Applied Radiology

In keeping with the prevalence and social impact of cardiac disease, the role of diagnostic techniques in nuclear medicine is rapidly growing in importance. These techniques have a proven utility based on their ability to offer unique insight into the pathophysiology of any system being imaged. More importantly, with the trend toward managed healthcare and cost capitation, the cost-effectiveness of nuclear medicine is increasingly being realized. This article reviews the two major classes of nuclear medicine techniques, namely, blood pool imaging, in which a radiotracer in the blood is used for assessment of contractile function, and perfusion imaging, in which the analysis of the delivery of radiotracers to the myocardium is used to assess myocardial perfusion. Other less widespread, but nonetheless important, techniques such as imaging using positron emitters and infarct avid imaging also are discussed. Finally, future developments are described in view of their potential to further the applicability of these techniques.

Blood pool imaging

First-pass studies-In this type of study, the information is gained by tracking the initial transit of a radiotracer through the heart. Thus, a proper injection technique is critical. The antecubital vein must be cannulated with at least a 20-gauge needle, properly secured, and attached to a three way stopcock and two syringes, one containing the radiotracer in a small volume (<1 ml) and the other containing a saline flush (10 ml to 20 ml). With the arm abducted and resting comfortably, the radiotracer is then injected, immediately followed by the saline, flushed strongly, as the tourniquet is released and the acquisition started.

Although any non-particulate radiotracer can be used, those that rapidly clear from the blood such as Tc-99m-DTPA and Tc-99m-sulphur colloid are preferred because they allow multiple studies to be performed with little residual from any preceding study. The data are recorded at a rate of at least 20 frames per second. The passage of the bolus normally can be seen in the superior vena cava, the right atrium and ventricle, the lungs, the left atrium and ventricle, and finally out the aorta (figure 1). This sequence is disrupted in right-to-left shunts by the appearance of the tracer in the left side of the heart before passage through the lungs.

The magnitude may be estimated by comparing the counts in appropriately drawn regions of interest. A left-to-right shunt may be recognized as a reduction in the peak of the time-activity curve, and the magnitude is best quantified by the pulmonary-to-systemic flow (Qp:Qs) ratio. This ratio is determined by a curve-fitting analysis of the pulmonary time-activity curve, with a Qp:Qs ratio of greater than 1.2 indicating significant left-to-right shunting.1 Reflux of activity into the jugular vein may be seen in tricuspid valve disease. Other valvular diseases manifest themselves by their effects on chamber size and function. The ejection fractions of the right and left sides of the heart can be measured separately using a composite curve for a region of interest over each ventricle that is calculated by summing in phase the six or seven beats with maximum activity (figure 2).

Rapid imaging and temporal separation of the two sides of the heart with assessment of the transit times are two advantages of first-pass studies. The disadvantage is that, with a single-headed camera, only one projection can be recorded per injection.

Equilibrium studies-These studies involve imaging following the radiolabelling of the blood pool with an agent such as an RBC or albumin. Here, the goal is to obtain the "average" heart beat. Using an R-wave trigger, the computer records an image of the heart by splitting the R-R interval into 16 to 32 electronic bins. Only a few events are recorded into each successive bin with each beat, but repeating the process for 500 to 1000 beats yields sufficient counts to form a representative time-activity curve for the left ventricle (best seen in the left anterior oblique projection).

This curve can be Fourier analyzed to yield the end systolic and diastolic volumes, the regional and global ejection fractions, the peak emptying and filling rates, and the amplitude and phase of the contraction (figure 3).2,3 While the parameters yielded by such an analysis can be reported from a printout, the interpretation of these studies must include viewing all the data on a computer screen, especially in a cinematic loop. The serial assessment of these parameters is important in the management of cardiomyopathy and the use of cardiotoxic chemotherapeutic agents.

Nuclear probe studies-A non-imaging variant of equilibrium studies uses a probe detector instead of a gamma camera. This permits ambulatory measurements of the ejection fraction.4

Perfusion imaging

Choice of stress-Many forms of stress have been shown to be adequate for myocardial perfusion imaging. Recently, however, pharmacological stress has been gaining acceptance because of its ease, reproducibility, and freedom from patient motivation and/or infirmities.

Physical stress-This remains the physiological standard. It may be performed on an ergometer, starting at 25 watts per minute and increasing by 25 watts per minute every three minutes thereafter, or a calibrated treadmill with an increasing speed and gradient to impose a similar workload.5 The endpoints are a) achievement of 70% to 80% of the age-predicted maximum heart rate; b) the appearance of serious symptoms such as progressive angina, a drop in blood pressure or heart rate, severe dyspnea or faintness, or ventricular arrhythmias; or c) symptoms such as marked ST segment depression, atrial arrhythmias, AV blocks, marked elevation of blood pressure, chest pain with no electrocardiographic changes, or fatigue.

