**Noninvasive Imaging for the Assessment of Coronary Artery Disease**

Punitha Arasaratnam and Terrence D. Ruddy

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61502

#### **Abstract**

Noninvasive cardiac imaging is a cornerstone of the diagnostic work-up in patients with suspected coronary artery disease (CAD), cardiomyopathy, heart failure, and congenital heart disease. It is essential for the assessment of CAD from functional and anatomical perspectives, and is considered the gate-keeper to invasive coronary angiography. Cardiac tests include exercise electrocardiography, single photon emission computed tomography myocardial perfusion imaging, positron emission tomography myocardial perfusion imaging, stress echocardiography, coronary computed tomography angiography, and stress cardiac magnetic resonance. The wide range of imaging techniques is advantageous for the detection and manage‐ ment of cardiac diseases, and the implementation of preventive measures that can affect the long-term prognosis of these diseases. However, clinicians face a chal‐ lenge when deciding which test is most appropriate for a given patient. Basic knowledge of each modality will facilitate the decision-making process in CAD as‐ sessment.

**Keywords:** Noninvasive, imaging, coronary artery disease, assessment, diagnosis

#### **1. Introduction**

Noninvasive cardiac imaging is crucial for coronary artery disease (CAD) assessment. The increasing global burden of CAD is a major contributor to the marked growth in the use of noninvasive imaging [1]. In recent years, the development of state-of-the-art hardware and software technologies has broadened the perspective and dimension of noninvasive imaging. This is advantageous to hybrid imaging in CAD assessment, with the introduction of anatom‐

© 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

ical, physiological, or combined approaches. These techniques allow clinicians to move beyond the dichotomous concept of the presence or absence of CAD by increasing their understanding of the unique pathophysiologic processes in CAD, including subclinical atherosclerosis, plaque vulnerability, myocardial blood flow (MBF), and scar detection.

Invasive coronary angiography (ICA), an anatomical test, is considered the gold standard method for the diagnosis of CAD. Nevertheless, the risk of complications precludes the routine use of ICA and it is only indicated in patients with a high pre-test probability of the disease [2]. Because most patients have low or intermediate pre-test probabilities of disease, noninvasive testing should be considered first, serving as a selection process for ICA. Clinicians can choose from a wide range of noninvasive tests, including exercise electrocardiography (ECG), single photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI), positron emission tomography (PET) MPI, stress echocardiography (SE), coronary computed tomography angiography (CCTA), and stress cardiac magnetic resonance (CMR). Therefore, clinicians are frequently faced with the apparently difficult clinical question: "What is the right test?" However, there are no right tests! The test to be used should be selected for each patient after considering the patient's characteristics, genetic and environmental factors, predisposi‐ tion, risk factors, and comorbidities. Cardiac testing is generally unnecessary in asymptomatic patients except in high-risk occupations or before starting antiarrhythmic drugs.

A basic understanding of the principles, diagnostic, and prognostic accuracy, and the strengths and limitations of each imaging technique is essential. The clinician must then adopt a structured approach, which will help choose the appropriate test to use after considering the risk and benefit profile of each test. The establishment of a diagnosis of CAD will influence the perceived likelihood of a future cardiac event and warrant secondary prevention to slow or prevent disease progression. The absence of CAD on imaging will reassure the patient, and encourage the clinician to adopt a primary prevention strategy. Hence the ultimate goal is for the chosen test to address the clinical question with a high level of certainty. The theme of this chapter is to provide a comprehensive guide to selecting the appropriate imaging test in patients with suspected CAD.

#### **2. Basic concepts for choosing cardiac imaging tests**

#### **2.1. Classification of chest pain**

There can be varied presentation of chest pain, including jaw pain, epigastric pain, indigestion, shortness of breath, or reduced effort tolerance. Atypical presentations are commonly seen among diabetics, female, and the elderly. Chest pain can be classified using the criteria below [3]:

Criteria:

**1.** Substernal chest pain or discomfort


Typical angina: (1) + (2) + (3)

ical, physiological, or combined approaches. These techniques allow clinicians to move beyond the dichotomous concept of the presence or absence of CAD by increasing their understanding of the unique pathophysiologic processes in CAD, including subclinical atherosclerosis,

Invasive coronary angiography (ICA), an anatomical test, is considered the gold standard method for the diagnosis of CAD. Nevertheless, the risk of complications precludes the routine use of ICA and it is only indicated in patients with a high pre-test probability of the disease [2]. Because most patients have low or intermediate pre-test probabilities of disease, noninvasive testing should be considered first, serving as a selection process for ICA. Clinicians can choose from a wide range of noninvasive tests, including exercise electrocardiography (ECG), single photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI), positron emission tomography (PET) MPI, stress echocardiography (SE), coronary computed tomography angiography (CCTA), and stress cardiac magnetic resonance (CMR). Therefore, clinicians are frequently faced with the apparently difficult clinical question: "What is the right test?" However, there are no right tests! The test to be used should be selected for each patient after considering the patient's characteristics, genetic and environmental factors, predisposi‐ tion, risk factors, and comorbidities. Cardiac testing is generally unnecessary in asymptomatic

plaque vulnerability, myocardial blood flow (MBF), and scar detection.

36 Coronary Artery Disease - Assessment, Surgery, Prevention

patients except in high-risk occupations or before starting antiarrhythmic drugs.

**2. Basic concepts for choosing cardiac imaging tests**

patients with suspected CAD.

**2.1. Classification of chest pain**

**1.** Substernal chest pain or discomfort

[3]:

Criteria:

A basic understanding of the principles, diagnostic, and prognostic accuracy, and the strengths and limitations of each imaging technique is essential. The clinician must then adopt a structured approach, which will help choose the appropriate test to use after considering the risk and benefit profile of each test. The establishment of a diagnosis of CAD will influence the perceived likelihood of a future cardiac event and warrant secondary prevention to slow or prevent disease progression. The absence of CAD on imaging will reassure the patient, and encourage the clinician to adopt a primary prevention strategy. Hence the ultimate goal is for the chosen test to address the clinical question with a high level of certainty. The theme of this chapter is to provide a comprehensive guide to selecting the appropriate imaging test in

There can be varied presentation of chest pain, including jaw pain, epigastric pain, indigestion, shortness of breath, or reduced effort tolerance. Atypical presentations are commonly seen among diabetics, female, and the elderly. Chest pain can be classified using the criteria below Atypical angina: (1) + (2) or (1) + (3)

Nonanginal chest pain: (1) or none.

#### **2.2. Pre-test and post-test probabilities**

Bayes' theorem proposes the use of a combined pre-test probability and test result to determine the post-test probability of disease [4]. This will help the clinician to determine whether a positive test result is a "true positive" or a "false positive" and whether a negative test result is a "true negative" or a "false negative". For example, a positive test result is likely to be a "true positive" result in a 70-year-old patient with typical angina or a "false positive" result in a 40-year-old female with nonanginal chest pain. Meanwhile, a negative test result is likely to be a "true negative" result in a 35-year-old man with atypical chest pain or a "false negative" result in a 60-year-old man with prior myocardial infarction (MI) and typical angina. Pretest probability can be estimated using the Diamond and Forrester classification [5] (Table 1).


**Table 1.** Diamond and Forrester Pre-Test Probability of Coronary Artery Disease by Age, Sex, and Symptoms. High: >90% pre-test probability. Intermediate: between 10% and 90% pre-test probability. Low: between 5 and 10% pre-test probability. Very low: <5% pre-test probability [5].

#### **2.3. Appropriate use criteria**

The appropriate use criteria (AUC) defines appropriate imaging for the different clinical indications. The AUC for test selection among symptomatic patients with suspected CAD [2] (Table 2).


**Table 2.** Appropriate Use Criteria for noninvasive testing in symptomatic patients for CAD assessment. A= appropriate, M= maybe appropriate, R= rarely appropriate. Uninterpretable ECG refers to resting abnormalities such as ST-segment depression (≥ 0.10 mV), complete left bundle branch block (LBBB), pre-excitation, digoxin use, or ventricular paced rhythm [2].

#### **3. Non-invasive imaging tests**

This section will focus on the unique principles, diagnostic and prognostic accuracy, strengths, limitations, representative cases and clinical pearls for each imaging modality.

#### **3.1. Exercise electrocardiography**

#### *3.1.1. Background*

Exercise ECG is a well-established and validated functional test used for CAD assessment. It has been used for more than 50 years, despite the increasing use of other imaging modalities. It is the first-line test in patients with suspected CAD who are able to exercise and who have an interpretable resting ECG [2]. Studies have demonstrated a lower diagnostic accuracy of exercise ECG in women because of their lower prevalence of CAD. However the risk of major adverse cardiac events (MACE) in women with good functional capacity and a normal resting ECG was not different between those who underwent exercise ECG compared with exercise SPECT MPI, and exercise ECG was considered a cost-effective strategy [6].

#### *3.1.2. Principles*

Indication for noninvasive testing in Exercise Stress Stress Stress CCTA

A R M R R

A A A M M

M A A A M

A A M M

A A A A

A A A M

**symptomatic patients ECG MPI Echo CMR**

**Table 2.** Appropriate Use Criteria for noninvasive testing in symptomatic patients for CAD assessment. A= appropriate, M= maybe appropriate, R= rarely appropriate. Uninterpretable ECG refers to resting abnormalities such as ST-segment depression (≥ 0.10 mV), complete left bundle branch block (LBBB), pre-excitation, digoxin use, or

This section will focus on the unique principles, diagnostic and prognostic accuracy, strengths,

Exercise ECG is a well-established and validated functional test used for CAD assessment. It has been used for more than 50 years, despite the increasing use of other imaging modalities. It is the first-line test in patients with suspected CAD who are able to exercise and who have an interpretable resting ECG [2]. Studies have demonstrated a lower diagnostic accuracy of exercise ECG in women because of their lower prevalence of CAD. However the risk of major adverse cardiac events (MACE) in women with good functional capacity and a normal resting ECG was not different between those who underwent exercise ECG compared with exercise

limitations, representative cases and clinical pearls for each imaging modality.

SPECT MPI, and exercise ECG was considered a cost-effective strategy [6].

Low pretest probability of CAD ECG interpretable AND able to exercise

Low pretest probability of CAD

ECG uninterpretable OR unable to exercise

38 Coronary Artery Disease - Assessment, Surgery, Prevention

Intermediate pretest probability of CAD ECG interpretable AND able to exercise

Intermediate pretest probability of CAD ECG uninterpretable OR unable to exercise

High pretest probability of CAD ECG interpretable AND able to exercise

High pretest probability of CAD

ventricular paced rhythm [2].

ECG uninterpretable OR unable to exercise

**3. Non-invasive imaging tests**

**3.1. Exercise electrocardiography**

*3.1.1. Background*

Exercise ECG evaluates the physiological response of the heart to a controlled level of exercise. The latter can be prescribed using specific exercise protocols such as Bruce and Naughton. Exercise ECG and hemodynamic-specific variables are shown to have diagnostic and prog‐ nostic value in the assessment of CAD. These variables include ST deviation, exercise capacity, percentage of the maximum age-predicted target heart rate (HR), heart rate recovery (HRR), blood pressure (BP) response, and the Duke Treadmill Score (DTS) [7].

Abnormal ST deviation is defined as ≥1 mm (0.1 mV) of downsloping or horizontal ST-segment depression (J point + 80 ms); or ≥1 mm of ST segment elevation in leads without pathological Q waves (except aVR). The J-point is defined as the junction of the QRS complex and the STsegment. The ST deviation should be seen in three or more consecutive beats in the same lead to be considered significant [8, 9]. An upsloping ST-segment depression is considered an "equivocal" response and is not suggestive of myocardial ischemia [10]. High risk features include ST-segment depression ≥ 2mm at < 5 metabolic equivalents (METs) in ≥5 leads and ≥5 minutes into recovery [11]. ST-segment elevation in two or more contiguous leads can help localize the site of significant ischemia, unlike ST-segment depression [12]. In the presence of prior Q waves, ST-segment elevation of > 1.0 mm (J point +60 ms) is considered abnormal. This could represent reversible ischemia in the peri-infarct zone or ventricular dyskinesis or akinesis of a segment of the left ventricle. This finding has been demonstrated among patients with anterior (~30%) and inferior (~15%) infarctions [13, 14].

Exercise capacity (a marker of cardiorespiratory fitness) is an estimate of the maximal oxygen uptake for a given workload, and is measured in METs [15]. The prevalence of significant ischemia was 0.4% and 7.1%, based on the workload achieved (≥10 METs and < 7 METs), respectively, on exercise SPECT MPI [16]. Hence, patients who are able to achieve a high workload (≥10 METs) on exercise ECG, may not require additional functional imaging.

The maximum age-predicted HR is usually described as "220-age". The inability to achieve 85% of the maximum age-predicted HR was associated with decreased survival [17].

HRR is calculated as the peak HR achieved (HR at 1 min) [18]. An abnormal HRR is defined as a decrease in the HR of < 12 bpm in the first minute of recovery, and is predictive of mortality.

A normal blood pressure (BP) response is defined as an increase in systolic BP and an increase or decrease in diastolic BP during exercise. A decrease in systolic BP of >10 mmHg may suggest the presence of acute left ventricular dysfunction owing to ischemia [7]. An abnormal BP response may be a specific marker for left main (LM) or triple vessel disease (TVD) [19].

The DTS is calculated as exercise time (minutes) – (5 x ST depression in mm) – (4 x angina index) (0= no angina; 1= nonlimiting angina; 2= limiting angina) [20]. DTS can be categorized into low risk (≥ +5), intermediate risk (-10 to +4) and high risk (≤ -11).

The absolute and relative contraindications for undergoing and termination of an exercise ECG, respectively, is illustrated in Table 3 and Table 4 [9].


**Table 3.** Absolute and relative contraindications for undergoing exercise ECG [9]


**Table 4.** Absolute and relative indications for termination of exercise ECG [9]

#### *3.1.3. Diagnostic and prognostic accuracy*

A meta-analysis evaluating the accuracy of exercise ECG reported a sensitivity of 68% and specificity of 77% for the detection of CAD [21]. The discriminatory cut-off point of 1 mm (0.1 mV) of horizontal or downsloping ST-segment depression had a sensitivity of 68% and specificity of 77% [22]. The frequency of significant CAD in patients with low, intermediate, and high DTS was 19.1%, 34.9%, and 89.2%, respectively. In patients with LM or TVD, the frequency of significant CAD was 3.5%, 12.4%, and 46% in patients with low, intermediate, and high DTS, respectively [23]. The 5-year survival rates in patients with a DTS of ≤ -11 and ≥ +7, were 67% and 93%, respectively[24]. In a recent study of 58,020 adults without CAD, the peak METs and the percentage of the maximum predicted HR were highly predictive of survival [25].

In relation to other imaging modalities, significant risk predictors for hard cardiac events in asymptomatic or symptomatic low-risk patients without CAD included abnormal SPECT findings (hazard ratio [HR] = 1.83), ischemia detected by exercise ECG (HR = 1.70), decreasing exercise capacity (HR = 1.11), decreasing DTS (HR = 1.07), and increasing severity of the coronary calcium score (CS) (HR = 1.29). The CS improved the long-term risk prediction for CAD when stratified according to the Framingham Risk Score [26].

*3.1.4. Strengths and limitations*

**Absolute Relative**

40 Coronary Artery Disease - Assessment, Surgery, Prevention

Physical disability that precludes safe and adequate testing

ST segment elevation (> 1.0 mm) in leads without preexisting Q waves (other than aVR, aVL and V1)

workload, in the presence of ischemia

AV block

Drop in systolic BP >10 mmHg, despite an increase in

Sustained ventricular tachycardia (VT), 2nd or 3rd degree

*3.1.3. Diagnostic and prognostic accuracy*

**Absolute Relative**

Active endocarditis

Acute myocardial infarction (<48 hours) Obstructive left main stenosis Unstable angina Moderate aortic stenosis

Uncontrolled cardiac arrhythmias High degree atrioventricular (AV) block Severe symptomatic aortic stenosis Recent stroke or transient ischemic attack

Acute myocarditis or pericarditis Uncontrolled BP >200/100 mmHg

**Table 3.** Absolute and relative contraindications for undergoing exercise ECG [9]

Moderate to severe angina Worsening chest pain

Technical difficulties in monitoring the ECG or BP BP >250/115 mmHg

**Table 4.** Absolute and relative indications for termination of exercise ECG [9]

Decompensated heart failure Hypertrophic obstructive cardiomyopathy with severe

Acute pulmonary embolism Mental impairment with limited ability to cooperate

Acute aortic dissection Uncorrected medical conditions, e.g. significant anemia,

Central nervous system symptoms (e.g. ataxia, dizziness) Fatigue, shortness of breath, wheezing, leg cramps or

Patient's request to stop Development of bundle branch block which is

Signs of poor perfusion Tachyarrhythmias, including multifocal ectopy, ventricular

A meta-analysis evaluating the accuracy of exercise ECG reported a sensitivity of 68% and specificity of 77% for the detection of CAD [21]. The discriminatory cut-off point of 1 mm (0.1

claudication

resting gradient

Significant tachyarrhythmias or bradyarrhythmias

important electrolyte imbalance, and hyperthyroidism

Marked ST displacement (horizontal or downsloping of >2

Drop in systolic BP >10 mmHg, despite an increase in

Bradyarrhythmias that potentially become more complex

mm measured 60 to 80 ms after the J point

workload, in the absence of ischemia

triplets, supraventricular tachycardia

or result in hemodyamic instability

indistinguishable from VT

The strengths and limitations of an exercise ECG are shown in Table 5.


**Table 5.** Strengths and limitations of exercise ECG

#### *3.1.5. Case example 1*

A 71-year-old man presents with atypical chest pain, able to exercise, and a normal resting ECG. Risk factors include ex-smoker and hypertension. Pretest probability of CAD is inter‐ mediate. Based on the AUC, exercise ECG is considered an appropriate test. Resting ECG revealed sinus rhythm at a HR of 61 bpm (Figure 1A). He underwent exercise ECG using Bruce protocol. He exercised for 7 minutes 31 seconds and achieved 10.1 METS. At 6 minutes into the exercise and at a HR of 121 bpm, ECG demonstrated 1 mm horizontal ST-segment depression in leads II, III, aVF, V3 to V6 and reached a maximum of 3 mm at peak stress at HR of 151 bpm in leads II, III, and aVF (Figure 1B). The changes resolved at 1 minutes 50 seconds during recovery. Test was terminated due to fatigue. He developed chest pain at recovery. The calculated DTS = -11.5. Conclusion: Abnormal exercise ECG with a high DTS. He was referred for ICA.

FIGURE 1A: Normal resting ECG. Figure 1B: Maximum of 3 mm horizontal ST segment depression seen ȱ ȱ ȱ ȱ ȱ ȱȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ **Figure 1.** (a) Normal resting ECG. (b) Maximum of 3 mm horizontal ST segment depression seen in leads II, III and aVF, 2 mm ST segment depression in leads V3 to V6, at peak stress.

#### *3.1.6. Clinical pearls*


#### **3.2. Single photon emission computed tomography**

#### *3.2.1. Background*

calculated DTS = -11.5. Conclusion: Abnormal exercise ECG with a high DTS. He was referred

(a)

(b) FIGURE 1A: Normal resting ECG. Figure 1B: Maximum of 3 mm horizontal ST segment depression seen ȱ ȱ ȱ ȱ ȱ ȱȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ ȱ **Figure 1.** (a) Normal resting ECG. (b) Maximum of 3 mm horizontal ST segment depression seen in leads II, III and

**1.** The J-point is defined as the junction of the QRS complex and the ST segment.

**4.** Exercise duration and METs achieved are strong predictors of prognosis.

**2.** Achieving 85% of age predicted HR should not be an indication for test termination.

**5.** The modification in ECG lead placement during exercise ECG compared to standard ECG may result in shift of the frontal axis to the right, increasing the voltage in the inferior leads, disappearance of Q waves in a patient with prior inferior infarct, or produce

aVF, 2 mm ST segment depression in leads V3 to V6, at peak stress.

