Principles of Nuclear Cardiac Imaging and Performing Single-Photon Emission Computed Tomography and Positron Emission Tomography Studies
Cardiac imaging using nuclear techniques plays a critical role in the diagnostic and therapeutic decision-making process in management of patients with coronary artery disease (CAD). Nuclear cardiac imaging involves the administration of a radionuclide agent (radiolabeled isotope) that is that is bound to a radiopharmaceutical. The chemically unstable isotope decays and emits energy in the form of gamma radiation or charged particles that are detected by the scanner to form an image of the heart. Cardiac single-photon emission computed tomography (SPECT) imaging is widely used in clinical practice. Cardiac positron emission tomography (PET) imaging, previously used primarily in research, is gaining wider acceptance as an important clinical diagnostic tool. Inherently, PET has higher spatial, temporal, and contrast resolution than SPECT and great sensitivity for quantifying physiologic processes in the heart. One of the essential requirements for producing high-quality SPECT or PET images is understanding the basis of nuclear cardiac imaging techniques so that imaging data can be acquired with consistency and reported with accuracy. This chapter focuses on the techniques, and Chapter 29 focuses on the clinical applications of nuclear cardiac imaging.
Scanners and Advances in Technology
Four types of scanners are most commonly used in nuclear cardiac imaging:
Conventional SPECT scanners, based on the Anger gamma camera
Newer scanners with solid-state and semiconductor detectors
Conventional PET scanners
Hybrid SPECT or PET scanners combined with computed tomography (CT) scanners
A conventional SPECT system or a scintillation scanner (Anger gamma camera) consists of the following elements: a collimator, with lead septa to localize the source of the emitted gamma rays; a sodium iodide crystal, which scintillates when gamma rays interact with it and produces visible light; photomultiplier tubes, which amplify the emitted light signal and convert energy from visible light into an electric signal; pulse height analyzers and positioning circuitry to localize the signal; and an analog or digital computer designed to determine the location and energy of a photon striking the crystal. Patients are imaged in the supine or prone position by using a step and shoot acquisition mode, in which the scanner starts imaging in the right anterior oblique position, moves to the next position (3 degrees) and takes another projection image, and so on until a 180-degree rotation is completed (left posterior oblique) and tomographic images are calculated. The images obtained are processed using postprocessing filters and reconstructed into three cardiac views (short-axis, vertical long-axis, and horizontal long-axis views). The spatial resolution of the SPECT system is approximately 9 to 12 mm.
Advances in scintillation scanner technology have resulted in the development of solid-state detectors using cesium iodide coupled to photodiodes or novel semiconductor-based detectors using cadmium zinc telluride (CZT). The latest of these newer technologies is the CZT detector, which directly converts gamma radiation to an electronic pulse and thereby eliminates the need for a scintillating crystal and photomultiplier tubes. Some of the detector systems are open and L-shaped or C-shaped with an array of detector elements, as opposed to the circular gantry ( Fig. 5-1 ). Some of the newer scanners also offer imaging with the heart at the focal point of the detector array (cardiofocal imaging), thereby improving image resolution. Because the heart volume is imaged simultaneously, rather than in a step and shoot mode, dynamic images can be acquired in a tomographic mode. With some of these systems, imaging is performed with the patient in an upright (sitting) position in a chair, rather than in the supine position in bed. In addition, high-speed imaging is possible such that acquisition of the stress and rest data takes only 4 and 2 minutes, respectively, and the entire stress-rest SPECT protocol can be completed within 30 minutes.
