Myocardial Nuclear Scintigraphy

SPECT uses the principles of radioactive decay to evaluate the myocardium and its blood supply. It is able to detect the presence of flow-limiting coronary artery stenosis, as well as myocardial infarction. The stability of the nucleus for emitting radiation depends on the ratio of neutrons to protons and on the nuclide’s atomic number (Z). The sources used for this are known as radionuclides, which are nuclides with neutron-proton ratios that are not on the stable nuclei curve and are unstable and, therefore, radioactive. There are several types of radioactive decay. The least penetrating radiation is called an alpha particle (α), which corresponds to the heaviest radiation. An alpha particle is composed of the nuclei of a helium atom (2 protons + 2 neutrons) with positive charge. A second type of radioactive decay is known as beta (β) particle emission, which is moderate penetrating radiation. Beta particles are lighter than alpha particles and are actually electrons emitted from the nucleus. Positron (β⁺) particles, which are positive electrons, have similar penetration to beta particles but are made of antimatter and emitted from positron tracers. Lastly, the highest energy emission particles are known as gamma (γ) rays and are the same as particles emitted from an X-ray tube.

The radionuclides that are used in SPECT are technetium-99m (Tc^99m) and thallium-201 (Tl²⁰¹). Tc^99m is a large radionuclide that emits a single photon or γ-ray per radioactive decay, with a half-life of 6 hours. The energy of the emitted photon is 140,000 electron volts, or keV. Thallium-201 is less commonly used and decays by electron capture. It has a much longer half-life than Tc^99m of 73 hours, and the energy emitted is between 69 and 83 keV. To obtain images, the gamma rays that are released by decay from the body must be captured and modified by a detector or gamma camera. The standard camera is composed of a collimator, scintillating crystals, and photomultiplier tubes. When a radionuclide emits gamma rays, it does so in all directions. A collimator made of lead with small, elongated holes is used as a filter to accept only those gamma rays traveling from the target organ toward the camera. Once the selected gamma rays have reached the scintillating crystals, they are converted to visible light and then into electrical signals by the photomultiplier tubes. These electrical signals are then processed by a computer to form images. Myocardial regions that are infarcted or ischemic after stress will have relatively decreased tracer uptake and, therefore, decreased signal or counts in the processed images.

PET is similar to SPECT in that it uses radioisotopes and the properties of radioactive decay to produce and acquire images. The most common radioisotopes used for cardiac evaluation are rubidium-82, N-ammonia-13, and fluorine-18 (F¹⁸). F¹⁸ is a much smaller radionuclide than Tc^99m. It emits a positron (β⁺) antiparticle. This ionized antiparticle travels until it interacts with an electron. The electron and the positron are antiparticles of each other, meaning they have the same mass but are opposite in charge. When this occurs, both particles disintegrate and are converted into energy in the form of two photons traveling in opposite directions. Both photons have the same energy, 511 keV. This phenomenon is known as pair annihilation, which is used to create the images in PET. PET cameras also differ from SPECT cameras in that they capture only incoming photons that travel in opposite directions and arrive at a circular detector around the body at precisely the same time. PET detectors have much higher sensitivity than SPECT cameras because they do not require a collimator. Like in SPECT, PET cameras also use scintillating crystals and photomultiplier tubes. Recently, PET systems have been combined with computed tomography (CT) and magnetic resonance imaging (MRI) systems to simultaneously display PET metabolic images with their corresponding anatomic information.

Cardiac Computed Tomography

CCT has grown significantly in clinical use since the early 2000s with the advent of multidetector CT scanners with submillimeter resolution allowing evaluation of the coronary anatomy. The X-ray tube produces beams that traverse the patient and are received by a detector array on the opposite side of the scanner. The X-ray tube and detector array are coupled to each other and rotate around the patient at a velocity of 250 to 500 msec/rotation. Initially, in 1999, the first multidetector CT scan used for coronary imaging had four rows of detectors and had a scanning coverage of 2 cm per slice rotation. Breath-holds on the order of 10 to 20 seconds were required to cover the entire heart. Artifacts produced by the patient’s respiration and heart rate variability rendered many studies nondiagnostic for the assessment of coronary stenosis. Technology has advanced at a rapid pace to the point that 64-slice systems are standard, and 320-slice systems with 16 cm of coverage are able to capture the entire heart in one heartbeat and rotation.

CCT utilizes ionizing radiation for the production of images. Concern over excessive medical radiation exposure has been raised in recent years. Although several techniques, such as prospective electrocardiogram (ECG)-gated acquisition, may be implemented^1–3 to reduce radiation dose, a risk-benefit assessment must be done for the selection of patients who have appropriate indications for CCT. The patient’s heart rate must be lowered to less than 65 beats/min to achieve adequate results imaging the coronaries with CCT. This usually requires the administration of oral or intravenous β-blockers. After the scan has been completed, images are reconstructed at different intervals of the cardiac cycles and analyzed in a computer workstation.

Cardiovascular Magnetic Resonance Imaging

Cardiovascular magnetic resonance is a robust and versatile imaging modality. It is able to evaluate multiple elements of cardiac status: function, morphology, flow, tissue characterization, perfusion, angiography, and/or metabolism. CMR is able to do this using its unique ability to distinguish morphology by taking advantage of the different molecular properties of tissues. This is achieved without the use of any radiation, by using the influence of magnetic fields on the abundance of hydrogen atoms in the human body. This is one of the main advantages of CMR over other imaging modalities. Multicontrast CMR uses the intrinsic properties of organs and takes advantage of the three imaging contrasts: T1, T2, and proton density without the need for gadolinium contrast. T1-weighted imaging is utilized for the imaging of lipid content and fat deposition appears bright or hyperintense. T2-weighted imaging is used for the evaluation of edema⁴ and fibrous tissue,⁵ which also appears hyperintense. Dynamic contrast-enhanced CMR uses the paramagnetic contrast agent gadolinium, which enhances the magnetization (T1) of protons of nearby water and creates a stronger signal. In addition, gadolinium contrast permeates through the intercellular space in necrotic or fibrotic myocardium, which is the basis for myocardial scar detection seen on late gadolinium enhancement.

CMR is able to evaluate both ventricular and valvular function. It also can evaluate atherosclerosis⁶ in large vessels and is capable of imaging morphology and distinguishing between different elements of atherosclerotic plaque composition including fibrous tissue, lipid core, calcification, and hemorrhage.⁷ In addition to vascular plaque assessment, CMR may be used for the evaluation of ischemia after the administration of gadolinium contrast agents. First-pass perfusion is evaluated at rest and after the administration of a pharmacologic stressor such as adenosine or dobutamine for the evaluation of myocardial infarction and ischemia.

Vascular Ultrasound

Vascular ultrasound has been in existence clinically since the 1950s. It is versatile and relatively inexpensive when compared with other imaging modalities. It is one of the few imaging techniques that may be performed at the patient’s bedside. In addition, there is no use of ionizing radiation, as opposed to CT or nuclear cardiology. For these reasons, vascular ultrasound can never be replaced in the clinical setting.

