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