Chapter 9: SPECT Myocardial Perfusion Imaging Protocols

Single photon-emission computerized tomographic (SPECT) myocardial perfusion imaging (MPI) remains the dominant noninvasive functional imaging perfusion method for the diagnosis as well as prognosis of epicardial coronary artery disease (CAD). The advent and advances of other methods used for similar purposes (cardiac positron emission tomography [PET], stress echocardiography, coronary computerized tomography [CTA], and magnetic resonance imaging [MRI]) have all contributed to recent re-examination of traditional MPI protocols to optimize its use.¹ This task was facilitated by introduction of high-efficiency cadmium zinc telluride (CZT) nuclear cameras and innovative software. With changes in society and concern of radiation exposure, emphasis has shifted from "one size fits all" to patient-centered imaging with individualized approach to each patient's unique constellation of reasons and urgency of testing, comorbidities, age, body habitus, physical ability, and results of previous tests and procedures.² In-depth knowledge of the advantages and disadvantages of available radionuclide tracers and stressors by those who perform stress testing and imaging is paramount. Patient participation in decision making becomes desirable, as well.

With an acknowledgment of possibly harmful effects of low radiation doses used at times frequently over an extended period of time due to the chronic nature of CAD, attention has shifted to dose reduction³ and potentially mandatory tracking of all received radiation doses.⁴

In view of competing noninvasive imaging modalities, cost effectiveness has also been addressed. The length of "traditional" MPI is almost ½ a day, which poorly compares to on average 1 hour to completion and diagnosis using CTA, stress echo, or PET. Many of the newer imaging protocols therefore address the need for increased throughput and improved productivity of a Nuclear Cardiology Laboratory. This chapter will describe protocols for SPECT MPI for the two primary imaging agents, thallium-201 and technetium-based products.

The primary indication for a SPECT MPI study is the assessment of the relative distribution of coronary flow in patients with suspected or known CAD. Since this distribution of coronary flow both at rest and stress is equal in all segments of the left ventricle, the presence of perfusion defects suggests intraluminal coronary obstruction, and if worse at stress than rest, ischemia. An increase in coronary flow is needed for the detection of significant coronary artery stenosis (>50% of luminal narrowing) since rest flow distribution is even unless prior infarction is part of the history. Coronary flow can be increased most physiologically with physical effort (treadmill exercise), or in patients who are unable to exercise adequately, using coronary vasodilators (adenosine, dipyridamole, and regadenoson) or dobutamine (for a more complete description, see Chapter 8).

Evaluation of left ventricular size and function became possible with the development of gating algorithms used in conjunction with MPI. The combination of perfusion and function data improved both the diagnostic and the prognostic value of SPECT studies. ECG-gated SPECT imaging is a powerful tool for evaluating fixed attenuation artifact. Ventricular function assessment and interpretation is described in Chapter 11.

The availability of more than one perfusion tracer and different modes of stress provides a multitude of imaging protocols. Ideally, the imaging protocol should be tailored for the individual patient, taking into account the patient's age, gender, size, physical ability, various comorbidities, and particularly the clinical question to be answered. Laboratory logistics, test urgency, and cost effectiveness also dictate imaging sequences. Knowledge of tracer and stressor characteristics is critical for the right choice and best results.

Thallium-201 Tracer Protocols

Tl-201 (clinically used since the 1970s) is a monovalent cation, analogous to potassium, with a physical half-life of 73.1 hours. Decay is by electron capture to Hg-201, with principal emission of 68- to 80-keV x-rays. First-pass extraction is high (approximately 85%). The tracer is actively transported to the myocyte as well as to other organs and washed out (redistributed) beginning 10 to 15 minutes after an IV injection. The relationship between flow and uptake is almost linear at physiologic flows and even during vasodilator-induced hyperemia. Recommended Tl-201 dose, injected at peak stress is 2.5 to 3.5 mCi. Lower Tl-201 doses are recommended with use of high-efficiency cameras (as low as 1 mCi). The standard effective radiation dose for Tl-201 is approximately 4.4 mSv per 1 mCi of Tl-201, or 10.9 to 15.3 mSv per patient. For a more detailed description, refer to Chapter 3.