Vascular ultrasound is composed of several techniques or modes, which include grayscale imaging (also known as B-mode), pulsed- and continuous-wave Doppler imaging, and color Doppler imaging. Each of these provides different information. Duplex ultrasound uses both B-mode and pulsed-wave Doppler to acquire vessel anatomy, as well as hemodynamic data. This includes peak and mean velocities of blood flow in addition to pressure gradients caused by stenosis. Duplex is also used for the evaluation of aneurysms and dissections. Color-flow Doppler allows for the visualization and direction of blood flow through vessels. Typically, the color scale is from red (flow toward transducer) to blue (flow away from transducer; see Chapter 12). Many times it aids in the localization and identification of vessels when duplex is inadequate. Vascular ultrasound is used for the evaluation of the aorta; carotid, renal, celiac, and mesenteric arteries; the lower extremity arterial system; and the peripheral venous system. More recently, it also has come into clinical use for the evaluation of atherosclerosis by measuring carotid intima-media thickness.

Left Ventricular Systolic Function

Perhaps the most important factor that contributes to surgical outcome is cardiac function, specifically left ventricular (LV) systolic function. Systolic dysfunction is directly related to patient outcome after surgery. Preoperative knowledge of LV systolic dysfunction is crucial for the anesthesiologist to prepare and anticipate perioperative and postoperative complications. Patients with systolic dysfunction who undergo coronary artery bypass graft (CABG) surgery require more inotropic support after cardiopulmonary bypass (CPB).^8,9 In addition, systolic dysfunction is a good prognosticator for postsurgical mortality.^10–12 In patients who are known to have CAD and are scheduled to have CABG surgery, the cause of systolic dysfunction is, most often than not, ischemic heart disease. In patients who are scheduled to have elective noncardiac surgery and are found to have newly diagnosed systolic dysfunction, it is important to do further testing to find the cause and exclude critical coronary stenosis and ischemia.

Transthoracic echocardiography (TTE) is the most widely used modality for this evaluation because it is inexpensive, portable, and readily available. However, limited acoustic windows may limit the accuracy of echocardiographic assessment of global and regional LV function in a significant number of patients.¹³

Nuclear scintigraphic methods, including both SPECT and PET myocardial perfusion imaging, can be used to evaluate global and segmental LV systolic function. This is achieved by implementing ECG gating during data acquisition. Most often, eight frames or phases are acquired per cardiac cycle. The left ventricular ejection fraction (LVEF) is measured using absolute end-diastolic (EDV) and end-systolic volumes (ESV), where LVEF = LVEDV − LVESV/LVEDV.

Gated images can be acquired at both rest and after stress; however, rest images typically have less radiation dose and the images may be noisy. In most institutions, gated imaging is done using poststress images because of the higher radioisotope dose and, thus, less noise. This does have its limitation for accurate LV systolic analysis in the circumstance of stress-induced ischemia, in which myocardial stunning can transiently reduce the LVEF. Another limitation of ECG-gated SPECT or PET is arrhythmias, specifically frequent premature ventricular contractions (PVCs) or atrial fibrillation.¹⁴ In patients who have extensive myocardial infarction, assessment of LV function also may be inaccurate because there is absence of isotope in the scar regions; thus, the endocardial border cannot be defined. Gated-blood pool scans (multiple gated acquisition; MUGA) image the cardiac “blood pool” with high resolution during the cardiac cycle. Ventricular function, as well as various temporal parameters, can be measured using this technique.¹⁵ There is good correlation between echocardiography and MUGA for the evaluation of LVEF. However, MUGA has demonstrated better intraobserver and interobserver reproducibility than echocardiography.¹⁶

CCT, with its excellent spatial and temporal resolution, allows for an accurate assessment of LV function when compared with echocardiography, invasive ventriculography, and cardiac MRI.^17–19 CCT also uses real three-dimensional volumes to calculate the LV systolic function. Functional analysis can be evaluated only when retrospective scanning is used because the entire cardiac cycle (both systole and diastole) is necessary. The raw dataset must be reconstructed in intervals or cardiac phases of 10%, from 0% (early systole) to 90% (late diastole). Advanced computer workstations allow cine images to be reconstructed and displayed in multiple planes (Figure 2-1). Segmental wall motion analysis may be performed using the 17-segment model recommended by the American Heart Association/American College of Cardiology (AHA/ACC)²⁰ (Figure 2-2).

Figure 2-1 Computed tomography angiography: left ventricular (LV) functional analysis in three orthogonal planes using specialized workstation. It allows for the evaluation of LV end-diastolic and end-systolic volumes, mass, and ejection fraction.

Figure 2-2 American Heart Association/American College of Cardiology (AHA/ACC)–recommended 17-segment model for left ventricular segmental wall motion analysis.

(From Cerqueira MD, Weissman NJ, Dilsizian V, et al: Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 105:539–542, 2002.)

The main limitation to using CCT for LV systolic function assessment is the required radiation exposure. Because retrospective ECG gating is required to image the entire cardiac cycle, radiation exposure is relatively high. In comparison, CCT studies performed with prospective ECG gating expose the patient to radiation during only 10% to 15% of the cardiac cycle. Thus, in most clinical scenarios, LV functional information usually is not acquired to reduce radiation exposure.

CMR is considered the gold standard for the quantitative assessment of biventricular volumes, EF, and mass, whereas also offering excellent reproducibility.²¹ CMR also has excellent spatial and temporal resolution allowing for cine imaging. Typically, a stack of 10 to 14 contiguous two-dimensional slices are acquired and used for LV functional analysis.²² The acquisition of each of these images generally requires a breath-hold of at least 10 to 20 seconds. In a computer workstation, the endocardial and epicardial contours of the LV can be traced in each short-axis slice at the phases of maximal and minimal ventricular dimensions. The software then calculates the volume of ventricular cavity per slice as the product of the area enclosed within the endocardial contour multiplied by the slice thickness. The data are then combined to calculate EDV and ESV and EF. In addition, cine images may be acquired in the four-, three-, and two-chamber views for LV segmental wall analysis (Figure 2-3).

Figure 2-3 Cardiac magnetic resonance demonstrating (A) short-axis, (B) two-chamber, (C) four-chamber, and (D) three-chamber views.

Left Ventricular Diastolic Function

Diastolic dysfunction is the most common abnormality found in patients with cardiovascular disease.^23,24 Patients with diastolic dysfunction may be asymptomatic²⁵ or may have exercise-induced dyspnea or overt heart failure.²⁶ Until recently, the profound impact of diastolic dysfunction on perioperative management and postoperative outcome has been underestimated. In fact, the prevalence of diastolic dysfunction in patients undergoing surgery is significant. A recent study demonstrated that in more than 61% of patients with normal LV systolic function undergoing surgery, diastolic filling abnormalities were present.²⁷ This is critical information for the anesthesiologist because patients with diastolic dysfunction who undergo CABG require more time on CPB, as well as more inotropic support up to 12 hours after surgery.²⁸ This may be because of deterioration of diastolic dysfunction after CABG, which may persist for several hours.^29–31 Taking all this into account, diastolic dysfunction increases the risk for perioperative morbidity and mortality.³²

In 85% of patients with diastolic dysfunction, hypertension is the primary cause. Diastolic function requires a complex balance among several hemodynamic parameters that interact with each other to maintain LV filling with low atrial pressure, including LV relaxation, LV stiffness, aortic elasticity, atrioventricular and intraventricular electrical conduction, left atrial contractility, pericardial constraint, and neurohormonal activation. Changes in preload, afterload, stroke volume, and heart rate can upset this delicate balance.^33–35

