Magnetic Resonance and Computed Tomographic Imaging of Myocardial Perfusion

Published on 24/02/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1201 times

CHAPTER 53 Magnetic Resonance and Computed Tomographic Imaging of Myocardial Perfusion

Ricardo C. Cury, Anand Soni, Ron Blankstein

Ischemic heart disease is a broad term that includes many different disease processes related to coronary atherosclerosis and its downstream consequences. On a continuum, it encompasses silent myocardial ischemia, chronic stable angina, acute coronary syndromes (ACS), and ischemic cardiomyopathy. All of these disease processes typically share the same underlying mechanism of reduced myocardial perfusion from obstructive coronary plaque, which results in myocardial ischemia or myocardial infarction (MI) from oxygen deprivation.

Data from the National Health and Nutrition Examination Survey (1999-2004, National Heart, Lung and Blood Institute) estimate the prevalence of coronary heart disease to be 16 million individuals in the United States, and incidence of new and recurrent coronary events to be 1.2 million per year.¹ Data from 44 years of follow-up in the original Framingham Heart Study cohort and 20 years of their offspring surveillance show the lifetime risk of developing coronary heart disease for individuals 40 years old to be 49% in men and 32% in women.² With mortality statistics claiming that one out of every five deaths is the result of coronary artery disease (CAD),¹ there is a clear need for improved diagnostic imaging strategies for detecting coronary heart disease and myocardium vulnerable to scenarios of reduced perfusion.

Myocardial ischemia is caused by inadequate coronary perfusion as a result of either an increase in myocardial oxygen requirements or a reduced supply of oxygen-carrying blood. Typically, myocardial ischemia is the result of atherosclerotic CAD. Initially, atherosclerotic vascular disease starts as atheromatous fatty streak buildup within the arterial wall, but in which there is no reduction in coronary blood flow (also known as “positive” remodeling) and patients remain asymptomatic. With further disease progression, there is mild to moderate buildup in plaque, which poses no limits to coronary perfusion at rest or stress, but poses the risk of being “unstable” and abruptly rupturing. If these plaques rupture, the thrombotic consequences can result in near-complete or complete coronary occlusion with severe acute reduction of myocardial perfusion; this is the framework for ACS. Lastly, atherosclerotic plaque may be large enough nearly to obstruct the coronary lumen (also known as “negative” remodeling). This lesion tends to cause reduced myocardial perfusion at times of stress and provides the basis for stable angina.

The clinical presentation of ischemic heart disease is incredibly diverse and includes asymptomatic or silent ischemia, chronic angina, unstable angina or infarction (i.e., ACS), new-onset or chronic heart failure, or sudden cardiac death. Although symptomatic patients are often identified by symptoms related to angina or heart failure, patients who have silent ischemia represent a diagnostic dilemma because they are less likely to be referred for testing. Patients who are particularly susceptible to silent ischemia include diabetics, elderly patients, and patients with prior MI or surgical revascularization.³

Noninvasive imaging is a crucial component in the evaluation of patients with suspected ischemic heart disease. Figure 53-1 illustrates an algorithm for the management of patients with potential ischemic heart disease. For patients presenting with a possible ACS who have a negative ECG and biomarkers (e.g., troponin), a stress test can be used to identify high-risk patients who would benefit from being admitted to the hospital for further evaluation and testing. The American College of Cardiology/American Heart Association guidelines for the management of unstable angina/non–ST segment elevation MI suggest that coronary CT angiography is a reasonable alternative to stress testing in patients with low to intermediate probability of CAD in whom initial ECG and initial biomarkers are unremarkable.⁴

FIGURE 53-1 Algorithm for management of patients with suspected ischemic heart disease. CCS, Canadian Classification System; CHF, congestive heart failure; CTA, CT angiography; DC, discharge; EF, ejection fraction; HF, heart failure; LBBB, left bundle branch block; OMT, optimal medical therapy; Pharm stress, pharmacologic stress testing; PTP, pretest probability; SE, stress echocardiogram; UA, unstable angina.