Myocardial count rates and image quality are higher than in conventional SPECT systems. Studies are currently under way to determine whether this improvement in image quality may translate into more accurate assessment of patients with CAD. A multicenter trial by Maddahi et al validated a half-time acquisition protocol and demonstrated that the rapid gated rest-stress upright system acquired perfusion and functional information comparable to that obtained with the conventional system. Clinical validation of the D-SPECT system (Spectrum Dynamics, Haifa, Israel) was performed by comparing the newer camera with a conventional SPECT camera. The newer high-speed camera produced images with higher counts and excellent linear correlation between the extent of perfusion abnormality at stress and rest compared with the conventional system. Taking into account the more linear relationship between myocardial uptake and blood flow at high flow rates of thallium-201 (Tl-201) than technetium-99m (Tc-99m), a fast sequential stress Tl-201 and rest Tc-99m sestamibi dual isotope protocol was accomplished within 20 minutes with image quality and dosimetry similar to those of a rest-stress Tc-99m protocol. Furthermore, CZT detectors have higher spatial and energy resolution when compared with sodium iodide systems. This superiority in energy resolution allows discrimination of photons of different energies and hence provides the opportunity to perform simultaneous acquisition of Tl-201 and Tc-99m. A comparison of simultaneous dual tracer rest Tl-201 and stress Tc-99m sestamibi on a solid-state D-SPECT scanner with a conventional SPECT camera by Ben-Haim et al showed that fast and high-quality imaging is feasible and diagnostically comparable to conventional SPECT and separate rest Tl-201 and stress Tc-99m acquisition.
PET scanners have higher spatial (4 to 6 mm) and temporal resolution, and they detect paired photons of 511 keV energy produced by annihilation of a positron. A positron is an unstable particle with a positive charge that can travel some distance (positron range) before it interacts with an electron and is annihilated, thus releasing two 511-keV photons at 180-degrees angle from each other. Photons striking diametrically opposing detectors within a short time interval (coincidence interval of approximately 20 nsec) are considered true coincidences and are imaged by the scanner. Higher-energy positrons have a longer positron range, which can limit image resolution ( Table 5-1 ). PET imaging is performed in a two-dimensional (2-D) mode with septa or in a three-dimensional (3-D) mode without septa. Because the 3-D mode offers higher count sensitivity, including more scatter and randoms, it requires more efficient crystals with shorter dead time and high computing capabilities. PET detectors are made up of bismuth germanate (BGO), lutetium oxyorthosilicate (LSO), gadolinium oxyorthosilicate (GSO), or lutetium yttrium orthosilicate (LYSO). BGO has limited energy and timing resolution compared with LSO, LYSO, and GSO crystals. Hence, BGO is more commonly used in 2-D PET systems, which have lower count sensitivity, and LSO, LYSO, and GSO are more commonly used in 3-D PET systems.
|SPECT Radionuclides||Production||Decay||Emission (keV)||Half-life||Role|
|Thallium-201||Cyclotron||Electron capture||68–80 (x-ray); |
167 (10%; gamma ray)
|73 hr||Flow tracer|
|Iodine-123||Cyclotron||Electron capture||159 (gamma ray)||13 hr||Metabolic tracer|
|Technetium-99m||Generator||Internal transition||140 (gamma ray)||6 hr||Flow tracer|
|PET Radionuclides||Production||Positron Energy (keV)||Half-life||Role|
|Oxygen-15||Cyclotron||735||122 sec||Flow tracer|
|Nitrogen-13||Cyclotron||491||9.96 min||Flow tracer|
|Carbon-11||Cyclotron||385||20.3 min||Flow tracer|
|Fluorine-18||Cyclotron||248||110 min||Metabolic tracer|
|Rubidium-82||Generator||1523||1.3 min||Flow tracer|
Most PET scanners have built-in attenuation correction. Attenuation from soft tissues is estimated using a transmission scan based on a radionuclide source (germanium-68, gallium-68, or cesium-137) or a CT image. Typically, the transmission and emission scans are obtained sequentially while the patient is kept in the same position. The CT in PET-CT systems allows for a rapid transmission scan (10 seconds versus 3 to 6 minutes for a radionuclide transmission scan), that permits improved laboratory throughput and reduced patient motion. Time of flight PET technology may improve the noise quality in the images by better delineating the origin of the photons, but it is not yet widely used in cardiac applications.