LV diastolic function is most easily and commonly assessed with echocardiography; however, different aspects of diastolic function also can be evaluated by SPECT and CMR. At least 16 phases of the cardiac cycle need to be acquired to evaluate diastolic dysfunction using SPECT. This is because diastolic functional analysis, as opposed to systolic function, is dependent on heart rate changes during acquisition and processing. The two main parameters that can be measured by SPECT are LV peak filling rate and time to peak filling rate. It is measured in EDV/sec, and is normally more than 2.5. The normal time to peak filling rate is less than 180 milliseconds. Heart rate, cardiovascular medications, and adrenergic state may alter these parameters.³⁶

Velocity-encoded (phase-contrast) cine-CMR is capable of measuring intraventricular blood flow accurately and is able to quantify mitral valve (MV) and pulmonary vein flow, which are hemodynamic parameters of diastolic function. It has been shown that in patients with amyloidosis, echocardiography and velocity-encoded cine imaging correlate significantly in estimating pulmonary vein systole/diastole ratios, LV filling E/A ratio, and E deceleration times, which are all diastolic functional indices.³⁷ In addition to measuring blood flow and velocity through the MV and pulmonary vein, CMR-tagging is able to measure myocardial velocities of the walls and MV similar to strain rate and tissue Doppler in echocardiography. CMR-delayed enhancement imaging also is used for the diagnosis of diastolic dysfunction. The presence and severity of fibrosis seen on delayed-enhancement imaging correlate significantly with severity of diastolic dysfunction.³⁸

Right Ventricular Function

In preoperative evaluation, knowledge of right ventricular (RV) dysfunction is critical for intraoperative management of the patient. RV dysfunction is an independent risk factor for clinical outcomes in patients with cardiovascular disease.^39–41 Patients with RV dysfunction in the presence of LV ischemic cardiomyopathy who undergo CABG surgery have increased risk for postoperative and long-term morbidity and mortality.⁴² Patients with RV dysfunction often require postoperative inotropic and mechanical support, resulting in longer surgical intensive care unit and hospital stays.⁴² In patients who undergo mitral and mitral/aortic valve surgery, RV dysfunction is a strong predictor of perioperative mortality.⁴³ In addition, RV dysfunction is associated with postoperative circulatory failure.⁴⁴ If RV dysfunction is detected before or after surgery, further evaluation is necessary. In the case of preoperative RV dysfunction, pulmonary hypertension (PH) is a common cause that negatively impacts perioperative and postoperative outcome. PH significantly increases morbidity and mortality in patients undergoing both cardiac^45,46 and noncardiac surgery.^47,48 Patients with acute onset of RV dysfunction without an explained cause must be evaluated for pulmonary emboli. Recent studies have demonstrated that the incidence rate of pulmonary emboli after CABG surgery can be as high as 3.9%.^49–51

The RV is designed to sustain circulation to the pulmonary system while preserving a low central venous pressure. Patients with RV dysfunction can maintain relatively normal functional capacity unless pulmonary vascular resistance is increased, at which point RV function is critical for pulmonary circulation. RV failure is characterized by venous congestion (i.e., hepatomegaly, ascites, edema), as well as decreasing LV preload and cardiac output. There is also an interdependence between the RV and LV imposed by the pericardium that can negatively affect LV filling. There are several mechanisms for RV dysfunction including primary causes like RV infarction and RV dysplasia, as well as secondary causes because of LF dysfunction. The severity of RV dysfunction may be difficult to evaluate by TTE at times because of suboptimal acoustic windows. Furthermore, the ability to derive accurate and reproducible estimations of RVEF by echocardiography is limited by the complex changes in RV geometry that occur as the right ventricle dilates.

CMR is the most accurate method for the assessment of RVEF and volumes.^52,53 The RV is evaluated in a similar manner to the LV by CMR, where short-axis cine slices from ventricular base to apex are obtained and measured in a computer workstation. CMR is the gold standard for the diagnosis of RV dysplasia, providing assessment of global and regional function, as well as detecting the presence of myocardial fat infiltration and scarring.^54,55

Global and segmental RV function also may be evaluated using first-pass radionuclide angiography (FPRNA). RVEF obtained by FPRNA has been shown to have good correlation with CMR.⁵⁶

CCT also is very accurate for RV functional assessment when compared with CMR.^57,58 The protocol used to acquire RV data is different from that used for coronary artery evaluation. A biphasic contrast injection is used to opacify the RV. In addition, retrospective ECG gating must be utilized to acquire the entire cardiac cycle for functional evaluation. CCT is, therefore, not frequently used primarily for RV functional assessment because the radiation dose is generally higher than for FPRNA and CMR.

RV dysfunction is a common cause of post- and perioperative hypotension and is associated with poor outcomes, regardless of its cause. New onset of RV dysfunction may be caused by RV infarction, pulmonary embolism, or acute respiratory failure (cor pulmonale). Echocardiography is more suitable than other imaging modalities in these cases because it is a portable imaging technique. Moreover, echocardiography allows estimation of RV systolic pressure, which is usually elevated in pulmonary embolism and respiratory failure, and low or normal in RV infarction.

Exercise versus Pharmacologic Testing

Preoperative assessment for ischemic burden in patients with CAD or those at risk for CAD who are to have elective noncardiac surgery is important. Figure 2-4 indicates the ACC/AHA algorithm for preoperative cardiac evaluation and care before noncardiac surgery. Nuclear myocardial perfusion imaging is the most common test used in the United States for preoperative evaluation. Patients can be stressed using exercise or pharmacologic agents. The preferred modality is exercise, which is most often done on a treadmill and less commonly on a stationary bike.⁵⁹ For an exercise stress test to be adequate, a patient must exercise for at least 6 minutes and reach at least 85% of their maximum predicted heart rate (MPHR) adjusted for their age (MPHR= 220 − age). Uniform treadmill protocols are used to compare with peers and serial testing. The most common protocols used are Bruce and modified Bruce. In addition, exercise stress tests are symptom limited. Exercise as a stressor has robust prognostic data for the risk for future cardiac events. There are several types of scores that predict a patient’s risk for cardiovascular disease. The most commonly used score is known as the Duke treadmill score, which uses exercise time in minutes, maximum ST-segment deviation on the ECG, and anginal symptoms during exercise. Heart rate recovery to baseline after exercise is also a strong predictor for cardiovascular disease. In general, exercise stress testing is safe as long as testing guidelines are followed carefully. The risk for a major complication is 1 in 10,000.

Figure 2-4 American Heart Association/American College of Cardiology (AHA/ACC) algorithm for preoperative evaluation for patients planning to go for noncardiac surgery. HR, heart rate; LOE, level of evidence.

(From Fleisher LA, Beckman JA, Brown KA, et al: ACC/AHA 2007 Guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: Executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines [Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery] developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. J Am Coll Cardiol 50:1707–1732, 2007.)

For myocardial perfusion imaging, a radioisotope must be injected during exercise. When using Tc^99m, it must be injected once the patient has reached peak heart rate and the patient must exercise for at least 1 minute afterward to allow sufficient time for the radioisotope to circulate through the myocardium.