(Data from Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines [Writing Committee to Revise the 2002 Guidelines for the Management of Patients with Unstable Angina/Non ST-Elevation Myocardial Infarction]: developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons: endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. Circulation 2007; 116:e148-e304; and Gibbons RJ, Abrams J, Chatterjee K, et al. ACC/AHA 2002 guideline update for the management of patients with chronic stable angina—summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines [Committee on the Management of Patients with Chronic Stable Angina]. Circulation 2003; 107:149-158.)

Patients who are admitted with an ACS can be treated with an early invasive or early conservative strategy. High-risk patients have been shown to benefit from early interventions, and warrant early referral for invasive angiography. For low-risk patients, and in particular women, a conservative strategy that uses noninvasive testing is recommended.⁴

For patients with chronic stable angina, the American College of Cardiology/American Heart Association guidelines ⁵ suggest that ejection fraction should be measured for all patients with a history of MI or signs suggesting heart failure. In the presence of a systolic murmur suggestive of aortic stenosis, mitral regurgitation, or hypertrophic cardiomyopathy, an echocardiogram should be obtained.

As is shown in Figure 53-1, all patients with chronic angina (i.e., stable angina) should have an evaluation for the presence of ischemic heart disease. This evaluation can be accomplished by ECG exercise testing, noninvasive stress imaging, or invasive angiography. Higher risk patients or patients with abnormal baseline ECG should be referred for an imaging-based test. The goal of such a test is to identify the extent, severity, and location of ischemia and when possible provide information about prognosis.

The role of cardiac imaging for the detection of silent ischemia is controversial. Patients with diabetes may benefit from such a strategy because they have a high risk of cardiovascular-related mortality, are more likely to have silent ischemia, and are less likely to survive an MI than nondiabetic patients.⁶ Nevertheless, the cost-effectiveness of screening all such at-risk patients with nuclear perfusion imaging is controversial.⁷

The detection of myocardial perfusion abnormalities is mainly used to identify patients with CAD, to evaluate the hemodynamic significance of epicardial coronary stenosis, and to enhance clinical decisions regarding treatment options. Myocardial perfusion imaging (MPI) assessment is frequently performed using rest and stress studies to detect myocardial ischemia (reversible defects) and myocardial infarct (fixed defects).

At present, multiple noninvasive imaging modalities are used in evaluating myocardial perfusion, including single photon emission computed tomography (SPECT), positron emission tomography (PET), echocardiography, and MRI. Although SPECT has been well shown to assess rest and stress perfusion accurately, and provides useful prognostic data based on a patient’s burden of ischemia and infarct, it is limited by its low spatial resolution and attenuation artifacts, which can lead to equivocal studies. ¹⁸FDG-PET has the ability to evaluate myocardial ischemia using radiotracers such as ammonia N 13 or rubidium 82, but has limitations regarding availability. MRI shows promise in the evaluation of myocardial ischemia with first-pass perfusion imaging, and large-scale multicenter trials are just beginning to be released. CT has the best spatial resolution of any imaging modality, and many emerging studies are suggesting it may have an important role for the future evaluation of myocardial perfusion and infarct detection.

This chapter focuses on the evaluation of MRI and CT for the assessment of myocardial perfusion (Figs. 53-2 and 53-3). The advantages and disadvantages of other imaging modalities are discussed and compared with MRI and CT, and these imaging modalities are put in clinical perspective with descriptions of how and when they can facilitate the diagnosis and management of patients suspected to have CAD or with known CAD.

FIGURE 53-2 A 55-year-old man presented with ST segment elevation MI. A, Invasive coronary angiography revealed a significant high-grade stenosis of the distal right coronary artery (arrow). The patient underwent cardiac CT and MRI 3 days later. B, Short-axis CT image showed a perfusion defect (arrow) in mid-inferior and inferolateral wall of the left ventricle. C, MRI first-pass perfusion showed a resting perfusion defect in the same corresponding territory (arrow). D, MDE MRI showed a large area of hyperenhancement (arrows) in the mid-inferior, inferolateral, and inferoseptal walls of the left ventricle with a greater than 50% transmurality. The hyperenhancement pattern was based in the subendocardium and was consistent with a large MI. Within the infarct, a smaller region of no hyperenhancement (arrowhead) was noted that corresponds to a region of no reflow (also known as microvascular obstruction).