Software techniques have evolved to improve image quality, shorten image acquisition time and thus improve patient comfort and minimize motion, and reduce radiation dose to the patient. The major advances in software and reconstruction techniques include resolution recovery, noise reduction, wide beam reconstruction, and improved scatter and attenuation correction algorithms. For example, resolution recovery methods are used to improve spatial resolution and reduce noise. By using recovery algorithms from a database of known detectors and collimator characteristics, resolution recovery mathematically corrects for resolution degradation that occurs inherently with parallel-hole collimators and increasing distance from the scintillation crystal.
Commercial manufacturers have developed different versions of resolution recovery algorithms that include not only improved resolution but also reduced image noise. For noise reduction, filters are applied either to the original projection or after reconstruction. In addition, these newer algorithms also include Compton scatter and attenuation correction techniques. Using these newer software techniques, SPECT myocardial perfusion imaging (MPI) may be performed using half the conventional scan time with preserved or even improved image quality. In several studies, half-time processing (Astonish [Philips, San Jose, Calif.]) has been associated with superior image quality and interpretative certainty compared with full-time ordered subset expectation maximization (OSEM). This has the advantage of reducing scan acquisition time and, potentially, patient radiation exposure with half-dose full-time wide beam reconstruction.
Current methods for normal limits-based quantification of myocardial perfusion on SPECT use ungated normal limits and derive perfusion information from summed images. However, image blurring caused by motion may have a significant effect on quantification. Initial evaluation of a novel motion-frozen display and quantification method for gated MPI that potentially eliminates the problem of image blurring by cardiac motion has been performed. In this proposed method, gated perfusion images are analyzed after cardiac motion tracking and 3-D motion correction. These motion-frozen images visually resemble end-diastolic frames, but they contain counts from all cardiac cycles and hence are less noisy and of higher resolution than summed images. Initial results with this method show an improvement in image quality and good diagnostic performance.
The direct conversion of photons into electrical signal without the need for photomultiplier tubes and diodes results in improved efficiency and a smaller footprint of cardiac-specific scanners. The disadvantage of the newer cardiac-specific SPECT systems is that some of them do not allow for noncardiac nuclear medicine scanning, thus limiting their utility in combined practices.
The open detector configuration in some of the newer scanners can be helpful in patients with claustrophobia.
Upright imaging reduces attenuation from diaphragm and interference from subdiaphragmatic activity. However, imaging with the patient in the sitting position may not be feasible, particularly in hospitalized patients with comorbidities.
Attenuation may affect different regions of the heart in the upright position compared with the supine position. For instance, in female patients, breast tissue attenuation affects the anterior or lateral walls in the supine position, and it may affect the apex and the inferior wall with upright imaging.
Because of the high sensitivity of the newer systems, low-dose imaging with longer scan durations (8 to 10 minutes) is feasible. This imaging is still faster compared with conventional scanners with sodium iodide crystals. This approach may reduce the radiation dose to the patient significantly and decrease the risk of patient motion.
With the newer systems, the heart volume is imaged simultaneously, rather than in a step and shoot mode. This technique permits dynamic tomographic imaging, which enables noninvasive quantification of myocardial blood flow using SPECT.
Two photons are required for coincidence detection; hence, if one of them is attenuated or scattered, the scanner will ignore the event as a random event. Therefore, PET imaging is more susceptible to attenuation, and only attenuation-corrected PET images can be used in clinical interpretation.
PET tracers with higher-energy positrons have a longer positron range and a somewhat lower spatial resolution.
Advances in software reconstruction algorithms (resolution recovery, noise reduction, and scatter correction algorithms) allow for improved image quality, faster image acquisition, and lower radiation dose to the patients. The advantage of the software advances is that they can be used with conventional scanners to reduce dose and improve image quality and may be more economical than purchasing a dedicated cardiac scanner for low-dose imaging (e.g., in practices with conventional scanners for general nuclear medicine imaging).