Pharmacologic stress testing is a negative prognosticator in itself because patients who, for one reason or another, are not able to do sufficient physical activity to attempt an exercise stress test have greater incidences of cardiovascular disease and other comorbidities. Pharmacologic stress testing is also preferred in patients with a left bundle branch block, Wolf-Parkinson-White (WPW) pattern, and ventricular pacing on ECG. There are two types of pharmacologic agents available on the market today: vasodilators that include dipyridamole, adenosine, and regadenoson; and the chronotropic agent, dobutamine. They each have their advantages and disadvantages. Dipyridamole was the original stressor used for myocardial perfusion imaging. It is an indirect coronary vasodilator that prevents the breakdown and increases intravascular concentration of adenosine. It is contraindicated in those patients with asthma and those with chronic obstructive pulmonary disease (COPD) who have active wheezing. Adenosine is used more widely now because it produces fewer side effects compared with dipyridamole. It induces coronary vasodilation directly by binding to the A2A receptor. Adenosine has similar contraindications to dipyridamole. Known side effects include bronchospasm, as well as high-degree AV block; however, because the half-life is seconds, it is usually enough just to discontinue the adenosine infusion and symptoms resolve without further treatment. If the patient is able to walk slowly on the treadmill, adenosine is given while the patient walks at a constant slow pace to alleviate the severity of potential side effects. In addition, image quality is improved with low-level exercise because there is less tracer uptake in the gastrointestinal system. Regadenoson is a relatively new agent to the market. It is a selective adenosine analog. It is given as a single intravenous (IV) bolus and has less incidence of significant AV block. However, it also may cause bronchospasm in patients with asthma or active COPD.⁶⁰

Dobutamine is a chronotropic agent that is more often used during stress echocardiography. Dobutamine may be used as a stressor during myocardial perfusion imaging if the patient is not able to exercise or if the patient cannot use a vasodilator secondary to asthma or COPD exacerbation. It also should not be used in patients with left bundle branch block or WPW. Dobutamine causes the heart rate and blood pressure to increase. After the radioisotope is injected, when the patient reaches at least 85% of MPHR, dobutamine infusion must be continued for an additional 2 minutes. In case of ischemia or severe side effects, short-acting β-blockers (esmolol) should be given to counteract the effects.

Single-Photon Emission Computed Tomography versus Positron Emission Tomography Myocardial Perfusion Imaging

Myocardial perfusion imaging can be performed using both SPECT and PET. They are based on LV myocardial uptake of the radioisotope at rest and after stress. Myocardial uptake will be reduced after stress in corresponding myocardial regions where significant coronary artery stenosis is present. The images are displayed in three different orientations for proper LV wall-segment analysis. The three LV orientations are short-axis, horizontal long-axis, and vertical long-axis, with the stress images to the corresponding rest images directly above. Resting images are acquired to differentiate between normal myocardium and infarcted myocardium (Figure 2-5). PET scanners have inherently less attenuation and higher resolution, making them more desirable than SPECT.⁶¹ PET myocardial perfusion tests usually use pharmacologic stressors because of the very short half-life of PET radioisotopes. The sensitivity and specificity of SPECT for the detection of obstructive CAD is 91% and 72%, respectively. The use of PET improves the specificity of diagnosing obstructive CAD to 90%.⁶¹ Patients with normal SPECT and Rb PET have less than 1% and 0.4% probability of annual cardiac events, respectively. The use of myocardial perfusion tests is recommended in those patients with an intermediate risk based on CAD risk factors.

Figure 2-5 Tc^99m sestamibi stress myocardial perfusion demonstrating (A) normal left ventricular size and perfusion, (B) apical and anteroapical infarct, and (C) moderate-to severe ischemia involving the apical, septal, anterior, and anteroseptal walls.

Once the patient has completed the examination, a decision must be made about what to do with the results. If the stress test is normal, then the risk for cardiovascular events is low and the patient is considered ready for surgery. If the stress test demonstrates ischemia, but the patient requires nonelective surgery, data support better outcomes with medical management. Several trials have examined the benefit of revascularization compared with medical management in patients with CAD who require noncardiac surgery. The Coronary Artery Revascularization Prophylaxis (CARP) trial evaluated more than 500 patients with significant but stable CAD who were undergoing major elective vascular disease. Percutaneous intervention was performed in 59% and CABG in 41% of the revascularization group. At 30 days after surgery, there were no differences in postoperative myocardial infarction, death, or length of hospital stay between the revascularization group and the medical management group. At 2.7 years, there was still no difference in mortality between both groups.⁶² The DECREASE-V study showed similar results. In this study, 430 high-risk patients were enrolled to undergo revascularization versus medical management before high-risk vascular surgery. Among the high-risk patients, 23% had extensive myocardial ischemia on stress testing. Again at 30 days and at 1 year, there were no differences in postoperative myocardial infarction or mortality between the revascularization and medical management groups.⁶³

With respect to the use of perioperative β-blockers, they should be continued in those patients who are already taking them. In those patients who are at high risk because of known CAD or have ischemia on preoperative testing, β-blockers may be started and titrated to blood pressure and heart rate, while avoiding bradycardia and hypotension.^64,65

Magnetic Resonance Perfusion Imaging

CMR perfusion imaging is evaluated by the first pass of IV gadolinium contrast through the myocardium. ECG-gated images are acquired generally using three LV short-axis slices (base, mid, and apical) and, possibly, a four-chamber image depending on the heart rate. As the contrast is being injected, it is being tracked through the right side of the heart and, subsequently, the LV cavity and the LV myocardium. The assessment of perfusion requires imaging during several consecutive heartbeats during which the contrast bolus completes its first pass through the myocardium. This is done during a breath-hold. First-pass perfusion images are acquired at rest, then repeated during adenosine infusion. The same slice positions (between 3 or 4) are used for both rest and stress for comparison (Figure 2-6). Perfusion defects appear as areas of delayed and/or decreased myocardial enhancement and are interpreted visually.

Figure 2-6 Adenosine cardiac magnetic resonance perfusion stress test of a 45-year-old woman with chest pain who had a normal nuclear perfusion stress test and was found to have triple-vessel disease on catheterization. Figure demonstrates short-axis views of the (A) left ventricular (LV) base, (B) LV midcavity, and (C) LV apex at stress with corresponding segments below (D–F) at rest. Stress images show diffuse circumferential subendocardial decreased myocardial enhancement in the LV midcavity and apex and partial subendocardial decreased myocardial enhancement in the LV base, which are not present at rest. This corresponds to balanced ischemia caused by three-vessel disease.

The accuracy of stress MRI perfusion has been validated in several trials. In one trial, which evaluated 147 consecutive women with chest pain or other symptoms suggestive of CAD, MRI perfusion was compared with invasive angiography. The CMR perfusion stress test had a sensitivity, specificity, and accuracy of 84%, 88%, and 87%, respectively.⁶⁶ Another study comparing stress perfusion MRI to invasive angiography examined 102 subjects. CMR demonstrated a sensitivity of 88% and specificity of 82% for the diagnosis of significant flow- limiting stenosis.⁶⁷ A negative MRI perfusion stress test also confers significant prognostic information. Patients with a normal stress MRI have a 3-year event-free survival rate of 99.2%.⁶⁸

Stunned and Hibernating Myocardium

Myocardial stunning occurs during acute ischemic injury in which the cardiac myocytes that are on the border of the myocardial infarction are underperfused and sustain temporary loss of function. In theory, function to these myocytes returns once the acute phase of injury resolves; however, this depends on duration of ischemic injury and time to recovery of blood flow to the artery. On rest perfusion imaging, this area would be normal.⁶⁹ If blood flow is not returned to normal levels or if repetitive stunning occurs, the myocardium enters a chronic state of hibernation. About 24% to 82% of hibernating myocardial segments can recover function after target-vessel revascularization; in different series, anywhere between 38% and 88% of patients with hibernating myocardium experience improvement in LVEF.^69,70 Several studies indicate meaningful improvement of LV systolic function occurs; at least 20% to 30% of the myocardium should be hibernating or ischemic.