Tl-201 Imaging Protocols

Tl-201 is injected approximately 1 minute prior to termination of the exercise or at peak effect of a coronary vasodilator. SPECT imaging must begin within 10 to 15 minutes (Fig. 9-1). The delay is needed for poststress monitoring, patient positioning in the camera, and avoidance of "upward creep," which is caused by cranial motion of the diaphragm due to hyperventilation during stress. Delay in imaging beyond this time may lead to missed ischemia as redistribution begins within 10 to 15 minutes. Tl-201 is by design "stress-first" imaging. Stress images should be reviewed, and, if there are no perfusion defects, rest imaging is unnecessary. The purpose of rest imaging is to ascertain reversibility (redistribution) of perfusion defects seen on stress images. The mechanism of Tl-201 "redistribution" is for the most part due to differential washout from the myocardium. In segments with high initial tracer uptake, that is, in segments supplied by nonobstructed coronary flow and with functional myocytes, washout rate is high. At the time of initial equilibrium (shortly after peak stress tracer injection), intravascular tracer concentration is negligible. In segments supplied by an obstructed epicardial coronary artery, coronary flow is limited and initial tracer uptake is decreased; the intracellular:intravascular tracer gradient is lower and washout rate is slower. After 3 to 4 hours of injection, a second SPECT imaging is obtained (rest scan). If there are no abnormalities, the study is considered normal. If a stress defect appears less prominent or absent on rest, the defect is considered due to ischemia and is consistent with viable but hypoperfused myocardium. If no stress defect reversibility is seen on rest images, the defect is assumed to be scar, possibly due to prior myocardial infarction. If there is concern for missed ischemia, additional imaging can be done either with additional delay (up to 24 hours) or after 1-mCi Tl-201 reinjection. (See strategies as outlined by ASNC guidelines in Fig. 9-1.)

Apart from using Tl-201 for diagnostic or prognostic purposes, Tl-201 is also indicated for detection of myocardial "viability" (Fig. 9-2). The testing should be limited to patients with resting LVEF ≤35% and who are also candidates for myocardial revascularization (surgical or percutaneous). In this case, a rest injection is performed and imaging at 15 to 20 minutes and 3 to 4 hours later. The presence of viability is judged by 50% or greater uptake in the region or vascular territory under consideration (left anterior descending, right or circumflex arteries). If viability is not demonstrated on 3- to 4-hour imaging, a 24-hour image can be obtained (Fig. 9-2).

Advantages of Tl-201 Imaging

1. Extensive experience, evidence-based data.

2. No need for delay after stress injection, possible "stress-only" protocol.

3. Good flow-uptake linearity, high first-pass extraction.

4. Absence of liver uptake.

5. Assessment of myocardial viability.

Disadvantages of Tl-201 Imaging

1. Low photon energy (68–80 keV) leads to more prominent soft tissue attenuation artifacts. Also, gated images may be count poor, requiring at times extended imaging time.

2. The long half-life of Tl-201 limits the dosage that can be safely administered to a patient (2.5–4 mCi). This affects the image quality and interpretive certainty due to relatively low counts, making accurate diagnosis more difficult.

3. High radiation dose (>15 mSv) compared to other available tracers.

Technetium-Based SPECT Protocols

Two Tc-99m–based perfusion tracers, Tc-99m-sestamibi (Cardiolite) and Tc-99m-tetrofosmin (Myoview™), are available for clinical use in the United States. Tc-99m agents are lipid soluble, cationic, with a physical half-life of 6 hours, and decay with the emission of 140-keV photons. The first-pass extraction of sestamibi and tetrofosmin is lower (45–50%) compared to Tl-201; therefore, stress needs to continue for at least 1 minute (preferably 2 minutes) after peak stress injection. The tracers are excreted via the hepatobiliary system to the gastrointestinal tract. Tetrofosmin's GI clearance is slightly faster than that of sestamibi. Considerable gut and liver uptake is frequently seen with exercise and pharmacologic stress, although liver activity is more prominent with pharmacologic stress.