The typical clinically radiotracers used in nuclear cardiac imaging can be characterized broadly as tracers to study myocardial blood flow (perfusion tracers) and tracers to study metabolic processes in the heart (see Table 5-1 ).
Single-Photon Emission Computed Tomography Radiotracers
For SPECT MPI, Tc-99m compounds are frequently used as radiotracers. Tc-99m (the m indicates that it is a metastable nuclear isomer of Tc-99) is produced from beta decay of molybdenum-99 (half-life of 66 hours), through a molybdenum-99–Tc-99m generator. Generators are devices that allow separation of a radionuclide from a long-lived parent. This allows for the continuous production of Tc-99m, the daughter radionuclide, inside the generator at a location remote from a nuclear reactor. Tc-99m emits 140-keV gamma rays and has a half-life of 6 hours. Tc-99m enters the cells passively through its lipophilic properties. Once it enters the myocyte, Tc-99m binds to the mitochondria as a free cationic complex and redistributes minimally. For Tc-99m sestamibi, as opposed to Tc-99m tetrofosmin, hepatobiliary uptake is high, with minimal redistribution.
Tl-201 is similar to potassium (K + ) and enters the myocytes through the sodium-K + adenosine triphosphatase pump. Tl-201 is produced by a cyclotron, emits low-energy mercury x-rays at 69 to 83 keV, and has a half-life of 73 hours. Once Tl-201 enters the cell proportional to myocardial blood flow, constant redistribution across the cell membrane reflects myocardial viability.
Iodine I-123 (I-123) beta-methyl-iodophenyl-pentadecanoic acid (BMIPP) is a tracer used to image fatty acid metabolism with SPECT. I-123 BMIPP emits 159 keV gamma rays and has a half-life of 13 hours. BMIPP is a methyl branched-chain fatty acid that does not readily undergo beta oxidation. This fatty acid tracer is trapped by normal myocardium. Thus, imaging fatty acid metabolism with BMIPP permits identification of regions of myocardial ischemia that have reduced fatty acid uptake, which manifests as a defect on these images.
Positron Emission Tomography Radiotracers
Rubidium-82 (Rb-82) and nitrogen-13 (N-13) ammonia are clinically used PET perfusion tracers, whereas oxygen-15 (O-15) water is a research perfusion tracer. Rb-82 is the most commonly used PET perfusion tracer, with a half-life of 76 seconds, and is produced by a strontium-83 generator. The strontium generator has a half-life of 23 days and is replaced every 4 to 6 weeks. Rb-82 is similar to K + and is taken up by the myocytes in relation to myocardial blood flow with a first-pass extraction fraction of approximately 65%. Because of kinetics similar to those of K + and Tl-201, delayed retention of Rb-82 may provide information about myocardial viability.
N-13 ammonia is produced by a cyclotron and has a physical half-life of 9.96 minutes. Therefore, an on-site cyclotron is necessary for its use. N-13 ammonia diffuses freely across the cell membranes and becomes incorporated into glutamine in the myocardium by the enzyme glutamine synthetase. The uptake of N-13 ammonia is related to the capillary blood flow and to the myocardial extraction characteristics. N-13 ammonia has a high extraction fraction of approximately 85%.
O-15 water is not approved by the U.S. Food and Drug Administration (FDA) and is primarily a research radiotracer. It is a freely diffusible radiotracer produced by a cyclotron and has a short half-life of 2.1 minutes. The uptake of O-15 water is linearly related to myocardial blood flow even at high flow rates.
Fluorine-18 (F-18) fluorodeoxyglucose (FDG) is a glucose analogue used to study myocardial glucose metabolism and to assess myocardial viability. F-18 FDG is a cyclotron-produced radiotracer with a half-life of 109 minutes. Uptake of F-18 FDG is through the glut-4 receptors. Once F-18 FDG enters the myocyte, it is converted by glucose-6-phosphate into FDG-6-phosphate, which is not metabolized any further and remains trapped in the myocyte. F-18 FDG uptake is exquisitely dependent on myocardial substrate use. The uptake of F-18 FDG is high when plasma glucose and insulin levels are high, and uptake is low when plasma free fatty acid levels are high.