Thallium-201 is used frequently for viability assessment with SPECT imaging, taking advantage of this isotope’s long half-life (73 hours). Thallium uptake is dependent on several physiologic factors, including blood flow and sarcolemmal intercellular integrity. Thallium is taken up in a short time in normal myocardium, but may take up to 24 hours in hibernating myocardium that still has metabolic activity. Patients are injected with thallium radioisotope and imaged the same day for baseline images. They are brought back after 24 hours without any further injection and reimaged. Baseline images are compared with the 24-hour images. Defects that are present at baseline and fill in at 24 hours represent viability (Figure 2-7). Technetium radioisotopes also can be used for the evaluation of viable myocardium using different protocols.

Figure 2-7 Thallium rest-redistribution scan demonstrating hibernating myocardium involving apical-basal anteroseptum, midbasal inferior, midbasal inferoseptum, and midbasal inferolateral wall segments. There is infarction of the apex, inferoapical, and apical-lateral wall segments.

PET imaging is more sensitive than SPECT and is considered by many experts as the gold standard for assessment of viability. PET has the ability to identify the presence of preserved metabolic activity in areas of decreased perfusion using 18-fluorodeoxyglucose (FDG). PET imaging uses both FDG and either rubidium or ammonia radioisotopes for quantification of energy utilization by the myocardium, as well as for evaluating patterns of blood flow. Areas with reduced blood flow and reduced FDG uptake are considered scar and infarcted. Areas with reduced blood flow (> 50%) and normal FDG uptake are considered viable.⁶⁹ A recent meta-analysis analyzing more than 750 patients demonstrated a sensitivity of 92% and specificity of 63% for regional functional recovery with positive and negative predictive values of 74% and 87%.⁷¹ When viable myocardium is detected by PET, it is important to revascularize as soon as possible because recovery of function decreases as revascularization is delayed.^72,73

Myocardial Scar Imaging

Myocardial viability is unlikely to occur in the presence of extensive scarring because scar is necrotic tissue that cannot regain function. The importance of identifying scar in hypokinetic areas will determine whether revascularization will benefit the patient.

CMR has taken over as the gold standard for evaluation of myocardial scarring. Delayed-enhancement (DE) imaging is achieved by administering gadolinium contrast intravenously and imaging 5 to 10 minutes later. Gadolinium contrast accumulates extracellularly; however, in normal myocardium, there is not sufficient space for gadolinium deposition. In the setting of chronic scar, the volume of gadolinium distribution increases because of an enlarged interstitium in the presence of extensive fibrosis.⁷⁴ Hence, normal or viable myocardium appears as nulled or dark, whereas scar appears bright (Figure 2-8). The advantage of delayed enhancement imaging is that it allows for the assessment of transmural extent of the scar. The percentage of scar-to-wall thickness is the basis for prognosis of viability and segmental functional recovery. Generally, identical LV short-axis images used for function are acquired for DE imaging. This allows for side-by-side comparison of function and DE evaluation. DE imaging is analyzed visually, and the thickness of scarring is quantified as percentages (none, 1–25%, 26–50%, 51–75%, 75–100%). A wall segment is considered to be viable and has a high probability of functional recovery if the scar thickness is ≤ 50% of the wall.⁷⁵

Figure 2-9 Noncontrast computed tomography (CT) demonstrating a severely calcified aortic valve (AoV). LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.

Autonomic Innervation

Myocardial infarction causes denervation of the scar and subsequent interruption of sympathetic nerves induces denervation of adjacent viable myocardium.^76,77 Sympathetic nerves are very sensitive to ischemia and usually become dysfunctional after repeated episodes of ischemia that do not result in irreversible myocyte injury.^78,79 Matsunari et al.⁸⁰ demonstrated that the area of denervation is larger than the area of scar and corresponds to the area at risk for ischemia. In addition, Bulow et al.⁸¹ showed that denervation of myocytes occurs in the absence of previous infarction. Myocyte sympathetic innervation is measured by PET using the radioisotope ¹¹C-hydroxyephedrine (HED). This is compared with PET resting perfusion to determine the area of the scar. Areas of normal resting perfusion and reduced HED retention indicate viable myocardium. In addition, SPECT imaging of myocardial uptake of ¹²³I-mIBG, which is an analog of the sympathetic neurotransmitter norepinephrine, provides an assessment of β-receptor density. Reduced ¹²³I-mIBG uptake is associated with adverse outcomes in patients with heart failure and has been proposed as a marker of response to treatment.⁸²

Aortic Valve Disease

Transthoracic and transesophageal echocardiography (TEE) are the principal imaging modalities for valvular heart disease; however, on several occasions, additional imaging adds important information. Aortic stenosis (AS) is a common cause for valve replacement. There are several different mechanisms for AS. For patients younger than 75, congenital bicuspid aortic valve (BAV) is the most common cause. They have a high incidence of calcification and stenosis. In patients older than 75, senile degenerative calcification of the aortic valve is the leading cause, which is most frequently seen in men.⁸³ Patients with degenerative aortic valve disease typically have concurrent CAD because they both have common risk factors including hypertension, active tobacco smoking, increased low-density lipoprotein (LDL), and lipoprotein (a) levels. In addition, patients with metabolic syndrome have increased incidence of aortic calcification.⁸⁴ Aortic calcification is directly related to the development of AS. CCT is an excellent tool for the evaluation of aortic valve calcification (Figure 2-9). This can be achieved by noncontrast CCT using the same protocol as calcium scoring of the coronary arteries. Coronary artery calcium is measured using the Agatston method. An aortic valve calcium score of ≥ 1100 has a 93% sensitivity and 82% specificity for severe AS.⁸⁵ Contrast-enhanced CCT allows for excellent visualization of the aortic valve and accurately differentiates between bicuspid and tricuspid aortic valves⁸⁶ (Figure 2-10). Aortic valve area (AVA) also can be evaluated by CCT using planimetry. AVAs measured by CCT have a strong correlation with valve areas and transvalvular gradients obtained by echocardiography.^87–91

Figure 2-8 Cardiac magnetic resonance (CMR) demonstrating delayed enhancement imaging of (A) four-chamber view with transmural scars (arrows) appearing bright in the septum and apex; (B) short-axis view shows partial scar with viability (arrowheads) of the anterior wall. LV, left ventricle.

Figure 2-10 Computed tomographic (CT) angiography. Bicuspid aortic valve (BAV) and ascending aortic aneurysm in orthogonal views displaying the BAV in short-axis for the evaluation of the valve area by planimetry.

CCT also can be used for the evaluation of aortic regurgitation (AR). CCT can elucidate the potential mechanism for the AR, including inadequate leaflet coaptation during diastole, leaflet prolapse, cusp perforation, or interposition of an intimal flap in cases of type A aortic dissection.