Myocardial uptake of Tc-99m-sestamibi and Tc-99m-tetrofosmin is almost linear with increased coronary flow at moderate increases in physiologic coronary flows (such as moderate exercise), but not so with high flows (200–300% increase over rest) at the peak of the vasodilator effect (roll-off phenomenon) or at high exercise level, which could affect sensitivity, and ability to identify multivessel ischemia. The tracers enter the myocytes by passive distribution and are retained in the mitochondria, with negligible tracer redistribution. Therefore, two injections are required: a rest and a separate stress dose. Imaging is usually begun 30 to 45 minutes after the rest injection while poststress injection imaging time is shorter after exercise (15–30 minutes) and longer after coronary vasodilation (45–60 minutes). The reason for delay is subdiaphragmatic tracer uptake, which interferes with image interpretation and lessens over time, and generally is considered worse during pharmacologic stress. Different recommendations have been proposed to reduce these effects, but none have been uniformly successful or accepted. However, it is commonly understood that some level of exercise in conjunction with pharmacologic stress reduces liver uptake.

ECG gating for ventricular function should be done with all Tc-99m studies, when possible. Gating should be performed during both the rest and poststress acquisitions. If using a 1-day rest/stress protocol, the quality of the high-dose images is superior due to higher count rate than at rest, but still can be used for comparative purposes, especially wall motion abnormalities. Since imaging is done early after stress (15–45 minutes), the calculated left ventricular ejection fraction (LVEF) and ventricular volumes do not represent true resting values, but rather "early poststress EF/LV volumes" and should be reported as such. In patients with extensive ischemia, these obtained values can reflect transient stunning; underestimation of the true rest LVEF is possible and has consequences for therapeutic decisions (e.g., AICD implantation and ACE inhibitor initiation). For a more detailed description, refer to Chapter 11.

Knowledge of Tc-99m–based agent's characteristics provides considerable flexibility of imaging protocols. The test can be completed in 1 or 2 days and the study protocol can be tailored to individual patient needs with stress–rest, rest–stress, or stress-only sequences.

Given the 6-hour physical half-life of Tc-99m, a 24-hour separation between the two injections is optimal to minimize background radioactivity for the second set of images. In clinical practice, however, having patients undergo imaging on 2 separate days may be inconvenient or impractical. In addition, it is more efficacious to have all the information from both studies available on the same day. However, both 2- and 1-day protocols for rest and stress imaging have been developed (Fig. 9-3).

For 2-day protocols, the Tc-99m–labeled agent is injected at peak stress (20–30 mCi) and imaged within 15 to 45 minutes. This is followed at least 24 hours (approximately 4 half-lives) later by a second injection at rest and a second set of images. A 2-day protocol is often needed in patients with a high BMI (>30), since low-dose rest images, as used in 1-day protocols, are more likely nondiagnostic due to poor image quality. Alternatively, the order of the injections for a 2-day test can be reversed, with the stress study being performed first and rest injection at least 24 hours later, using high tracer dose (20–30 mCi) for both injections.

The more common protocol is that in which both the rest and stress are performed on the same day.⁵ It is estimated that this protocol is used 90% of the time. In general, the rest is performed first, followed by the stress portion. In order to offset the residual activity of the first injection, it has been determined that the rest: stress tracer dose should be 1:3. The suggested¹ rest dose is 8 to 12 mCi and stress dose is 24 to 36 mCi. A rest–stress sequence is preferable when using a 1-day protocol because the rest image performed initially represents a "true" rest study. This is not necessarily the case with the stress–rest sequence due to cross-talk from the stress study present in the rest images. If the stress–rest sequence is used, a longer time interval or a higher rest dose is needed. In addition, the first study is of lesser quality due to the lower dose and using standard equipment in most laboratories it is preferable that this be the rest study. For newer cameras such as with CZT detectors or OSEM processing, the image quality of a low-dose stress may be acceptable and the sequence preferable if considering a "stress-only" or "stress-first" protocol. As many new protocols are being suggested it is advised to follow the current literature.

