 hours of the treadmill stress-induced ischemia, 3 to 6 mCi [123I]-BMIPP containing up to 0.5 mg BMIPP was injected at rest. The timing of the BMIPP injection relative to the time of stress-induced ischemia was not dictated per protocol. Rather, sites were asked to schedule the research BMIPP SPECT study to result in a wide range of postischemia BMIPP SPECT imaging times.
SPECT Imaging
SPECT imaging was performed with double-headed or triple-headed gamma cameras equipped with low-energy high-resolution collimators. A 10% to 20% window was centered on the 159-keV photopeak of 123I. Ten minutes after administration of [123I]-BMIPP, SPECT image acquisition began for the "early" BMIPP image. The total acquisition time was 10 minutes. At 35 to 60 minutes after injection, a second BMIPP SPECT acquisition (the "delayed" image) was started, with a total acquisition time of 20 minutes. Images were acquired over 360° (20 projections per head in a 64x64 digital matrix) with a triple-head camera or 180° (32 projections per head in a 64x64 digital matrix) with a dual-head camera in the step-and-shoot mode. Reconstruction was performed with a conventional filtered back-projection algorithm to an in-plane resolution of 10-mm full-width half-maximum. Corrections for attenuation or scatter were not performed. Background medications were not altered between the imaging sessions.
SPECT Image Analysis
All SPECT images were interpreted by 3 expert nuclear cardiology readers after images were transferred to a core nuclear imaging laboratory. The readers interpreted the studies by consensus, using a 17-segment model of the myocardium and a semiquantitative visual analysis score on a 5-point scale, according to recommended guidelines.11 The stress/reinjection or redistribution thallium studies were read by readers blinded to patient clinical data and to the SPECT BMIPP data from the same patients. The early and delayed SPECT BMIPP studies were read at a distinct sitting from the thallium SPECT interpretation, separated by 4 weeks, and were interpreted by readers blinded to patient clinical data and to the thallium SPECT data. For the thallium studies, summed stress score (SSS) and summed rest score (SRS) were calculated, as was a summed difference score (SDS), which represents the extent and severity of reversible perfusion defects (SDS=SSS–SRS). For the BMIPP early and delayed images, SSS were calculated for each image set on the basis of the sum of the 17 segment scores for each image set.
Data and Statistical Analyses
Level of agreement between the early or delayed BMIPP data and the thallium stress data was calculated initially for the dichotomous classification of studies as normal or abnormal on a patient level (abnormal as a prespecified summed score >3). Agreement was also assessed on a vascular territory level, assigning individual segments to vascular territories as per guideline-recommended segmentation schema.11 Analyses were performed for all patients studied at all time points. A stratified analysis was also performed on the basis of whether the patients were studied with BMIPP on the same calendar day as the thallium study (n=21) or on the following day (n=11). Summed scores were compared by use of paired t tests. Correlations were also performed between the thallium stress data and the BMIPP early and delayed imaging using the Pearson correlation coefficients.
Results
Study Population
Most patients (24 of 32, 75%) were male. Patients ranged in age from 38 to 80 years; the mean age was 60.9 years. Most patients (24 of 32, 75%) were white. All patients had blood glucose levels <200 mg/dL at screening. A patient-reported history of myocardial infarction was present in 7 of 31 patients (23%). ECG evidence of myocardial infarction by the presence of pathological Q waves was present in 2 of 31 patients (6%). During stress testing, evidence of ECG ischemia was present in 11 of 31 patients (35%), and chest pain during exercise occurred in 9 of 31 patients (29%).
Distribution of Thallium Defects
Among the 32 patients studied, 31 (97%) were graded as having abnormal perfusion defect on the thallium stress images and 1 (3%) was graded as normal by the blinded reading panel. After thallium redistribution or reinjection at rest, 30 of the 31 patients (97%) with abnormal studies on stress demonstrated evidence of reversibility, and 1 patient (3%) demonstrated only an irreversible defect. The vascular distribution of thallium defects was 72% in the left anterior descending, 59% in the left circumflex, and 87% in the right coronary artery vascular territory.