Regurgitant orifice areas measured by CCT have an excellent correlation to AR severity parameters, including vena contracta width and regurgitant/left ventricular outflow tract (LVOT) height ratio obtained by TTE.^92,93

CMR, like CCT, allows for excellent evaluation of valvular morphology, but it also has advantages over CCT including blood-flow analysis, as well as no radiation exposure. CMR allows for differentiation between BAV and TAV using cine imaging. AS severity can be quantified using phase-encoding imaging. Similarly to echocardiography, phase-encoding imaging allows for the measurement of velocities through the AV, which, in turn, can be used to derive mean and peak AV gradients by implementing the modified Bernoulli equation (ΔP = 4V²). The effective AVA also can be obtained by measuring the LVOT area and using the continuity equation: Area_valve = Area_LVOT [VTI_LVOT/VTI_valve].⁹⁴ Another approach to calculation of the AVA is by direct planimetry of the AV using cine images⁹⁵ (Figure 2-11).

Figure 2-11 Cardiac magnetic resonance demonstrating a short-axis view of a stenotic bicuspid aortic valve. LA, left atrium; RA, right atrium; RV, right ventricle.

CMR also uses phase-encoded imaging for the evaluation of AR. Phase-encoded imaging is acquired just above the AV, and the velocity and the volume of blood per heartbeat are measured in the forward and reverse directions. This allows for measurement of the exact amount of blood that exits the AV, as well as the amount of blood that regurgitates back through the valve. From this the regurgitant amount and regurgitant fraction are obtained (Table 2-1).

TABLE 2-1 Aortic and Mitral Valve Regurgitant Fractions and Corresponding Severity

Mitral Valve Disease

The most common cause of mitral stenosis (MS) worldwide continues to be rheumatic heart disease. In the United States, it rarely is seen except for in the immigrant population. Visualization of the valvular apparatus in rheumatic valvular disease demonstrates retraction, thickening, and calcification of the mitral leaflets, chordae, and, occasionally, papillary muscles. Accurate assessment of MV morphology is performed by examining cine imaging using ECG-gated, contrast-enhanced CCT.^96,97 Calcium scoring of the MV also is possible, but it has lower reproducibility than for the AV.⁹⁸ The degree of MV calcification correlates significantly with the severity of stenosis seen on TTE.⁹⁹ MV areas obtained by planimetry also correlate significantly with TTE data of MS.¹⁰⁰

Mitral regurgitation is the most common cause for valve surgery. MV prolapse is a frequent cause of mitral regurgitation. It can be diagnosed by evaluating cine loops of the MV, and visualization of which scallops of the leaflets prolapsed can aid in the planning before surgery.

Severity of the mitral regurgitation using CCT can be assessed by planimetry of the regurgitant orifice, which in a recent study has been shown to correlate with TEE.¹⁰¹ In addition, the presence of calcification of the MV annulus and leaflets will determine whether the valve can be repaired or needs to be replaced.

CMR also allows for excellent morphologic evaluation of rheumatic MVs. Planimetry of the MV using cine images is also feasible. MV insufficiency can be quantified using phase-encoded imaging and the LV stroke volume calculated by functional analysis. Mitral regurgitant volume is measured by subtracting the volume of forward flow through the AV acquired by phase contrast (PC) imaging from the LV stroke volume. Once the regurgitant volume is calculated, the regurgitant fraction is easily obtained by dividing the regurgitant volume by the total stroke volume.

Tricuspid Valve Disease

The tricuspid valve (TV) is the atrioventricular valve on the right side of the heart. In general, pathology of the TV is not of clinical significance, unless it is congenital or involves endocarditis. TV pathology is best imaged using TTE and TEE, but on occasion a patient may have poor TTE windows and the TEE also may be insufficient. Tricuspid stenosis (TS) occurs in less than 1% of the population in the United States. In patients with rheumatic heart disease, TS becomes clinically significant only 5% of the time. In cases of congenital TS, either CCT or CMR should be done to evaluate for additional congenital abnormalities.

Mild tricuspid regurgitation (TR) is present in approximately 70% of the normal population. Mild TR is clinically insignificant, and clinically significant TR occurs in only 0.9% of the population. Functional TR is most often a result of PH, mitral disease, or severe LV dysfunction. The degree of TR severity can be quantified by using CMR. Similar to mitral regurgitation evaluation, by using PC imaging of the pulmonary artery (PA) just above the PV, RV outflow volume is measured. This can be subtracted from the RV systolic volume acquired by cine imaging, to give TR regurgitant volume and fraction. CMR also is important for morphologic valve evaluation in patients with Ebstein’s anomaly. In the instance in which CMR cannot be performed, CCT is also excellent for morphologic evaluation of the TV in Ebstein’s anomaly.

Pulmonic Valve Disease

The pulmonic valve is generally not well visualized on either TTE or TEE. Pulmonic stenosis (PS) usually occurs as isolated valvular, subvalvular, or supravalvular stenosis. It also may be associated with more complex congenital disorders. Significant PS in congenital heart disease presents in infancy or early childhood. Acquired PS affects morbidity and mortality only when it becomes severe. Both CMR and CCT are appropriate for anatomic evaluation of the PV apparatus. CMR has the advantage of measuring velocities and gradients across the stenosis using PC imaging.

Trivial or mild pulmonary regurgitation is physiologic and normal. Severe PR is rare and is typically secondary to PH or repair of congenital PS. PR can be calculated using CMR by PC imaging of the PA just above the PV and measuring forward and reverse flow through the PV.

Prosthetic Valves

The visualization of mechanical prosthetic valves is difficult with TTE and TEE because of metal-related artifacts. CCT has the ability to clearly depict the mechanical prosthesis and detect any abnormality including valve thrombosis. This is done by using retrospective scanning and acquiring the entire cardiac cycle to play the cine movie and visualize the leaflets through systole and diastole. The mechanical valves that are used today consist of two disks that open symmetrically (Figure 2-12). The valve function of the two-disk prosthesis, as well as opening and closing angles, was evaluated by CCT and then compared with fluoroscopy and echocardiography. CCT correlated significantly with both imaging modalities for two-disk mechanical valves.¹⁰² The role of CT in the assessment of bioprosthetic valves is similar because the metallic ring causes artifact on echo and often is difficult to assess.

Figure 2-12 Computed tomography angiography of an aortic mechanical valve in (A) short-axis view and (B) three-chamber view. LA, left atrium; LV, left ventricle; RA, right atrium.

In general, echocardiography is the gold standard for imaging valvular disease; however, when TTE or TEE is technically difficult or there are discrepancies between tests, advanced imaging is recommended. CMR offers more functional data than CCT; however, CCT may be used when further anatomic information about a valve is required. For evaluating prosthetic valves, CCT is usually superior to CMR because of metallic artifact from the valve, which is seen on CMR (see Chapters 12, 13, and 19).

Infective Endocarditis

Bacterial endocarditis is a cause for valve replacement of native and prosthetic valves and is a life-threatening disease. Valvular endocarditis is associated with a mortality of up to 40%.¹⁰³ Diagnosis is usually made by visualization of vegetations by TEE, which is the gold standard for diagnosis. In severe cases of endocarditis, perivalvular abscesses are present and are an indication for valve replacement. CCT is excellent for the diagnosis of abscesses. They appear as perivalvular fluid-filled collections on CCT and are imaged by acquiring a delayed scan approximately 1 minute after contrast is given. Contrast is retained within the abscess after the contrast washes out of the circulation¹⁰⁴ (Figure 2-13). A recent study comparing multidetector computed tomography (MDCT) with intraoperative TEE for the detection of suspected infective endocarditis and abscesses demonstrated excellent correlation. CCT correctly identified 96% of patients with valvular vegetations and 100% of patients with abscesses. In addition, CCT performed better than TEE in the characterization of abscesses.¹⁰⁵

Figure 2-13 Computed tomography angiography of a bioprosthetic aortic valve (arrowhead) with a perivalvular abscess (arrow) in the (A) short-axis view and (B) three-chamber view. LA, left atrium; LV, left ventricle; RA, right atrium.