Pharmacological stress-There are two main classes of drugs used to cause cardiac stress for perfusion imaging purposes. The choice depends primarily on whether or not the patient has significant pulmonary disease, such as asthma, which would require the use of inhalers. In this scenario, vasodilators would not be the agents of first choice.

Vasodilators-These agents are postulated to affect the intracellular cyclic AMP and GMP levels, along with calcium, via the A2 receptors, which results in coronary hyperemia that is indirect in the case of dipyridamole and direct in the case of adenosine. The use of caffeine and other methylxanthine compounds, such as those in tea, chocolate, several types of cola drinks, and certain over-the-counter medications, should be discontinued at least 24 hours before the test to avoid a reduction in sensitivity, which may require a 50% increase in the vasodilator dose used. It is useful to give each patient a prepared list of the agents to be avoided.

Dipyridamole-The dose is 0.56 mg/kg infused over 4 minutes, with infusion of the imaging agent 3 to 5 minutes later. Side effects of dipyridamole are mostly flushing and nausea, but occasionally bronchospasm requiring the use of aminophylline may be induced.6 The drug's duration of action is about 30 minutes, and the patient will require some supervision after the test.

Adenosine-A six-minute infusion at the rate of 140 mcg/kg per minute is given, with the injection of the radiotracer during the third minute. The incidence of minor side effects such as flushing is higher compared to dipyridamole, and the frequency of AV blocks, usually mild and transient, mandates close supervision during the infusion.7 Adenosine has a short duration of action: only 2 to 3 minutes. Two advantages arising from this, however, are that most of the side effects or blocks that occur during the test can be relieved by merely stopping the infusion, and prolonged supervision after the test is not necessary.

Inotropes- The prime candidates for the use of these agents are those with COPD, asthma, or allergy to the vasodilator agents. These drugs exert their effects via beta-1 agonist activity to increase myocardial contractility, thus increasing oxygen demand. Hypertension, especially in severe cases, and atrial fibrillation or flutter are important contraindications. If these agents are used in patients who are not asthmatics, beta blocker administration should be stopped at least 48 hours before the test.

Dobutamine-Depending on the patient's condition, the dose can be started at 5 mcg/kg per minute for five minutes, and increased in steps of 5 mcg/kg every five minutes, to a maximum infusion rate of 30 mcg/kg per minute. Alternatively, an infusion starting at 10 mcg/kg per minute, increased by the same every three minutes, to a maximum of 40 mcg/kg per minute, also can be given.8 Both protocols can be titrated according to the patient's response. The radiotracer is injected at the onset of significant symptoms, ECG changes, or achievement of the maximal rate of infusion or heart rate, and the infusion is maintained for an additional two minutes, with titration adjusted to the condition of the patient.

Arbutamine-This newer agent is more potent than dobutamine. Additionally, it is available in the market with its own computerized delivery system that titrates the dose rate automatically, thereby reducing the workload of the supervising physician.9

Choice of imaging agent-The two major radionuclides used for perfusion imaging are thallium-201 (T1-201) as a solution of its chloride salt, and technetium-99m (Tc-99m), coupled to one of three classes of agents: isonitriles (e.g. MIBI), boronic acid oximes (e.g. teboroxime), or phosphines (e.g. tetrofosmin). While the properties of each are further described below, it is important to recognize their major differences. Tl-201 is a low energy emitter; its administration dosage is limited by its relatively higher radiation level as compared to the optimal physical characteristics of Tc-99m. However, Tl-201 has a higher total accumulation in the myocardium, and can provide redistribution information which is relevant in assessing viability.

Depending mainly on the logistics at a particular center, combined protocols, such as rest Tl-201 followed by stress Tc-99m-MIBI imaging, can be chosen.10,11 This is made possible by the fact that while there is considerable downscatter from the Tc-99m into the Tl-201 energy window (about 30%), there is very little crosstalk between the Tl-201 and Tc-99m windows (about 3%), so that a lower activity, lower energy Tl-201 imaging scan can be performed first. However, the performance characteristics must be carefully evaluated with extensive phantom studies before comparing clinical images, due to the differences in processing (especially filtering) necessitated by the different energies.