Another F-18 PET perfusion tracer (F-18 flurpiridaz) is currently under development. This tracer is produced by a cyclotron and has a half-life of 109 minutes. It targets the mitochondria and shows rapid and high myocardial uptake with better myocardial extraction fraction than technetium compounds.
Technetium-99m versus Thallium-201
The higher energy of Tc-99m gamma rays and the shorter half-life translate into higher image resolution and lower radiation dose to patients when compared with Tl-201. With Tl-201, the low-energy mercury x-rays make them more susceptible to scattering, attenuation, and limited gated images. In addition, the longer physical half-life of Tl-201 (half-life, 73-hours) limits the ability to administer higher doses and hence results in lower image quality. However, Tl-201 has a high first-pass extraction (approximately 85%) compared with Tc-99m agents (approximately 65%). Therefore, Tl-201 images can provide a better measure of hyperemic myocardial blood flow and may be more sensitive than Tc-99m images for the detection of milder degrees of stenosis. This property has been used, and newer protocols using Tl-201 for stress and Tc-99m agents for rest have been developed for some of the newer higher-sensitivity cardiac SPECT scanners. Although the first-pass extraction of Tc-99m agents is lower than that of Tl-201, the shorter half-life of Tc-99m allows for the use of a higher radiotracer dose that emits higher-energy gamma rays (less prone to attenuation) and results in better image quality compared with Tl-201.
Because of the high hepatobiliary uptake of Tc-99m sestamibi, one must wait 45 to 60 minutes for optimal myocardial images and good clearance of hepatic uptake. The more rapid liver clearance of tetrofosmin compared with sestamibi significantly improves the ratios of cardiac to digestive activity, especially after exercise or at rest, and images can be obtained sooner after injection (10 to 15 minutes). Moreover, although tetrofosmin is reconstituted with Tc-99m, it can be allowed to stand at room temperature, unlike sestamibi, which requires boiling. This property, in addition to more rapid liver clearance and earlier image acquisition after tetrofosmin injection, is useful for performing studies in patients with acute chest pain, with higher patient throughput and image quality comparable to that with sestamibi.
Positron Emission Tomography Radiotracers
The main advantage of Rb-82 is the ability to perform PET MPI without an on-site cyclotron. However, the necessary strontium-83 generator is very expensive, and this factor should be considered, especially if patient volumes are limited. The ultrashort half-life of Rb-82 of 76 seconds requires that the generator be directly connected to the intravenous line of the patient so that dose can be delivered directly to the patient during PET imaging. Repeat imaging is uncomplicated, and preintervention and postintervention studies are simple to perform without a time lag. The short physical half-life of N-13 ammonia of just less than 10 minutes makes an on-site cyclotron necessary for its use. N-13 ammonia produces better-quality perfusion images with high signal-to-noise ratio, low background, and low liver activity. At high flow rates (>3 mL/g/minute), N-13 ammonia may result in some underestimation of myocardial blood flow. Because O-15 water is freely diffusible and metabolically inert, it is an ideal perfusion agent and does not underestimate flow even at high flow rates. However, its freely diffusible state can lead to poor image contrast resolution and requires correction of blood pool activity or myocardial activity with a separate scan to improve signal-to-noise ratio. F-18 flurpiridaz, the investigational PET perfusion tracer, would be suitable for exercise stress imaging because of its half-life of 109 minutes. Also because of its long half-life, this radiotracer can be transported to remote (but still accessible) locations and does not need an on-site cyclotron or a generator. The lower-energy positron to F-18 compared with Rb-82 and N-13 ammonia results in sharper images related to shorter positron range. For viability imaging, myocardial F-18 FDG uptake is imaged ( Fig. 5-2 , I ). Myocardial uptake of F-18 FDG highly depends on the substrate milieu. Therefore, good patient preparation is paramount to obtaining interpretable diagnostic images ( Tables 5-2 and 5-3 ; see Fig. 5-3 ) and may be time-consuming and difficult to implement without protocols in place. For more details, readers are referred to the excellent review on this topic by Machac et al.