Relation Between BMIPP Uptake and Thallium Ischemia
The thallium studies were rated as optimal/diagnostic quality in 84% of the stress studies and 75% of the redistribution or reinjection studies. Similarly, BMIPP studies were rated as optimal/diagnostic quality in 75% of the early studies and 97% of the delayed studies. The remainder were rated as suboptimal but diagnostic/interpretable.
Including all vascular territories and analyzing overall patient-level concordance, there was 91% agreement between the early BMIPP and the thallium data for the presence or absence of a scintigraphic abnormality (95% CI, 75 to 98). The data were similar for the delayed BMIPP, with 94% agreement (95% CI, 79 to 99). Agreement data on a patient level, by gender, and by vascular territory are summarized in Table 1.
Sensitivity and positive predictive values for detection of patients and vascular territories with reversible thallium defects are shown in Table 2.
Relation Between Same-Day Versus Next-Day BMIPP Studies and Thallium Ischemia
The protocol called for BMIPP injection and imaging to be performed within 30 hours of the stress-induced thallium ischemia. As the patients were recruited for the protocol, some were injected with BMIPP within the same calendar day as the thallium study, whereas others were injected on the next calendar day. Thus, there was a bimodal distribution with regard to the timing of the BMIPP injections and imaging: 21 patients were studied the same day (mean, 6.2±1.4 hours after exercise-induced ischemia; range, 4.4 to 9.2), and 11 patients were studied on the next calendar day (mean, 24.9±2.6 hours after exercise-induced ischemia; range, 21.7 to 30.5). The patient-level agreement of BMIPP with the thallium studies based on day of imaging is shown in Table 1. There appeared to be no significant difference between same-day and next-day BMIPP studies and thallium ischemia, with the caveats of relatively smaller number of patients studied the next day and thus relatively wide CIs.
Relation Between the Extent and Severity of BMIPP Metabolic Defect and Exercise-Induced Thallium Defect
The 32 studies graded by the blinded readers had a mean thallium SSS of 11.9±7.4, with no significant difference between early (12.3±7.3) and delayed (10.8±4.5) BMIPP SSS (P=0.76 and 0.41 for comparison of the early and delayed BMIPP with the thallium SSS, respectively). When the imaging data were analyzed by vascular territory, no significant difference was observed between the thallium SSS for each vascular territory and the BMIPP early or delayed SSS from that territory (Table 3). Early BMIPP SSS correlated with thallium SSS among all patients with abnormal thallium studies (r=0.61, P<0.001) (Figure 1). Similarly, there was good agreement between and delayed BMIPP SSS and thallium SSS (r=0.49, P=0.005). When a single patient with a large infarction in 1 territory was excluded, significant modest correlation remained (r=0.43, P<0.02).
These data suggest that the extent and severity of metabolic abnormality (as a reflection of preceding ischemia) increased in general proportion to the extent and severity of exercise-induced perfusion abnormality identified by the thallium imaging. An example of the similar extent and severity of rest metabolic BMIPP defect compared with exercise-induced thallium perfusion defects is shown in Figure 2.
Discussion
Depending on fuel availability and the physiological environment, the heart oxidizes the most efficient metabolic substrate for high-energy ATP production. In the setting of myocardial ischemia, suppression of fatty acid metabolism and a switch to glucose utilization represent metabolic adaptation of the stressed heart. Such metabolic adaptation has been well characterized in patients with hibernating myocardium. In this study, we demonstrate that suppression of fatty acid metabolism persists for up to 30 hours after exercise-induced myocardial ischemia, long after restoration of blood flow at rest, called metabolic stunning.