Regardless of which agent is used, it is important to remember that, as a rough estimate, the half-value thickness in tissue is only about 3 cm for the Tl-201, and only about 4.5 cm for Tc-99m. This means that the deeper parts of the heart, such as the inferior wall, or those overlain with dense/large tissue (i.e. breast tissue), such as the anterolateral wall, will appear less avid on images in spite of equal uptake. Techniques such as attenuation correction, which may reduce but not totally eliminate this disparity, may be employed to counteract this limitation.

Thallium-201-This radionuclide is produced from the conversion of stable thallium-203 into lead-201 by the (p,3n) reaction in a cyclotron. It has a half-life of 9.4 hours, after which it decays into thallium-201, which in turn decays with a half-life of 74 hours into stable mercury-201; both transformations occur by electron capture. The 69-83 KeV mercury-201 x-rays, along with a smaller abundance of 135 KeV (2%) and 167 KeV (8%) gamma photons, are available for imaging. The low energy x-rays and the relatively long half-life make for less than ideal imaging characteristics, and limit the total administered activity per procedure to 3 mCi to 4 mCi, either as a single dose with imaging immediately post-stress and at 3 hours, or split into 2 mCi for the stress study, followed 3 hours later by a 1 mCi re-injection and a rest study 30 minutes later. Either protocol can include delayed imaging at 24 hours for a viability study (figure 4).

The myocardium takes up the administered activity by the Na-K ATPase system with a first-pass extraction of 80% to 90%. The overall amount taken up varies from about 3% to 4% at rest up to about 8% to 10% with dipyridamole stress. The initial uptake is proportional to regional perfusion (figure 5), and differences may be displayed as polar plots, also called bull's eye views (figure 6). The distribution of Tl-201 is not static after this initial uptake. Clearance from the myocardium is proportional to the regional perfusion, so that areas with normal perfusion that took up relatively greater amounts initially also lose this activity at a rate greater than that from less well-perfused areas.12

Additionally, areas with greatly reduced perfusion continue to increase their accumulation slowly, so that by 24 hours they may show uptake, thus indicating their viability as compared to nonviable areas that persist as defects in uptake. It is this dynamic pattern of distribution that makes Tl-201 the current agent of choice in assessing viability. Additional information such as the calculation of the heart-to-lung ratios is of value in assessing triple-vessel disease. In this, increased activity in the lungs, seen despite a relatively normal pattern of distribution within the myocardium itself, may belie the severity of the disease.

Technetium-99m-This workhorse radionuclide of nuclear medicine is popular because of its low cost, low radiation dose, easy availability, and excellent imaging characteristics (figure 7). Other advantages include flexible scheduling and increased patient throughput; however, these advantages are somewhat offset by the characteristics of the agents it is coupled to for myocardial perfusion imaging. While gated SPECT, which can correlate perfusion with regional contractile dysfunction (figure 8), is possible with these agents, overall, they are less well suited to assess viability, and they have demonstrated a lower defect reversibility. More development and evaluation will be required before these agents, or their successors, can replace Tl-201. Some of the more important derivatives amongst these are described briefly:

Tc-99m-MIBI-The uptake of this cationic lipophilic isonitrile complex, which associates intracellularly with myocyte mitochondria, is proportional to the regional perfusion of the myocardium, albeit nonlinearly, with a fall-off in extraction at higher rates of flow. Both the first-pass extraction (about 40%) and overall accumulation (about 4%) of Tc-99m-MIBI are only half that of Tl-201 at stress, with significant amounts of hepatic activity that can cause artifacts and problems in interpretation.13 The subsequent washout is also much slower, necessitating the use of separate injections for the stress and rest studies. A particular advantage of this slow washout is that an injection can be made in the emergency room at admission for a suspected cardiac event that reflects the perfusion at initial presentation, with imaging possible for several hours after stabilization of the patient.

Tc-99m-teboroxime-This neutral bor- onic acid oxime complex has first-pass extraction and total accumulation levels comparable to that of Tl-201, and also exhibits a biexponential washout from the myocardium.14 However, there is higher background from lung and liver activity in contrast to Tl-201, negating some of this agent's advantages. Thus, the imaging protocol represents a compromise between the conflicting requirements of high background requiring a delay for clearance and the rapid component of redistribution requiring early onset of imaging. This component also causes reconstruction artifacts, which require the use of triple-headed cameras or rapid, repeated acquisitions on older machines for minimization.

Tc-99m-tetrofosmin-This diphosphine complex, as well as its related compounds Q12 (furifosmin) and Q3, have lower first-pass extraction and overall accumulation levels compared to Tl-201,15 but offer what is, perhaps, a better compromise, due to their slow myocardial washout with rapid background clearance. However, rapid redistribution studies for viability are not possible.