Preoperative Coronary Evaluation before Valve Surgery

Coronary computed tomography angiography (CCTA) has been used in many centers for the evaluation of CAD in patients with low-to-intermediate CAD risk before valve surgery to avoid invasive testing. CCTA has been well-studied in the diagnosis of CAD in patients without known ischemic heart disease, demonstrating a sensitivity of 94% and a negative predictive value of 99%¹⁰⁶ (Figure 2-14). Several studies have examined the use of CCTA for preoperative evaluation before valve surgery. One such study used 64-slice MDCT in 50 patients, who had a mean age of 54 years, undergoing valve replacement for AR. CCTA demonstrated a sensitivity of 100%, specificity of 95%, and a negative predictive value of 100%, respectively, when compared with invasive catheterization. In addition, it was determined that 70% of the patients could have avoided invasive catheterization.¹⁰⁷ Two further studies used preoperative 16- and 64-slice CCTA in patients with AS. The mean ages of the patients were 68 and 70 years, respectively. Both the sensitivity and negative predictive value for each study were 100% for the detection of significant stenosis.^108,109 These studies show that preoperative coronary evaluation with CCTA is safe and accurate. It is important that only patients with no known CAD or those with low-to-intermediate risk are referred for CCTA. In general, patients with degenerative AS are older and have greater risk for CAD.¹¹⁰ Patients who undergo valve surgery for mitral regurgitation because of MV prolapse are usually younger and are excellent candidates for CCTA (Table 2-2).

Figure 2-14 Computed tomography angiography demonstrating a long, nonobstructive, mixed eccentric plaque (arrows) in the proximal LAD artery. Ao, aorta; LAD, left anterior descending.

TABLE 2-2 Appropriate Indications for the Use of Computed Tomography Angiography¹⁴¹

CAD, coronary artery disease.

Carotid Artery Stenosis

Stroke is a severely debilitating disease, and extracranial atherosclerotic disease, specifically carotid artery stenosis, is the major cause. Atherosclerotic plaques most often form in the proximal internal carotid artery; however, the common carotid artery is also the culprit at times. In patients who have had a carotid endarterectomy, the distal common carotid artery is a frequent location for plaque formation. Generally, stroke occurs as the first symptom of the disease, and often a carotid bruit is the only sign that can be seen on physical examination. The two main predictors for stroke are previous symptoms (transient ischemic attack and recent stroke) and severity of stenotic lesions.¹¹¹ For this reason, diagnosis is critical for the prevention of stroke. Several imaging modalities can be used for diagnosis. CTA has excellent spatial and contrast resolution for plaque detection, as well as morphology. It is able to detect plaque at the bifurcation of the internal and external carotid arteries, and is used to define vascular anatomy proximal and distal to a stenotic plaque.

CT, however, is not used as the initial screening test. Vascular ultrasound is easily accessible and can be brought to the patient’s bedside. It is inexpensive, risk-free, and excellent for the evaluation of carotid anatomy and flow dynamics. B-mode ultrasound is used for the anatomic definition of the arteries, whereas severity of plaques are evaluated by Doppler, which measures the velocity and pressure gradients across a lesion. There are limitations of Doppler imaging, which can give false measurements. Anything that decreases the velocity of the blood from the heart to the carotid arteries can interfere with accurate estimation of carotid stenosis. Most commonly, severe LV dysfunction, valvular heart disease, and aortic disease are the culprits. Highly calcified plaques also may cause artifact on ultrasound that may interfere with accurate assessment.

Magnetic resonance angiography (MRA) is another tool for carotid artery assessment. It is more expensive than the previous two modalities, but it is relatively safe and provides anatomy, as well as plaque morphology. “Black-blood” imaging is a magnetic resonance sequence in which blood is black and vessel walls are enhanced to highlight and define plaque morphology (Figure 2-15). Angiography can be performed without gadolinium contrast by using “time-of-flight” sequence, which provides high-intensity signals for flowing blood. In addition, PC imaging also can give blood flow velocity pressure information across stenotic lesions. In general, CT and MRI are used only in the cases in which vascular ultrasound is limited or when a patient requires carotid endarterectomy for carotid artery stenosis.

Figure 2-15 Cardiac magnetic resonance demonstrating “black-blood” imaging of a left common carotid artery with significant atherosclerosis (arrow) and right common carotid artery with mild atherosclerosis (arrowhead).

Aortic Aneurysm and Dissection

The aorta is composed of three different layers: the intima, which is a thin delicate inner layer; the media, which is a thick middle layer; and the adventitia, which is a thin outer layer. Aortic aneurysm is a dilatation of a segment or various segments of the aorta. Aneurysm refers to a dilatation of more than 1.5 times the normal size. Ascending aortic aneurysms usually occur because of cystic medial degeneration. These aneurysms frequently involve the aortic root and cause AR. There are also several connective tissue diseases that predispose a patient to aortic aneurysms, including Marfan and Ehler–Danlos syndromes; in addition, patients with Turner syndrome or congenital BAV are also at greater risk (see Chapter 21).¹¹²

Descending aortic aneurysms are mostly caused by atherosclerosis. They are associated with the same risk factors as CAD. In addition, patients with a history of tobacco smoking are recommended to have prophylactic screening for abdominal aortic aneurysms. Abdominal aneurysms are more common than thoracic aortic aneurysms. Aortic aneurysms are generally diagnosed as accidental findings on examinations performed for other reasons.

Aortic dissection is one of the true emergencies and needs to be diagnosed and treated surgically when it involves the ascending or aortic arch. In aortic dissections there is a tear in the intima that forms a communication with the aortic true lumen. The media is exposed to blood flow and a false lumen typically forms, and the dissection extends antegradely or retrogradely.^113,114

On occasion, the blood in the false lumen coagulates and thromboses if there is not a reentry site or other communication at the distal portion of the dissection. Aortic dissections most commonly originate in one of two locations that experience greatest stress: in the ascending aorta just above the sinuses of Valsalva and in the descending aorta just distal to the subclavian artery. Aortic dissections take place most often in the ascending aorta, where they occur 65% of the time. Twenty percent occur in the descending aorta, 10% in the aortic arch, and 5% in the abdominal arch.

Computed tomography angiography (CTA) is most commonly used for the diagnosis of aortic aneurysms and dissections. Similar protocols used for CCTAs also can be used for the evaluation of the aorta. It is important to have the scan gated to the patient’s ECG because the ascending aorta has significant motion during the cardiac cycle. Nongated CTAs have inherent motion artifact that can be confused with a dissection. On some occasions, ascending aortic dissections can include the ostia of the coronary arteries, so visualization of the root and arteries is crucial. In addition, ECG-gated scans using prospective ECG-gating may be performed with low radiation exposure.

Once the images are on the specialized CT workstation, the aorta is evaluated and measured. The aorta is lined up in multiple orthogonal views to get a true short-axis at any point along the aorta to get correct measurements. The excellent spatial and contrast resolution is useful for the evaluation of dissection. Entry points of dissection, as well as intimal flap location, false lumen, and abdominal aortic circulation, are easily visualized.