FIGURE 53-3 A 40-year-old man was transferred from an outside hospital with MI. A, During invasive coronary catheter angiography, he was found to have a complete occlusion of the left circumflex coronary artery (arrow). B and C, Short-axis (B) and long-axis (C) CT images showed a large perfusion defect of the entire lateral wall (arrows). D, First-pass perfusion MRI confirms this and shows a large perfusion defect of the inferolateral wall (arrow). MDE MRI shows the extent of the MI. Within the area of hyperenhancement, there is a substantial central area that lacks hyperenhancement, representing microvascular obstruction (arrowhead).

MAGNETIC RESONANCE IMAGING

The goal of stress myocardial perfusion MRI is to visualize the first pass of gadolinium contrast agent within the left ventricle blood pool into the myocardium during rest and stress. Stress is achieved with pharmacologic vasodilation with either adenosine or dipyridamole. Under vasodilatory stress, myocardial blood flow should increase four to five times except in areas of obstructive epicardial CAD, where downstream vascular beds are already maximally vasodilated. These coronary territories obtain lower peak myocardial signal intensity on contrast administration compared with resting conditions, and a myocardial perfusion defect is visualized.

Preclinical and Clinical Evaluation

Numerous animal studies have shown good correlation of MRI perfusion with tissue perfusion as measured by radioactive microspheres.⁸^,⁹ Notable studies included work by Wilke and colleagues and Klocke and associates,⁸ in which porcine and canine models of left circumflex coronary artery stenosis were created that showed that MRI can detect different degrees of myocardial perfusion under adenosine stress.

Subsequently, Lee and colleagues ⁹ used a canine model of left circumflex coronary artery stenosis and compared perfusion MRI with Tc 99m sestamibi and thallium 201. They showed that MRI could detect perfusion defects with a left circumflex coronary artery stenosis of 50% or greater, whereas SPECT perfusion defects were detected only with left circumflex coronary artery stenosis of 85% or greater.

Multiple clinical human studies testing the diagnostic accuracy and performance of stress perfusion MRI were performed comparing MRI with nuclear imaging and conventional coronary angiography. Table 53-1 summarizes published data with x-ray angiography as the reference standard. Nandalur and associates¹⁰ performed a meta-analysis of all stress MRI studies with two main techniques in use, perfusion imaging and imaging of stress-induced wall motion abnormalities, from January 1990 to January 2007 with a total of 37 studies involving 2191 patients. All studies used catheter x-ray angiography as the reference standard. Fourteen studies (N = 754 patients) using stress-induced wall motion abnormalities imaging showed 83% sensitivity and 86% specificity on a patient level. Perfusion imaging showed a sensitivity of 91% and specificity of 81% on a patient level (disease prevalence 57.4%).

TABLE 53-1 Diagnostic Accuracy of Stress Perfusion Magnetic Resonance Imaging Studies, Having Invasive Coronary Angiography as the Reference Standard

Two more recent studies by Cury and associates ¹¹ and Klem and coworkers¹² (see Table 53-1) sought to improve the diagnostic accuracy of stress perfusion MRI by using a comprehensive imaging approach of first-pass contrast administration for myocardial perfusion at rest and stress with the addition of myocardial delayed enhancement (MDE) imaging for infarct detection and characterization. Our group showed that this combined approach has 87% sensitivity, 89% specificity, and 88% accuracy, and was superior to rest/stress perfusion MRI alone, which has 81% sensitivity, 87% specificity, and 85% accuracy, having invasive coronary angiography as the reference standard. Klem and coworkers¹² showed similar results—that stress perfusion and MDE MRI can better detect inducible ischemia and fixed defects compared with stress/rest perfusion MRI. MDE MRI also provides the benefit of improved identification of regions of no reflow or microvascular obstruction, which is readily identified on MDE as regions of no hyperenhancement within the hyperenhancing infarction (see Figs. 53-2 and 53-3).

