The worldwide burden of coronary artery disease (CAD) remains substantial. Annually, an estimated 1.5% of the US population visits the primary care services with symptoms of chest pain.¹ Each year over 660,000 patients in the United States present to hospital with either a first myocardial infarction (MI) or sudden cardiac death (SCD) due to CAD.² Of these 305,000 will have had a recurrent MI. An estimated 160,000 will also suffer a silent first MI.² CAD is the leading cause of death across the world and is predicted to remain so for the next 20 years. Annually, approximately 3.8 million men and 3.4 million women die from CAD. The number is estimated to rise to 11.1 million deaths globally by 2020.³ An estimated $108.9 billion is spent annually on treatment of CAD and is expected to exceed $320 billion by the year 2030.² Thus, means and methods to reduce the prevalence, morbidity, and mortality associated with CAD remain of great importance to health care providers.

For these reasons, a systematic approach for early diagnosis and risk stratification of CAD is important for patients to benefit from preventive and therapeutic strategies. Over the past two decades, radionuclide myocardial perfusion imaging (MPI) has become clinical mainstay for the noninvasive evaluation of CAD. In this chapter, the role of nuclear cardiac perfusion including both single-photon emission computed tomographic (SPECT) and positron emission tomographic (PET) imaging will be discussed in relation to the diagnosis, risk stratification, and patient management decisions in patients with suspected CAD.

To optimize the use of testing for suspected CAD, the initial step involves stratifying the likelihood of such a patient to have underlying disease. Although a variety of tools are available for this purpose, the evaluation of risk factors and the nature of presenting chest pain and associated symptoms are often used to estimate the pretest likelihood of CAD in patients (Table 14-1).⁴ Modifiable cardiac risk factors include hypercholesterolemia, tobacco smoking, hypertension, diabetes mellitus, physical inactivity, and obesity, while nonmodifiable risk factors are family history of CAD in first-degree relatives under the age of 60 years, advanced age, and male gender.^5,6 Multiple clinical prediction models have been developed to stratify asymptomatic patients into low, intermediate, and high risk for presence of CAD. The most commonly used risk score in the United States is the Framingham Risk Score (FRS) which predicts a 10-year risk for nonfatal MI and cardiac death based on the presence or absence of cardiac risk factors.⁷ Using the FRS, patients can be categorized into low-risk (age-specific risk level below average; 10-year absolute risk of CAD <6%), intermediate-risk (age-specific risk level average or above average; 10-year absolute risk of CAD between 6% and 20%), or high-risk (patients with coronary risk equivalents; 10-year absolute risk >20%) groups. A modified version of the FRS has been incorporated into the Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel [ATP] III).⁸ However, none of these models incorporated stroke, transient ischemic attack, claudication, and heart failure as outcomes, nor have the majority of models included family history of CAD. Additional models have also been developed including the following: (1) Framingham General CVD Score with a higher predictability for cardiovascular morbidity and mortality⁹; (2) Reynolds CVD Score which is gender specific and incorporates hemoglobin A1c, high-sensitivity CRP, and family history of MI¹⁰; (3) the ACC/AHA CVD risk calculator which is similar to the FRS¹¹; and (4) the MESA risk calculator¹² which incorporates coronary calcium score.

Patients who present for evaluation of suspected CAD may have chest symptoms suggestive of angina or have no symptoms but have increased risk for CAD. Patients may also present for primary cardiac evaluation due to an upcoming surgical procedure. For symptomatic patients, the role of nuclear MPI is an important one. At the present time, MPI is considered appropriate care only in those asymptomatic patients who are at high cardiovascular disease risk based on standard ATP III criteria, or are at high cardiovascular disease risk with a moderately abnormal (100–400) or a severely abnormal (>400) coronary artery calcium score.^13,14 Stress testing may also be used in asymptomatic persons in high-risk jobs such as airline pilots or public transportation workers. This chapter will primarily focus on evaluation of symptomatic subjects.