**3.** Exercise ECG with a high DTS may warrant ICA.

artifactual Q waves in normal subjects.

*3.1.6. Clinical pearls*

for ICA.

42 Coronary Artery Disease - Assessment, Surgery, Prevention

There is robust evidence supporting the use of SPECT MPI in the workup and risk stratification of patients with suspected or known CAD because of its high diagnostic and prognostic value [27]. The development of solid state detector cameras compared with conventional SPECT (Anger) cameras offer improved signal resolution and shorter image acquisition time, which increases laboratory throughput [28, 29]. The switch from the standard, filtered back projection reconstruction method to iterative algorithms, which include depth independent resolution recovery, noise regularization, and scatter and attenuation correction, has improved the signalto-noise ratio and image quality. This also allows for low-dose imaging using a standard acquisition time with reduced radiation exposure to the patient and operator,without com‐ promising image quality and accuracy in the detection of CAD [30–32]. This is consistent with the American Society of Nuclear Cardiology's goal to reduce the radiation dose to ≤9 mSv in 50% of MPI studies by 2014.

#### *3.2.2. Principles*

SPECT MPI uses radionuclide-labeled compounds that emit Y ray photons. SPECT perfusion radiotracers include 201-Thallium (Tl-201, half-life [t½] = 73 hours), and 99m-Technetium ( 99mTc, t ½ = 6 hours)-labeled sestamibi or tetrofosmin. Tl-201 is produced from a cyclotron and 99mTc is produced by a molybdenum-99-99mTc generator.

SPECT MPI can be performed using exercise or pharmacological stressors. Exercise is preferred for patients who can exercise at an adequate workload, aiming for a minimum of 85% of the maximal age-predicted HR and 5 METs [33]. A submaximal exercise workload decreases the sensitivity of exercise SPECT MPI for the detection of CAD. Exercise usually increases MBF by 2–3 times of resting flow [24, 34].

A pharmacological stressor should be used in patients who are unable to exercise or those with baseline ECG abnormalities (pre-excitation, paced ventricular rhythm, LBBB). Pharmacologi‐ cal stressors include intravenous vasodilators (adenosine, dipyridamole or regadenoson) or dobutamine, a β-adrenoceptor agonist. Vasodilators activate the adenosine A2A receptor and cause coronary arteriolar vasodilatation. Vasodilators increase MBF by 3–5 times in normal coronary vessels, an increase termed as the coronary flow reserve. Meanwhile, dobutamine increases MBF similar to that induced by exercise. In the presence of flow limiting stenosis, the coronary vessel is maximally vasodilated at baseline. Hence, the administration of a vasodilator is unable to augment coronary flow. MPI assesses the regional flow heterogeneity in normal and diseased coronary vessels [35]. Generally, vasodilators do not cause myocardial ischemia because MBF increases, albeit with some variability in all coronary artery beds with a minimal or no increase in the rate-pressure product, a measure of myocardial oxygen demand. In patients with extensive CAD, ischemia can be induced by the coronary steal phenomenon [33]. Vasodilators may also activate other adenosine receptors (A1, A2B, and A3) resulting in bronchospasm (A2B and, A3) or AV conduction delay (A1). Regadenoson is a selective A2A receptor agonist that is better than other vasodilators in patients with moderate chronic obstructive pulmonary disease (COPD) or asthma [36, 37].

For pharmacological stress SPECT MPI, contraindications include asthma or COPD with active wheezing, 2nd or 3rd degree AV block without a pacemaker or sick sinus syndrome, systolic BP <90mmHg,use ofmethylxantines (e.g., aminophylline or caffeine) <12hours, knownhypersen‐ sivity to the vasodilator, acute myocardial infarction (MI) and acute coronary syndrome [33].

#### *3.2.3. Diagnostic and prognostic accuracy*

The diagnostic accuracy of all noninvasive tests is subject to the post-test referral bias, also known as the verification bias. This increases the sensitivity and reduces the specificity of the test. A normalcy rate is used as a surrogate for specificity. It is defined as the percentage of normal perfusion scans in patients with a low likelihood (<10%) of CAD based on the results of clinical and ECG stress tests [38–40]. A pooled analysis of 4,480 patients with known or suspected CAD showed that exercise SPECT MPI had a mean sensitivity of 87% and a specificity of 73% for the detection of >50% stenosis [41]. The normalcy rate of SPECT MPI is around 84% [38]. Standard MPI studies include a combination of stress and rest protocols. The use of a stress-only protocol with a "normal" stress study can reduce the radiation dose by 40% and detected similar event rates [42–44]. A "normal" stress-only study is defined as homogenous perfusion, summed stress score (SSS) of <3, normal left ventricular (LV) cavity size, function, and wall motion [43].

The prognostic value of a normal pharmacological stress MPI test is independent of the radiopharmaceutical used [45]. A meta-analysis of 19 SPECT MPI studies comprising 39,000 patients demonstrated a low annual event rate of 0.6%, for hard cardiac events such as cardiac death or nonfatal MI in patients with a normal test result [46]. However, the prognostic value is based on the studied population. For example, the annual event rate among individuals with a normal test result was higher among diabetic patients than in non-diabetic subjects (0.5% vs. 1.7%, respectively, p < 0.005). Diabetic patients with a LV ejection fraction (LVEF) of ≤ 45% had the worst outcome [47]. High-risk MPI variables include large perfusion defect size and extent, transient ischemic dilatation (TID), post stress stunning, increased right ventricular uptake, and increased lung uptake especially with Tl-201. A normal perfusion on exercise SPECT MPI was associated with a low event rate (<1% per year) in subjects with a low or intermediate DTS, compared in subjects with an intermediate DTS and high-risk SPECT variables or those with a high DTS and a normal perfusion [48].

#### *3.2.4. Strengths and limitations*

The strengths and limitations of SPECT using 99mTc-labeled sestamibi or tetrofosmin are shown in Table 6.

#### *3.2.5. Case example 2*

A 60-year-old female with hypertension presented with atypical chest pain. Pre-test probabil‐ ity of CAD is intermediate. Pharmacological (dipyridamole) SPECT MPI was performed due to an uninterpretable resting ECG that showed a LBBB. The test is considered appropriate based on the AUC. SPECT MPI demonstrated normal perfusion (Figure 2).


**Table 6.** Strengths and limitations of 99mTc-labelled sestamibi or tetrofosmin SPECT MPI

For pharmacological stress SPECT MPI, contraindications include asthma or COPD with active wheezing, 2nd or 3rd degree AV block without a pacemaker or sick sinus syndrome, systolic BP <90mmHg,use ofmethylxantines (e.g., aminophylline or caffeine) <12hours, knownhypersen‐ sivity to the vasodilator, acute myocardial infarction (MI) and acute coronary syndrome [33].

The diagnostic accuracy of all noninvasive tests is subject to the post-test referral bias, also known as the verification bias. This increases the sensitivity and reduces the specificity of the test. A normalcy rate is used as a surrogate for specificity. It is defined as the percentage of normal perfusion scans in patients with a low likelihood (<10%) of CAD based on the results of clinical and ECG stress tests [38–40]. A pooled analysis of 4,480 patients with known or suspected CAD showed that exercise SPECT MPI had a mean sensitivity of 87% and a specificity of 73% for the detection of >50% stenosis [41]. The normalcy rate of SPECT MPI is around 84% [38]. Standard MPI studies include a combination of stress and rest protocols. The use of a stress-only protocol with a "normal" stress study can reduce the radiation dose by 40% and detected similar event rates [42–44]. A "normal" stress-only study is defined as homogenous perfusion, summed stress score (SSS) of <3, normal left ventricular (LV) cavity

The prognostic value of a normal pharmacological stress MPI test is independent of the radiopharmaceutical used [45]. A meta-analysis of 19 SPECT MPI studies comprising 39,000 patients demonstrated a low annual event rate of 0.6%, for hard cardiac events such as cardiac death or nonfatal MI in patients with a normal test result [46]. However, the prognostic value is based on the studied population. For example, the annual event rate among individuals with a normal test result was higher among diabetic patients than in non-diabetic subjects (0.5% vs. 1.7%, respectively, p < 0.005). Diabetic patients with a LV ejection fraction (LVEF) of ≤ 45% had the worst outcome [47]. High-risk MPI variables include large perfusion defect size and extent, transient ischemic dilatation (TID), post stress stunning, increased right ventricular uptake, and increased lung uptake especially with Tl-201. A normal perfusion on exercise SPECT MPI was associated with a low event rate (<1% per year) in subjects with a low or intermediate DTS, compared in subjects with an intermediate DTS and high-risk SPECT variables or those with

The strengths and limitations of SPECT using 99mTc-labeled sestamibi or tetrofosmin are shown

A 60-year-old female with hypertension presented with atypical chest pain. Pre-test probabil‐ ity of CAD is intermediate. Pharmacological (dipyridamole) SPECT MPI was performed due to an uninterpretable resting ECG that showed a LBBB. The test is considered appropriate

based on the AUC. SPECT MPI demonstrated normal perfusion (Figure 2).

*3.2.3. Diagnostic and prognostic accuracy*

44 Coronary Artery Disease - Assessment, Surgery, Prevention

size, function, and wall motion [43].

a high DTS and a normal perfusion [48].

*3.2.4. Strengths and limitations*

in Table 6.

*3.2.5. Case example 2*

**Figure 2.** Stress (top row) & Rest (bottom row) images in the short axis (SA), horizontal long axis (HLA) and vertical long axis (VLA) show normal homogenous tracer uptake. Gated images showed normal left ventricular ejection frac‐ tion (not shown). This is a normal SPECT MPI study.

#### *3.2.6. Case example 3*

 

͵ǣ

ǡͶǤ 

 

A 58-year-old man with a history of hyperlipidemia presented with typical angina (Canadian Cardiovascular Society Class 2). Pre-test probability of CAD: High. He underwent exercise SPECT MPI, which is considered an appropriate test based on the AUC. Baseline ECG showed sinus rhythm (Figure 3A) and resting BP of 140/90 mmHg. At 2 minutes in Bruce protocol, he developed significant ST-segment depression (Figure 3B). A significant BP drop from 140/90 mmHg to 80/60 mmHg during exercise was present and the test was terminated. The ECG changes persisted 5:50 minutes into recovery, and BP gradually returned to baseline. He achieved a total of 5 METS. He remained asymptomatic of chest pain. Calculated DTS was -17 (high risk). SPECT MPI findings as described in Figure 3C. He was referred for ICA (Figure 3D).

ȋȌǤȋȌȀ ȋȌǤȋͳǤʹͳȌǤ ǤͻͲΨǡͻͲΨǡͻͲΨǡ ͳͲͲΨǡͳͲͲΨǡͳͲͲΨǡ Ǥǡ ȋǤǤȌǡǡǡ Ǥ  ̴͵ǣȋȌȋ ȌǤ **Figure 3.** (a) Resting ECG showed normal sinus rhythm (b) ECG showed 3 mm ST segment depression in leads 1, aVF and V3, and a maximum of 4 mm ST segment depression leads II, V4 to V6. (c) Stress (top row), Rest (bottom row) in the SA slices: Mild reduction in tracer uptake in the mid to distal anterolateral walls (arrow) which normalized at rest. This was consistent with mild ischemia in the left circumflex (LCX) artery/ left anterior descending artery territory (LAD). TID is defined as the ratio of ungated LV volumes at stress and rest was present (1.21). TID is due to extensive subendocardial ischemia post stress that resolves on the rest images. ICA showed distal left main 90%, ostial LAD 90%, first diagonal 90%, first obtuse marginal 100%, second obtuse marginal 100%, right coronary artery 100%, and subse‐ quently referred for CABG. This example clearly illustrates, despite a mildly abnormal perfusion (i.e. mild reduction in myocardial tracer uptake), the associated presence of high risk variables such as TID, and a high DTS, identifies a high risk scan. (d) Significant stenosis in the ostial LAD (orange arrow) and occluded proximal RCA with diffuse disease (purple arrows).

͵ͳǡ ͵ǡͶ

̴͵ǣȋȌǡȋȌǣ

͵ǣ

## **4. Clinical pearls**

*3.2.6. Case example 3*

46 Coronary Artery Disease - Assessment, Surgery, Prevention

 

ȌǤ

͵ǣ

Ǥ 

(purple arrows).

ǡͶǤ 

 

A 58-year-old man with a history of hyperlipidemia presented with typical angina (Canadian Cardiovascular Society Class 2). Pre-test probability of CAD: High. He underwent exercise SPECT MPI, which is considered an appropriate test based on the AUC. Baseline ECG showed sinus rhythm (Figure 3A) and resting BP of 140/90 mmHg. At 2 minutes in Bruce protocol, he developed significant ST-segment depression (Figure 3B). A significant BP drop from 140/90 mmHg to 80/60 mmHg during exercise was present and the test was terminated. The ECG changes persisted 5:50 minutes into recovery, and BP gradually returned to baseline. He achieved a total of 5 METS. He remained asymptomatic of chest pain. Calculated DTS was -17 (high risk). SPECT MPI findings as described in Figure 3C. He was referred for ICA (Figure 3D).

(a) (b)

(c) (d)

**Figure 3.** (a) Resting ECG showed normal sinus rhythm (b) ECG showed 3 mm ST segment depression in leads 1, aVF and V3, and a maximum of 4 mm ST segment depression leads II, V4 to V6. (c) Stress (top row), Rest (bottom row) in the SA slices: Mild reduction in tracer uptake in the mid to distal anterolateral walls (arrow) which normalized at rest. This was consistent with mild ischemia in the left circumflex (LCX) artery/ left anterior descending artery territory (LAD). TID is defined as the ratio of ungated LV volumes at stress and rest was present (1.21). TID is due to extensive subendocardial ischemia post stress that resolves on the rest images. ICA showed distal left main 90%, ostial LAD 90%, first diagonal 90%, first obtuse marginal 100%, second obtuse marginal 100%, right coronary artery 100%, and subse‐ quently referred for CABG. This example clearly illustrates, despite a mildly abnormal perfusion (i.e. mild reduction in myocardial tracer uptake), the associated presence of high risk variables such as TID, and a high DTS, identifies a high risk scan. (d) Significant stenosis in the ostial LAD (orange arrow) and occluded proximal RCA with diffuse disease

ȋȌǤȋȌȀ ȋȌǤȋͳǤʹͳȌǤ ǤͻͲΨǡͻͲΨǡͻͲΨǡ

ȋǤǤȌǡǡǡ

̴͵ǣȋȌȋ

LCX

͵ͳǡ ͵ǡͶ

LAD

RCA

Ǥǡ

̴͵ǣȋȌǡȋȌǣ

͵ǣ

ͳͲͲΨǡͳͲͲΨǡͳͲͲΨǡ


#### **4.1. Positron emission tomography**

#### *4.1.1. Background*

PET MPI has superior temporal and spatial resolution, greater count sensitivity, and accurate attenuation correction compared with SPECT. These features translate into greater diagnostic image quality with fewer equivocal results, especially in obese subjects for example [49, 50]. The latest generation of PET cameras are combined with CT usually with ≥16 slices, and provide 3-dimensional imaging. Integrated PET (emission scan)/CT (transmission scan) systems facilitate sequential scanning, faster acquisition of transmission images, and enable functional and anatomic assessments in a single study [51–53]. These features make PET a useful clinical choice, but widespread utilization is limited by cost.

#### *4.1.2. Principles*

PET imaging is based on the detection of positron emission from radionuclide decay. After emission, the positron travels a short distance before it collides with an electron, resulting in mutual annihilation. This annihilation results in the production of 2 Y photons of 511 keV that travel in nearly opposite directions. The simultaneous detection of both photons by the detector of a PET camera is called coincidence detection [51–53].

Examples of cardiac PET radiotracers include rubidium-82 (82Rb; t½ = 76 s), nitrogen-13 ammonia (13N- Ammonia; t½ = 9.96 min), and oxygen-15 water (15O-H2O; t ½ = 2 min). A shorter t½ allows a lower radiation dose per test. Radiotracers are produced by a cyclotron, except for 82Rb, which is eluted from a strontium-82 (82Sr)/82Rb generator.

Ischemia is detected in the same way as for SPECT MPI. The fundamental difference in uptake kinetics of radiotracers used in SPECT and PET explain the superior sensitivity of PET for the detection of CAD. The ideal radiotracer is 15O-H2O, which has linear uptake by myocardial tissue with increasing blood flow and "no roll-off" phenomenon. 99mTc- labeled tracers have a much lower extraction fraction. At high MBF, 99mTc- labeled tracers are characterized by a rolloff phenomenon to a greater degree, unlike PET radiotracers [54]. At high MBF levels during stress, the relative difference in myocardial tracer uptake may be reduced leading to an underestimation of the regional flow heterogeneity between normal and diseased coronary arteries. With dynamic imaging of the tracer kinetics in PET MPI, it is feasible to measure the myocardial flow reserve (MFR), which is defined as the ratio of absolute MBF during stress compared to MBF at rest. An abnormal global MFR is indicative of diffuse atherosclerosis or microvascular dysfunction.

#### *4.1.3. Diagnostic and prognostic accuracy*

PET MPI showed superior sensitivity and diagnostic accuracy, compared to SPECT [55, 56]. A meta-analysis of three different stress perfusion modalities determined the pooled sensitiv‐ ity, specificity, and area under the curve of SPECT (88%, 61%, and 0.86, respectively), CMR (89%, 76%, and 0.90, respectively) and PET (84%, 81%, and 0.92, respectively) for the detection of CAD [57].

PET MPI can be used to determine prognosis, revealing annual event rates for cardiac death and nonfatal MI of 0.4%, 2.3%, and 7.0% for SSS values of <4 (normal), 4–7, and >8, respectively [58]. A similar trend was noted when subjects were stratified according to the percentage of ischemic LV myocardium, with a relative hazard for cardiac death of 2.3%, 4.2%, and 4.9% for the strata of 0.1–9.9%, 10–19.9%, and ≥ 20%, respectively [59]. Parameters that measure the extent of ischemia have been shown to guide the decisions regarding revascularization [60]. An abnormal global MFR (<2.0) was associated with an increased incidence of MACE com‐ pared with a normal MFR (1.3% vs. 4.7%, p = 0.03), independent of a normal perfusion and a CS of 0 [61].

#### *4.1.4. Strengths and limitations*

The strengths and limitations of PET MPI are shown in Table 7.


**Table 7.** Strengths and limitations of PET MPI

#### *4.1.5. Case example 4*

off phenomenon to a greater degree, unlike PET radiotracers [54]. At high MBF levels during stress, the relative difference in myocardial tracer uptake may be reduced leading to an underestimation of the regional flow heterogeneity between normal and diseased coronary arteries. With dynamic imaging of the tracer kinetics in PET MPI, it is feasible to measure the myocardial flow reserve (MFR), which is defined as the ratio of absolute MBF during stress compared to MBF at rest. An abnormal global MFR is indicative of diffuse atherosclerosis or

PET MPI showed superior sensitivity and diagnostic accuracy, compared to SPECT [55, 56]. A meta-analysis of three different stress perfusion modalities determined the pooled sensitiv‐ ity, specificity, and area under the curve of SPECT (88%, 61%, and 0.86, respectively), CMR (89%, 76%, and 0.90, respectively) and PET (84%, 81%, and 0.92, respectively) for the detection

PET MPI can be used to determine prognosis, revealing annual event rates for cardiac death and nonfatal MI of 0.4%, 2.3%, and 7.0% for SSS values of <4 (normal), 4–7, and >8, respectively [58]. A similar trend was noted when subjects were stratified according to the percentage of ischemic LV myocardium, with a relative hazard for cardiac death of 2.3%, 4.2%, and 4.9% for the strata of 0.1–9.9%, 10–19.9%, and ≥ 20%, respectively [59]. Parameters that measure the extent of ischemia have been shown to guide the decisions regarding revascularization [60]. An abnormal global MFR (<2.0) was associated with an increased incidence of MACE com‐ pared with a normal MFR (1.3% vs. 4.7%, p = 0.03), independent of a normal perfusion and a

microvascular dysfunction.

of CAD [57].