In another study evaluating patients presenting to the emergency department with acute chest pain, Ingkanisorn and colleagues ¹³ evaluated the diagnostic value of adenosine stress myocardial perfusion MRI in 135 patients who presented to the emergency department with chest pain and a negative initial troponin value. The main study outcome was the detection of any evidence of significant CAD. Patients were contacted at 1 year to determine the incidence of significant CAD, defined as significant coronary artery stenosis (>50%) on invasive coronary angiography, abnormal correlative stress test, new MI, or death. Adenosine myocardial perfusion MRI abnormalities had 100% sensitivity and 93% specificity for detection of significant CAD, and an abnormal MRI added significant prognostic value in predicting a future diagnosis of CAD, MI, or death over clinical risk factors. No patients with a normal adenosine myocardial perfusion MRI had a subsequent diagnosis of CAD or an adverse outcome.

The first multicenter trial involving 18 centers in Europe and the United States (MR-IMPACT trial) comparing stress myocardial perfusion MRI with SPECT imaging and invasive coronary angiography was published more recently.¹⁴ When comparing perfusion MRI at 0.1 mmol/kg versus the entire SPECT population, the receiver operating curve analysis showed a better performance for perfusion MRI (n = 42, area under the curve [AUC] 0.86 ± 0.06) versus SPECT (n = 212, AUC 0.67 ± 0.05; P = .013 vs. MRI). The MRI performance at 0.1 mmol/kg was also superior in patients with multivessel disease (n = 32 and n = 161 for MRI and SPECT, AUC 0.89 ± 0.06 vs. 0.70 ± 0.05; P =.006). Overall, the authors concluded that the comparison of stress perfusion MRI with the entire SPECT population suggests superiority of MRI over SPECT, which warrants further evaluation in larger trials. These results are in concordance with a prior study from Ishida and associates^14a that also showed superiority of MRI over SPECT in a single-center trial.

MRI offers an impressive range of information (i.e., function, anatomy, viability, perfusion, and advanced research applications) that is pertinent to the clinical evaluation and subsequent management of patients presenting with acute chest pain. The technique has inherent benefits in terms of its lack of exposure to ionizing radiation and use of nephrotoxic iodinated contrast agents. Initial studies have shown its utility in the setting of acute chest pain. In addition, it has become the gold standard for imaging in the setting of myocardial injury and MI. In the future, MRI may be the imaging study of choice in the evaluation of patients with acute chest pain or MI.

COMPUTED TOMOGRAPHY

Although multiple, more recent single-center and multicenter studies ¹⁵ have established the diagnostic accuracy of cardiac CT for the detection of coronary artery stenosis, the functional significance of many coronary artery lesions identified by such techniques (or by invasive coronary angiography) is often unknown.¹⁶ MPI and angiography have the potential to provide complementary information by imaging ischemia and atherosclerosis. The potential of obtaining this information from a single imaging modality is very attractive.

This section presents a brief overview of the animal and human studies showing the use of CT perfusion during rest and stress. Because most of the initial animal studies employed an infarct model to assess perfusion defects, they showed the use of CT in accurately characterizing rest perfusion defects that are found in infarcted myocardium. In the process of characterizing such resting perfusion defects further, the use of MDE imaging was found to be useful in distinguishing nonviable (i.e., typically infarcted) from viable myocardium. As a natural extension of the ability to assess perfusion, more recent human and animal studies have used adenosine stress perfusion to identify ischemia. Current animal studies in this field use a “stenosis model” to identify lesions that are non–flow-limiting at rest, but are significant during stress. Only more recently have small human studies been conducted showing the feasibility of using multidetector CT to characterize perfusion during adenosine stress.

Animal Studies: Rest Perfusion

The ability of CT to detect acute MI in explanted hearts and experimental animal models dates back to the late 1970s.¹⁷ More recently, multidetector CT has been used to assess myocardial perfusion in animal models of total coronary occlusion.¹⁸^,¹⁹ Table 53-2 lists some of the main studies that characterized myocardial perfusion in animal models under rest and stress conditions.

