SCOPE OF DIAGNOSTIC IMAGING: INTRODUCTION

The landmark discovery of x-rays by Roentgen in 1895 was made serendipitously following the fluorescence of a barium platinocyanide screen by an unknown radiation, after it had passed through a black cardboard-covered cathode-ray tube. By clever substitution of a photographic plate for the fluorescent screen, the existence of this radiation could be recorded. By placing a piece of platinum on the plate and exposing the plate to this radiation source, a light area appeared on the plate where the platinum absorbed the radiation. This unexpected finding was further confirmed after Roentgen placed his wife's hand on a cassette containing a photographic plate and made a 15-minute exposure! The bones appeared white on the developed plate, in contrast to the surrounding darker appearing flesh. Shortly thereafter, he prepared a short manuscript entitled "On a New Kind of Rays, a Preliminary Communication," which he presented to the Würzburg Physical Medical Society on December 28, 1895. In this document, he described the generation of these x-rays and discussed the relative transparency of almost all materials to this radiation, assuming the materials to be of equal thickness. Only the density of the material dictated the degree of transparency. He also noted that, in addition to barium platinocyanide, calcium compounds, uranium glass, ordinary glass, calcite, and rock salt fluoresced when exposed to x-rays. That photographic plates were sensitive to x-ray detection he felt was fortunate as "one is able to make a permanent record of many phenomena whereby deceptions are more easily avoided."

Since this discovery, the field of radiology has progressed rapidly, with successive advances made in the areas of catheter angiography (1950s), nuclear medicine (1960s), ultrasound and computed tomography (CT) (1970s), magnetic resonance imaging (MR imaging), positron emission tomography (PET), and interventional radiology (1980s), multidetector CT and ultrafast MR techniques (early to mid-1990s), and functional and molecular imaging (late 1990s to the present). The classification of radiologic subspecialties can be organ based, modality oriented, or organized according to subspecialty fields. Organ-based subspecialties include musculoskeletal, breast, neurologic, abdominal, thoracic, gastrointestinal, and genitourinary imaging. Ultrasound, PET, and MR imaging are classified as modality oriented. Subspecialty fields include pediatric imaging and women's imaging.

This chapter is intended to provide an overview of a variety of modalities in diagnostic radiology. The physics governing x-ray generation and image formation are described in Chapter 2. Subsequent chapters are organized using an organ-based approach, and are further subclassified in terms of specific imaging modalities. The proper selection of diagnostic studies is essential for ensuring the best possible patient care and for containing medical costs.

CONVENTIONAL RADIOGRAPHY

Conventional radiography remains a fundamental anatomic tool in the detection and diagnosis of disease presenting in the chest (Fig. 1–1), abdomen, pelvis, breasts, and bones, and continues to be utilized in the initial evaluation of the patient. Other modalities, including ultrasonography, CT, and MR imaging, have also been used to provide supplementary diagnostic information and, in some cases, have replaced plain radiograph examinations in the initial work-up of the patient.

Computed radiography (CR) or digital radiography has recently replaced conventional screen-film combination techniques. The most successful digital radiography method to date is photostimulable phosphor computed radiography (PPCR), initially developed in 1981. This technique uses a photostimulable phosphor plate within a cassette to capture and store transmitted x-rays in the form of trapped electrons. When the exposed cassette is scanned with a low-energy laser beam, electron release is stimulated, and the associated energy is used to encode the digital or gray-scale image. Advantages of this technique over conventional screen-film techniques include increased image quality, decreased patient dose, long plate lifetime, and a linear response of the storage phosphors over a wide range of exposures, permitting greater freedom in selecting exposure doses. Disadvantages include reduced spatial resolution due to optical scatter processes within the image plate during scanning. A newer technology that attempts to eliminate these signal losses employs materials such as amorphous selenium, a photoconductor that converts x-rays directly into electrical charges.