CS of 0 [61].

accuracy in the obese

SPECT study

Absolute MBF

Assessment of calcium on CT

**Table 7.** Strengths and limitations of PET MPI

*4.1.4. Strengths and limitations*

The strengths and limitations of PET MPI are shown in Table 7.

Robust attenuation correction Not widely available Improved diagnostic Radiation exposure

Short half life of tracers Pharmacological stress

Alternative for an equivocal Unable to assess exercise capacity

**Strengths Limitations** High spatial resolution Expensive

*4.1.3. Diagnostic and prognostic accuracy*

48 Coronary Artery Disease - Assessment, Surgery, Prevention

A 68-year-old female with obesity, hypertension and an ex-smoker presented with chest pain. Noncontrast CT showed a CS of 715. CCTA was not performed. A persantine 82Rb PET MPI was performed (Figure 4A) and gated images were acquired at stress and rest (Figure 4B). Coronary calcium was visualised on CT and MBF was abnormal (Figure 4C).

 Ͷǣ ǡȋȌǡǡ ǤǤ ȋͳǡ͵ǡͷȌȋʹǡͶǡȌǤ  Ͷǣ ȋ͵Ȍȋ͵ȌǤ Ǥ ǡȋȌȋ ȌǤǡǤ Ǧ͵ͲͶͷǤ  Ͷǣ ȋȌǤȋǡȌǡȋǡȌǤ ȋ αͲǤͶȀȀȌǡ Ǥ ͳǤͳǡͲǤͻͻȋȌǤ ȋα ͲǤͻͺǡαͳǤ͵ͺǡαͳǤʹͶȌǡǤ Ǥ ǦȋȌǡǡǤ **Figure 4.** (a) Following stress, there is mild to moderate reduction in tracer uptake in the apical (distal) segments of the septum and anterior wall, and apex, which predominantly improves at the rest. This is consistent with a moderate sized area of moderate ischemia in the distal LAD territory. Stress images (rows 1, 3, and 5) and rest images (rows 2, 4, and 6). (b) Gated stress (top 3 rows) and gated rest (bottom 3 rows) during PET MPI. Normal LVEF and wall motion at rest. Following stress, there is mild hypokinesis in the apical (distal) septum and apex consistent with stress induced ischemia (red arrows). Stress acquisition during PET MPI is acquired at peak stress, and hence presence of regional wall motion is specific for ischemia. This differs from post-stress acquisition during SPECT which is delayed by 30 to 45 minutes. (c) Polar map shows the extent of reversible ischemia in the LAD territory (red arrow). CT shows presence of coronary calcification in the RCA (top, blue arrow), LCX and LAD (bottom, blue arrow). MBF demonstrate marked reduction in stress flow in the areas of reduced perfusion (global MBF= 0.74 ml/min/g), as depicted by the polar map which is labelled as stress Rubidium. Global MFR is reduced at 1.16, and 0.99 (after correction for RPP). There is re‐ duced regional MFR in all three coronary artery territories (LAD= 0.98, LCX= 1.38, and RCA= 1.24), as seen on the polar map labelled as Reserve. The MFR finding is suggestive of underlying triple vessel disease. Measurement of MBF is corrected for rate-pressure product (RPP), and depicted in the shade of gray, next to the uncorrected values.

#### *4.1.6. Clinical pearls*


#### **4.2. Stress echocardiography**

#### *4.2.1. Background*

SE is a robust, versatile imaging modality for CAD assessment. Advances in digital image acquisition, strain imaging, tissue harmonics, and contrast agents have increased the use of SE. In particular, the advances in tissue harmonics and contrast agents have greatly improved the visualization of endocardial borders, and diagnostic accuracy of SE for the detection of wall motion abnormalities (WMA) [62–64]. Because SE is highly operator- dependent, those who perform the test should have adequate training and experience to meet the level of competency required for performing and interpreting this test [65].

#### *4.2.2. Principles*

The fundamental principle for the detection of myocardial ischemia is the development of new or worsening WMA during stress [65, 66]. Based on the ischemic cascade, WMA appear after perfusion abnormalities and precede the manifestation of ECG changes and symptoms. Thus, SE has decreased sensitivity and superior specificity compared to perfusion-based imaging for the detection of CAD. Images are acquired at rest and stress. The images are compared on a four-screen setup side-by-side for WMA, LV cavity size and LVEF.

The stress component may be exercise (treadmill or bicycle) or pharmacologically (dobutamine or dipyridamole). An adequate level of stress using exercise requires a minimum target of 85% of the age-predicted HR, and is preferably symptom-limiting, considering the additional prognostic value of the subject's exercise capacity. Bicycle SE (supine or upright ergometry) allows simultaneous image acquisition during peak stress, and the measurement of Doppler information. During treadmill exercise SE, the stress images are acquired within 60–90 seconds post stress. The contraindications and indications for terminating an exercise SE are similar to those of exercise ECG.

Pharmacological SE using dobutamine is preferred over dipyridamole for wall motion assessment [65], although either stressor can be used [67]. The dose for dipyridamole in SE is higher at 0.84 mg/kg over 10 minutes, compared to the dose used in nuclear perfusion imaging at 0.14 mg/kg/min over 4 minutes. A typical dobutamine infusion rate is at 5 micrograms/kg/ min and increasing at a 3-minute interval to 10, 20, 30, and 40 micrograms/kg/min, aiming to achieve 85% of the age-predicted target HR. Atropine can be used to achieve the desired target HR. The induction of ischemia is due to an increase in myocardial oxygen demand. Deforma‐ tion analysis using the tissue velocity index (TVI) and two-dimensional speckle tracking (ST) based strain imaging have been proposed using dobutamine SE [68, 69]. However, TVI- and ST-based analysis conferred no additional diagnostic value over WMA for detection of ischemia [70].

Indications for terminating the test include the de novo or worsening of WMA, significant arrhythmias, hypotension, severe hypertension, and intolerable symptoms. For image interpretation, normal wall motion is defined as normal wall thickening and endocardial excursion. Visual assessment of wall motion can be categorized and scored as follows: 1 = normal, 2 = hypokinetic, 3 = akinetic, 4 = dyskinetic or aneursymal. Please note that a score of "5" is no longer applicable. Each segment using the 16 segment LV model (i.e., apical cap not included) is scored and the wall motion score index (WMSI) is calculated. A normal WMSI equals to "1". WMSI = Total wall motion score/ Total number segments visualised. Unlike in MPI that utilizes the 17 segment model, for WMA assessment the 16 segment model is preferred. A WMSI of >1.7 corresponds to a perfusion defect of >20% on MPI. If feasible, the right ventricular wall motion should be assessed and presence of WMA suggest greater extent of CAD.

Normal: normal at rest; hyperdynamic at stress

Ischemia: normal at rest; inducible new or worsening WMA (hypokinesis, akinesis or dyski‐ nesis) at stress

Infarction: fixed abnormality at rest and stress

#### *4.2.3. Diagnostic and prognostic accuracy*

*4.1.6. Clinical pearls*

CAD.

keV Y photons.

50 Coronary Artery Disease - Assessment, Surgery, Prevention

transmission scans.

**4.2. Stress echocardiography**

*4.2.1. Background*

*4.2.2. Principles*

those of exercise ECG.

**1.** PET imaging is based on positron annihilation and coincidence detection of paired 511

**2.** PET has higher diagnostic accuracy compared to SPECT in the assessment of obstructive

**3.** PET is an excellent choice for imaging in the obese and those with an equivocal SPECT. **4.** Abnormal global MFR suggest diffuse atherosclerosis or microvascular dysfunction.

**5.** Patient motion can result in significant misregistration artifact between the emission and

SE is a robust, versatile imaging modality for CAD assessment. Advances in digital image acquisition, strain imaging, tissue harmonics, and contrast agents have increased the use of SE. In particular, the advances in tissue harmonics and contrast agents have greatly improved the visualization of endocardial borders, and diagnostic accuracy of SE for the detection of wall motion abnormalities (WMA) [62–64]. Because SE is highly operator- dependent, those who perform the test should have adequate training and experience to meet the level of

The fundamental principle for the detection of myocardial ischemia is the development of new or worsening WMA during stress [65, 66]. Based on the ischemic cascade, WMA appear after perfusion abnormalities and precede the manifestation of ECG changes and symptoms. Thus, SE has decreased sensitivity and superior specificity compared to perfusion-based imaging for the detection of CAD. Images are acquired at rest and stress. The images are compared on a

The stress component may be exercise (treadmill or bicycle) or pharmacologically (dobutamine or dipyridamole). An adequate level of stress using exercise requires a minimum target of 85% of the age-predicted HR, and is preferably symptom-limiting, considering the additional prognostic value of the subject's exercise capacity. Bicycle SE (supine or upright ergometry) allows simultaneous image acquisition during peak stress, and the measurement of Doppler information. During treadmill exercise SE, the stress images are acquired within 60–90 seconds post stress. The contraindications and indications for terminating an exercise SE are similar to

Pharmacological SE using dobutamine is preferred over dipyridamole for wall motion assessment [65], although either stressor can be used [67]. The dose for dipyridamole in SE is

**6.** Normal PET perfusion and MFR confer an excellent prognosis.

competency required for performing and interpreting this test [65].

four-screen setup side-by-side for WMA, LV cavity size and LVEF.

The development of WMA depends on the extent of stenosis detected by ICA, and occurs at a cut-off relative diameter of 54% for exercise SE, 58% for dobutamine SE, and 60% for dipyri‐ damole SE [71]. The sensitivities and specificities for the detection of CAD were 85% and 77%, respectively, for exercise SE, 80% and 86%, respectively for dobutamine SE, and 78% and 91%, respectively, for dipyridamole SE [71, 72]. SE had a similar diagnostic accuracy to that of SPECT MPI for the detection of CAD [73, 74]. However, because of its greater specificity, SE showed a better discriminatory capacity for the diagnosis of LM and TVD [75]. For patients presenting at an emergency department with chest pain, nondiagnostic ECG, and negative cardiac biomarkers, the overall sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of SE for the diagnosis of CAD were 90%, 92%, 78% and 97%, respec‐ tively. Moreover, exercise SE had a better specificity than dobutamine SE [76].

A normal exercise SE result is associated with an excellent prognosis with an annual event rate of cardiac events of 0.54% [77]. The best discriminator of increased risk of cardiac events was WMSI ≥ 1.25 and ≤ 6 METs in both genders [78]. There was no difference in the prognostic value of dobutamine SE and dipyridamole SE [79, 80]. Therefore, the choice between these techniques may depend on institutional practices. With regards to LV cavity size, an abnormal stress LV end-systolic volume (LVESV) (i.e., no change or an increase) was associated with an increase risk of cardiac events compared with a decrease in the LVESV (2.9% vs. 1.6%) [81].

#### *4.2.4. Strengths and limitations*

The strengths and limitations of SE are shown in TABLE 8.


**Table 8.** Strengths and limitations of stress echocardiography

#### *4.2.5. Case example 5*

65-year-old woman with rheumatoid arthritis and hypertension presented with heartburn and indigestion of increasing severity for 1 month. An exercise SE with contrast was performed at rest. She experienced a similar episode of heartburn and nausea at 2 minutes into exercise. The test was stopped and stress images were acquired immediately (Figure 5A). Rest and stress images are shown as still frames captured at end-systole (Figure 5B i-iv). The patient was admitted in view of a positive SE at low workload. ICA showed ostial LAD 70%, mid LAD 90%, proximal LCX 80%, and mid RCA 100%. She was referred for CABG.

#### *4.2.6. Clinical pearls*


Noninvasive Imaging for the Assessment of Coronary Artery Disease http://dx.doi.org/10.5772/61502 53

value of dobutamine SE and dipyridamole SE [79, 80]. Therefore, the choice between these techniques may depend on institutional practices. With regards to LV cavity size, an abnormal stress LV end-systolic volume (LVESV) (i.e., no change or an increase) was associated with an increase risk of cardiac events compared with a decrease in the LVESV (2.9% vs. 1.6%) [81].

COPD)

ischemia

65-year-old woman with rheumatoid arthritis and hypertension presented with heartburn and indigestion of increasing severity for 1 month. An exercise SE with contrast was performed at rest. She experienced a similar episode of heartburn and nausea at 2 minutes into exercise. The test was stopped and stress images were acquired immediately (Figure 5A). Rest and stress images are shown as still frames captured at end-systole (Figure 5B i-iv). The patient was admitted in view of a positive SE at low workload. ICA showed ostial LAD 70%, mid LAD

90%, proximal LCX 80%, and mid RCA 100%. She was referred for CABG.

**1.** Normal wall motion at rest does not rule out the presence of obstructive CAD.

**2.** Abnormal regional wall motion can be seen in the presence of ischemic or nonischemic

detection of posterior wall

*4.2.4. Strengths and limitations*

52 Coronary Artery Disease - Assessment, Surgery, Prevention

Contrast echo improves endocardial definition

stress

Exercise and pharmacological

Medium procedural time

*4.2.5. Case example 5*

*4.2.6. Clinical pearls*

etiology.

The strengths and limitations of SE are shown in TABLE 8.

**Strengths Limitations**

Cheap Poor echo window (obese,

Widely available Reduced sensitivity in

No radiation Foreshortened LV apex Portable Operator dependent

**Table 8.** Strengths and limitations of stress echocardiography

**Figure 5.** (a) ECG during episode of chest pain demonstrated diffuse 2 mm ST segment depression in leads II, III, aVF, V2 to V6 and ST elevation in aVR. (b) SE using contrast showing images captured at end-systole during rest and stress. The LV in different views are illustrated in the following sequence from top to bottom (i) parasternal long axis, (ii) 2 chamber, (iii) short axis, (iv) apical, demonstrate regional WMA in the anterior, anteroseptum, apex and lateral walls. Note the stress induced LV cavity dilatation, at peak stress which is suggestive of underlying triple vessel disease.


#### **4.3. Coronary computed tomography angiography**

#### *4.3.1. Background*

CCTA has come a long way from electron beam CT to the present multidetector CT system, with continually increasing detector rows (i.e., = slices) from 4-slices to 320-slices in some systems. A 64-slice CT system is the minimum requirement for coronary imaging. Other developments include iterative reconstruction, dual-source CT, prospective scanning, tube current modulation, and a z-flying focal spot [82]. These innovations have led to significant improvements in spatial and temporal resolution, radiation dosimetry, and image quality, which are prerequisites for coronary imaging. The ability to visualize subclinical atheroscle‐ rosis and characterize plaques has led to techniques for quantifying the extent of the athero‐ sclerotic burden [83].

#### *4.3.2. Principles*

CCTA can determine the extent of coronary atherosclerosis and estimate the severity of coronary artery stenosis. The major components of a CT scanner include the table, X-ray tube, detector array and the gantry that rotates the X-ray tube and detector array around the patient. The ability to image a beating heart requires a high temporal resolution, which is defined as the time taken to obtain a complete data set for image reconstruction. The typical temporal resolution of CT is 280–420 ms. Data covering 180° are needed to construct one image, termed "half scan reconstruction". Accordingly, the temporal resolution of a single-source CT system is half the time required for the gantry to rotate 360°. For a dual-source CT (DSCT), the temporal resolution is one-quarter of the gantry rotation time, because data covering 90° are sufficient for image acquisition. A DSCT has a temporal resolution of 75 ms (compared to 20–30 ms for ICA) [84].

The ability for CT to discriminate two structures is called the spatial resolution, which is measured in line pairs per centimeter (lp/cm). The typical spatial resolution of CT is 10 lp/cm, which is equivalent to 0.5 mm (compared to 0.1 mm for ICA). CT data are acquired as isotropic voxels enabling the images to be viewed in multiple planes with similar spatial resolutions [85]. The spatial resolution of CT is less than ideal to accurately quantitate the degree of stenosis, hence a grading method is used [86]:

Normal: 0%; Minimal: <25%; Mild: 25–49%; Moderate: 50–69%; Severe: 70–99%; Occluded: 100%.

ECG gating is essential to minimize cardiac motion by synchronizing image acquisition to the cardiac cycle. There are two types of scanning modes. In prospective ECG-triggered scanning, data acquisition is triggered by the R waves on the ECG. Its advantage is the low radiation dose, typically 3–5 mSv. Disadvantages include image reconstruction limited to the desired phase and functional evaluation is not possible. In retrospective ECG-gated scanning, data acquisition are acquired throughout the cardiac cycle and the ECG signal is simultaneously recorded with the raw data. Advantages include image reconstruction can be performed at any point in the cardiac cycle and it is useful in patients with arrhythmias. Its main disad‐ vantage is the high radiation dose, typically 10–12 mSv because the tube current remains "on" throughout image acquisition. The application of tube current modulation can reduce the radiation dose. The best phase for image acquisition is mid-diastole, when the heart is moving the least.

The images to be acquired are the initial scout image, which defines the scan length, followed by a non-contrast calcium scan and finally the contrast-enhanced CT images. The average scan length for native coronary artery imaging is 12–14 cm. Using a 64-slice scanner, images are acquired over a few cardiac cycles, as opposed to a 320-slice scanner where one cardiac cycle is sufficient because of the larger volume covered. Data can be reconstructed in two or three dimensions, although the two-dimensional axial views should serve as a reference for image interpretation.

Contrast enhancement in the ascending aorta can be tracked using a test bolus or automated bolus tracking. Both methods are acceptable and the choice between them may depend on the institutional protocol. About 80–100 ml of iodinated contrast is typically used. The target HR of 50–65 bpm can be achieved by oral or intravenous β-adrenoceptor blockers. Nitroglycerin spray is recommended to improve image quality by inducing coronary vasodilatation.

Calcium is measured using the Agatston score (i.e., CS), which correlates with the extent of the atherosclerotic plaque burden. Screening for CS are recommended in two groups of asymptomatic patients: (1) patients with low global coronary heart disease (CHD) risk and a family history of premature CAD and (2) patients with intermediate global CHD risk [87]. CT perfusion is still research-based, and will not be discussed in this review.

#### *4.3.3. Diagnostic and prognostic accuracy*

**3.** The absence of radial motion of the mitral valve annulus can result in a reduction in motion of the basal inferior or inferoseptal segments resulting in a false-positive study.

CCTA has come a long way from electron beam CT to the present multidetector CT system, with continually increasing detector rows (i.e., = slices) from 4-slices to 320-slices in some systems. A 64-slice CT system is the minimum requirement for coronary imaging. Other developments include iterative reconstruction, dual-source CT, prospective scanning, tube current modulation, and a z-flying focal spot [82]. These innovations have led to significant improvements in spatial and temporal resolution, radiation dosimetry, and image quality, which are prerequisites for coronary imaging. The ability to visualize subclinical atheroscle‐ rosis and characterize plaques has led to techniques for quantifying the extent of the athero‐

CCTA can determine the extent of coronary atherosclerosis and estimate the severity of coronary artery stenosis. The major components of a CT scanner include the table, X-ray tube, detector array and the gantry that rotates the X-ray tube and detector array around the patient. The ability to image a beating heart requires a high temporal resolution, which is defined as the time taken to obtain a complete data set for image reconstruction. The typical temporal resolution of CT is 280–420 ms. Data covering 180° are needed to construct one image, termed "half scan reconstruction". Accordingly, the temporal resolution of a single-source CT system is half the time required for the gantry to rotate 360°. For a dual-source CT (DSCT), the temporal resolution is one-quarter of the gantry rotation time, because data covering 90° are sufficient for image acquisition. A DSCT has a temporal resolution of 75 ms (compared to 20–30 ms for

The ability for CT to discriminate two structures is called the spatial resolution, which is measured in line pairs per centimeter (lp/cm). The typical spatial resolution of CT is 10 lp/cm, which is equivalent to 0.5 mm (compared to 0.1 mm for ICA). CT data are acquired as isotropic voxels enabling the images to be viewed in multiple planes with similar spatial resolutions [85]. The spatial resolution of CT is less than ideal to accurately quantitate the degree of

Normal: 0%; Minimal: <25%; Mild: 25–49%; Moderate: 50–69%; Severe: 70–99%; Occluded:

ECG gating is essential to minimize cardiac motion by synchronizing image acquisition to the cardiac cycle. There are two types of scanning modes. In prospective ECG-triggered scanning, data acquisition is triggered by the R waves on the ECG. Its advantage is the low radiation

**4.** A common cause of a false negative test is a suboptimal stress level.

**4.3. Coronary computed tomography angiography**

54 Coronary Artery Disease - Assessment, Surgery, Prevention

*4.3.1. Background*

sclerotic burden [83].