Patients with classic angina have retrosternal chest pain, pressure, or discomfort which comes on with emotional or physical stress, may radiate to the shoulders, arms or jaw, and rarely below the epigastrium or above the mandible. The discomfort usually lasts for minutes, is relieved by cessation of activity or with sublingual nitroglycerin. When all three characteristics of angina are present (chest pain, exertion related, relieved with cessation or nitroglycerin), it is denoted as “typical.” When only two out of the three characteristics are present, chest pain is considered “atypical” and when only one of the three characteristics is present the chest pain is likely “noncardiac.” Using the characteristic of the presenting chest discomfort and the patient’s age and gender, patients can thus be placed into low-, intermediate-, or high-risk categories for having significant CAD (Table 14-1).¹⁵ It should be noted that many elderly and female patients who are eventually diagnosed with significant CAD present with atypical symptoms.

The choice of diagnostic testing of patients with suspected CAD is based on a multitude of factors. The objective is to determine whether the patient has CAD as accurately as possible with the least harm, least discomfort, and cost involving the tests and procedures needed to do so. Other factors which are equally important in deciding the type of stress test include the patient’s ability to perform adequate exercise, the presence of disqualifying abnormalities on the resting ECG and the patient’s body habitus. In general, symptomatic patients who are in the category of lower pretest likelihood of CAD are referred for exercise tolerance testing (ETT) alone, while those in higher-risk categories are candidates for imaging procedures.

Exercise Tolerance Testing

The simplest form of noninvasive stress testing is the ETT. While this chapter primarily focuses on nuclear imaging, it can be said that ETT should be a consideration in many patients who meet the criteria. This form of testing involves the patient exercising on a treadmill or stationary bicycle and can be an adequate test alone for certain patients without ancillary imaging. This procedure is discussed in detail in Chapter 8. For ETT, the patients should be able to achieve an adequate workload defined by peak heart rate and exercise duration to derive meaningful diagnostic and prognostic value from the test. In addition to the inability to exercise, the presence of left bundle branch block (LBBB), a paced rhythm, ST-T changes on ECG due to LVH or digoxin use, ventricular pre-excitation and ST depression of ≥1 mm on the resting ECG in such patients can render the testing ECGs nondiagnostic, and therefore not useful in such a patient.

Currently, exercise is the preferred stress modality with or without perfusion imaging since it provides prognostic information in the form of hemodynamic data, functional capacity, and evaluation of symptoms. The use of the Duke Treadmill Score which incorporates exercise duration, exercise-induced ST-segment changes, and stress-induced angina can risk stratify patients into low, intermediate, and high risk for likelihood of coronary artery stenosis of ≥75%, multivessel disease, and 1-year all-cause mortality.¹⁶

Various studies have shown the sensitivity and specificity of ETT is approximately 68% and 77%, respectively.¹⁷ Thus, the diagnostic accuracy of ETT alone is moderate, indicating that the health care provider ordering the test needs to assume a modicum of inaccuracy. While a false-positive study can be resolved, a false-negative result is more difficult, since it is a normal result and dependent upon the clinical situation for further testing. Moreover, in women who tend to present with atypical symptoms, ETT has a lower accuracy to diagnose CAD as compared to men due to a lower prevalence of epicardial CAD in women, an increased likelihood of inadequate exercise especially in the elderly, and an increased incidence of false-positive ST depression on ECG during exercise in women.^17,18 The value of ETT in the elderly has been questioned. In a meta-analysis by Rai et al.¹⁸ in patients ≥65 years of age, patients with abnormal and normal ETT results had similar cardiac event rates (OR 3.1, 95% confidence interval [CI] 0.8–11.5) following the procedure. In contrast, an abnormal stress MPI compared to normal stress MPI accurately stratified risk in these patients (OR 11.8, 95% CI 7.5–18.7). Thus in elderly patients, ETT may not be beneficial. A diagnostic study by Froelicher et al.¹⁹ minimized referral bias by performing ETT and coronary angiography on all patients, irrespective of indication. In this predominantly male population, the ETT sensitivity was 45% and the specificity was 85%. Thus, ETT is primarily useful for patients at a lower risk of CAD in whom a true-negative test is much more likely than a false-negative test, and will be a reassuring finding.