Fluoroscopy uses a continuous beam of x-rays to evaluate dynamic processes, such as bowel peristalsis and diaphragmatic excursion, and it is often employed for angiographic or other interventional procedures, including feeding tube or drainage catheter placement. It continues to be a frequently utilized modality in the evaluation of the upper and lower gastrointestinal tract (e.g., barium enema), joint spaces, and spinal cord in the case of a lumbar puncture or myelogram. The recently introduced digital detector systems, such as charged couple devices (CCDs), may begin to replace many of the fluoroscopic system components.

Conventional tomography produces an image in which the desired plane of interest is maintained in sharp focus, while structures located on either side of this plane are intentionally blurred. To achieve this end, the x-ray tube and film move in opposite directions concomitantly during the exposure. An 8- to 40-degree arc may be traversed during a tomographic exposure. The wider the arc used, the thinner the slice thickness obtained. Twenty-degree arcs are routinely used in renal studies (nephrotomograms). In addition to simple linear motions, the x-ray tube can perform more complex motions, which can increase blurring, thereby improving overall image quality. Although CT has replaced many conventional tomographic studies, particularly those relating to chest and musculoskeletal applications, it continues to be utilized for nephrotomograms.

The radiographs for screen-film mammography must have excellent resolution, contrast, and film density for optimal detection of breast lesions, particularly early breast carcinoma. A mammographic unit is installed with a special x-ray tube and a plastic breast compression device. Low voltages are utilized to enhance contrast resolution. Two views, craniocaudal (CC) and mediolateral oblique (MLO), are routinely obtained with the breasts maintained in compression. This permits reduced radiation exposure and improves overall image quality. Magnification views are typically acquired for further characterization of microcalcifications, whereas spot compression views are useful for resolving overlapping soft tissue structures, thus facilitating lesion detection. Ultrasonography (US) and, more recently, MR imaging have been used as complementary modalities for more detailed lesion characterization. Image-guided breast interventions are commonly performed and include preoperative needle localization, stereotactic core needle biopsy of microcalcifications, and US-guided procedures (core biopsy or fine-needle aspiration). Digital mammography units have recently been introduced for clinical use.

Contrast Studies

Decreased natural contrast between adjacent structures of roughly similar radiographic density mandates the use of contrast material. Contrast media are commonly utilized for the evaluation of the gastrointestinal tract, urinary tract, and vascular system; less frequent applications involve studies of the biliary tree for carcinoma or obstruction, spinal canal, joints for ligamentous or cartilaginous tears, potential fistulous tracts arising from abscesses, and the uterine cavity and fallopian tubes for uterine anomalies and tubal patency, respectively (hysterosalpingography). Interventional procedures involving catheter or percutaneous tube insertion (nephrostomy, gastrostomy, and biliary) are usually guided by administration of a small volume of contrast.

Barium suspensions, high-density compounds mixed with water, are commonly used in the examination of the gastrointestinal tract. The high radiographic density achieved using these agents is ideal for the standard upper gastrointestinal (UGI) series and for evaluation of the small bowel and colon. Both single- and double-contrast techniques can be performed in an UGI series after oral administration of a barium suspension or during a barium enema, following rectally administered barium (Fig. 1–2). In the single-contrast study, only a barium suspension is used, whereas double-contrast studies use both barium and air to delineate mucosal irregularities and superficially located lesions. Air can be introduced directly by inflating a rectal catheter during a barium enema study, or it can be intentionally generated after ingestion of an effervescent agent during a double-contrast UGI series. Small-bowel studies can be performed in several ways. A barium suspension can be administered orally and followed as it opacifies the small bowel (peroral approach). A second approach, enteroclysis, requires initial placement of a catheter in the proximal jejenum prior to barium infusion. Enteroclysis is preferred for evaluating focal small-bowel lesions or to determine the cause of a small-bowel obstruction. The small bowel can also be assessed via a retrograde approach, either secondary to reflux of barium during a barium enema or to direct infusion of contrast into an ileostomy.