*4.3.2. Principles*

ICA) [84].

100%.

stenosis, hence a grading method is used [86]:

There is extensive literature describing the diagnostic and prognostic value of CCTA. The main advantage of CCTA is the ability to exclude disease from the differential diagnosis since CCTA has an excellent NPV. The ACCURACY trial demonstrated sensitivity, specificity, PPV, and NPV of 95%, 83%, 64%, and 99%, respectively, for the detection of ≥ 50% stenosis, and 94%, 83%, 48%, and 99%, respectively, for the detection of ≥70% stenosis. The NPV was high in patient- and vessel-level analyses [88]. The low PPV is due to its tendency to overestimate stenosis, while the presence of artefacts lead may lead to a false positive test. CCTA is considered appropriate for patients with low or intermediate pre-test probabilities of CAD, and a negative scan reliably indicates the absence of significant CAD. However, CCTA is of limited clinical value and functional imaging tests are more appropriate in patients with a high pre-test probability [89]. The presence of severe coronary calcification (CS >400) can reduce the diagnostic accuracy by overestimating the severity of stenosis owing to blooming artefacts [88]. Although there is no specific cut-off level to cancel a CCTA, CS of > 600–1000 is typically used for this purpose, considering the high likelihood of a nondiagnostic study.

The use of CCTA in the emergency department resulted in a shorter hospital stay, increased discharge rate [90, 91], and reduced time to CAD diagnosis [90], while patients with a negative scan had an excellent prognosis [92]. The all-cause mortality rate was 0.65% for normal CCTA, 1.99% for <50% stenosis, 2.9% for ≥50% stenosis, and 4.95% for LM ≥50%, TVD ≥70%, or two vessel disease with proximal LAD disease [93]. The excellent prognosis of a negative CCTA result was seen in other large series of patients [94]. The "warranty period" of a normal CCTA is ~7 years [95]. Coronary calcium has prognostic value beyond traditional risk factors with a hazard ratio of 3.89, 7.08, and 6.84 for CS of 1–100, 101–300, and ≥300, respectively, for coronary events [96].

#### *4.3.4. Strengths and limitations*


The strengths and limitations of CCTA are shown in Table 9.

**Table 9.** Strengths and limitations of CCTA

#### *4.3.5. Case example 6*

A 71-year-old female with hypertension, hyperlipidemia, and an ex- smoker presented with atypical chest pain and LBBB. CCTA was performed. Findings included a CS of 269, and ≥70% stenosis (calcified and noncalcified plaque) with positive remodeling in the mid LAD (Figure 6). There was mild disease in ostium of the RCA. She was referred for ICA.

#### *4.3.6. Case example 7*

A 57-year-old man with hypertension. CCTA showed a CS of 524 with an occluded proximal RCA. He was referred for ICA and underwent percutaneous coronary intervention to the RCA (Figure 7 A-E).

ǃ

The use of CCTA in the emergency department resulted in a shorter hospital stay, increased discharge rate [90, 91], and reduced time to CAD diagnosis [90], while patients with a negative scan had an excellent prognosis [92]. The all-cause mortality rate was 0.65% for normal CCTA, 1.99% for <50% stenosis, 2.9% for ≥50% stenosis, and 4.95% for LM ≥50%, TVD ≥70%, or two vessel disease with proximal LAD disease [93]. The excellent prognosis of a negative CCTA result was seen in other large series of patients [94]. The "warranty period" of a normal CCTA is ~7 years [95]. Coronary calcium has prognostic value beyond traditional risk factors with a hazard ratio of 3.89, 7.08, and 6.84 for CS of 1–100, 101–300, and ≥300, respectively, for coronary

events [96].

*4.3.4. Strengths and limitations*

56 Coronary Artery Disease - Assessment, Surgery, Prevention

**Table 9.** Strengths and limitations of CCTA

*4.3.5. Case example 6*

*4.3.6. Case example 7*

(Figure 7 A-E).

The strengths and limitations of CCTA are shown in Table 9.

value of assessing stenosis High temporal resolution Radiation exposure High spatial resolution Morbidly obese Visualize coronary anatomy Arrhythmias

Excellent negative predictive High calcium limits accuracy

Short procedural time Allergy to iodinated contrast

Follow breath hold instruction Renal impairment (>2.0 mg/dl)

No functional assessment

A 71-year-old female with hypertension, hyperlipidemia, and an ex- smoker presented with atypical chest pain and LBBB. CCTA was performed. Findings included a CS of 269, and ≥70% stenosis (calcified and noncalcified plaque) with positive remodeling in the mid LAD (Figure

A 57-year-old man with hypertension. CCTA showed a CS of 524 with an occluded proximal RCA. He was referred for ICA and underwent percutaneous coronary intervention to the RCA

6). There was mild disease in ostium of the RCA. She was referred for ICA.

Assessment of calcium score Heart rate control

**Strengths Limitations**

**Figure 6.** First two images of the curved multiplanar reformatted (MPR) and 2D-axial views demonstrate ≥70% steno‐ sis of the LAD (blue arrows). The third image (curved MPR) of the RCA showing a calcified plaque with minimal stenosis at the ostium (blue arrow).

**Figure 7.** Axial slices from cranial to caudal demonstrate (A) contrast enhanced lumen in the proximal RCA, (B) absent of contrast enhancement and (C) reappearance of the contrast in the RCA. (D) Axial maximum intensity projection of the RCA demonstrate an occluded vessel and correlates with ICA (E).

#### *4.3.7. Clinical pearls*


#### **4.4. Stress cardiac magnetic resonance**

#### *4.4.1. Background*

CMR has recently emerged clinically as a highly versatile technique with superior spatial and temporal resolution. The development of high field strength magnets (3T) and rapid imaging techniques such as gradient echo, echo-planar, and balanced steady-state free precession, have contributed to the feasibility of MR perfusion. The type of pulse sequence or hybrid sequences affect the contrast-to-noise ratio and the suspectibility to artifacts, which can affect image quality. In the context of MPI, CMR is a promising tool in parallel with well-established and validated modalities such as SPECT and PET MPI. The advantage of MR is its superior sensitivity for the detection of subendocardial perfusion defects without radiation exposure. The paramagnetic property of gadolinium (Gd)-based contrast agents can alter the local magnetic field in the tissue, which enables differentiation between normally and abnormally perfused myocardium. Arterial spin labeling and blood oxygen level-dependent techniques are new advances in perfusion imaging, but are still research-based. The use of coronary magnetic resonance angiography is not well established, except in some highly specialized centers, and will not be discussed here.

#### *4.4.2. Principles*

The fundamental basis of CMR perfusion imaging is the first-pass imaging of contrast transit through the LV myocardium. This exploits the effect of Gd on the T1 relaxation time of myocardial tissue [98]. Gd-based contrast agents are paramagnetic, extracellular agents that are rapidly taken up and rapidly washed out of the normal myocardium, but accumulate in damaged tissues with slower washout kinetics. Gd is highly toxic in its native state. Therefore, Gd chelators (e.g., Gd-DTPA) are used clinically. These agents shorten the T1 and T2 relaxation time constants that represent the decay of the MR signal. However, at low doses, T1 shortening is predominant. During first-pass perfusion, the normal myocardium (i.e., normal perfusion) shows substantial Gd uptake, and appears bright (i.e., hyperintense) owing to a short T1. Ischemic myocardium (i.e., reduced perfusion) shows diminished Gd uptake and appears dark (i.e., hypointense) owing to a long T1 [98].

Similar to pharmacological stress SPECT or PET MPI, stress CMR requires the use of a pharmacological stressor, such as a vasodilator (e.g., adenosine, dipyridamole, and regade‐ noson) or dobutamine. Diseased coronary arteries, exhibit a lower peak myocardial signal intensity and increases in myocardial contrast transit time (e.g., signal upslope, arrival time, time-to-peak signal, and mean transit time)[99]. The difference in signal intensity can be quantitatively, semiquantitatively, or visually evaluated to identify possible perfusion defects. The use of adenosine and visual interpretation are common, and these approaches are discussed in further detail.

Historically, inducible ischemia was only assessed using stress and rest perfusion cine images. This method demonstrated a sensitivity, specificity, and diagnostic accuracy of 88%, 90%, and 89%, respectively, for the detection of significant CAD [100]. The caveat being, in patients with prior MI, the resultant perfusion deficit may include areas of prior infarct and may not reflect true inducible ischemia. Imaging with late Gd-enhancement (LGE) was used to detect prior infarction. A method combining first-pass stress and rest imaging with LGE demonstrated an overall accuracy of 0.88, or 0.96 for one-vessel disease, 0.75 for two-vessel disease, and 0.9 for prior coronary artery bypass graft, in the detection of significant stenosis [101].

A stress CMR study can be interpreted using the following algorithm [99]:

#### **Step 1. Assess for LGE**

*4.3.7. Clinical pearls*

quality.

*4.4.1. Background*

*4.4.2. Principles*

**4.4. Stress cardiac magnetic resonance**

58 Coronary Artery Disease - Assessment, Surgery, Prevention

centers, and will not be discussed here.

**1.** An asymptomatic patient with a CS of 0 has a very low event rate of 0.1% per year [97].

**2.** Prospective ECG-triggered acquisition is preferred in view of the lower radiation dose.

**3.** Regular, low HR and obeying breath-hold instructions are essential for diagnostic image

**4.** Appropriate timing of contrast injection is crucial for optimal enhancement as contrast

**5.** Volume coverage in the z-axis for a 64-slice CT (0.625 mm detector width) is 4 cm (64 x

CMR has recently emerged clinically as a highly versatile technique with superior spatial and temporal resolution. The development of high field strength magnets (3T) and rapid imaging techniques such as gradient echo, echo-planar, and balanced steady-state free precession, have contributed to the feasibility of MR perfusion. The type of pulse sequence or hybrid sequences affect the contrast-to-noise ratio and the suspectibility to artifacts, which can affect image quality. In the context of MPI, CMR is a promising tool in parallel with well-established and validated modalities such as SPECT and PET MPI. The advantage of MR is its superior sensitivity for the detection of subendocardial perfusion defects without radiation exposure. The paramagnetic property of gadolinium (Gd)-based contrast agents can alter the local magnetic field in the tissue, which enables differentiation between normally and abnormally perfused myocardium. Arterial spin labeling and blood oxygen level-dependent techniques are new advances in perfusion imaging, but are still research-based. The use of coronary magnetic resonance angiography is not well established, except in some highly specialized

The fundamental basis of CMR perfusion imaging is the first-pass imaging of contrast transit through the LV myocardium. This exploits the effect of Gd on the T1 relaxation time of myocardial tissue [98]. Gd-based contrast agents are paramagnetic, extracellular agents that are rapidly taken up and rapidly washed out of the normal myocardium, but accumulate in damaged tissues with slower washout kinetics. Gd is highly toxic in its native state. Therefore, Gd chelators (e.g., Gd-DTPA) are used clinically. These agents shorten the T1 and T2 relaxation time constants that represent the decay of the MR signal. However, at low doses, T1 shortening is predominant. During first-pass perfusion, the normal myocardium (i.e., normal perfusion) shows substantial Gd uptake, and appears bright (i.e., hyperintense) owing to a short T1.

non-uniformity in the distal coronary vessels can simulate stenoses.

0.625), and a 320-slice CT (0.5 mm detector width) is 16 cm.

Negative: Move to step 2.

Positive: CAD present.

#### **Step 2. Assess stress perfusion**

Negative: No CAD

Positive: Move to step 3.

#### **Step 3. Assess rest perfusion**

Negative: Inducible ischemia suggestive of CAD.

Positive: Likely artifact\*

\* Common being the Gibbs artifact, which is more pronounced on a 3T scanner. This usually occurs in the phase encoding direction and tends to be transient. If the segment of perfusion defect is also positive for LGE, then inducible ischemia cannot be assessed in the same segment.

An abbreviated adenosine stress CMR protocol [102]:


contrast bolus has transited the LV myocardium, adenosine is stopped. Stress images are acquired for 40 to 50 heart beats.


(Some centers omit the rest perfusion module, if the first pass stress perfusion study is normal).

Adenosine (t ½ = 10 s) is safe and well-tolerated. The potential adverse effects include flushing, chest pain, palpitations, breathlessness, transient episodes of heart block, hypotension, sinus bradycardia, and bronchospasm [102]. The contraindication to adenosine stress CMR (in addition to the general contraindication of any MR study) include known hypersensitivity to adenosine, known or suspected bronchoconstrictive or bronchospastic disease, 2nd or 3rd degree AV block, sinus bradycardia (HR <45 bpm), and systolic BP <90 mmHg [102].

#### *4.4.3. Diagnostic and prognostic accuracy*

Diagnostic accuracy of stress CMR in a population with high prevalence of CAD (57%) showed on an overall sensitivity of 89% and specificity of 80% for the diagnosis of significant obstruc‐ tive CAD. Adenosine-based stress demonstrated better sensitivity than dipyridamole (90% vs. 86%), but with similar specificity (81% vs. 77%) [103]. Adenosine and dobutamine stress CMR have similar sensitivity and specificity [104]. Stress CMR showed no difference in diagnostic accuracy when compared to SPECT MPI for CAD detection [105]. The concordance and accuracy of stress CMR with 320-detector row CT, showed an excellent agreement (92%, kappa value = 0.81) in an intermediate risk cohort [106]. Adenosine perfusion was the most accurate component of the stress CMR study in predicting which patients had significant CAD, compared with resting WMA and LGE [107].

Negative findings on stress CMR are reassuring and associated with annualized event rates of 0.4% for MI and 0.3% for cardiovascular death. In patients with inducible ischemia, the annual event rates for MI and cardiovascular death were 2.6% and 2.8%, respectively. The concomitant presence of LGE was associated with a worse prognosis [108]. Other predictors of cardiac events were resting WMA, inducible WMA and LGE. Patients with inducible WMA (i.e., ischemia) experienced significant benefits from revascularization, compared with patients without inducible WMA (i.e., without ischemia) [109].

#### *4.4.4. Strengths and limitations*

The strengths and limitations of stress CMR are shown in Table 10.


**Table 10.** Strengths and limitations of stress CMR

#### *4.4.5. Case example 8*

contrast bolus has transited the LV myocardium, adenosine is stopped. Stress images are

**3. Rest perfusion module***–* performed after 10 minutes to ensure sufficient clearing of Gd from the blood pool. Reinjection of a second dose of Gd. Rest images are acquired. Slice

**5. Analysis***–* visual interpretation using the 17 segment LV model. The stress and rest cine images are viewed side-by-side using equivalent slices, in addition to the LGE images.

(Some centers omit the rest perfusion module, if the first pass stress perfusion study is normal).

Adenosine (t ½ = 10 s) is safe and well-tolerated. The potential adverse effects include flushing, chest pain, palpitations, breathlessness, transient episodes of heart block, hypotension, sinus bradycardia, and bronchospasm [102]. The contraindication to adenosine stress CMR (in addition to the general contraindication of any MR study) include known hypersensitivity to adenosine, known or suspected bronchoconstrictive or bronchospastic disease, 2nd or 3rd degree

Diagnostic accuracy of stress CMR in a population with high prevalence of CAD (57%) showed on an overall sensitivity of 89% and specificity of 80% for the diagnosis of significant obstruc‐ tive CAD. Adenosine-based stress demonstrated better sensitivity than dipyridamole (90% vs. 86%), but with similar specificity (81% vs. 77%) [103]. Adenosine and dobutamine stress CMR have similar sensitivity and specificity [104]. Stress CMR showed no difference in diagnostic accuracy when compared to SPECT MPI for CAD detection [105]. The concordance and accuracy of stress CMR with 320-detector row CT, showed an excellent agreement (92%, kappa value = 0.81) in an intermediate risk cohort [106]. Adenosine perfusion was the most accurate component of the stress CMR study in predicting which patients had significant CAD,

Negative findings on stress CMR are reassuring and associated with annualized event rates of 0.4% for MI and 0.3% for cardiovascular death. In patients with inducible ischemia, the annual event rates for MI and cardiovascular death were 2.6% and 2.8%, respectively. The concomitant presence of LGE was associated with a worse prognosis [108]. Other predictors of cardiac events were resting WMA, inducible WMA and LGE. Patients with inducible WMA (i.e., ischemia) experienced significant benefits from revascularization, compared with patients

geometry, scan setting, and Gd dose should be similar to step (2).

AV block, sinus bradycardia (HR <45 bpm), and systolic BP <90 mmHg [102].

**4. LGE module***–* performed after 5 minutes of completion of step (3).

acquired for 40 to 50 heart beats.

60 Coronary Artery Disease - Assessment, Surgery, Prevention

*4.4.3. Diagnostic and prognostic accuracy*

compared with resting WMA and LGE [107].

without inducible WMA (i.e., without ischemia) [109].

The strengths and limitations of stress CMR are shown in Table 10.

*4.4.4. Strengths and limitations*

A 65-year-old man with prior coronary artery bypass surgery presents with chest pain. An adenosine stress CMR was performed. For study interpretation, follow the steps as described in the text starting with LGE, stress and rest images. There is inducible inferior wall perfusion defect with no LGE (Figure 8).

**Figure 8.** Left to right: 1st image: LV short axis LGE image show no evidence of scar. 2nd image: Stress first pass perfu‐ sion image demonstrate an inducible inferior wall perfusion defect. 3rd image: The corresponding rest image demon‐ strate no perfusion defect in the inferior wall.

#### *4.4.6. Clinical pearls*

**1.** Avoid caffeinated food and beverages, theophylline, and dipyridamole for 24 hours prior to stress CMR.


We have come to the end of our review on the essentials of noninvasive imaging modalities for the assessment of CAD. A proposed algorithm for test selection in suspected CAD is included (Figure 9). The following are four clinical scenarios commonly encountered in clinical practice and the suggested answers:

**1.** Would you perform a test in an asymptomatic 35-year-old woman who plans to partici‐ pate in a marathon next month? She has a normal resting ECG with no cardiovascular risk factor.

Answer: No. Asymptomatic patients generally do not warrant cardiac testing. Her pretest probability of CAD is very low (<5%).

**2.** What is the next suitable test in a 60-year-old woman with morbid obesity who presents with chest pain? She completed 4 METs and achieved 70% of the maximum age-predicted HR following an exercise ECG.

Answer: Inability to achieve an adequate stress level reduces the sensitivity of an exercise ECG. Her pretest probability of CAD is intermediate. In the presence of morbid obesity, PET MPI or CCTA would be considered a suitable alternative.

**3.** What test would you perform in a 75-year-old man with a sedentary lifestyle who presents with chest pain on exertion that is relieved with nitroglycerin spray? He has underlying COPD, diabetes, and hypertension. Baseline creatinine is 300 micromol/l.

Answer: Pretest probability for CAD is high (≥90%) and the diagnosis of CAD is not in question. He has stable angina. First step would be to inititate medications such as aspirin, beta-blockers, calcium channel blockers, and statin therapy. Pharmacological SPECT or PET MPI for risk stratification is reasonable. Pharmacological SE can be performed if a good echo window can be obtained. In the presence of moderate ischemic burden, high risk variables on MPI, or worsening of symptom despite on optimal medical therapy, ICA is warranted.

**4.** Would you repeat another CCTA in a 50-year-old man with an active lifestyle presenting with atypical chest pain? He had a calcium score of 0 and a normal CCTA a year ago.

Answer: No. No testing is required. If symptoms persist, consider an exercise ECG.