Rest–stress and stress–rest protocols for Tc-99m-sestamibi SPECT imaging have been compared and demonstrated that the former sequence provides better image contrast and an increased ability to detect reversibility of perfusion defects. The 1-day stress–rest protocol offers advantages that must be taken into consideration: it allows for the elimination of the rest study if the stress study is found to be normal similar to a planned 2-day study (see section on stress-only imaging).

The estimated radiation dose to a 70-kg adult using ICRP 60 calculation is 15.7 mSv for a 2-day high-dose study, 11.3 mSv for a 1-day rest–stress study, and 7.9 mSv for a stress-only study.⁶

Tc-99m–labeled agents, compared to Tl-201, allow more flexibility and more "tailoring" of the test sequences to patient needs. Other advantages include a higher photon flux (higher injected dose) and higher photon energy (140 keV), which result in better image quality and clarity. Images are of higher quality, attenuation artifacts are slightly less prominent, and gated images are also of higher quality than thallium-201. The disadvantages in relation to Tl-201 include considerably more gut and liver uptake and less linearity of blood flow, potentially reducing diagnostic accuracy.

Advantages of Tc-99m Agents

1. Flexibility of imaging protocols.

2. Higher image quality, reduced soft tissue attenuation compared to Tl-201.

3. Well-documented accuracy and risk stratification data.

4. Lower radiation exposure compared to Tl-201.

Disadvantages of Tc-99m Agents

1. Nonlinearity of flow and uptake may affect detection of moderate coronary lesions and multivessel ischemia.

2. Considerable gut and liver uptake resulting in reduced specificity.

3. Modest diagnostic accuracy.

Dual-Isotope Imaging

A dual-radionuclide imaging protocol was introduced, validated, and popularized in the early 1990s.⁷ This protocol consists of an injection of 3.0 to 3.5 mCi of Tl-201 at rest, imaging and of 25 to 30 mCi of Tc-99m-sestamibi at peak stress (Fig. 9-4).⁷ SPECT imaging begins 10 to 15 minutes after the initial injection of Tl-201 at rest. Immediately following Tl-201 imaging, the patient performs the stress part of the test (exercise or pharmacologic). At peak stress, a dose of 25 to 30 mCi of Tc-99m-sestamibi or Tc-99m-tetrofosmin is injected. Imaging starts 15 to 45 minutes later depending upon whether exercise or pharmacologic stress (generally 15 minutes after exercise stress, 45 minutes after pharmacologic stress). The separate acquisition dual-radionuclide imaging procedure can be completed in approximately 2 to 3 hours.

The dual-radionuclide imaging protocol is shorter than a 1-day Tc-99m-sestamibi/tetrofosmin protocol. Although imaging time does not change, time is saved since Tl-201 rest images are acquired shortly after the tracer injection as there is minimal liver or gut uptake. Sequential imaging is performed without contribution of the rest counts to the stress image (as is the case of low-dose–high-dose Tc-99m imaging). However, this protocol also presents some disadvantages. The physical characteristics of the two radionuclides involved are quite different, resulting in a different count density. This may affect the evaluation of the degree of defect reversibility, especially in patients with prior myocardial infarction and an abnormal Tl-201 rest study. Furthermore, the quality of the rest Tl-201 studies is sometimes suboptimal. Visual interpretation of transient ischemic cavity dilation (TID) is difficult to assess due to inherent differences in thallium and technetium images.

A major concern and disadvantage is the radiation exposure of this protocol. The estimated radiation dose of a dual-isotope study could be very high: >20 mSv. Thus, as with the Tl-201 only study, routine use of this protocol is not recommended by national societies and guidelines.^1,8 It may be justified in the evaluation of patients with severe congestive heart failure, when viability detection is of importance for therapeutic decisions (revascularization, medical therapy, or cardiac transplantation).