Alterations in myocardial fatty acid metabolism were first evaluated noninvasively in humans with the positron-emitting radiotracer C-11 palmitate, which requires an onsite cyclotron and a PET camera.12 Because most nuclear cardiology laboratories are equipped with SPECT cameras, investigators subsequently focused their attention on developing gamma-emitting fatty acid tracers.13 In contrast to palmitate, which is a straight-chain fatty acid that is rapidly catabolized, BMIPP is an iodine-labeled, methyl branched-chain fatty acid that is trapped predominantly in myocardial cells with limited catabolism.5 Uptake of BMIPP from the plasma into myocytes occurs via the CD36 transporter protein present on the sarcolemmal membrane.14 Once in the cell, BMIPP will back-diffuse to the plasma, accumulate in the lipid pool, or undergo limited - and -oxidation. Enzymatic conversion of BMIPP to BMIPP-CoA or triacylglycerol in the myocyte is ATP dependent and is an irreversible step. Such conversion prevents back-diffusion of BMIPP to the plasma and facilitates its cellular retention.6 The prolonged retention of BMIPP in the myocardium, combined with rapid clearance from the blood and diminished uptake in the liver and lung, results in excellent visualization and imaging of the myocardium by SPECT techniques. Thus, BMIPP provides a means of measuring myocardial fatty acid utilization in vivo.
The data from this study suggest that BMIPP can successfully image the suppression of fatty acid metabolism after exercise-induced ischemia at least up to 30 hours after an ischemic episode. There was excellent patient-level agreement (> 90%) between BMIPP and thallium data for the presence or absence of a scintigraphic abnormality on both the early and delayed BMIPP imaging. Furthermore, there was also good correlation between the extent and severity of the stress thallium perfusion abnormality (reflected by the summed segmental scores) and the extent and severity of the BMIPP metabolic abnormality on both the early and delayed BMIPP SPECT images. This is important because the extent and severity of the stress SPECT perfusion abnormality (often evaluated by the SSS) is the most powerful predictor of natural history outcome and is a more complete descriptor of both the extent of coronary artery disease and prognosis than simply classifying the scan as positive or negative.
Our findings suggest that disturbances in cellular fatty acid metabolism may persists for up to 30 hours after demand ischemia, as evidenced by the presence of decreased BMIPP uptake despite restoration of regional blood flow at rest after thallium reinjection (3 to 4 hours after exercise). Long after restoration of blood flow at rest, BMIPP images (injected and acquired at rest) showed regional defects that reflected the extent and severity of exercise-induced thallium defects rather than thallium uptake at redistribution or reinjection. These data support the concept that BMIPP imaging can successfully demonstrate the metabolic imprint of a stress-induced ischemic episode, also known as ischemic memory. These data lay the groundwork for subsequent studies to assess more directly the clinical scenarios in which BMIPP data may provide important diagnostic and prognostic information to assist in clinical decision making. The findings of metabolic stunning after demand ischemia (exercise on a treadmill) extend the previous observation by Kawai and associates,8 who studied patients with "supply-type" ischemia (acute coronary syndrome) in the emergency room. All patients underwent myocardial perfusion SPECT at rest, BMIPP metabolic images within 2 days after the perfusion study, and coronary angiography within 1 to 4 days. Among patients with documented coronary artery stenosis or spasm (on ergonovine provocation), metabolic defects were present in 74% of patients on BMIPP imaging, whereas only 38% of patients showed perfusion defects on rest SPECT imaging (P<0.001). Both BMIPP and perfusion studies were normal in >90% of patients without coronary stenosis or spasm. Thus, BMIPP may have an important role in detecting myocardial ischemia remote from symptoms and in the absence of resting myocardial perfusion abnormalities. Although a myocardial perfusion study at rest can also identify patients with acute chest pain resulting from myocardial ischemia, myocardial perfusion may appear normal at rest in the subacute phase, particularly if the chest pain was >2 to 3 hours earlier.15,16
In this study, the distribution of timing of the rest BMIPP injection fell into a bimodal distribution. Twenty-one patients were studied with BMIPP on the same calendar day as the thallium study, and 11 patients had BMIPP imaging on the subsequent calendar day. When these data were analyzed separately, there did not seem to be any influence of the "same day/next day" injection time on the overall patient-level agreement between the BMIPP and the stress thallium studies. In a separate subgroup analysis, there also did not seem to be any influence of gender on the overall patient-level agreement. Exactly how long the "window" extends after an exercise-induced ischemic insult for BMIPP to identify suppression of fatty acid metabolism cannot be determined from the data in this study. Whether the window extends beyond 30 hours requires future studies with longer time intervals between the different isotopes.