CMR is also an excellent tool for the evaluation of aortic aneurysms and dissections. It has no radiation and is ideal for serial evaluation of the aorta. Black-blood imaging provides great morphologic information of the aortic wall. CMR is also ECG gated to compensate for the cardiac movement. Bright blood cine sequences provide alternative anatomic assessment. Delayed enhancement imaging also aids in the diagnosis of false lumen thrombosis. Three-dimensional images also can be acquired and transferred to a workstation for evaluation, and measurements, similar to CTA analysis.

Renal Artery Stenosis

Renal artery stenosis (RAS) is the most common cause of secondary hypertension. It can be caused by atherosclerosis, fibromuscular dysplasia, or systemic disease, which affects the renal arteries. Atherosclerosis is responsible for approximately 90% of all RAS cases.^115,116 Fibromuscular dysplasia is the most common cause in young and middle-aged women and is responsible for 10% of all cases. Atherosclerotic RAS is associated with similar CAD risk factors including diabetes, hypertension, and dyslipidemia. The clinical presentation can appear as renal involvement or extrarenal involvement. RAS can cause renovascular hypertension in addition to systemic hypertension and causes renal damage, renal atrophy, and the creatinine level to increase. Extrarenal effects range from angina, myocardial infarction, to hypertension-induced stroke and flash pulmonary edema.

The initial diagnostic tool used is vascular ultrasound because of its advantages mentioned previously. Using B-mode and Doppler ultrasound, renal artery anatomy and flow velocities can be accurately analyzed. Ultrasound is a good tool to monitor the renal artery after percutaneous or surgical intervention. Common limitations to ultrasound for the visualization of renal arteries are patient obesity and gas in the gastrointestinal system. This affects 15% to 20% of all studies. In addition, mild stenosis and accessory renal arteries may be completely missed.

CTA of the renal arteries has the same advantages as seen for coronary evaluation. Data can be reconstructed and visualized on workstations that allow two-dimensional analysis of the renal arteries in any desired plane. One of the main disadvantages is that patients with RAS often have abnormal renal function and iodine contrast is contraindicated.

MRA is an excellent tool for the diagnosis of RAS. Using multicontrast and contrast-enhanced magnetic resonance, the sensitivity and specificity for the diagnosis of RAS are 100% and 99%, respectively.^117–124 In addition, the renal artery assessment, anatomic, and perfusion evaluation of the kidneys are also performed.

Peripheral Arterial Disease

Peripheral arterial disease (PAD) refers to noncoronary atherosclerosis but is considered a CAD equivalent. Cerebrovascular and renovascular disease are generally considered separate entities, and PAD usually refers to lower extremity disease. Because atherosclerosis is a systemic disease, patients with coronary atherosclerosis should be assumed to have PAD as well and vice versa. However, a history of cigarette smoking confers two to three times more risk for PAD than CAD.¹²⁵ Eighty percent of all patients with PAD are active smokers or have smoked cigarettes in the past.^126,127 In the PARTNERS study, almost 7000 patients were evaluated for the prevalence of PAD. Ankle–brachial indices (ABIs) were used for PAD diagnosis. The study included subjects older than 70 years of age or subjects between the age of 50 and 69 with either history of tobacco smoking or diabetes. PAD was found in 29% of this population.¹²⁸ PAD most often is asymptomatic, with a relatively small percentage of patients experiencing intermittent claudication.^129–131

Vascular ultrasound is generally the first modality used once PAD has been diagnosed or suspected clinically. It has very high sensitivity and specificity (90% and 95%) for the detection of a ≥ 50% stenosis from the iliac artery to the popliteal artery.

CTA and MRA may be the preferred modalities in the cases in which percutaneous or surgical intervention is planned. CTA because of its excellent spatial resolution has a sensitivity and specificity of greater than 92.9% and greater than 96.2%, respectively, for the detection of obstructions greater than 50%.^132,133

MRA also is accurate for the detection of PAD (Figure 2-16). It has a sensitivity and specificity between 90% and 100% for the detection of greater than 50% stenosis when compared with conventional angiography.¹³⁴ When MRA is compared with CTA, MRA demonstrates greater interobserver agreement.^135,136

Figure 2-16 Magnetic resonance angiography demonstrating abdominal aorta (arrowhead) and common iliac arteries (arrows) with severe atherosclerosis.

Pulmonary Arterial Disease

Pulmonary arterial disease is important for preoperative evaluation and postoperative care. The two principal entities are PH and pulmonary embolus (PE). PH is a very complex disease and increases the risk for perioperative morbidity and mortality. It is defined as a chronic elevation of mean pulmonary arterial pressure to greater than 25 mm Hg at rest or greater than 30 mm Hg with exercise. Patients who require CABG are increasingly sicker people who often have several comorbidities including significant PH. It commonly is diagnosed by echocardiography or by invasive right-heart catheterization. CTA also can evaluate signs of PH by analyzing RV function, RV and RA volumes, RV hypertrophy, enlarged proximal pulmonary vessels, and pruning of distal ones. ECG-gated CTA is required to assess RV function and volumes. CMR is the gold standard for RV functional analysis; however, 64-MDCT was recently compared with CMR for RV function and RV volumes and was found to have excellent correlation.¹³⁷ CMR, in addition to its analysis of the RV, PC imaging of the PA can be used to evaluate severity of PH. This is performed by measuring the velocity of blood in the PA, as well as the elasticity of the PA.

PE is usually caused by migration of a deep venous thrombosis (DVT) to the pulmonary arterial system. DVTs occur more frequently after surgery, and 80% of the time PEs are caused from lower extremity DVTs. In the United States, 2.5 million cases of DVT occur annually. Approximately 25% of all untreated DVTs will embolize and cause a PE. Vascular ultrasound is the imaging modality of choice for the diagnosis of DVT. The sensitivity and specificity for the detection of lower extremity DVTs are 90.6% and 94.6%, respectively.¹³⁸

The test of choice for the diagnosis of PE is MDCT angiography (Figure 2-17). It has a sensitivity and specificity of 83% and 96%, respectively, for the detection of acute PE. Including a lower extremity CT venogram increased the sensitivity and specificity for PE diagnosis to 90% and 95%, respectively. However, this is accompanied by a much higher level of ionizing radiation exposure¹³⁹ (see Chapter 24).

Figure 2-17 Computed tomography angiography showing large pulmonary emboli (arrows). PA, pulmonary artery.

CTA has largely replaced nuclear ventilation/perfusion imaging (also known as lung scintigraphy or V/Q scan) because the latter has limited use in patients with chronic lung disease and a high number of V/Q scans ( > 72%) are found to have intermediate probability, with a 20% to 80% likelihood of PE. When CTA cannot be performed because of an increased creatinine level, a V/Q scan may be used alternatively.

Peripheral Venous Insufficiency

Chronic venous insufficiency includes a large array of symptoms. It occurs more often with increased age and also has a greater incidence in women than men. Common clinical symptoms include limb pain, swelling, stasis skin changes, itching, restless legs, nocturnal leg cramps, and ulceration. In general, most cases of deep venous disease have either a nonthrombotic or post-thrombotic cause. Both types can involve reflux, obstruction, or a combination. Vascular inflammation, most notably by way of several cytokine mechanisms, causes tissue damage and, thus, chronic venous insufficiency.¹⁴⁰ Vascular ultrasound is commonly used for diagnosis of venous disease. In addition to previously mentioned DVT diagnosis, it also is accurate for the detection of venous post-thrombotic changes, patterns of obstructive flow, and reflux.