 ̴ͻǣǤ αǢαǢαǢα ǢαǤȗη͵ͲȀʹǣǢδ͵ͲȀʹǣǡǤ ECG= electrocardiography; CCTA= coronary computed tomography angiography; MPI= myocardial perfusion imag‐ ing; CMR= cardiac magnetic resonance; BMI= body mass index. \* BMI ≥30 kg/m<sup>2</sup> : PET MPI is preferred; BMI <30 kg/m<sup>2</sup> : SPECT or PET MPI, depending on resource.

ȋεʹǤͲȀǡǢ δ͵ͲȀǡǤ ǡȋͳǤǡͲȌǡǤ Patients with renal impairment (Creatinine >2.0 mg/dl, avoid CCTA; if GFR <30 mls/min, stress CMR is contraindicat‐ ed.

For pharmacological MPI, use of low level exercise (1.7 mph, grade 0) is an option, in the absence of LBBB.

**Figure 9.** Proposed algorithm for test selection in suspected coronary artery disease.

#### **5. Conclusion**

**2.** Stress CMR should be avoided in patients with a glomerular filtration rate (GFR) < 30 mls/

**3.** The presence of infarction on LGE (subendocardial or transmural in a coronary artery

**4.** Criteria for a perfusion defect is a persistent delay in enhancement pattern during first

**6.** Dark-rim artifact typically appear as dark lines at the blood pool-myocardium interface,

We have come to the end of our review on the essentials of noninvasive imaging modalities for the assessment of CAD. A proposed algorithm for test selection in suspected CAD is included (Figure 9). The following are four clinical scenarios commonly encountered in clinical

**1.** Would you perform a test in an asymptomatic 35-year-old woman who plans to partici‐ pate in a marathon next month? She has a normal resting ECG with no cardiovascular risk

Answer: No. Asymptomatic patients generally do not warrant cardiac testing. Her pretest

**2.** What is the next suitable test in a 60-year-old woman with morbid obesity who presents with chest pain? She completed 4 METs and achieved 70% of the maximum age-predicted

Answer: Inability to achieve an adequate stress level reduces the sensitivity of an exercise ECG. Her pretest probability of CAD is intermediate. In the presence of morbid obesity, PET MPI

**3.** What test would you perform in a 75-year-old man with a sedentary lifestyle who presents with chest pain on exertion that is relieved with nitroglycerin spray? He has underlying

Answer: Pretest probability for CAD is high (≥90%) and the diagnosis of CAD is not in question. He has stable angina. First step would be to inititate medications such as aspirin, beta-blockers, calcium channel blockers, and statin therapy. Pharmacological SPECT or PET MPI for risk stratification is reasonable. Pharmacological SE can be performed if a good echo window can be obtained. In the presence of moderate ischemic burden, high risk variables on MPI, or

**4.** Would you repeat another CCTA in a 50-year-old man with an active lifestyle presenting with atypical chest pain? He had a calcium score of 0 and a normal CCTA a year ago.

COPD, diabetes, and hypertension. Baseline creatinine is 300 micromol/l.

worsening of symptom despite on optimal medical therapy, ICA is warranted.

Answer: No. No testing is required. If symptoms persist, consider an exercise ECG.

distribution) favors CAD, irrespective of perfusion findings.

pass observed in at least 3 consecutive temporal images.

**5.** Perfusion defects should be graded according to transmurality.

and can mimic a perfusion defect (typically Gibbs artifact).

practice and the suggested answers:

62 Coronary Artery Disease - Assessment, Surgery, Prevention

probability of CAD is very low (<5%).

HR following an exercise ECG.

or CCTA would be considered a suitable alternative.

min.

factor.

Understanding the merits and limitations of each noninvasive imaging modality, together with local expertise and resources, will increase the clinician's confidence in selecting imaging tests for the assessment of CAD. This is crucial to avoid layered testing, unnecessary radiation exposure, and maintain a cost-effective approach. Functional and anatomical noninvasive tests are associated with similar cardiovascular outcomes in patients with low to intermediate risk [110]. The use of stress MPI to serve as a gatekeeper for ICA is well validated [111]. A detailed history and physical examination with sound clinical judgement and the integration of evidence-based guidelines are vital for selecting the right test. In summary, noninvasive tests are the cornerstone of CAD assessment. Although not covered in this chapter, such tests also serve as a guide for ischemia-driven ICA strategies.

#### **6. Conflict of interest**

The authors have no conflict of interest to disclose pertaining to the contents of this book chapter.

#### **Acknowledgements**

All images used in this chapter were obtained from the University of Ottawa Heart Institute.

#### **Author details**

Punitha Arasaratnam1,2 and Terrence D. Ruddy1\*

\*Address all correspondence to: TRuddy@ottawaheart.ca

1 University of Ottawa Heart Institute, Ottawa, Canada

2 Ng Teng Fong General Hospital, Singapore

#### **References**


mal Method for Ischemia Evaluation in Women (WOMEN) trial. Circulation. 2011;124:1239-1249.

[7] Kohli P, Gulati M. Exercise stress testing in women, going back to the basics. Circula‐ tion. 2010;122:2570-2580.

**Acknowledgements**

64 Coronary Artery Disease - Assessment, Surgery, Prevention

**Author details**

**References**

Punitha Arasaratnam1,2 and Terrence D. Ruddy1\*

2 Ng Teng Fong General Hospital, Singapore

Img. 2010;3(7):789-794.

No. 4, 2014.

diol. 1983;1:574-5.

J Cardiol. 1984;54(1):43-9.

\*Address all correspondence to: TRuddy@ottawaheart.ca

1 University of Ottawa Heart Institute, Ottawa, Canada

All images used in this chapter were obtained from the University of Ottawa Heart Institute.

[1] Shaw L, Marwick T, et al. Why all the focus on cardiac imaging? J Am Coll Cardiol

[2] Wolk, et al. ACCF/AHA/ASE/ASNC/HFSA/HRS/SCAI/ SCCT/ SCMR/ STS, 2013 Multimodality appropriate use criteria for the detection and risk assessment of stable ischemic heart disease. A Report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, American Heart Association, American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Failure Society of America, Heart Rhythm Society, Society for Cardiovascular Angiography and In‐ terventions, Society of Cardiovascular Computed Tomography, Society for Cardio‐ vascular Magnetic Resonance, and Society of Thoracic Surgeons. JACC 2014; Vol. 63,

[3] Diamond GA. A clinically relevant classification of chest discomfort. J Am Coll Car‐

[4] Weintraub WS, Madeira SW, et al. Critical analysis of the application of Bayes' theo‐ rem to sequential testing in the noninvasive diagnosis of coronary artery disease. Am

[5] Diamond GA, Forrester JS. Analysis of probability as an aid in the clinical diagnosis

[6] Shaw L, Mieres J, Hendel R, et al. Effectiveness of exercise electrocardiography with or without myocardial perfusion single photon emission computed tomography in women with suspected coronary artery disease. Results form the What Is the Opti‐

of coronary artery disease. N Engl J Med. 1979:300:1350-8.


[31] Envoldsen LH, Menashi CAK, et al. Effects of acquisition time and reconstruction al‐ gorithm on image quality, quantitative parameters, and clinical interpretation of my‐ ocardial perfusion imaging. J Nucl Cardiol 2013. doi:10.1007/s12350-013-97752.

[19] Sanmarco ME, Pontius S, Selvester RH. Abnormal blood pressure response and marked ischemic ST-segment depression as predictors of severe coronary artery dis‐

[20] Mark DB, Shaw L, Harrell FE Jr, et al. Prognostic value of a treadmill exercise score in outpatients with suspected coronary artery disease. N Eng J Med. 1991;325(12):

[21] Gianrossi R, Detrano R, Mulvihill D, et al. Exercise-induced ST depression in the di‐ agnosis of coronary artery disease. A meta-analysis. Circulation. 1989;80:87-98. [22] Gibbons RJ, Balady GJ, Bricker JT, et al. ACC/AHA 2002 guideline update for exer‐ cise testing: summary article: A report of the American College of Cardiology/Ameri‐ can Heart Association Task Force on Practice Guidelines (Committee to Update the

[23] Alexander KP, Shaw LJ, Shaw LK, et al. Value of exercise treadmill testing in women.

[24] Mark DB, Hlatky MA, Harrell FE Jr, et al. Exercise treadmill score for predicting prognosis in coronary artery disease. Ann Intern Med. 1987;106(6):793-800.

[25] Ahmed H, Al-Mallah M, McEvoy J, et al. Maximal exercise testing variables and 10 year survival: Fitness risk score derivation from the FIT project. Mayo Clin Proc.

[26] Chang SM, Nabi F, Xu J, et al. Value of CACS compared with ett and myocardial per‐ fusion imaging for predicting long-term cardiac outcome in asymptomatic and symptomatic patients at low risk for coronary disease: Clinical implications in a mul‐

[27] Klocke FJ, Baird MG, et al. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging- executive summary: A report of the American College of Car‐ diology/ American Heart Association Task Force on practice guidelines (ACC/AHA/ ASNC Committee to revise the 1995 guidelines for the clinical use of cardiac radionu‐

[28] Patton J, Sandler M, et al. D-SPECT: A new solid state camera for high speed molecu‐

[29] Patton J, Slomka PJ, et al. Recent technologic advances in nuclear cardiology. J Nucl

[30] Marcassa C, Campini R, et al. Wide beam reconstruction for half-dose or half-time gated SPECT acquisitions: Optimization of resources and reduction in radiation ex‐

1997 Exercise Testing Guidelines). Circulation. 2002;106:1883-1892.

timodality imaging world. J Am Coll Cardiol Img. 2015;8(2):134-44.

clide imaging). J Am Coll Cardiol. 2003;42:1318-33.

posure. Eur J Nucl Med Mol Imaging. 2011;38:499-508.

lar imaging. J Nucl Med. 2006;47:189.

Cardiol. 2007;14:501-13.

ease. Circulation. 1980;61:572-578.

66 Coronary Artery Disease - Assessment, Surgery, Prevention

J Am Coll Cardiol. 1998;32:1657-1664.

2015;90(3):346-55.

849-53.


[56] McArdle B, Dowsley T, et al. Does Rubidium-82 PET have superior accuracy to SPECT perfusion imaging for the diagnosis of obstructive coronary artery disease? A systematic review and met-analysis. J Am Coll Cardiol. 2012;60(18):1828-1837.

[43] Chang SM, Nabi F, Xu J, Raza U, Mahmarian JJ. Normal stress only versus standard stress/rest myocardial perfusion imaging: Similar patient mortality with reduced ra‐

[44] Duvall WL, Wijetunga MN, Klein TM, Razzouk L, Godbold J, Croft LB, et al. The prognosis of a normal stress-only Tc-99m myocardial perfusion imaging study. J

[45] Shaw LJ, Hendel R, Borges-Neto S, et al. Prognostic value of normal exercise and ad‐ enosine (99m)Tc-tetrofosmin SPECT imaging: Results from the multicenter registry

[46] Shaw LJ, Iskandrian AE. Prognostic value of gated myocardial perfusion SPECT. J

[47] Acampa W, Petretta M, et al. Warranty period of normal stress myocardial perfusion imaging in diabetic patients: A propensity score analysis. J Nucl Cardiol.

[48] Hachamovitch R, Berman DS, Kiat H, et al. Exercise myocardial perfusion SPECT in patients without known coronary artery disease: Incremental prognostic value and

[49] Arasaratnam P, Ayoub C, et al. Positron emission tomography myocardial perfusion imaging for diagnosis and risk stratification in obese patients. Curr Cardiovasc Imag‐

[50] Lim SP, Arasaratnam P, et al. Obesity and the challenges of noninvasive imaging for the detection of coronary artery disease. Canadian Journal of Cardiology.

[51] Machac J. Cardiac positron emission tomography imaging. Sem Nucl Med.

[52] Machac J, Bacharach S, et al. Quality assurance committee of the american society of nuclear cardiology. Positron emission tomography myocardial perfusion and glucose

[53] Dilsizian V, Bacharach S, Beanlands R, et al. PET myocardial perfusion and metabo‐

[54] Baggish AL, Boucher CA. Radiopharmaceutical agents for myocardial perfusion

[55] Parker M, Iskander A, et al. Diagnostic accuracy of cardiac positron emission tomog‐ raphy versus single photon emission computed tomography for coronary artery dis‐

ease, a bivariate meta-analysis. Circ Cardiovasc Imaging. 2012;5:700-707.

diation exposure. J Am Coll Cardiol. 2010 Jan 19;55(3):221-30.

of 4,728 patients. J Nucl Med. 2003 Feb;44(2):134-9.

use in risk stratification. Circulation. 1996;93:905-914.

metabolism imaging. J Nucl Cardiol. 2006;13:e121-51.

lism clinical imaging. J Nucl Cardiol. 2009;16:651.

imaging. Circulation. 2008;118:1668-1674.

Nucl Cardiol. 2010 Jun;17(3):370-7.

68 Coronary Artery Disease - Assessment, Surgery, Prevention

Nucl Cardiol. 2004;11:171-185.

2014;21:50-6.

ing Re. 2015;8:9304.

2015;31:223-226.

2005;35:17-36.


tive summary: The Task Force on the Management of Stable Angina Pectoris of the European Society of Cardiology. Eur Heart J. 2006;27:1341-81.


prospective, large-scale, multicenter, head-to-head comparison between dipyrida‐ mole and dobutamine test. Echo-Per- santine International Cooperative (EPIC) and Echo- Dobutamine International Cooperative (EDIC) Study Groups. J Am Coll Cardi‐ ol. 1999;34:1769-7.

tive summary: The Task Force on the Management of Stable Angina Pectoris of the

[68] Bjork Ingul C, Rozis E, Slordahl SA, et.al. Incremental value of strain rate imaging to wall motion analysis for prediction of outcome in patients undergoing dobutamine

[69] Ingul CB, Stoylen A, Slodahl SA, et al. Automated analysis of myocardial deforma‐ tion at dobutamine stress echocardiography: An angiographic validation. J Am Coll

[70] Nagy AI, Sahlen A, Manouras A, et al. Combination of constrast-enhanced wall mo‐ tion analysis and myocardial deformation imaging using dobutamine stress echocar‐

[71] Beleslin BD, Osojic M, Djordjevic-Dikie A, et al. Integrated evaluation of relation be‐ tween coronary lesion features and stress echocardiography results: The importance

[73] Picano E, Bedetti G, Varga A, et al. The comparable diagnostic accuracies of dobuta‐ mine stress and dipyridamole-stress echocardiographies: a meta-analysis. Coron Ar‐

[74] Ho FM, Huang PJ, Liau CS, et al. Dobutamine stress echocardiography compared with dipyridamole thallium-201 single-photon emission computed tomography in

[75] Mahajan N, Polavaram N, Vankalaya et al. Diagnostic accuracy of myocardial perfu‐ sion imaging and stress echocardiography for the diagnosis of left main and triple vessel coronary artery disease: A comparative meta-analysis. Heart. 2010;96:956-966.

[76] Innocenti F, Lazzeretti D, Conti A, et al. Stress echocardiography in the ED: Diagnos‐ tic performance in high-risk subgroups. Am J Emerg Med. 2013;31:1309-1314.

[77] Metz LD, Beattie M, Hom R, et al. The prognostic value of normal exercise myocar‐ dial perfusion imaging and exercise echocardiography: A meta- analysis. J Am Coll

[78] Arruda-Olson AM, Juracan EM, Mahoney DW, et al. Prognostic value of exercise echocardiography in 5, 798 patients: Is there a gender difference: J Am Coll Cardiol.

[79] Schroder K, Wieckhorst A, Voller H. Comparison of the prognostic value of dipyri‐ damole and dobutamine stress echocardiography in patients with known or suspect‐

[80] Pingitore A, Picano E, Varga A, et al. Prognostic value of pharmacological stress echocardiography in patients with known or suspected coronary artery disease: A

of coronary lesion morphology. J Am Coll Cardiol. 1999;33:717-26.

[72] Marwick TH. Stress echocardiography. Heart. 2003;89:113-8.

detecting coronary artery disease. Eur Heart J. 1995;16:570e5.

ed coronary artery disease. Am J Cardiol. 1997;79:1516-8.

European Society of Cardiology. Eur Heart J. 2006;27:1341-81.

stress echocardiography. Circulation. 2007;115:1252-9.

Cardiol. 2007;49:1651-9.

70 Coronary Artery Disease - Assessment, Surgery, Prevention

tery Dis. 2000;11:151e9.

Cardiol. 2007;49(2):227-37.

2002;39(4):625-31.

diography. Eur Heart J. 2015;16:88-95.


[103] Hamon M, Fau G, Nee G, et al. Meta-analysis of the diagnostic performance of stress perfusion cardiovascular magnetic resonance for detection of coronary artery dis‐ ease. J Cardiovasc Magn Reson. 2010;12:29.

[90] Litt HI, Gatsonis C, Snyder B, et al. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med. 2012;366:1393-403.

[91] Hoffmann U, Truong QA, Schoenfeld DA, et al. Coronary CT angiography versus

[92] Goldstein JA, Chinnaiyan KM, Abidov A, et al. The CT-STAT (coronary computed tomographic angiography for systematic triage of acute chest pain patients to treat‐

[93] Chow BJ, Small G, Yam Y, et al. Incremental prognostic value of cardiac computed tomography in coronary artery disease using CONFIRM: COroNary computed to‐ mography angiography evaluation for clinical outcomes: an InteRnational Multicen‐

[94] Habib PJ, Green J, Butterfield RCA, et al. Association of cardiac events with coronary artery disease detected by 64-slice or greater coronary CT angiography: A systematic

[95] Ostrom MP, Gopal A, Ahmadi N, et al. Mortality incidence and the severity of coro‐ nary atherosclerosis assessed by computed tomography angiography. J Am Coll Car‐

[96] Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictor of coronary

[97] P. Greenland, J. S. Alpert, G. A. Beller, et al. ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: A report of the American College of Car‐ diology Foundation/American Heart Association task force on practice guidelines.

[98] Shehata ML, Basha TA, Hayeri MR, et al. MR myocardial perfusion imaging: Insights on techniques, analysis, interpretation, and findings. RadioGraphics. 2014;34:1636-57.

[99] Kim HW, Klem I, Kim R. Detection of myocardial ischemia by stress perfusion cardi‐

[100] Nagel E, Klein C, Paetsch I, et al. Magnetic resonance perfusion measurements for

[101] Cury RC, Cattani CA, Gabure LA, et al. Diagnostic performance of stress perfusion and delayed enhancement MR imaging in patients with coronary artery disease. Ra‐

[102] Kramer C, Barkhausen J, Flamm S, et al. Standardized cardiovascular magnetic reso‐ nance imaging (CMR) protocols, society for cardiovascular magnetic resonance: Board of trustees task force on standardized protocols. J Cardiovasc Magn Reso‐

events in four racial or ethnic groups. N Engl J Med. 2008;358:1336-1345.

standard evaluation in acute chest pain. N Engl J Med. 2012;367:299-308.

ment) trial. J Am Coll Cardiol. 2011;58:1414-22.

72 Coronary Artery Disease - Assessment, Surgery, Prevention

ter registry. Circ Cardiovasc Imaging. 2011;4:463-72.

diol. 2008;52:1335-43.

Circulation. 2010;122:584-636.

diology. 2006;240:39-45.

review and meta-analysis. Int J Cardiol. 2013; 169:112-120.

ovascular magnetic resonance. Cardiol Clin. 2007:25:57-70.

nance. 2008;10:35. doi:10.1186/1532-429X-10-35.

the noninvasive detection of CAD. Circulation. 2003;108:432-437.


**Non-Invasive Imaging of Coronary Artery Disease — The Expanding Role of Coronary Computed Tomographic Angiography in the Management of Low- to Intermediate-Risk Patients and Dealing with Intermediate Stenosis**

Michael Campbell, Stephen Lyen, Jonathan Rodrigues, Mark Hamilton and Nathan Manghat

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61837

#### **Abstract**

Non-invasive anatomic imaging modalities play a crucial role in the diagnosis of coro‐ nary artery disease (CAD), particularly in the case of the symptomatic patient presenting in the emergency department.