This analysis has several limitations. Patients were eligible for this study on the basis of having a stress-induced ischemic defect on a clinically indicated thallium study. Thus, bias may have been introduced into the image analysis by the readers’ knowledge of the protocol design. However, the BMIPP data were read at a temporally distinct sitting, and the correlation between the summed scores of the BMIPP and the thallium data, as well as the similarity of the summed scores in paired comparisons (Table 3), suggests that bias could not fully explain the findings. The sequence of the imaging studies could not be randomized because, by design, patients were required to have demonstrated a reversible thallium defect on SPECT imaging to be eligible for enrollment. Although this may have engendered a sequence bias, it should be pointed out that the territories of interest, the ischemic areas with reversible thallium defects, already had essentially normal uptake of thallium on redistribution or reinjection imaging by definition; thus, the BMIPP defects are unlikely to have been even in part "caused" by the stress thallium defects. The relatively small number of patients renders preliminary any conclusions with regard to the similarity of same day/next day or early versus delayed data pending larger data sets in which a type 2 error can be more clearly ruled out. Moreover, patients in this study did not consistently undergo cardiac catheterization. Thus, data correlating the metabolic abnormalities and the site and severity of coronary stenoses are not available from this data set but will be an important focus for future studies. The lower agreement between the BMIPP and the thallium studies in the per-vascular-territory analysis relative to patient-level agreement is similar to data reported in other studies comparing 2 tracers for myocardial uptake.17 Because BMIPP assesses metabolic consequence of myocardial ischemia and thallium assesses myocardial perfusion, the differences observed in summed score in specific vascular territories could be secondary to actual differences in the underlying pathophysiology of perfusion versus metabolism rather than simple misregistration from the blinded readings. Although fully quantitative analysis of the thallium and BMIPP images would have been desirable to confirm our findings, quantitative normal databases are not yet available for BMIPP. Hence, quantification with comparison to normal limits was not possible for the BMIPP images.
Hence, rest imaging with [123I]-BMIPP, performed up to 30 hours after a stress-induced ischemic episode, can identify regions of suppressed fatty acid metabolism, also known as ischemic memory. Despite restoration of blood flow at rest (as evidenced by thallium reinjection), prolonged and persistent disturbances in fatty acid metabolism after an ischemic event represent a scintigraphic marker of metabolic stunning. Application of such a metabolic tracer, as opposed to a perfusion tracer, potentially extends the time window for noninvasive imaging an ischemic event beyond the resolution of symptoms.
Acknowledgments
Disclosure
Drs Dilsizian, Bateman, Bergmann, Des Prez, Magram, and Udelson and/or their institution received financial support for this study from Molecular Insight Pharmaceuticals, Inc. Drs Goodbody and Babich are employees of and shareholders in Molecular Insight Pharmaceuticals, Inc.
Footnotes
Guest Editor for this article was Robert O. Bonow, MD.
References
Taegtmeyer H, Woods A. Metabolic modulation as a principle for myocardial protection. Curr Hypertens Rep. 2003; 5: 443–444.
Braunwald E, Kloner RA. The stunned myocardium: prolonged, postischemic ventricular dysfunction. Circulation. 1982; 66: 1146–1149.