TABLE 53-2 Multidetector Computed Tomography Rest and Stress Perfusion: Animal Studies

Hoffmann and coworkers ¹⁸ used four-slice multidetector CT and performed a quantitative analysis of CT attenuation and compared that with microsphere-determined blood flow and triphenyltetrazolium chloride (TTC)–stained tissue samples. The quantitative analysis by multidetector CT showed significant differences in the mean CT attenuation of infarct and reference areas (32.1 ± 8.5 Hounsfield units [HU] vs. 75.6 ± 16.7 HU; P < .001), and this correlated with changes in microsphere-determined blood flow. The volume of perfusion defect was similar to volume of tissue that lacked TTC staining (17 ± 6.4% vs. 13.6 ± 6%), with slight overestimation of infarct size by multidetector CT.

Mahnken and colleagues ¹⁹ used a similar porcine model to assess the ability of multidetector CT to evaluate rest myocardial perfusion versus first-pass perfusion MRI with TTC staining serving as the gold standard. In their protocol, they used dynamic multidetector CT imaging by acquiring 64 scans at the apical level with a prospectively acquired ECG triggered examination protocol. Hypoenhanced regions on multidetector CT corresponded directly to perfusion defects visualized on MRI and areas of MI seen on TTC staining. The hypoenhanced regions detected by multidetector CT were again slightly larger than areas of acute MI as detected by MRI and TTC staining.

These studies exploited the ability of multidetector CT to visualize areas of myocardial hypoattenuation, indicative of decreased myocardial perfusion. Areas of decreased myocardial perfusion in the setting of MI can represent either myocardial necrosis with microvascular obstruction or areas of infarction with preserved microvascular obstruction. Direct comparisons of multidetector CT delayed enhancement with TTC showed excellent correlation with infarct morphology, transmurality, and infarct volume ratios. Additionally, analogous to MRI, multidetector CT hypoenhanced identified regions of microvascular obstruction at 5 minutes after contrast agent injection in the delayed enhancement images that compared well with thioflavin S–derived measurements.

Human Studies of Rest Perfusion

Table 53-3 summarizes several key human studies of stress and rest myocardial perfusion using multidetector CT. Nikolaou and colleagues²⁰ attempted to correlate rest multidetector CT to stress perfusion MRI and MDE MRI in 30 patients with chronic infarcts or suspected CAD or both. In this retrospective study, all patients previously underwent multidetector CT (16 detectors) and MRI within 10 ± 16 days. Multidetector CT was able to detect 13 of 17 perfusion defects correctly (sensitivity 76%, specificity 92%, accuracy 83%); however, when considering only the 6 perfusion defects not associated with chronic MI, the sensitivity decreased to 50%—not surprising given that multidetector CT was performed under resting conditions, whereas stress perfusion MRI used vasodilator-induced hyperemic blood flow. Comparing multidetector CT versus MDE MRI for detection of infarct resulted in a sensitivity of 91%, specificity of 79%, and accuracy of 83%. The attenuation values in the 10 infarcted areas correctly detected by multidetector CT were significantly lower than in noninfarcted areas of myocardium (53.7 ± 33.5 HU vs. 122.3 ± 25.5 HU; P < .01). In the volumetric assessment of infarct size, a strong correlation between the volumes of 16-multidetector CT and MDE MRI was found (r = 0.98), but 16-multidetector CT tended to underestimate the infarct volume as assessed by cardiac MRI by 19% (P < .01).

TABLE 53-3 Multidetector Computed Tomography Rest and Stress Perfusion: Human Studies

More recently, Nieman and coworkers ²¹ retrospectively tested the hypothesis that 64-multidetector CT can differentiate recent (<7 days) versus old (>12 months) MI. They found significantly lower CT attenuation values in patients with long-standing MI (−13 ± 37 HU) than patients with acute MI (26 ± 26 HU) and normal controls (73 ± 14 HU; P < .001). The attenuation difference between infarcted and remote myocardium was larger in patients with long-standing MI than in patients with recent MI (89 ± 41 HU and 55 ± 33 HU; P < .001), probably owing to fatty replacement. As anticipated, long-standing MI was associated with wall thinning and ventricular dilation, whereas recent MI was not (P > .05).