Some of the key issues of discussion will be the appropriate use of coronary computed tomography (CT) in the anatomical assessment of CAD, the prognostic information that this assessment holds and how the role of CT may evolve in the coming years.

The aim of this chapter is to summarise and evaluate the current best non-invasive ana‐ tomical strategies of CAD imaging, notably in those with a low to intermediate pre-test risk of CAD and those with an intermediate luminal stenosis.

**Keywords:** Coronary artery disease, computed tomography, imaging, intermediate stenosis

#### **1. Introduction**

The British physician William Harvey described the heart as, "the household of Divinity which, discharging its function, nourishes, cherishes, quickens the whole book and is indeed the foundation of life."[1] In Harvey's words, together with his wider body of work, he en‐

© 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

capsulates three simple truths; it is the heart's role to pump the blood that it is the role of the blood vessels to circulate the blood and it is the blood that sustains living tissue. These as‐ sertions may seem obvious, but they are just as pertinent now as when they were first pen‐ ned in the 17th century. The reason for this is clear, the leading cause of mortality in the developed world is coronary artery disease(CAD). CAD is a narrowing of the lumen of the arteries that supply blood to the heart resulting in a 'failure' of the circulation to deliver blood to the heart.[2] Modern medical imaging allows the physician to accurately assess the extent to which the coronary arteries are narrowed (anatomical imaging), and to what extent this narrowing results in a 'failure' to circulate the blood (functional imaging).

It is an undisputed fact that the current best method for imaging the burden of CAD is inva‐ sive coronary angiography (ICA). This fact is reflected in its ubiquitous use in clinical prac‐ tice today; figures from 2004 show that 201,000 of these procedures were undertaken in the United Kingdom, an increase of 7% from the previous year. Whilst the risks may be small – a mortality of 0.07% and a radiation exposure risk lower than other imaging procedures [3] – it would not be safe, practical or cost effective to offer every patient with symptoms of CAD to ICA. Furthermore, up to 40% of the patients who do go on to have elective ICA are found to have sub-clinical stenosis in all coronary arteries.[4,5] This highlights the inadequate and inaccurate information provided by orthodox first-/second-line investigations in the assess‐ ment of symptomatic patients with suspected CAD.

There is an increasing body of evidence which suggests that non-invasive anatomical imag‐ ing modalities have a crucial role to play in the diagnosis of CAD, particularly, in the case of the symptomatic patient presenting in the emergency department. That is, not to say that non-invasive imaging will completely supplant ICA, but it may, and in some instances al‐ ready has been proven to be clinically useful in the right patient group, at the right stage of the patient pathway [6, 7, 8]. One such promising imaging modality, which will be evaluat‐ ed in explicit detail in the course of this review, is CT. Some of the key issues of discussion will be the appropriate use in anatomic assessment of CAD, the prognostic information that this assessment holds and how the role of CT may evolve in the coming years.

The aim of this chapter is to summarise and evaluate the current best non-invasive anatomi‐ cal strategies of imaging CAD. The main focus will be to unveil the best to approach the two less well understood (and sometimes overlapping) cohorts of CAD symptomatic patients: those with a low to intermediate pre-test risk (10–29%) of CAD and those with CAD, who have an intermediate amount of luminal stenosis (40–69% of the luminal cross-sectional area).

#### **2. What is CAD? Blood haemodynamics and myocardial demand**

Coronary artery disease has one main consequence: it limits blood flow to the myocardium. When this happens, CAD can cause an imbalance between the myocardial oxygen demand and the rate of delivery of oxygen by the blood, leading to tissue hypoxia. The overall proc‐ ess is a broad spectrum of conditions known as ischaemic heart disease. In 1772, when Wil‐ liam Heberden described an 'uncomfortable sensation' while walking and called it angina, he was describing a manifestation of ischaemic heart disease (Figures 1 and 2).[2] **Figure 1**

**Figure 1 –** Darcy's law and Poiseuilleʹs law **Figure 1.** Darcy's law and Poiseuille's law

capsulates three simple truths; it is the heart's role to pump the blood that it is the role of the blood vessels to circulate the blood and it is the blood that sustains living tissue. These as‐ sertions may seem obvious, but they are just as pertinent now as when they were first pen‐ ned in the 17th century. The reason for this is clear, the leading cause of mortality in the developed world is coronary artery disease(CAD). CAD is a narrowing of the lumen of the arteries that supply blood to the heart resulting in a 'failure' of the circulation to deliver blood to the heart.[2] Modern medical imaging allows the physician to accurately assess the extent to which the coronary arteries are narrowed (anatomical imaging), and to what extent

It is an undisputed fact that the current best method for imaging the burden of CAD is inva‐ sive coronary angiography (ICA). This fact is reflected in its ubiquitous use in clinical prac‐ tice today; figures from 2004 show that 201,000 of these procedures were undertaken in the United Kingdom, an increase of 7% from the previous year. Whilst the risks may be small – a mortality of 0.07% and a radiation exposure risk lower than other imaging procedures [3] – it would not be safe, practical or cost effective to offer every patient with symptoms of CAD to ICA. Furthermore, up to 40% of the patients who do go on to have elective ICA are found to have sub-clinical stenosis in all coronary arteries.[4,5] This highlights the inadequate and inaccurate information provided by orthodox first-/second-line investigations in the assess‐

There is an increasing body of evidence which suggests that non-invasive anatomical imag‐ ing modalities have a crucial role to play in the diagnosis of CAD, particularly, in the case of the symptomatic patient presenting in the emergency department. That is, not to say that non-invasive imaging will completely supplant ICA, but it may, and in some instances al‐ ready has been proven to be clinically useful in the right patient group, at the right stage of the patient pathway [6, 7, 8]. One such promising imaging modality, which will be evaluat‐ ed in explicit detail in the course of this review, is CT. Some of the key issues of discussion will be the appropriate use in anatomic assessment of CAD, the prognostic information that

The aim of this chapter is to summarise and evaluate the current best non-invasive anatomi‐ cal strategies of imaging CAD. The main focus will be to unveil the best to approach the two less well understood (and sometimes overlapping) cohorts of CAD symptomatic patients: those with a low to intermediate pre-test risk (10–29%) of CAD and those with CAD, who have an intermediate amount of luminal stenosis (40–69% of the luminal cross-sectional

Coronary artery disease has one main consequence: it limits blood flow to the myocardium. When this happens, CAD can cause an imbalance between the myocardial oxygen demand and the rate of delivery of oxygen by the blood, leading to tissue hypoxia. The overall proc‐ ess is a broad spectrum of conditions known as ischaemic heart disease. In 1772, when Wil‐

this assessment holds and how the role of CT may evolve in the coming years.

**2. What is CAD? Blood haemodynamics and myocardial demand**

this narrowing results in a 'failure' to circulate the blood (functional imaging).

ment of symptomatic patients with suspected CAD.

76 Coronary Artery Disease - Assessment, Surgery, Prevention

area).

**Figure 2.** Schematic representation to show Darcy's law (volume flow rate) and Poiseuille's law (resistance to flow) in a rigid tube, assuming laminar flow. *adapted without permission*[42]

The situation presented by traditional thinking, derived from animal experimental models in the 1970s, is that CAD is not deemed to be flow limiting at all, when 60% of the lumen cross-sectional area is occluded and is only deemed to be obstructive to flow during stress when 70% of the luminal cross-sectional area is blocked (Figure 3).[10] The relationship be‐ tween the percentage of obstruction and flow forms the basis for much of the findings in anatomical imaging, such as CT coronary angiography (CTCA). For example, it is a well-re‐ ported fact that sub-clinical stenosis (<50%). CAD is associated with a very low myocardial infarction event rate.[7]

**Figure 3.** A graph derived from animal experiments in the 1970s, showing the relationship between the amount of ves‐ sel occlusion and flow at rest and during stress. It can be seen that flow is affected at a lower threshold of diameter narrowing during exercise than during stress. This was originally cited as the cause of angina described by Heberden in the 18th century. However, it is now known that the relationship between percentage of vessel narrowing and coro‐ nary flow is a more complicated process than this experiment implied [10] *graph adapted without permission*

The problem with the traditional thinking and the experimental models, from which it was derived, is that the animal experiments were done on healthy coronary arteries, which were externally compressed at one point along the coronary artery. Equating the flow demon‐ strated in an externally compressed coronary artery to that in an equally stenosed, diseased coronary is wrong for two main reasons. Firstly, the pathophysiological process that causes CAD does far more than physically 'block' the vessel. Secondly, CAD can exist 'diffusely' along a great portion of the vessel and still have profound haemodynamic consequences. For example, one can use ICA to demonstrate that diffuse disease with narrowing as little as 38% can cause as much as a 65% decrease in coronary flow reserve (Figure 4). The implica‐ tions of this for imaging are clear; an anatomical assessment of coronary artery stenosis is useful information, however, this test alone cannot comprehensively assess the extent or be the best approach to tackle CAD.[11]

#### **3. Atherosclerosis (Figures 5–7)**

As highlighted earlier, CAD results in a narrowing of the coronary arteries. The narrowing of these coronary arteries can cause a limitation of blood flow and therefore oxygen supply Non-Invasive Imaging of Coronary Artery Disease — The Expanding Role of Coronary Computed... http://dx.doi.org/10.5772/61837 79

when 70% of the luminal cross-sectional area is blocked (Figure 3).[10] The relationship be‐ tween the percentage of obstruction and flow forms the basis for much of the findings in anatomical imaging, such as CT coronary angiography (CTCA). For example, it is a well-re‐ ported fact that sub-clinical stenosis (<50%). CAD is associated with a very low myocardial

**Figure 3.** A graph derived from animal experiments in the 1970s, showing the relationship between the amount of ves‐ sel occlusion and flow at rest and during stress. It can be seen that flow is affected at a lower threshold of diameter narrowing during exercise than during stress. This was originally cited as the cause of angina described by Heberden in the 18th century. However, it is now known that the relationship between percentage of vessel narrowing and coro‐

The problem with the traditional thinking and the experimental models, from which it was derived, is that the animal experiments were done on healthy coronary arteries, which were externally compressed at one point along the coronary artery. Equating the flow demon‐ strated in an externally compressed coronary artery to that in an equally stenosed, diseased coronary is wrong for two main reasons. Firstly, the pathophysiological process that causes CAD does far more than physically 'block' the vessel. Secondly, CAD can exist 'diffusely' along a great portion of the vessel and still have profound haemodynamic consequences. For example, one can use ICA to demonstrate that diffuse disease with narrowing as little as 38% can cause as much as a 65% decrease in coronary flow reserve (Figure 4). The implica‐ tions of this for imaging are clear; an anatomical assessment of coronary artery stenosis is useful information, however, this test alone cannot comprehensively assess the extent or be

As highlighted earlier, CAD results in a narrowing of the coronary arteries. The narrowing of these coronary arteries can cause a limitation of blood flow and therefore oxygen supply

nary flow is a more complicated process than this experiment implied [10] *graph adapted without permission*

infarction event rate.[7]

78 Coronary Artery Disease - Assessment, Surgery, Prevention

the best approach to tackle CAD.[11]

**3. Atherosclerosis (Figures 5–7)**

**Figure 4.** A schematic representation that illustrates the limitations of anatomic imaging by Artgm (arteriogram) and IVUS (intravascular ultrasound) during invasive coronary angiography. The recorded value of blood flow, measured by CFR, is severely at odds with the findings using only anatomic measurements. *Adapted without permission*[11]

(ischaemia) to the myocardium. The cause of this 'narrowing' is a poorly understood chron‐ ic inflammatory process called atherosclerosis.[2]. Atherosclerosis, literally meaning 'hard gruel', is a more complicated process than the development of an atherosclerotic plaque, which thickens the vessel wall, intruding and obstructing the vessel lumen. The reason for this atherosclerosis is that the healthy coronary artery cannot be thought of as simply an in‐ ert tube through which blood flows. [2, 9] In fact, the healthy artery must be thought of as three layers of differing function: the tunica intima, tunica media and tunica adventitia.

The intima must be recognised as more than simply a mechanical barrier which encloses the blood; it is involved in metabolism, signalling and through a combination of both, and it plays a crucial role in haemodynamics of blood flow. The way it does this is twofold:


During the process of atherosclerosis, for reasons beyond the scope of this chapter, the inti‐ ma lining cells fail to perform their normal role in regulating the flow of blood at a local lev‐ el. As a result, endothelial cell dysfunction arising from atherosclerosis causes both inappropriate vasoconstriction and loss of normal anti-thrombotic properties (thus increas‐ ing blood viscosity).[2] This is relevant to imaging because the relationship between how much of the luminal cross-sectional area of a coronary artery is occluded is not always di‐ rectly proportional to flow limitation. Therefore, imaging modalities that merely acquire anatomical information cannot provide haemodynamic information specific to the individu‐ al.

Another reason why a good understanding of the pathophysiology of atherosclerosis is im‐ portant is because the composition of the atherosclerotic plaque itself is having an incremen‐ tal role in interpreting the significance of imaging findings. [4, 7, 8, 12] During the process of atherosclerosis, 'bad' cholesterol, low-density lipoproteins, accumulate under the intimal wall; these lipids quickly become oxidised and are engulfed by tissue macrophages forming 'foam cells'. Another thing that happens, not necessarily in a sequential order, is that smooth muscle cells are moved away from their native sites in the media and these migrate into the intima. These smooth muscle cells secrete extracellular matrix forming a fibrous plaque, which often have high calcium concentrations.[2] Misleadingly, this calcium does not al‐ ways concentrate at the site of maximal stenosis and calcified plaque only represents ap‐ proximately 20% of the total coronary atherosclerotic burden. Therefore, calcium can be considered a good marker of CAD and its absence is an excellent marker of no CAD [13], but it cannot diagnose obstructive CAD. [6] [12] The relative proportions of fibrous tissue to lipid also determine the vulnerability of the plaque to rupture. This is crucial in risk stratifi‐ cation when assessing CAD.

**Figure 5.** The three layers of artery wall: tunica intima, media and adventitia. The media contains smooth muscle cells. During the process of atherosclerosis, these can migrate into the intima and elaborate fibrous ECM. The adventitia is composed of connective tissue, nerves and lymph. *adapted without permission*[2]

Non-Invasive Imaging of Coronary Artery Disease — The Expanding Role of Coronary Computed... http://dx.doi.org/10.5772/61837 81

**DisplayText cannot span mo**

inappropriate vasoconstriction and loss of normal anti-thrombotic properties (thus increas‐ ing blood viscosity).[2] This is relevant to imaging because the relationship between how much of the luminal cross-sectional area of a coronary artery is occluded is not always di‐ rectly proportional to flow limitation. Therefore, imaging modalities that merely acquire anatomical information cannot provide haemodynamic information specific to the individu‐

Another reason why a good understanding of the pathophysiology of atherosclerosis is im‐ portant is because the composition of the atherosclerotic plaque itself is having an incremen‐ tal role in interpreting the significance of imaging findings. [4, 7, 8, 12] During the process of atherosclerosis, 'bad' cholesterol, low-density lipoproteins, accumulate under the intimal wall; these lipids quickly become oxidised and are engulfed by tissue macrophages forming 'foam cells'. Another thing that happens, not necessarily in a sequential order, is that smooth muscle cells are moved away from their native sites in the media and these migrate into the intima. These smooth muscle cells secrete extracellular matrix forming a fibrous plaque, which often have high calcium concentrations.[2] Misleadingly, this calcium does not al‐ ways concentrate at the site of maximal stenosis and calcified plaque only represents ap‐ proximately 20% of the total coronary atherosclerotic burden. Therefore, calcium can be considered a good marker of CAD and its absence is an excellent marker of no CAD [13], but it cannot diagnose obstructive CAD. [6] [12] The relative proportions of fibrous tissue to lipid also determine the vulnerability of the plaque to rupture. This is crucial in risk stratifi‐

**Figure 5.** The three layers of artery wall: tunica intima, media and adventitia. The media contains smooth muscle cells. During the process of atherosclerosis, these can migrate into the intima and elaborate fibrous ECM. The adventitia is

composed of connective tissue, nerves and lymph. *adapted without permission*[2]

al.

cation when assessing CAD.

80 Coronary Artery Disease - Assessment, Surgery, Prevention

**Figure 6.** A micrograph showing that atherosclerotic plaques are not homogeneous in their composition. The figure illustrates two stenotic plaques of differing morphology a) plaque consisting of hard, collagen rich sclerotic tissue b) A plaque many comprised lipid-rich atheromatous core, separated from the vessel lumen by a thin fibrous cap. *Adapted without permission*[45]

**Figure 7.** Schematic showing the pathological development of Atherosclerosis and the factors that contribute during each phase. One of the key aims going forward in anatomic imaging is to identify plaques which are at high risk of rupture as seen in C *adapted without permission*[2]

#### **4. Anatomic imaging: The evolving role of coronary CT in the management of chest pain (Figures 8–12)**

CT has come a long way since the first clinical scans in 1971. It is now estimated that over 4 million scans are performed annually in the UK and one of the recent successful applica‐ tions of CT has been the emergence of CT coronary angiography (CTCA) in the manage‐ ment of chest pain.[4] A moving object, such as the heart, was thought impossible to image by CT as the discrete nature of the imaging process meant early scanners had poor temporal resolution. However, with the advancement in CT technology, particularly the development of greater number of detector row and more rapid gantry rotation, CT can now accurately evaluate structures in constant motion like the heart and coronary arteries due to its superi‐ or spatial and temporal resolution. [4, 14] CTCA involves visualising the coronary arteries directly with sub-millimetre isotropic spatial resolution and using this anatomic information on stenosis severity to determine the true extent of CAD.

**Figure 9 Figure 8.** A brief summary of the Physics of CT

$$HU = 1000 \times \frac{\mu\_X - \mu\_{water}}{\mu\_{water}}$$

**Figure 9.** CT number (Hounsfield) equation – the equation used to calculate the voxel intensity

#### **5. A Mandate for CTCA (Figure 13)**

Traditionally, a patient presenting with chest pain in the emergency department with a clini‐ cal history indicating CAD, but with no ECG changes or troponin-plasma irregularities, would be sent for an exercise treadmill test (ETT). The weaknesses of ETT are encapsulated Non-Invasive Imaging of Coronary Artery Disease — The Expanding Role of Coronary Computed... http://dx.doi.org/10.5772/61837 83

of greater number of detector row and more rapid gantry rotation, CT can now accurately evaluate structures in constant motion like the heart and coronary arteries due to its superi‐ or spatial and temporal resolution. [4, 14] CTCA involves visualising the coronary arteries directly with sub-millimetre isotropic spatial resolution and using this anatomic information

> CT became possible in 1917 when the mathematician, J.H. Radon proved that the 2D distribution of an object can be determined exactly, if the integral values along any number of lines passing

 The practical implications of this in medicine were not realised until the English engineer, Hounsfield showed that by measuring attenuation x‐rays fired from different angles in a plane perpendicular to the scanning subject (although theoretically any arbitrary plane can be used) one

Modern CT scanners use multiple x‐ray beam collimators, which convert the x‐ray beam into a

 Multiple detectors measure the difference between the known primary intensity of the x‐ray sheets and their attenuated intensity having passed through different tissues. [14] X‐ray beam generators and detectors spin 360 degrees around the scanning subject collecting many thousands of attenuation intensities from different angles within the same plane. These attenuation intensities are used to create a discrete number of cubic volume elements or voxel. Each individual voxel is described by its individual CT number or Hounsfield unit. [15] The intensity of radiation recorded at all the detectors during the scan is equal to the

 This is called the ray sums. The ray sums can be solved simultaneously to obtain the co‐ordinates of each voxel in three dimensions and pair it to its exact attenuation properties, thus eliminating

Each individual voxel is then combined to produce a thin sliced, three‐dimensional image.

Traditionally, a patient presenting with chest pain in the emergency department with a clini‐ cal history indicating CAD, but with no ECG changes or troponin-plasma irregularities, would be sent for an exercise treadmill test (ETT). The weaknesses of ETT are encapsulated

on stenosis severity to determine the true extent of CAD.