Taegtmeyer H. Metabolism: the lost child of cardiology. J Am Coll Cardiol. 2000; 36: 1386–1388.
Apstein CS, Eberli FR. Critical role of energy supply and glycolysis during short-term hibernation. Basic Res Cardiol. 1995; 90: 2–4.
Goodman MM, Kirsch G, Knapp FF Jr. Synthesis and evaluation of radioiodinated terminal p-iodophenyl substituted a- and b-methyl-branched fatty acids. J Med Chem. 1984; 27: 390–397.
Knapp FF Jr, Ambrose KR, Goodman MM. New radioiodinated methyl branched fatty acids for cardiac studies. Eur J Nucl Med. 1986; 12: S39–S44.
Hosokawa R, Nohara R, Fujibayashi Y, Okuda K, Ogino M, Hata T, Fujita M, Tamaki N, Konishi J, Sasayama S. Myocardial kinetics of iodine-123-BMIPP in canine myocardium after regional ischemia and reperfusion: implications for clinical SPECT. J Nucl Med. 1997; 38: 1857–1863.
Kawai Y, Tsukamoto E, Nozaki Y, Morita K, Sakurai M, Tamaki N. Significance of reduced uptake of iodinated fatty acid analogue for the evaluation of patients with acute chest pain. J Am Coll Cardiol. 2001; 38: 1888–1894.
Pohost GM, Zir LM, Moore RH, McKusick KA, Guiney TE, Beller GA. Differentiation of transiently ischemic from infarcted myocardium by serial imaging after a single dose of thallium-201. Circulation. 1977; 55: 294–302.
Dilsizian V, Rocco TP, Freedman NM, Leon MB, Bonow RO. Enhanced detection of ischemic but viable myocardium by the reinjection of thallium after stress-redistribution imaging. N Engl J Med. 1990; 323: 141–146.
Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan TJ, Verani MS. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: AHA Writing Group on Myocardial Segmentation and Registration for Cardiac Imaging. Circulation. 2002; 105: 539–542.
Ter-Pogossian MM, Klein MS, Markham J, Roberts R, Sobel BE. Regional assessment of myocardial metabolic integrity in vivo by positron emission tomography with 11C-labeled palmitate. Circulation. 1980; 61: 242–255.
Feinendegen LE. Measurement of myocardial fatty acid metabolism by iodine-123 heptadecanoic acid kinetics. J Nucl Med. 1987; 28: 924–925.
Tanaka T, Kawamura K. Isolation of myocardial membrane long-chain fatty acid binding protein: homology with a rat membrane protein implicated in the binding or transport of long-chain fatty acids. J Mol Cell Cardiol. 1995; 27: 1613–1622.
Udelson JE, Beshansky JR, Ballin DS, Feldman JA, Griffith JL, Handler J, Heller GV, Hendel RC, Pope JH, Ruthazer R, Spiegler EJ, Woolard RH, Selker HP. Myocardial perfusion imaging for evaluation and triage of patients with suspected acute cardiac ischemia; a randomized controlled trial. JAMA. 2002; 288: 2693–2700.
Fram DB, Azar RR, Ahlberg AW, Gillam LD, Mitchel JF, Kiernan FJ, Hirst JA, Mather JF, Ficaro E, Cyr G, Waters D, Heller GV. Duration of abnormal SPECT myocardial perfusion imaging following resolution of acute ischemia: an angioplasty model. J Am Coll Cardiol. 2003; 41: 452–459.
Acampa W, Cuocolo A, Sullo P, Varrone A, Nicolai E, Pace L, Petretta M, Salvatore M. Direct comparison of technetium 99m-sestamibi and technetium 99m-tetrofosmin cardiac single photon emission computed tomography in patients with coronary artery disease. J Nucl Cardiol. 1998; 5: 265–274.(Vasken Dilsizian, MD; Tim)