Of note, despite data suggesting less accurate results in women, the most recent ACC/AHA guidelines recommend having a similar approach to the utilization of ETT for diagnosis of CAD regardless of gender.²⁰

Exercise Stress Testing with SPECT Myocardial Perfusion Imaging

A major limitation of ETT is its less than optimal diagnostic accuracy for detection of significant CAD. Studies have shown that the diagnostic accuracy of stress MPI is significantly higher than that of ETT alone and provides greater risk stratification for predicting future cardiac events. Thus, in a population of patients in which CAD is prevalent, stress MPI is a better choice than ETT alone. In an aging population, it is important to note that the accuracy of the ETT depends on a patient’s ability to reach a predicted maximum heart rate. Patients with medical illness, debilitation, deconditioning, or musculoskeletal problems may be unable to perform an adequate ETT. MPI with pharmacologic stress using vasodilators (dipyridamole and adenosine) or dobutamine can be implemented in such patients. Pharmacologic stress with MPI is most advantageous in older patients who are at the highest risk of CAD, yet are likely to be the population least able to exercise adequately.

The addition of SPECT imaging to exercise stress testing is used to improve diagnostic accuracy and to provide valuable risk stratification. SPECT MPI is based on the principle of visualizing the effects of relative blood flow between the resting and stressed myocardium. If the perfusion abnormalities occur only on stress as compared to rest, it indicates ischemia in that territory whereas the presence of matched perfusion abnormalities in stress and rest images suggests areas of scar or prior infarction. Moreover, by incorporating ECG gating into the test, additional diagnostic and prognostic data in the form of left ventricular function and volume as well as wall thickening and motion can be obtained. The procedural aspects of SPECT imaging are described in Chapter 9.

The sensitivity of stress testing with SPECT imaging (87%) is substantially higher than that of exercise ECG testing without imaging (68%), as is specificity. A meta-analysis of SPECT accuracy reported a sensitivity of 88.3% and a specificity of 75% in analysis of 108 published studies.²¹ A recent meta-analysis by Iskandar et al.²² evaluating the effect of gender on the diagnostic accuracy of SPECT MPI yielded a mean sensitivity and specificity of 84.2% (95% CI 78.7–88.6%) and 78.7% (CI 70.0–85.3%) for SPECT MPI in women; and 89.1% (CI 84.0–92.7%) and 71.2% (CI 60.8–79.8%) for SPECT MPI in men. There was no significant difference in the sensitivity (p = 0.15) or specificity (p = 0.23) between male and female subjects, thus this test is equally effective in both genders (Fig. 14-1). The prognostic implications of stress MPI are discussed in Chapter 15. In summary, current literature demonstrates that diagnostic accuracy with SPECT is much higher than with ETT alone, and particularly useful in intermediate- to high-risk patients. In elderly patients, SPECT is particularly important in comparison to ETT.

Pharmacologic Stress Testing with SPECT Myocardial Perfusion Imaging

One of the important advantages of MPI in assessing patients with suspected CAD is that patients who are unable to achieve adequate exercise for a diagnostic result can be studied using pharmacologic stress. It has been estimated that 40% to 50% of all stress MPI in the US is performed with pharmacologic agents, and the percentage increases annually. A detailed discussion regarding the use of pharmacologic stress testing may be found in Chapter 8. The two classes of pharmacologic agents are vasodilator (adenosine, dipyridamole, and more recently, a selective A2a agent, regadenoson) or inotropic stress (dobutamine). In brief, vasodilators, given intravenously, markedly increase coronary blood flow. This increased flow is less pronounced in arteries that are stenotic (flow restricted) due to atherosclerosis, as they are already maximally vasodilated. This causes heterogeneous myocardial perfusion, which can be identified using a radionuclide tracer that follows coronary blood flow. As an alternative to vasodilator stress in patients with contraindications to vasodilators, dobutamine works by increasing myocardial oxygen demand (through increased heart rate, systolic blood pressure, and myocardial contractility). This creates increased coronary blood flow, which results in heterogeneous blood flow due to stenotic arteries. An important marker of inotropic stress success is reaching at least 85% maximally predicted heart rate, similar to exercise stress. As in exercise MPI, scintigraphy images obtained at rest are compared to those obtained during peak pharmacologic stress to distinguish myocardial ischemia from infarction and normal.