Stress Multidetector Computed Tomography to Identify Ischemia

George and colleagues ²² performed rest and adenosine-mediated stress multidetector CT on a canine model of left anterior descending artery stenosis and were able to achieve lesions that were non–flow-limiting at rest but flow-limiting during pharmacologic stress. They were able to identify perfusion defects in noninfarcted myocardium. By using microspheres to measure myocardial blood flow, they were able to show that the myocardial signal density ratio (myocardial signal density/left ventricular blood pool signal density) corresponded well with microsphere-derived myocardial blood flow.

Preliminary human studies investigating the feasibility and accuracy of multidetector CT stress myocardial perfusion have been presented only more recently. In 19 patients with an abnormal SPECT study, George and colleagues ²³ used 256-detector multidetector CT to assess for subendocardial perfusion defects during rest and intravenous adenosine infusion. When compared with 50% or greater stenosis by CT angiography, the sensitivity and specificity of multidetector CT stress MPI were 85% and 77% compared with 69% and 74% sensitivity and specificity of SPECT. When compared with a gold standard combining 50% or greater stenosis by CT angiography with a SPECT perfusion defect, the 256-row multidetector CT was 78% sensitive and 90% specific. Although these findings represent preliminary work, they suggest that ischemia can be detected with a modest sensitivity and reasonably high specificity.

Most of the multidetector CT studies reviewed in this section are small, and many of the imaging approaches are continuously being improved. Currently, there are no widely accepted multidetector CT protocols for properly imaging the myocardium in patients with suspected myocardial ischemia.

INDICATIONS

Magnetic Resonance Imaging

The indication for stress perfusion MRI is the evaluation of myocardial ischemia in the assessment of chest pain syndrome in patients with low to moderate pretest probability of having CAD. Patients with a high pretest likelihood of significant disease should proceed directly to invasive coronary angiography. These indications are similar to nuclear MPI with SPECT or PET and stress echocardiography. The improved spatial and temporal resolution of MRI compared with SPECT makes MRI a superior imaging modality with fewer false-positive results.

Computed Tomography

Although clinical data are currently too limited to advocate use of CT as a primary modality to assess MI or to evaluate for perfusion defects, in patients already undergoing CT for evaluation of coronary anatomy, rest perfusion defects—when present—can be very helpful in identifying areas of infarcted myocardium.

CONTRAINDICATIONS

Magnetic Resonance Imaging

The contraindications for myocardial perfusion MRI are similar to contraindications for other MRI examinations. The most common contraindications include periorbital metallic fragments or intracranial aneurysm clips, magnetically activated implanted devices, insulin pumps, neurostimulators, and cochlear implants. Because stress MRI involves gadolinium administration, decreased renal function is also a contraindication because of the rare occurrence of nephrogenic systemic fibrosis. Other contraindications include the inability to cooperate with breath-holding exercises and claustrophobia.

Computed Tomography

Contraindications for CT include renal insufficiency and known allergy to iodinated contrast agents. Because CT is associated with a low level of exposure to ionizing radiation, individuals who may potentially be more susceptible to adverse effects of radiation (e.g., young women) may be viewed as having a relative contraindication.

TECHNIQUE DESCRIPTION

Magnetic Resonance Imaging

Patients are monitored during the entire MRI scan with continuous ECG recording, vital signs, and pulse oximetry. A standard stress perfusion MRI protocol consists of the following steps; the entire examination takes approximately 40 minutes.

1 Localization. Localization of the heart position in the axial, sagittal, and coronal planes of the chest are obtained using a nongated breath-hold steady-state free precession (SSFP) pulse sequence. After localization of the left ventricle, paraseptal long-axis, short-axis, and four-chamber cine views are obtained.