Iodinated contrast agents are water soluble and can be classified in several ways, namely, whether the agent is ionic/nonionic, monomeric/dimeric, or hyperosmolar/hypo-osmolar in solution. First, contrast materials can be ionic or nonionic. An ionic compound dissociates into both anions and cations in water, while nonionic agents do not. The number of iodine atoms per milliliter of solution influences the degree of x-ray attenuation. High- and low-osmolality agents are defined in terms of both the number of iodine atoms in the molecule and the number of osmotically active particles generated in solution. Nonionic compounds were developed to decrease the osmolality of the agent while maintaining optimal contrast characteristics. Based on the above classifications, subcategories of agents can be identified, namely, hyperosmolar ionic (diatrizoate and its derivatives), low-osmolarity ionic (meglumine ioxaglate), nonionic monomers (iohexol, iopamidol, ioversol, iopromide), and nonionic dimers (iodixanol). Side effects, while uncommon, encompass a spectrum, ranging from mild reactions (flushing, tachycardia, and metallic taste in the mouth) to life-threatening effects (hypotension, severe bronchospasm, cardiac arrest). At the present time, the use of low-osmolality contrast agents is advocated, because these agents have a significantly lower incidence of drug reactions. However, because these agents cost considerably more than high-osmolality agents, criteria have been established for their use in more selective patient populations, such as patients who have had a previous reaction to contrast material.

Water-soluble contrast agents are used in the gastrointestinal tract for suspected perforation, prior to surgical procedures, for confirming the position of a percutaneously placed catheter, or for contraindications to barium suspensions. If these agents leak from the gastrointestinal tract, they may be absorbed by the peritoneum, unlike barium suspensions. The use of certain hyperosmolar agents in the gastrointestinal tract will cause fluid to enter the bowel lumen, thereby promoting a hypovolemic state. The likelihood of inducing these conditions with a lower osmolality agent is reduced. Because CT examinations utilize dilute water-soluble agents, the risks of hypovolemia are diminished.

Intravenous urography (IVU), the most common contrast study of the urinary tract, may utilize ionic or nonionic agents. After injection of contrast material into a peripheral vein, sequential filming of the kidneys is performed during the time the agent is concentrated by the kidneys (nephrographic phase, typically 1–3 minutes postinjection), as well as during contrast excretion into the renal collecting system and ureters (pyelographic phase, typically 3–10 minutes) (Fig. 1–3). Tomographic images are usually obtained prior to contrast administration to determine which images optimally depict the entire kidney. Primary indications for this study include hematuria, urinary tract calculi, ureteral obstruction, and evaluation of a suspected congenital anomaly. Nonenhanced helical CT is rapidly superceding the standard IVU examination in cases of stone disease and ureteral obstruction. However, IVU remains the procedure of choice for evaluation of suspected uroepithelial neoplasms, such as transitional cell carcinoma, because it more sensitively detects subtle mucosal irregularities. Retrograde studies of the renal collecting system (retrograde pyelography), ureter (retrograde ureterography), and bladder (cystography) can additionally be performed by direct instillation of water-soluble contrast material into the urinary bladder. These studies are typically performed for evaluation of vesicoureteral reflux, calculi, or tumor. Dynamic retrograde urethrography permits urethral assessment while the urethra is being distended by infusion of water-soluble contrast. This study is typically performed for suspected urethral injury or stricture.

Hysterosalpingography is a radiographic method for evaluation of the endometrial cavity and the fallopian tubes following direct instillation of water-soluble contrast material (some institutions prefer oil-based iodinated contrast agents) into the cervical canal. This procedure is principally used as part of a primary or secondary infertility work-up, but other applications include detection of suspected uterine anomalies, assessment of tubal patency and morphology following tubal reconstructive surgery, or determination of the location of a uterine leiomyoma within the endometrial cavity. Two spot films are routinely taken during contrast opacification of the uterus and fallopian tubes, both before and after spillage into the peritoneal cavity. Transcervical recanalization of an obstructed fallopian tube has been introduced in recent years to improve the fertility rate.