|Fast patient||6-12 hr|
|Load oral glucose||If fasting BG <~250 mg/dL (13.9 mmol/L), give 25-100 g glucose orally; then monitor BG based on Table 5-3 |
If fasting BG >~250 mg/dL, notify physician
|Inject FDG||After monitoring BG (see Table 5-3 ), when BG is ~100-140 mg/dL (5.55-7.77 mmol/L), inject FDG intravenously|
|Begin PET imaging||Start PET imaging ~90 min after FDG injection|
|Blood Glucose 45 to 60 Min After Administration||Possible Restorative Measure (Intravenous)|
|130-140 mg/dL (7.22-7.78 mmol/L)||1 unit regular insulin|
|140-160 mg/dL (7.78-8.89 mmol/L)||2 units regular insulin|
|160-180 mg/dL (8.89-10 mmol/L)||3 units regular insulin|
|180-200 mg/dL (10-11.11 mmol/L)||5 units regular insulin|
|>200 mg/dL (>11.11 mmol/L)||Notify physician|
In addition, F-18 FDG PET can be used to image cardiac sarcoid (see Fig. 5-2 , J ). F-18 FDG PET is emerging as a promising diagnostic modality in identifying cardiac involvement of sarcoidosis. In these studies, localized uptake of F-18 FDG is a hallmark of active inflammatory change and potentially may also prove to be a useful indicator of response to steroid therapy (see Chapter 29 for more information).
The higher energy of PET radiotracers compared with SPECT radiotracers can result in greater penetration through the subject and can thereby lower the radiation dose to the patient. However, the radiation burden to the staff may be higher because more gamma rays are emitted from the patient and may reach the staff. The shorter the half-life of the radiotracer, the faster the protocols and lower the radiation dose to the patient will be. Extremely short half-life radiotracers such as O-15–labeled water (and sometimes Rb-82) are challenging to use for exercise stress testing because of decay of the radiotracer before the patient can be transferred from the treadmill to the scanner. Therefore, pharmacologic stress testing is preferable with short half-life radiotracers, and exercise or pharmacologic stress testing can be performed with longer half-life radiotracers. The short half-life radiotracers have a significant advantage in that rest and stress imaging can be performed using equal doses of radiotracer injection. In addition, if the dose has inadvertently infiltrated subcutaneously, rather than being injected intravenously, another scan can be obtained immediately after reinjection of another dose (rest imaging with repeat stress imaging is determined based on the half-life of the stress agent). Finally, for short half-life radiotracers produced by a cyclotron, the clinical schedule must be closely coupled with the cyclotron production schedule, thus making it less suitable for high-volume imaging. High-volume imaging using PET MPI is best performed with generator-produced radiotracers such as Rb-82 or with unit dose tracers such as the novel F-18 flurpiridaz (when it is approved by the FDA).
How to Decide Which Radioisotope to Use
This decision is based on a balance of several factors:
The type of stress protocol
The clinical question: ischemia assessment, viability assessment, assessment of ischemic memory
The availability for routine clinical use
The physical property of the radioisotope
The decision to use SPECT or PET can be based on a simple algorithm ( Fig. 5-3 ). Exercise is the preferred modality for stress imaging. Hence, any patient able to exercise is scheduled for an exercise MPI study (SPECT study for normal weight and N-13 ammonia PET when available, for obese individuals). When resources are not constrained and both SPECT and PET imaging techniques are available, PET is preferred for all pharmacologic studies. For SPECT imaging, Tc-99m agents are the most widely used; less than 10% of laboratories use Tl-201. Tc-99m sestamibi and Tc-99m tetrofosmin appear to be equally efficient, except for the differences mentioned earlier. Tl-201 has superior extraction characteristics and can be used to improve the test sensitivity for detection of ischemia, particularly with the high-sensitivity scanners and improved image quality. Moreover, Tl-201 has less hepatic uptake and can occasionally be used in patients with excessive subdiaphragmatic activity from Tc-99m tracers that cannot be resolved with delayed imaging, feeding, or prone imaging. When the clinical question involves assessment of myocardial viability when PET is not available, Tl-201 perfusion imaging with delayed imaging can be performed.