Infarct avid imaging

The fact that acute myocardial infarction may be silent in up to one-fourth of patients, and the ability to use reperfusion therapy to improve outcomes, if given early in the course of the disease, make the detection of myocardial necrosis increasingly important. While several agents have been used for this purpose, the most widely used are Tc-99m-pyrophosphate (PYP) and Tc-99m-antimyosin Fab fragments.

Tc-99m-pyrophosphate-PYP is thought to localize in damaged myocardial tissue by combining with the calcium that is known to accumulate in these areas. The largest levels of accumulation occur with a 60% to 70% reduction in flow, with a reduction in accumulation from this peak seen with greater levels of occlusion. This may lead to a doughnut-shaped pattern of uptake, which indicates a worse prognosis. The accumulation can be detected as early as 6 hours after an infarction, but is usually better seen at 24 hours, with the peak occurring at around 48 to 72 hours and persisting for several days thereafter.16

Problems with the use of PYP include overlying activity in the ribs, significantly reduced sensitivity for nontransmural infarcts compared to transmural infarcts (which are usually detectable by conventional modalities anyway), and a false-positive incidence of about

10%; these have a wide variety of causes, the more important being blood pool activity, recent cardioversion, cardiac contusion, previous infarcts, valvular calcifications, and cardiomyopathy.

Tc-99m-antimyosin Fab fragments-In order to provide a specific marker of irreversible myocyte damage, Fab fragments of an antibody, raised against the water-insoluble heavy chains of cardiac myosin that are exposed due to necrosis, have been used with some success.17 (Radioisotopes of iodine, and indium-111 have also been used as radiolabels.) Localization occurs only in acute infarcts, and the intensity decreases as the infarcts heal. The sensitivity achievable with this agent is around 95%.

Positron emission tomography (PET)

The use of positron emitters to assess damage to the heart offers the advantages of 1000-fold improvement in sensitivity of event detection, and a two-fold improvement in resolution, albeit at greater cost and more limited availability. PET can provide information both about perfusion and metabolism.

Perfusion-Regional perfusion can be imaged with nitrogen-13-ammmonia, oxygen-15-water, and rubidium-82. This last agent has the advantage of being available from a strontium-82 generator, enabling it to be used without an on-site cyclotron. Overall, while the accuracy of perfusion for detecting fixed lesions is similar to that of Tl-201, it is reportedly higher for reversible ischemia.18

Metabolism-Several agents, each looking at a particular metabolic pathway, have been evaluated with good results. Fluorine-18-deoxyglucose correlates with the rate of glycolysis and shows a focal increase in areas of mild reduction in perfusion. This mismatch, that is, an area of reduced perfusion as seen on a Tl-201 scan (or with a Tc99m agent) showing a focal increase in F-18-DG uptake, is an important predictor of improvement in regional function after surgical intervention, with over 90% of matching defects on perfusion and F-18-DG images not improving after bypass.19

Carbon-11-palmitate assesses beta-oxidation, and reduction of uptake correlates with the severity of ischemia. Carbon-11-acetate has been suggested to evaluate tricarboxylic acid cycle activity, with delayed clearance indicating impairment, and may offer the ability to estimate regional oxygen consumption.

Receptor and inflammation imaging

Receptor imaging-Various receptors play a role in the regulation of cardiac activity and the pathogenesis of important diseases of the heart. While these can be imaged with use of appropriate agents, such as iodine-123-MIBG, carbon-11-propanolol, carbon-11-prazosin, flourine-18 metaraminol, and hydrogen-3-methylscopolamine, it is iodine-123-MIBG that has been evaluated the most due to its single-photon emission and wider availability. Reductions in MIBG uptake correlate with the severity of heart failure and idiopathic cardiomyopathy, with the heart-to-mediastinum activity ratios serving as better predictors of survival than ejection fraction and echocardiography.20 This agent also has been used to assess neuronal damage after infarction, which correlates with wall motion abnormalities and arrhythmias.

Inflammation-In addition to the use of PYP and antimyosin fragments for the asessment of myocarditis, non-specific agents such as gallium-67-citrate and Tc99m-antigranulocyte antibodies may have a limited role in assessment of pericarditis and subacute infective endocarditis, respectively.