Angiography, the study of the vascular system, is performed by injection of water-soluble contrast media intra-arterially (or intravenously, for venous studies) through a percutaneously placed catheter under fluoroscopic guidance. Pathologic processes involving both the vasculature and parenchyma can be characterized by fluoroscopic monitoring using either digital or conventional image recording. Specific diagnostic angiographic procedures can be performed, such as peripheral angiography, coronary angiography, thoracic/abdominal aortography, pulmonary angiography, cerebral angiography, and central venography or superior/inferior venocavography. Common applications include assessment of vascular disease (atherosclerosis, aneurysm, vasculitis), determination of vascular injury, vascular mapping for preoperative purposes (organ transplantation) or prior to therapeutic interventions (stent placement), characterization of tumor vascularity prior to endovascular procedures (embolization), and evaluation for pulmonary embolus.

Thoracic aortography is typically performed for the evaluation of suspected traumatic injury (Fig. 1–4), dissection, aneurysm, vasculitis, thromboembolic disease, tumors, arteriovenous malformations (AVMs), or compression syndromes. This procedure is currently rivaled by both CT angiography and MR angiography, which have replaced conventional angiographic studies in various instances, such as in the determination of vessel dissection following blunt trauma. Coronary angiographic studies identify areas of stenosis or occlusion involving the coronary arteries, in addition to identifying aberrant vascularity. In patients suspected of having a pulmonary embolus, pulmonary angiography may be performed following cannulation of the femoral vein. This procedure is generally used when the results of a ventilation-perfusion (V/Q) scan or pulmonary CT angiography are equivocal, when a low-probability V/Q scan is obtained in the presence of continued high clinical suspicion, or when the performance of either of these studies is clinically impractical. Inferior venocavography is performed prior to filter placement for venous mapping and may additionally be used for evaluation of caval disease, including occlusion, obstruction, or extrinsic compression secondary to fibrosis or retroperitoneal lymphadenopathy.

Less commonly performed contrast studies include myelography for evaluation of spinal cord compression; fistulography for detection of epithelialized tracts arising from an inflammatory, infectious, or neoplastic process; sialography for the assessment of ductal obstruction or tumor involving the salivary glands; galactography for detection of masses within large breast ducts; and cholangiography for detection of biliary ductal masses or strictures. Lymphangiography, a rarely utilized procedure in which water-soluble contrast is injected into the lymphatic system, permits lymph nodes and lymph channels to be evaluated for the presence of malignancy.

Computed Tomography

Computed tomography, an axial tomographic technique, results in source images perpendicular to the long axis of the body (Fig. 1–5). Multiple generations of CT scanners have traditionally used a single-slice fan-beam system. In this system, x-ray photons are initially collimated into a thin fan-shaped beam that is attenuated by the patient being imaged. The attenuation profile of this fan-beam is recorded by a single row of detectors (array) roughly containing a 1000 detector elements. These attenuation values, which reflect the density and atomic number of various tissues, are usually expressed as relative attenuation coefficients, or Hounsfield units (HUs). By definition, the HU of water is zero, and that for air is –1000. Typically, the HU of soft tissues ranges from 10 to 50. Fat also has a negative HU, whereas bone is at least 1000 HU. The contrast resolution of vascular structures, organs, or hypervascular neoplasms can be enhanced following intravenous infusion of water-soluble contrast media. The type and amount of contrast agent used, the rate of administration, and the scan delay time will vary with the study indication. Additionally, oral contrast material, namely, water-soluble agents or barium suspensions, can be administered for improved bowel visualization. Artifacts may be produced by patient motion or high-density foreign bodies, such as surgical clips.

Conventional CT scanners have traditionally operated in a step-and-shoot mode, defined by data acquisition and patient positioning phases. During the data acquisition phase, the patient is kept in a stationary position, while the x-ray tube rotates around the patient. A complete set of projections is acquired at a prescribed scanning location, prior to the patient positioning phase. During this latter phase, the patient is transported to the next prescribed scanning location.