Stress-Only or Stress-First Imaging to Reduce Patient Radiation

The most common stress protocol is that of rest imaging first, followed by stress MPI, as previously described. Despite this, it has been estimated that only 8% to 10% of nuclear cardiology studies demonstrate ischemia.⁹ Thus, in a majority of studies, a rest study is of no clinical value, particularly when the stress study is completely normal. It has been proposed for several years that performing the stress first, and rest only when necessary would be a reasonable means of reducing patient radiation. The "stress-only" or "stress-first" protocols have been recommended by ASNC for several years.^10,11 Using this approach, the number of rest studies can be substantially reduced, thus shortening the protocol for the patient and reducing radiation exposure. Since decisions are made with only one study, the stress, the use of attenuation correction (AC) or prone imaging is important in reducing the number of unnecessary rest studies.¹⁰

An important consideration of stress-only imaging is the administered stress dose. In older camera systems, a low stress dose of 8 to 15 mCi of Tc-99m may yield an image that is difficult to interpret due to poor image quality and attenuation artifact. Thus, with these systems, this protocol may require a high-dose stress study (25–30 mCi), but will still avoid the rest radiation exposure in most circumstances. The radiation reduction is even greater when low-dose stress only (8–15 mCi) protocols are utilized with appropriate and newer camera systems (see below). Significant reduction (35–85%) in the need for rest images is seen when AC is applied to image processing.¹⁰ In a study by Mathur et al. using stress only with AC in a chest pain center environment, only 8% required a rest study, and at least ½ of these patients had evidence of scar.¹¹

The optimal candidates for stress-only imaging are patients with no known history of CAD and secondarily those with CAD but no history of myocardial infarction or bypass surgery. In addition, patient with CAD who had a recent stress MPI study for comparison could also be considered candidates. Less optimal candidates are those with prior MI, CABG, or ventricular dysfunction. Ideally an interpreting physician should be available to decide on whether a rest study is indicated in order to complete the study in an efficacious manner.

Use of Newer Technologies

There have been many recent advancements in camera technology and software which impact protocols and radiation exposure.¹²

Solid state cameras employ the use of different detector technology from the more common Anger cameras. Solid-state cameras use CZT or cesium-iodide (CsI) semiconductor detectors. CZT cameras are also called high-efficiency cameras,^1,12 and hold the potential to dramatically decrease tracer doses and reduce imaging times. Laboratory efficiency is improved particularly for high-volume laboratories, thus potentially offsetting higher purchase prices of these new cameras. The downside of a CZT camera includes lack of efficient motion detection and AC, presence of new imaging artifacts unique to these systems, limited utility for morbidly obese and immobile patients and lack of versatility compared to traditional NaI cameras (only cardiac imaging is feasible). A more complete discussion of this technology is in Chapter 4.

Novel processing software is offered for most newly purchased traditional SPECT cameras; most <10 years old NaI Anger cameras can be retrofitted. Proprietary variation of resolution recovery/noise reduction is implemented into iterative reconstruction software. Software is offered by Philips (Astonish^TM), Siemens (Flash 3D), GE Healthcare (Evolution), Digirad (Nspeed); wide beam reconstruction (WBR) is offered by a third-party vendor (Ultra SPECT), as well as ImagenSPECT from Cardiovascular Imaging Technologies (CVIT). Each of the algorithms has been clinically validated and will allow ½ dose and/or ½ time acquisition.

This chapter has described the current protocols for SPECT MPI with Tl-201 and Tc-99m. Careful attention is given to the time required for each protocol as well as radiation exposure. Stress-only or stress-first imaging should be considered for selected patients as a means of reducing radiation exposure and improving laboratory efficiency. Newer camera technologies and software innovations offer the potential of radiation dose reduction as well as more time-efficient protocols.