Future developments

The role of nuclear medicine presented here is likely to be enhanced

in the future due to the recent progress in equipment and reconstruction techniques. Of particular interest is the ability to perform attenuation correction, which employs a transmission source (for example, gadolinium-153) to estimate the distribution of attenuation within each patient, and then uses this information to correct the emission data. Another interesting development is the use of F-18-DG as a single-photon emitter with specially adapted gamma cameras in an attempt to obviate the need for expensive dedicated PET scanners. The use of iterative reconstruction techniques, with their improvement in image quality, is also increasing due to the affordability of powerful computers. Initial results for these are encouraging, but further experience is required before they are incorporated into mainstream technologies. AR

Acknowledgment

The authors thank Dr. Ronald G. Schwarz and Ms. Marie Mackin for providing some of the figures used in this article.

References

1. Maltz DL, Treves S: Quantitative radionuclide angiography: Determination of Qp:Qs in children. Circulation 47:1049-1056, 1973.

2. Burwo RD, Strauss HW, Singleton R, et al: Analysis of left ventricular function from multiple-gated acquisition cardiac blood pool imaging: Comparison to contrast angiography. Circulation 56:1024-1028, 1977.

3. Botvinick EH, Sceinman M, Hattner RS, et al: An accurate means of detecting and characterizing abnormal patterns of ventricular activation by phase image analysis. Am J Cardiol 50:289-298, 1982.

4. Tamaki N, Yasuda T, Moore R, et al: Continuous monitoring of left ventricular function by an ambulatory radionuclide detector in patients with coronary artery disease. J Am Coll Cardiol 12:669-679, 1988.

5. Bruce RA, DeRouen TA, Hossack, KF: Value of maximal exercise tests in risk assessment of primary coronary heart disease events in healthy men: Five years' experience of the Seattle heart watch study. Am J Cardiol 46:371-378, 1980.

6. Leppo JA: Dipyridamole myocardial perfusion imaging. J Nucl Med, 35:730-733, 1994.

7. Iskandrian AS: Adenosine myocardial perfusion imaging. J Nucl Med 35:734-736, 1994.

8. Verani M: Dobutamine myocardial perfusion imaging. J Nucl Med 35:737-739, 1994.

9. Kiat H, Iskandrian AS, Villegas BJ, et al: Arbutamine stress thallium-201 single photon emission computed tomography using a computerized closed-loop delivery system: Multicenter trial for evaluation of safety and diagnostic accuracy: The International Arbutamine Study Group. J Am Coll Cardiol 26:1159-1167, 1995.

10. Kiat H, Germano G, Friedman J, et al: Comparative feasibility of separate or simultaneous rest thallium-201/stress technetium-99m-sestamibi dual-isotope myocardial perfusion SPECT. J Nucl Med, 35:542-548, 1994.

11. Heo J, Wolmer I, Kegel J, et al: Sequential dual-isotope SPECT imaging with thallium-201 and technetium-99m sestamibi. J Nucl Med 549-553, 1994.

12. Pohost GM, Alpert NM, Ingwall JS, et al: Thallium redistribution: Mechanism and clinical utility. Semin Nucl Med 10:70-93, 1980.

13. Leppo JA, Meerdink D: Comparison of the myocardial uptake of a technetium-labelled isonitrile analogue and thallium. Circ Res 65:632-639, 1989.

14. Glover D, Okada R, Hebert C: Tc-99m-teboroxime (SQ30217, Cardiotec): Kinetics in normal and ischemic canine myocardium. Circulation 82:486, 1990.

15. Kelly J, Forester A, Higley B, et al: Technetium-99m-tetrofosmin as a new radiopharmaceutical for myocardial perfusion imaging. J Nucl Med 34:222-227, 1993.

16. Parkey RW, Bonte FJ, Buja LM, et al: Myocardial infarct imaging with technetium-99m phosphate. Semin Nucl Med 7:15-28, 1977.

17. Khaw BA, Narula J: Focus on antimyosin imaging. In: Wagner HN, Szabo Z, Buchanan JW (eds.): Principles of Nuclear Medicine, pp 819-825. WB Saunders Company, Philadelphia, 1995.

18. Tamaki N, Yonekura Y, Senda M, et al: Is stress N-13 ammonia PET imaging superior to Tl-201 SPECT? J Nucl Med 28:567, 1987.

19. Tillisch J, Bunken R, Marshall R, et al: Reversibility of cardiac wall-motion abnormalities predicted by positron tomography. N Engl J Med 314:884-888, 1986.

20. Merlet P, Valette H, Dubois-Rande JL, et al: Prognostic value of cardiac MIBG imaging in patients with congestive heart failure. J Nucl Med 33:471-477, 1992.