Imaging of ischemic memory is being explored with I-123 BMIPP. This technique enables identification of ischemic episodes approximately 24 hours before rest imaging with I-123 BMIPP and may play a role in imaging of patients with acute chest pain and a nondiagnostic electrocardiogram (ECG).
When to Consider Positron Emission Tomography Myocardial Perfusion Imaging
PET offers several advantages compared with SPECT ( Box 5-1 ). PET yields higher-resolution images than SPECT. The unique opportunity to measure peak stress left ventricular ejection fraction (LVEF) during PET is also an advantage over current SPECT protocols, which acquire poststress LVEF. LVEF reserve (peak stress minus rest LVEF) has been shown to be an important diagnostic tool for identifying multivessel disease, and a low LVEF reserve appears to have adverse prognostic implications. Images are acquired simultaneously in multiple projections that allow quantification of absolute blood flow at rest, and peak stress is a distinct advantage of PET. Because conventional SPECT and PET techniques rely on the concept of flow heterogeneity and compare relative uptake of radiotracer within the myocardium, they may potentially underdiagnose multivessel disease.
Higher spatial and temporal resolution
Accurate and depth-independent attenuation correction
Lower radiation dose
Faster protocols and laboratory throughput
Peak stress left ventricular ejection fraction
Quantitative myocardial blood flow assessment
Assessment of myocardial viability
Lack of wide availability
Difficulty of exercise stress
Need for cyclotron or rubidium generator
If all segments of the myocardium are ischemic to similar degrees, a situation known as balanced ischemia can result. In PET, perfusion tracers such as Rb-82, N-13 ammonia, and oxygen-15 water allow for absolute quantification of myocardial blood flow in which resting and peak stress myocardial blood flow is measured. This approach may be useful for detecting the presence of multivessel CAD. In addition, absolute quantification of PET tracer uptake permits the detection of abnormal flow reserve in preclinical states before the development of significant epicardial coronary stenoses in which abnormal microvascular or endothelial function causes vascular dysfunction. Quantitative blood flow estimation can allow the study of progression or regression of atherosclerosis and appears to add incremental prognostic value to relative perfusion imaging. The indications and clinical applications for PET MPI and quantitative PET MPI are discussed in Chapter 29 .
How to Choose Among Perfusion Imaging Protocols
The choice of imaging protocol depends on the following:
The clinical question
Considerations for the appropriate radiation dose for each patient
The availability of radiopharmaceutical agents
The logistics for patients and laboratory personnel
The throughput of the laboratory
Single-Photon Emission Computed Tomography Myocardial Perfusion Imaging
Conventional protocols for SPECT acquisition studies using Anger camera technology and filtered back projection or OSEM (iterative) reconstruction are shown in Figure 5-2 , A to E. The advantages and disadvantages of each of the protocols are listed in Table 5-4 . The reader is also referred to the American Society of Nuclear Cardiology (ASNC) imaging guidelines for nuclear cardiology procedures, available in the article by Machac et al, for detailed protocols and image acquisition parameters.