2 Adenosine stress scan. Adenosine infusion is started at 0.14 mg/kg/min over 6 minutes. Five minutes into the adenosine infusion, 0.1 mmol/kg of gadolinium-chelate contrast agent is injected at 5 mL/s followed by 15 mL of saline. Five seconds after initiation of contrast agent injection, the stress perfusion MRI sequence is acquired during one breath-hold in the short-axis view in 8 to 10 locations in the left ventricle to monitor the wash-in of the gadolinium through the myocardium.

3 Left ventricular functional images. Left ventricular function is assessed over 8 to 10 short-axis slices using ECG-gated breath-holding SSFP pulse sequences.

4 Resting perfusion scan. Another 0.1 mmol/kg of gadolinium-chelate contrast agent is injected at 5 mL/s followed by 15 mL of saline. Five seconds after initiation of contrast injection, a rest perfusion MRI sequence is acquired during one breath-hold in the short-axis view in 8 to 10 locations in the left ventricle to monitor the wash-in of the gadolinium through the myocardium.

5 Myocardial delayed enhancement. MDE imaging is obtained 7 to 10 minutes after the aforementioned scans in the left ventricular short-axis and long-axis views to determine the presence and extent of MI.

Computed Tomography

There are no standard techniques for the assessment of myocardial perfusion using CT. In our center, an 18-gauge intravenous catheter is placed for contrast injection for all CT examinations. If stress perfusion is planned, a second intravenous catheter (20-gauge) is placed for adenosine administration. The patient is brought to the CT suite, and standard ECG monitoring leads are connected. For studies involving rest or stress perfusion assessment, we do not administer any intravenous nitroglycerin or β blockers.

PITFALLS AND SOLUTIONS

Magnetic Resonance Imaging

Dark rings that occur mainly in the septum (Gibbs artifacts) are the main artifact in stress perfusion MRI, and are due to high susceptibility of gadolinium contrast agent in the right and left ventricles. Otherwise, myocardial perfusion MRI is a reliable examination. The major issue with execution of the protocol is patient participation and cooperation with breath-holding. There are pulse sequence–specific imaging artifacts that an experienced operator needs to recognize and correct, but generally this modality is faster, more reliable, and less subject to imaging artifact and false-positive results compared with nuclear perfusion imaging.

Computed Tomography

CT-based evaluation of myocardial perfusion offers the advantage of superb spatial resolution, but it has lower image contrast resolution, especially compared with MRI. In addition, artifacts may appear as areas of hypoenhancement and be falsely interpreted as perfusion defects. CT artifacts may occur because of cardiac motion or image noise. A particular artifact that can lead to an area of decreased attenuation is beam hardening—decreased mean energy of x-ray beams when they pass through a dense object. Although beam hardening is usually caused by metallic objects such as clips, it may be caused by areas of dense contrast enhancement in natural structures such as the ventricles.

When a multiphase series is available, a true perfusion defect persists throughout multiple different phases of the cardiac cycle and is visualized in systole and diastole. Correlation of perfusion data with functional images is also helpful in improving the specificity of the CT examination.

IMAGE INTERPRETATION

Postprocessing

Magnetic Resonance Imaging

Comprehensive MR image analysis first needs to assess perfusion defects on stress imaging. Comparison with resting perfusion is mandatory to assess reversibility, similar to nuclear imaging. A perfusion defect under vasodilator stress in a coronary territory that appears normal at rest indicates a significant epicardial coronary artery stenosis. A stress perfusion defect that remains “fixed” at rest may indicate a chronic infarction or artifact. MDE images then become vital because a fixed perfusion defect owing to infarction would have a matched area of hyperenhancement on delayed images.

The absence of significant hyperenhancement on MDE (Fig. 53-4) is consistent with viable myocardium, and suggests that the patient would likely benefit from myocardial revascularization (i.e., percutaneous coronary intervention or coronary artery bypass graft surgery). Lastly, left ventricular systolic function and wall motion under stress are assessed. A wall motion abnormality in a coronary territory that matches perfusion defects is further evidence of ischemic myocardium. The diagnosis of CAD is made in the presence of a stress-induced perfusion defect or if an MI is detected on MDE images.