82 Coronary Artery Disease - Assessment, Surgery, Prevention

through the same layer are known.

He called this technique CT.

It was realised that the same principle was true of 3D objects.

CT visualises the body as composed of a series of finite slices.

'fan' beam, covering the individual body section or slice.

integration of all the CT numbers of the voxels.

the superimposition that occurs in conventional radiography.

**Figure 9.** CT number (Hounsfield) equation – the equation used to calculate the voxel intensity

can gather the line integrals necessary to reconstruct an image in 3D.

**Figure 9**

**Figure 8.** A brief summary of the Physics of CT

**5. A Mandate for CTCA (Figure 13)**

**DisplayText cannot span more than one line!**

**Figure 10.** Schematic demonstrating how the ray sums are acquired in the simplest image matrices possible (2 × 2 and 3 × 3). There are N squared unknown values of attenuation for an N × N image matrix. The ray sums can then be solved as simultaneous linear equations to find the attenuation values of each voxel. *Image adapted without consent*[14]

**Figure 11.** In a single plane, perpendicular to the scanning subject, the x-ray tube, and detector acquire attenuation values 360 degrees around the scanning subject (thus the scanning subject is the axis of rotation). This allows a single 'volume' slice to be obtained. Slices can be combined to form a 3D image visualised from any angle. *image adapted with‐ out consent*[14]

in a meta-analysis conducted by Patel et al, reviewing a sample of 398,978 cases of chest pain admissions with unknown CAD. The results show that 59% of the positive tests had no ob‐ structive CAD and 28% were false negatives when ICA was performed. This level of inac‐ curacy is unacceptable, regardless of the fact that ETT is inexpensive and readily available; it is no better than flipping a coin at positively predicting flow-limiting CAD. The findings of Patel et al also illustrate that in the established patient pathway only 37.6% of patients refer‐ red to ICA were found to have obstructive CAD. The overall conclusion of the study was that "better strategies for risk stratification...to increase diagnostic yield of cardiac catheteri‐ sation in routine practice" are necessary [17]

**Figure 12.** A flow diagram to show the procedure of calcium scoring and full-blown, contrast-enhanced CTCA. *Origi‐ nal figure: information adapted*[7]

**Figure 13.** Flow diagram illustrating CT's current role in the management of chest pain in the UK does not extend be‐ yond a 'rule out' in the intermediate stenosis group. It currently has no role in suspected ACS in the UK. *original figure, information adapted*[7]

The mandate for CT, if proven to work, is clear; too many patients, who do not need it are being sent for ICA. This is expensive and unnecessarily increases the patient's risk of com‐ plication. ICA has a serious complication rate of about 1/1,000[12]. Importantly, indecisive testing also increases the amount of time a patient spends in the waiting room. The CT coro‐ nary angiography for systematic triage of acute chest pain patients to treatment (CT-STAT) and randomised controlled trial (RCT) showed that as compared with the conventional pathway, a patient evaluated with CTCA can expect to wait on average less than 4 h at a cost, which is on average \$1,500 cheaper compared with the healthcare provider.[18] Anoth‐ er large-scale RCT (*n* = 1,365), which compared the traditional care group with that who re‐ ceived CTCA after first line tests, found that there were 26.8% fewer (95% CI 21.4–32.2) unnecessary admissions in the CCTA group. In addition, there were fewer negative invasive angiograms and a greater number of patients who were correctly identified as having ob‐ structive CAD. [19] In fact, the health economic model, using ICA as the reference standard, shows that at a pre-test probability of 50% or lower, CTCA results in a lower cost per patient with a true positive diagnosis.[15] Annually, there are 6 million presentations for chest pain, only 20% of these receive a diagnosis of CAD, and a large number are hospitalised unneces‐ sarily. In 2006, the bill for non-specific chest pain in the UK came to £11.2 billion.[8] On the issue of assessing chest pain for patients in the emergency department, medicine can and should do better than ETT. The evidence shows CTCA is far superior on the basis of time and expense [18].

#### **6. How can CTCA be used and does it work?**

**Figure 12.** A flow diagram to show the procedure of calcium scoring and full-blown, contrast-enhanced CTCA. *Origi‐*

**Figure 13.** Flow diagram illustrating CT's current role in the management of chest pain in the UK does not extend be‐ yond a 'rule out' in the intermediate stenosis group. It currently has no role in suspected ACS in the UK. *original figure,*

*nal figure: information adapted*[7]

84 Coronary Artery Disease - Assessment, Surgery, Prevention

*information adapted*[7]

CTCA has advanced to the stage that it is now advocated by 'National Institute for Clinical Excellence (NICE) Guidance 95' for patients with a low–intermediate risk of CAD (10–29%). [7] The main reason CTCA has taken on this new role is because of its high negative predic‐ tive value and thus the exceptional ability to rule out CAD (Figures 14–16). [20, 8, 4, 6, 21]

**Figure 14 –** A note on methodology of the trials **Figure 14.** A note on methodology of the trials

**Figure 15–** Strengths and weaknesses of the EVASCAN study methodology **Figure 15.** Strengths and weaknesses of the EVASCAN study methodology

**Figure 15**


**Figure 16.** A table summarising the data concerning the accuracy of 64 MDCT at various disease prevalence and de‐ grees of stenoses, considered flow limiting

One of the most definitive RCTs conducted to evaluate the accuracy of CTCA at detecting or ruling out >50% stenosis is the EVAluation of CT SCANner (EVASCAN) study, which in‐ cluded the largest sample of intermediate–high risk stable symptomatic patients, to date. Considering the low–intermediate group forms, the greatest proportion of individuals pre‐ senting with symptomatic CAD in the emergency department (ED) is 50–70% [19], EVAS‐ CAN's study sheds light on a previously under-investigated and crucial patient group. EVASCAN examined population with a prevalence of CAD of 54%, in a study sample of 757, using a 64 slice multi-detector row CT scanner. The sensitivity, specificity, positive pre‐ dictive value (PPV) and negative predictive value (NPV) compared with ICA as the refer‐ ence standard were 91%, 50%, 68% and 83%, respectively.[20] At 83%, EVASCAN's negative predictive value concords with the CORE 64 RCT [22], which also examines CTCA's effec‐ tiveness at detecting CAD severity compared with the gold standard of ICA. At a lowly 68%, the positive predictive value (PPV) of CTCA is badly affected by CT's facility to sys‐ tematically over-estimate stenosis severity due to local accumulation of non-obstructive lo‐ cal foci of calcium.

The results of EVASCAN and CORE 64 must be contextualised. Both look at the ability of CT to determine haemodynamically obstructive CAD, based on a 'binary', > or <50% steno‐ sis severity. For reasons highlighted earlier, 50% stenosis is quite a low boundary to be con‐ sidered flow limiting across the board. Whilst the inability to account for other physiological factors in flow limitation is an inherent weakness of anatomical CTCA, ruling out CAD ex‐ clusively on the basis of a 50% stenosis severity is bound to be less accurate than using the 70% threshold. For this reason, NICE guidance advocates the use of the >70% threshold in the assessment of flow limitation.[23]

The challenging question of, "what extent of luminal obstruction should be assumed to be flow limiting?" has recently been tackled by a few small-scale RCTs. These RCTs call for a re-evaluation of the binary classification of stenosis severity as flow limiting/not flow limit‐ ing according to a single cut-off threshold. These studies demonstrate the potential clinical usefulness of a more quantitative, tiered approach to evaluating stenosis severity. These studies also illustrate that stenosis severity can be accurately 'graded' by CTCA and this 'grading' shows excellent correlation with flow limitation (*r* = 0.82), assessed by ICA.[24] Further large-scale RCTs are needed to evaluate the usefulness of this graded approach in risk stratification and prognostic outcome. More evidence is also needed to determine the exact effectiveness of anatomical CTCA in assessing flow limitation in the intermediately stenosed group (40–69%).

**Figure 15–** Strengths and weaknesses of the EVASCAN study methodology

Disease Prevalence

specific patient groups rather than using ICA

predictive value (NPV) and specificity.

 A key strength of EVASCAN, which is not seen in many other studies of this kind, is that non evaluable segments were not excluded from the analysis, or as they are in other studies, assumed to be stenosed; falsely elevating the negative

 One of the EVASCAN study limitations was that currently, stenosis analysis is more often performed by more crude qualitative assessment in clinical practice rather than the quantitative approach taken in research.[20] For this reason, the findings of EVASCAN should be evaluated with caution when applied to the

 A weakness of CTCA pointed out by all major RCT study findings is that diagnostic accuracy of CTCA is limited in distal vessels or in vessels that are narrower than 1.5 mm due to its limited spatial resolution (0.5 mm) which is still

by far inferior to that of the invasive angiograph (0.2 mm)[27]. All studies pointed toward the need for further study on the 'real life' management of patients with chest pain. It is hoped that this could help to define a precise role for CCTA in symptomatic patients with CAD and determine exactly what benefit it can provide to

Study

CORE 64[22] >50% 51% N=291 85 90 91 83 ACCURACY [25] >70% 13.9% N=231 91 84 51 99

SLICE CTCA [26] >50% 68% N=360 <sup>99</sup> <sup>64</sup> <sup>86</sup> <sup>97</sup>

**Figure 16.** A table summarising the data concerning the accuracy of 64 MDCT at various disease prevalence and de‐

One of the most definitive RCTs conducted to evaluate the accuracy of CTCA at detecting or ruling out >50% stenosis is the EVAluation of CT SCANner (EVASCAN) study, which in‐ cluded the largest sample of intermediate–high risk stable symptomatic patients, to date. Considering the low–intermediate group forms, the greatest proportion of individuals pre‐ senting with symptomatic CAD in the emergency department (ED) is 50–70% [19], EVAS‐ CAN's study sheds light on a previously under-investigated and crucial patient group. EVASCAN examined population with a prevalence of CAD of 54%, in a study sample of 757, using a 64 slice multi-detector row CT scanner. The sensitivity, specificity, positive pre‐ dictive value (PPV) and negative predictive value (NPV) compared with ICA as the refer‐ ence standard were 91%, 50%, 68% and 83%, respectively.[20] At 83%, EVASCAN's negative predictive value concords with the CORE 64 RCT [22], which also examines CTCA's effec‐

Number Sensitivity Specificity

Positive Predictive Value

Negative Predictive value

**Figure 15**

ACCURACY OF 64

grees of stenoses, considered flow limiting

[Doubleclick to insert a new figure]

severity assessed

Study Name Degree of stenosis

**Figure 15.** Strengths and weaknesses of the EVASCAN study methodology

clinical setting.

86 Coronary Artery Disease - Assessment, Surgery, Prevention

The ACCURACY RCT illustrates CTCA's excellent ability to rule out CAD in populations of low prevalence of CAD with a negative predictive value of 99%.[25] At the other end of the spectrum, ACCURACY of 64 shows that CTCA can accurately rule out CAD in populations with a high prevalence of CAD with an NPV of 98%.[26] These results demonstrate that it is feasible to use CTCA in populations at high and low risk of the disease. Something that needs to be considered from a research perspective is that 99% rule out is extremely impres‐ sive and far more encouraging than ETT. However, from a clinical perspective 1 mistake in every 100, given the high turnover of chest pain patients in the ED, could prove to be very costly in terms of lives lost. Therefore, while it is feasible to use CTCA in the rule out of ob‐ structive CAD in both of these patient populations, it does not necessarily provide clear clin‐ ical benefit over ICA, the gold standard to which CTCA is being compared with, in terms of accuracy of the 'rule out'.

The overall findings of EVASCAN, and associated RCTs, illustrate that in populations with a low prevalence of disease and using a 70% stenosis as a threshold for flow limitation, CTCA can be used to a great effect to rule out flow-limiting CAD. This is why EVASCAN and other studies call for clinicians to recognise the importance of the pre-test probability. In low to intermediate pre-test probability of CAD (10–29%), where the disease prevalence of obstructive CAD will be low, it is more cost effective, safer than and almost equally as accu‐ rate as the gold standard ICA at ruling out CAD. The reason CTCA's use cannot be extend‐ ed to high pre-test risk population (>61%) is because in spite of the high NPV, a patient will receive no benefit from receiving CTCA.[7] In this high-risk group, a rule-in test is required, and CCTA's unimpressive PPV and ICA's added benefit of being able to revascularise straight away, if the culprit lesion is shown to be flow limiting, makes it more cost effective than CCTA in this patient group.

#### **7. Prognostic value of CTCA**

The focus thus far has been to assess to what extent CT can accurately identify obstructive CAD and the monetary and time-saving benefit CT can offer. It is also crucial to follow up patients after diagnosis by CTCA and record their outcomes that is. given a negative finding on CT, what proportion of patients still go on to have major adverse coronary event (MACE)?

A recent meta-analysis (*n* = 9,592), with a median 20-month follow-up, showed that the risk of MACE following a negative test on CTCA is 0.17% per year, a figure that is comparable with the baseline rate (0.15% per year). In patients with abnormal findings on CTCA, there is a risk of MACE of 8.8%, 40 times more than the risk in the general population.[27] Another meta-analysis (*n* = 3,670), which performed similar analysis over a longer mean follow-up period (21.6 months), found a tenfold higher risk in patients with "any detectable coronary stenosis by CTCA compared with subjects without coronary stenosis".[28]

A recent case–control study went into more detailed analysis at the potential for CT to strati‐ fy risk. It was found that, by grading coronary stenosis using a segmental stenosis score, and following up after 52 months, those with a score up to 5 (less severe stenosis) had an eventfree survival of 85%, whereas those who had a score >5 had an event-free survival of just 20%. The same study also found that those with an intermediate degree of the stenosis (40– 69% ruled out as flow limiting) showed an event-free survival between those with normal coronary arteries and those with obstructive CAD.[29] This study conceded that it could not be sure that the higher number of deaths in this group was because of the non-obstructive CAD, developing into obstructive CAD, or because of non-obstructive CAD plaque rupture, and highlighted, "early identification of non-obstructive CAD with CTCA is clinically im‐ portant because it may lead to a more aggressive strategy of cardiovascular risk factor con‐ trol and modification of clinical follow-up."[29] It is for exactly this reason that prognostic information obtained by CTCA is so useful; rather than widespread distribution of primary and secondary interventions, targeted aggressive treatment can be handed to the patients who need it the most.

#### **8. Prognostic value of CTCA: Calcium scoring**

Calcium scoring is a tool available for cardiac CT (Figures 17 and 18), which can be per‐ formed immediately, without the use of contrast and without the use of high doses of radia‐ tion, in order to stratify for cardiac risk. It works on the basis that coronary artery calcium (CAC) has excellent x-ray attenuation properties and is a quantifiable marker of atheroscler‐ otic plaque. However, the quantity of CAC is poorly correlated with the degree of stenosis, so its presence should not be extrapolated to be a good predictor of flow limitation.[8]

**Figure 17 – NICE Recommendations on the calcium score Figure 17.** NICE Recommendations on the calcium score

**Figure 17**

ed to high pre-test risk population (>61%) is because in spite of the high NPV, a patient will receive no benefit from receiving CTCA.[7] In this high-risk group, a rule-in test is required, and CCTA's unimpressive PPV and ICA's added benefit of being able to revascularise straight away, if the culprit lesion is shown to be flow limiting, makes it more cost effective

The focus thus far has been to assess to what extent CT can accurately identify obstructive CAD and the monetary and time-saving benefit CT can offer. It is also crucial to follow up patients after diagnosis by CTCA and record their outcomes that is. given a negative finding on CT, what proportion of patients still go on to have major adverse coronary event

A recent meta-analysis (*n* = 9,592), with a median 20-month follow-up, showed that the risk of MACE following a negative test on CTCA is 0.17% per year, a figure that is comparable with the baseline rate (0.15% per year). In patients with abnormal findings on CTCA, there is a risk of MACE of 8.8%, 40 times more than the risk in the general population.[27] Another meta-analysis (*n* = 3,670), which performed similar analysis over a longer mean follow-up period (21.6 months), found a tenfold higher risk in patients with "any detectable coronary

A recent case–control study went into more detailed analysis at the potential for CT to strati‐ fy risk. It was found that, by grading coronary stenosis using a segmental stenosis score, and following up after 52 months, those with a score up to 5 (less severe stenosis) had an eventfree survival of 85%, whereas those who had a score >5 had an event-free survival of just 20%. The same study also found that those with an intermediate degree of the stenosis (40– 69% ruled out as flow limiting) showed an event-free survival between those with normal coronary arteries and those with obstructive CAD.[29] This study conceded that it could not be sure that the higher number of deaths in this group was because of the non-obstructive CAD, developing into obstructive CAD, or because of non-obstructive CAD plaque rupture, and highlighted, "early identification of non-obstructive CAD with CTCA is clinically im‐ portant because it may lead to a more aggressive strategy of cardiovascular risk factor con‐ trol and modification of clinical follow-up."[29] It is for exactly this reason that prognostic information obtained by CTCA is so useful; rather than widespread distribution of primary and secondary interventions, targeted aggressive treatment can be handed to the patients

Calcium scoring is a tool available for cardiac CT (Figures 17 and 18), which can be per‐ formed immediately, without the use of contrast and without the use of high doses of radia‐

stenosis by CTCA compared with subjects without coronary stenosis".[28]

**8. Prognostic value of CTCA: Calcium scoring**

than CCTA in this patient group.

88 Coronary Artery Disease - Assessment, Surgery, Prevention

**7. Prognostic value of CTCA**

(MACE)?

who need it the most.

**Figure 18.** An unenhanced calcium score showing low degree calcification in proximal left anterior, descending artery. The patient was thus referred to CTCA. *Figure adapted without consent*[7]

A calcium score of zero has a 12-year-survival of 99.4% and a score of 100–400 Agatston units has a lifetime risk ratio of myocardial infarction of 4.3 compared with those with a cal‐ cium score of zero. Despite the excellent prognosis of a CAC score of zero, its use is not indi‐ cated for the purpose of screening as the likelihood of finding stenosis in low-risk patients (using the Framingham score) is too low to warrant imaging.[12] The CAC score is best de‐ ployed in the intermediate-risk population.[7] A serious consideration which requires the physician's meticulous attention is the calcific distribution; a relatively low overall calcium score may be taken more seriously if it is found in the 'spotty distribution' of calcium.[30]

Another aspect of the calcium score in risk stratification is the patient's age. Whilst no coro‐ nary calcium is an excellent marker of prognosis, the true false negative rate is not really known and controversy surrounds the ability of CAC to rule out non-calcified fibrolipid pla‐ que.[7] In a meta-analysis of 10,355 symptomatic patients, testing the ability of CAC scoring to detect significant CAD compared with ICA, results showed as high as 2% of the patients had significant CAD with no detectable calcium and CAC scoring had a poor overall specif‐ icity of just 40%.[31] These individuals (significant CAD with no CAC) tend to be younger than 50 years of age and particular diligence must be taken with patients in this age group. Worryingly, the presence of this non-calcified plaque is higher in patients with serious acute coronary syndrome (ACS) rather than stable angina.[12]

#### **9. Prognostic value of CTCA: Plaque composition**

An advantage of CTCA, and an area of great promise, is the ability to provide more infor‐ mation about the coronary artery than just the luminal information offered by ICA. CTCA can offer insight into the degree of mural plaque burden and the plaque sub-type, which is beyond ICA without the use of intravascular ultrasound.[4]

Broadly speaking, CTCA can identify three types of plaque: calcified, non-calcified and mixed. Comparison with intravascular ultrasound during ICA shows that CTCA can cor‐ rectly identify 95% of the calcified plaque, 83% of the non-calcified plaque and 84% of the mixed plaque. The accuracy for the identification of non-calcified plaque is lower for the same reason; CAC scoring is a poor predictor of obstructive CAD: CTCA systematically overestimates CAC due to the high attenuation properties of calcium and the partial volume effect. However, it is hoped that one day CTCA will be able to unlock useful prognostic in‐ formation about the chance of CAD plaque rupture, and therefore detect and direct medical intervention.[12]

It has already been highlighted that 90% of the ACS is caused by plaque rupture [2] and up to two-thirds of MIs occur from disruption of plaque that causes less than 50% stenosis. [12] Currently, CTCA would result in the discharge a patient with chest pain at this intermediate degree of stenosis and any further functional testing that would be performed (stress testing or ICA) is unlikely to show any signs of ischaemia. [12]

Plaques at the risk of rupture have a specific morphology called 'thin-capped fibroathero‐ ma'. These have a lipid-rich, necrotic core with a thin fibrous cap. The most important prop‐ erty in the risk of rupture of plaques is the thickness of the fibrous cap; the thinner the fibrous cap, the greater the risk of rupture. Currently, the spatial resolution of CT is limited to 330 μm and, by definition, the fibrous cap is less than 65 μm in thickness. The current limits of CT suggest that being able to visualise 'at risk' fibrous caps is impossible. However, these 'fibroatheromas' have slightly different attenuation properties to more stable, fibrous lesions. Although there is still much progress to be made, promise remains in the ability of CTCA to distinguish the attenuation properties of the lipid-rich core in the hope of recognis‐ ing and quantifying the risk of rupture in vulnerable plaque. [30, 12]

#### **10. CT in the assessment of Fractional Flow Reserve(FFR): The future**

ICA has been referred to many times throughout the course of this chapter as the 'gold standard' in the assessment of CAD (Figure 19). In order to define a precise role for anatom‐ ic CTCA, in the assessment of CAD, one must first understand the process of ICA, what use‐ ful information it gathers and whether it is within the capability of CTCA to gather similar, useful information.