|Single-day rest-stress||1-day protocol||Suboptimal contrast between stress defect and rest|
|Enhanced stress image quality||Long procedure time (~3-4 hr)|
|Well-validated quantitative programs|
|2-day stress-rest||Ideal protocol for image quality||Long procedure time (2 days)|
|Improved defect and reversibility detection (no shine through of rest radiotracer activity)|
|If stress test result normal, rest test can be safely avoided, with resultant time and radiation savings|
|Single-day stress-rest with or without reinjection-rest-redistribution Tl-201||Potential to improve the assessment of myocardial ischemia and viability of patients with apparent irreversible defects||Rapid peak of myocardial concentration within 5 min of Tl-201 injection (rapid redistribution means that imaging must be done quickly and scanner must be readily available)|
|Higher radiation dose to the patient|
|Single-Day Dual Isotope Thallium-201 and Technetium-99m|
|Tl-201 at rest and Tc-99m after stress||Reduced waiting time between rest and stress image acquisition||Higher radiation dose to the patient|
|Rest Tl-201 study can be reimaged after a 4-hr or 24-hr delay for viability assessment||Differences in defect resolution and photon scatter may exist when comparing two different radiotracers|
|Can be used during Tc-99m shortages||Images may demonstrate apparent transient ischemic dilatation from inherent resolution differences between Tl-201 and Tc-99m|
The most commonly used SPECT MPI protocol is a single-day, low-dose rest study followed by a high-dose stress Tc-99m study, acquired as a gated study and reconstructed using filtered back projection or OSEM reconstruction. The images are acquired using standard parameters, as listed in the ASNC guidelines ( Table 5-5 ). Protocols for newer scanners are not yet well defined, but laboratories with newer CZT scanners may elect to use lower-dose imaging (see Fig. 5-2 , F ), rather than faster imaging with the usual dose of radiotracer (see Fig. 5-2 , G ).
|Parameter||Rest Study||Stress Study|
|Dose||8-12 mCi||24-36 mCi|
|Delay time to imaging |
From injection to imaging
From rest to stress imaging
30 min-4 hr
|Energy window||15%-20% symmetric||15%-20% symmetric|
|Orbit||180 degrees (45 degrees RAO to 45 degrees LPO)||180 degrees (45 degrees RAO to 45 degrees LPO)|
|Orbit type||Circular or |
|Circular or |
|Pixel size||6.4 ± 0.4 mm||6.4 ± 0.4 mm|
|Acquisition type||Step and shoot or |
|Step and shoot or |
|Number of projections||60-64||60-64|
|Matrix||64 × 64||64 × 64|
|Time/projection||25 sec||20 sec|
|ECG gating |
|8 or 16 |
|8 or 16 |
For viability assessment, at sites using SPECT technology, a rest-redistribution Tl-201 study and a dual isotope (Tc-99m stress, Tl-201 rest) study acquired as gated images are commonly used. When PET is available, perfusion and viability imaging can be performed using FDG PET. Typical protocols for assessment of myocardial viability include myocardial perfusion assessment using gated SPECT or PET (preferred) and imaging myocardial glucose metabolism using F-18 FDG. Assessment of myocardial viability is combined with assessment of ischemia, unless stress testing is contraindicated (critical CAD or hemodynamically unstable patient) or ischemia evaluation is deemed not necessary and a decision is made to proceed with revascularization of critical coronary stenoses pending viability assessment.
Gated Myocardial Perfusion Imaging
ECG-gated SPECT is an accurate and reproducible technique used to measure LVEF, volumes, and regional wall motion. At each step of a conventional Anger scanner, several images (8 to 16 images) are acquired in relation to the surface ECG ( Fig. 5-4 ). The newer SPECT scanners and PET scanner allow for list mode image acquisition in which image frames are acquired in relation to the ECG, as well as a time signal (see Fig. 5-4 ). The list mode files can be unlisted into a static or a gated image file. Gating with a higher frame rate improves temporal resolution for assessment of regional wall motion, and the overall LVEF may be comparable or slightly higher. The gated images are reconstructed using parameters similar to those used for the static images, and they are reviewed using commercial software packages for computation of LVEF and volumes. In patients who have irregular heart rates, gated information may be inaccurate, and acceptance windows can be specified to reject cycle lengths that do not fall within the predetermined R-R interval range.