**Figure 19 Figure 19.** The procedure of ICA

physician's meticulous attention is the calcific distribution; a relatively low overall calcium score may be taken more seriously if it is found in the 'spotty distribution' of calcium.[30]

Another aspect of the calcium score in risk stratification is the patient's age. Whilst no coro‐ nary calcium is an excellent marker of prognosis, the true false negative rate is not really known and controversy surrounds the ability of CAC to rule out non-calcified fibrolipid pla‐ que.[7] In a meta-analysis of 10,355 symptomatic patients, testing the ability of CAC scoring to detect significant CAD compared with ICA, results showed as high as 2% of the patients had significant CAD with no detectable calcium and CAC scoring had a poor overall specif‐ icity of just 40%.[31] These individuals (significant CAD with no CAC) tend to be younger than 50 years of age and particular diligence must be taken with patients in this age group. Worryingly, the presence of this non-calcified plaque is higher in patients with serious acute

An advantage of CTCA, and an area of great promise, is the ability to provide more infor‐ mation about the coronary artery than just the luminal information offered by ICA. CTCA can offer insight into the degree of mural plaque burden and the plaque sub-type, which is

Broadly speaking, CTCA can identify three types of plaque: calcified, non-calcified and mixed. Comparison with intravascular ultrasound during ICA shows that CTCA can cor‐ rectly identify 95% of the calcified plaque, 83% of the non-calcified plaque and 84% of the mixed plaque. The accuracy for the identification of non-calcified plaque is lower for the same reason; CAC scoring is a poor predictor of obstructive CAD: CTCA systematically overestimates CAC due to the high attenuation properties of calcium and the partial volume effect. However, it is hoped that one day CTCA will be able to unlock useful prognostic in‐ formation about the chance of CAD plaque rupture, and therefore detect and direct medical

It has already been highlighted that 90% of the ACS is caused by plaque rupture [2] and up to two-thirds of MIs occur from disruption of plaque that causes less than 50% stenosis. [12] Currently, CTCA would result in the discharge a patient with chest pain at this intermediate degree of stenosis and any further functional testing that would be performed (stress testing

Plaques at the risk of rupture have a specific morphology called 'thin-capped fibroathero‐ ma'. These have a lipid-rich, necrotic core with a thin fibrous cap. The most important prop‐ erty in the risk of rupture of plaques is the thickness of the fibrous cap; the thinner the fibrous cap, the greater the risk of rupture. Currently, the spatial resolution of CT is limited to 330 μm and, by definition, the fibrous cap is less than 65 μm in thickness. The current limits of CT suggest that being able to visualise 'at risk' fibrous caps is impossible. However, these 'fibroatheromas' have slightly different attenuation properties to more stable, fibrous

coronary syndrome (ACS) rather than stable angina.[12]

90 Coronary Artery Disease - Assessment, Surgery, Prevention

**9. Prognostic value of CTCA: Plaque composition**

beyond ICA without the use of intravascular ultrasound.[4]

or ICA) is unlikely to show any signs of ischaemia. [12]

intervention.[12]

**Figure 19 –** The procedure of ICA One of the major advantages of ICA is the ability to perform percutaneous coronary inter‐ vention (PCI) in an attempt to restore normal flow to the obstructed coronary artery. This assumption was called into question by the COURAGE RCT trial, which compared the out‐ comes of elective, stable angina patients with those who received optimised medical therapy (anti-platelet therapy and statins) when compared with PCI. The findings of this study were that, "as an initial management strategy in patients with stable coronary artery disease, PCI did not reduce the risk of death, myocardial infarction or other major cardiovascular events when added to optimal medical therapy."[32]

The COURAGE trial was followed up by the COURAGE nuclear substudy, which took into account the extent of ischaemia on perfusion imaging, using the same two treatment alloca‐ tions. The substudy found that revascularisation did lead to a decrease in ischaemia and a decrease in adverse cardiac events.[27] The results of the two trials were not incongruent. When taken together, they imply that there is a lack of revascularisation benefit, identified on anatomical grounds exclusively. In order to effectively identify patients who require re‐ vascularisation, some other test, which directly measures haemodynamic consequences of stenosis, is required. The test that has been developed and proven to be clinically useful is the fractional flow reserve (FFR). It has been found that if FFR > 0.75, PCI can be deferred without increased patient risk, despite an angiographic appearance of significant stenosis. Moreover, the cardiac event rates were lower in patients with FFR > 0.75 who did not have PCI than patients who did have PCI. [27, 33, 34]Finally, the evidence supports the fact that there is no benefit in revascularising, unless the haemodynamic consequences of stenosis are known.

The relevance to CTCA is that with modern 320 detector row CT scanners, one can also measure FFR (Figure 20) by applying fluid mechanical modelling to CCTA images, with no extra radiation exposure and no change to the normal CCTA procedure.[35]

**Figure 20.** A) An invasive coronary angiogram of LAD artery. QCA – Quantified coronary angiography (maximal quantified stenosis) is 50.68% showing it is an intermediately stenosed lesion. Invasively measured fractional flow re‐ serve (FFR) is 0.71, which indicates this lesion is causal of ischaemia B) Conventional CCTA concurs with the findings on ICA that the lesion is in the intermediately stenosed group and C) Combined function and anatomical image of LAD artery. The shading corresponds with the FFR at that point along the coronary artery. FFR measured by CCTA is 0.78, which by this study's definition (ischaemia if FFR<0.8) is an ischaemia causing lesion. This highlights the com‐ bined ability of anatomic and, new, functional CCTA to diagnose a lesion as flow limiting despite the fact it is only modestly stenosed (50.68% by QCA) [35]

How best could we use the anatomical information obtained by CCTA and apply it to the haemodynamic consequences of CAD on blood flow? The group of patients this has posed a particular problem for is the intermediate stenosis group (40–69%). Within this group CTCA has, as yet, been unable to unlock the haemodynamic consequences of CAD through an ana‐ tomical approach.

Recent evidence has shown early promising signs; using CTCA measured FFR can produce a diagnosis of ischaemia in lesions of intermediate stenosis severity with a PPV (compared with ICA measured FFR) of 82.4% and a NPV of 90.6%.[35] CTCA measured FFR, in combi‐ nation with anatomical imaging, has been shown to increase the accuracy of the diagnosis of ischaemia in lesions of all types by 25%. Another potential benefit is that in instances of mul‐ ti-segment stenoses, the culprit lesion(s) can be correctly identified. [35] The findings of these studies are based on a very small cohort (*n* = 60) and need to be shown to be reprodu‐ cible on a larger scale. Furthermore, the potential to use CCTA measured FFR should not be seen as a challenge to ICA and its potential benefits, compared or in conjunction with stresstesting modalities, need to be fully evaluated before its precise role can be defined.

## **11. Conclusion**

tions. The substudy found that revascularisation did lead to a decrease in ischaemia and a decrease in adverse cardiac events.[27] The results of the two trials were not incongruent. When taken together, they imply that there is a lack of revascularisation benefit, identified on anatomical grounds exclusively. In order to effectively identify patients who require re‐ vascularisation, some other test, which directly measures haemodynamic consequences of stenosis, is required. The test that has been developed and proven to be clinically useful is the fractional flow reserve (FFR). It has been found that if FFR > 0.75, PCI can be deferred without increased patient risk, despite an angiographic appearance of significant stenosis. Moreover, the cardiac event rates were lower in patients with FFR > 0.75 who did not have PCI than patients who did have PCI. [27, 33, 34]Finally, the evidence supports the fact that there is no benefit in revascularising, unless the haemodynamic consequences of stenosis are

The relevance to CTCA is that with modern 320 detector row CT scanners, one can also measure FFR (Figure 20) by applying fluid mechanical modelling to CCTA images, with no

**Figure 20.** A) An invasive coronary angiogram of LAD artery. QCA – Quantified coronary angiography (maximal quantified stenosis) is 50.68% showing it is an intermediately stenosed lesion. Invasively measured fractional flow re‐ serve (FFR) is 0.71, which indicates this lesion is causal of ischaemia B) Conventional CCTA concurs with the findings on ICA that the lesion is in the intermediately stenosed group and C) Combined function and anatomical image of LAD artery. The shading corresponds with the FFR at that point along the coronary artery. FFR measured by CCTA is 0.78, which by this study's definition (ischaemia if FFR<0.8) is an ischaemia causing lesion. This highlights the com‐ bined ability of anatomic and, new, functional CCTA to diagnose a lesion as flow limiting despite the fact it is only

How best could we use the anatomical information obtained by CCTA and apply it to the haemodynamic consequences of CAD on blood flow? The group of patients this has posed a particular problem for is the intermediate stenosis group (40–69%). Within this group CTCA has, as yet, been unable to unlock the haemodynamic consequences of CAD through an ana‐

Recent evidence has shown early promising signs; using CTCA measured FFR can produce a diagnosis of ischaemia in lesions of intermediate stenosis severity with a PPV (compared with ICA measured FFR) of 82.4% and a NPV of 90.6%.[35] CTCA measured FFR, in combi‐ nation with anatomical imaging, has been shown to increase the accuracy of the diagnosis of

extra radiation exposure and no change to the normal CCTA procedure.[35]

known.

92 Coronary Artery Disease - Assessment, Surgery, Prevention

modestly stenosed (50.68% by QCA) [35]

tomical approach.

This chapter has aimed to show that the process and manifestations of CAD are nuanced; therefore, what is required is a far more detailed analysis than the current diagnosis of, or the 'ruling out' of, the ACS.

CTCA has been shown to be a cost effective, quick and accurate means of managing patients with acute chest pain, and it has been established that for CTCA to be used effectively, it must be targeted at the right patient group (10–29% pre-test probability). The prognostic in‐ formation that can be garnered by CTCA is useful; however, the role of the prognostics in the direction of primary and secondary intervention requires further study in order for its precise use in risk stratification and direction of primary, secondary and tertiary interven‐ tions.

The findings demonstrate that the role of CTCA is not to supplant ICA as the gold standard in the investigation of CAD in a similar way that CT pulmonary angiography has its inva‐ sive counterpart. The current role of CTCA must be viewed as a means to avoiding unneces‐ sary invasive angiography and the associated risks that come with it in the right patient population. There is no longer a role or a need for ETT in the assessment of patients with acute chest pain.[36]

Whilst the focus of this chapter is purely upon the merits of anatomical assessment of CAD, it must be noted that the role of cardiac CT in the management of chest pain in the ED is evolving.[37] Growing evidence is emerging about the possibility of a 'one-stop shop' ap‐ proach where the anatomy, physiology and perfusion of the heart and coronary arteries, as well as assessment of pulmonary embolism and aortic aneurysm, are all used to diagnose the cause of chest pain in one study.[38] This has been reflected by a readiness of centres in the USA to use the modern 320 MDCT scanners in the emergency department assessment of chest pain.

The potential of using CTCA to determine FFR in the diagnosis of ischaemic heart disease is also an exciting development and has the potential to expand the role of CTCA further by elucidating the haemodynamic significance of the intermediate stenosis. Another avenue of much research lies in the attempt to increase the accuracy of CAD assessment of stable, symptomatic, intermediate-risk patients, by combining the anatomical approach of CTCA alongside myocardial perfusion imaging by CT.[27] What is clear is that CTCA, and CT more broadly, has and will continue to have an expanding role in safeguarding the function of the heart that William Harvey breathlessly outlined for the first time.

#### **Author details**

Michael Campbell1 , Stephen Lyen2,3, Jonathan Rodrigues2,3, Mark Hamilton2,3 and Nathan Manghat2,3\*

\*Address all correspondence to: nathan.manghat@uhbristol.nhs.uk

1 University of Bristol, UK

2 Bristol Heart Institute, University Hospitals Bristol NHS Foundation Trust, Bristol Royal Infirmary, Bristol, UK

3 National Institute for Health Research, Bristol Cardiovascular Biomedical Research Unit, Bristol Royal Infirmary, Bristol, UK

#### **References**


[9] E. P. Widmaier, H. Raff and K. T. Strang, Vander's Human Physiology: The Mecha‐ nisms of Body Function, New York: McGraw-Hill, 2008.

**Author details**

Michael Campbell1

Nathan Manghat2,3\*

1 University of Bristol, UK

Bristol Royal Infirmary, Bristol, UK

94 Coronary Artery Disease - Assessment, Surgery, Prevention

tion, McGraw-Hill Medical, 2007.

Infirmary, Bristol, UK

**References**

, Stephen Lyen2,3, Jonathan Rodrigues2,3, Mark Hamilton2,3 and

2 Bristol Heart Institute, University Hospitals Bristol NHS Foundation Trust, Bristol Royal

3 National Institute for Health Research, Bristol Cardiovascular Biomedical Research Unit,

[1] V. Fuster, R. A. O'Rourke, R. Walsh and P. Poole-Wilson, Hurst's the Heart, 12th Edi‐

[2] Leonard S. Lilly, Pathophysiology of Heart Disease:A collaborative project of medical

[3] A. H. Gershlick et al, "Role of non-invasive imaging in the management of coronary artery disease: An assessment of likely change over the next 10 years. A report from the British Cardiovascular Society Working Group," Heart, vol. 93, pp. 423-431, 2007.

[4] M. C. Williams, J. H. Reid, G. McKillop, N. W. Weir, E. J. van Beek, N. G. Uren and D. E. Newby, "Cardiac and coronary CT comprehensive imaging approach in the as‐

[5] M. R. Patel, E. D. Peterson, D. Dai, J. M. Brennan, R. F. Redberg, R. G. Brindis and P. S. Douglas, "Low diagnostic yield of elective coronary angiography", The New Eng‐

[6] M. Michail, "Feasibility of using coronary computed tomography angiography (CCTA) in patients with acute low-to-intermediate likelihood chest pain," British

[7] A. Wallis, N. Manghat and M. Hamilton, "The role of coronary CT in the assessment and diagnosis of patients with chest pain," Clinical Medicine, vol. 12, pp. 222-229,

[8] G. Bastarrika, C. Thilo, G. F. Headden, P. L. Zwerner, P. Costello and U. J. Schoepf, "Cardiac CT in the assessment of acute chest pain in the emergency department,"

students and faculty. 4th Edition, Lipincott William and Wilkins, 2007.

sessment of coronary heart disease," Heart, vol. 97, pp. 1198-1205, 2011.

land Journal of Medicine, vol. 362, pp. 886-895, 2010.

American Journal of Roentgenology, vol. 2, pp. 397-409, 2009.

Journal of Cardiology, vol. 20, 2013.

2012.

\*Address all correspondence to: nathan.manghat@uhbristol.nhs.uk


[30] J. Sun, Z. Zhang, B. Lu, W. Yu, Y. Yang, Y. Zhou, Y. Weng and Z. Fan, "Identification and quantification of coronary atherosclerotic plaques: A comparison of 64 MDCT and intravascular ultrasound," American journal of Roentgenology, no. 190, pp. 748-754, 2008.

[21] E. A. Hulten, S. Carbonaro, S. P. Petrillo, J. D. Mitchel and T. C. Villines, "Prognostic value of cardiac computed tomography angiography: A systematic review and meta-

analysis," American Journal of Cardiology, vol. 10, no. 57, pp. 1237-1247, 2011.

row CT,"The New England Journal of Medicine, vol. 359, pp. 2324-2336, 2008.

95)," National Institute for Health and Clinical Excellence, London, 2010.

referring clinician.," JACC Cardiovascular Imaging, pp. 460-471, 2008.

ican College of Cardiologists, vol. 52, no. 21, pp. 1724-1732, 2008.

College of Cardiology, vol. 52, no. 25, pp. 2135-2144, 2008.

cine, vol. 52, no. 7, pp. 1079-1086, 2011.

96 Coronary Artery Disease - Assessment, Surgery, Prevention

ogy, vol. 24, no. 57, pp. 2426-2436, 2011.

690–701, 2012.

[23] N. I. f. H. a. C. Excellence, "Chest pain of recent onset:Assessment and diagnosis of recent onset chest pain or discomfort of suspected cardiac origin (NICE guidance

[24] C. V, A. Gutstein, A. Wolak, Y. Suzuki, D. Dey, H. Gransar, L. Thomson, S. Hayes, J. Friedman and D. Berman, "Moving beyond binary grading of coronary arterial sten‐ oses on coronary computed tomographic angiography. Insights for the imager and

[25] Matthew J. Budoff, David Dowe,. James G. Jollis, Michael Gitter, John Sutherland, Edward Halamert, Markus Scherer, RayeBellinger, Arthur Martin, Robert Benton, Augustin Delago and James K. Min, "Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease," Journal of the Amer‐

[26] Meijboom WB, Meijs MF, Schuijf JD, Cramer MJ, Mollet NR, van Mieghem CA, Nie‐ man K, van Werkhoven JM, Pundziute G, Weustink AC, de Vos AM, Pugliese F, Re‐ nsing B, Jukema JW, Bax JJ, Prokop M, Doevendans PA, Hunink MG, Krestin GP, de Feyter PJ"Diagnostic accuracy of 64-slice computed tomography coronary angiogra‐ phy: a prospective, multicenter, multivendor study," Journal of the Journal American

[27] B. Tamarappoo and R. Hachamovitch, "Myocardial perfusion imaging vs computed tomography coronary angiography: When to use which?," Journal of Nuclear Medi‐

[28] F. Bamberg, W. H. Sommer, V. Hoffman, S. Achenbach, K. Nikolaou, D. Conen, M. F. Reiser, U. Hoffman and C. R. Becker, "Meta-analysis and systematic review of longterm predictive value of assessment of coronary atherosclerosis by contrast enhanced coronary computed tomography angiography," Journal American College of Cardiol‐

[29] D. Andreini, G. Pontone, S. Mushtaq, A. L. Bartorelli, E. Bertella, L. Antonioli, A. For‐ menti, S. Cortinovis, F. Veglia, A. Annoni, P. Agostoni, P. Montorsi, G. Ballerini, C. Fiorentini and M. Pepi, "A long-term prognostic value of coronary CT angiography in suspected coronary artery disease," JACC: Cardiovascular Imaging, vol. 5, no. 7, p.

[22] J. M. Miller, C. E. Rochitte, M. Dewey, A. Arbab-Zadeh, H. Niinuma, I. Gottlieb, N. Paul, M. E. Clouse, E. P. Shapiro, J. Hoe, A. C. Lardo, D. E. Bush, A. de Roos, C. Cox, J. Brinker and J. A. Lima, "Diagnostic performance of coronary angiography by 64-