The first helical (spiral) CT scanner was introduced for clinical applications in the early 1990s. Helical CT is characterized by continuous patient transport through the gantry while a series of x-ray tube rotations simultaneously acquires volumetric data. These dynamic acquisitions are typically obtained during a single breath-hold of about 20 to 30 seconds. Currently available spiral scanners generally use a single-row detector array that acquires data from a tissue thickness prescribed by the width of the x-ray beam collimator (1–10 mm). Higher spatial resolutions can be achieved with narrower collimations. The advantages of helical CT technology include reduced scan times, improved speeds at which the volume of interest can be adequately imaged, increased ability to detect small lesions that may otherwise change position in non–breath-hold studies. In addition, gains in scan speed permit less contrast material to be administered for the same degree of vessel opacification.

The evolution of multidetector CT scanners (MDCTs) has resulted from the combination of helical scanning with multislice data acquisition. In this CT system, a multiple-row detector array is employed, as opposed to the single-row detector array utilized in single-slice helical scanners. Current state-of-the-art models are capable of acquiring 16, and soon 32, channels of helical data simultaneously. For a given length of anatomic coverage, MDCT can reduce scan time, permit imaging with thinner collimation, or both. The use of thinner collimation (1–2 mm), in conjunction with high-resolution reconstruction algorithms, yields images of higher spatial resolution (high-resolution CT), a technique commonly used for evaluation of diffuse interstitial lung disease or detection of pulmonary nodules. Multidetector CT offers the additional advantages of decreased contrast load, reduced respiratory and cardiac motion artifacts, and enhanced multiplanar reconstruction capabilities. Furthermore, these innovations have had a significant impact on the development of CT angiography.

CT ANGIOGRAPHY

CT angiography protocols combine high-resolution, volumetric helical CT acquisitions with intravenous bolus administration of iodinated contrast material. Using a MDCT scanner, images are acquired during a single breath-hold, ensuring that data acquisition will commence during times of peak vascular opacification. This has permitted successful imaging of entire vascular distributions, in addition to minimizing motion artifact and increasing longitudinal spatial resolution, potentially lowering administered contrast doses. The time between the start of contrast injection and the commencement of scanning can be tailored in response to a particular clinical question, permitting image acquisition during the arterial, venous, and/or equilibrium phases. Exquisite anatomic detail of both intra- and extraluminal structures is revealed using this technique, including detection of intimal calcification and mural thrombosis. CT angiography has become an important tool for assessment of the abdominal and iliac arteries and their branches, the thoracic aorta, the pulmonary arteries, and the extra- and intracranial carotid circulation.

CT COLONOGRAPHY

CT colonography (virtual colonoscopy), introduced in 1994, is a relatively new, noninvasive method of imaging the colon in which thin-section, helical CT data are used to generate two- or three-dimensional images of the colon. This technology has been used primarily in the detection and characterization of colonic polyps, rivaling traditional colonoscopic approaches and conventional barium enema examinations. These images display the mucosal surface of the colon and internal density of the detected lesions; they also directly demonstrate the bowel wall and extracolonic abdominal and pelvic structures.

ULTRASONOGRAPHY

Diagnostic ultrasound is a noninvasive imaging technique that uses high-frequency sound waves greater than 20 kilohertz (kHz). A device known as a transducer is used to emit and to receive sound waves from various tissues in the body. The transducer is placed against the patient's skin with a thin layer of coupling gel. This gel displaces the air that would otherwise reflect virtually all of the incident ultrasound beam. As sound travels into the patient, wave fronts spread out, diminishing the overall beam intensity. Beam attenuation also occurs secondary to partial tissue absorption with associated heat conversion. At tissue interfaces, the beam is partially reflected and transmitted. The reflected sound waves, or echoes, travel back to the transducer, are converted into electric signals, and then amplified. The amplitude of the returning wave partially depends on the degree of beam absorption. A shade of gray is then assigned to each amplitude, with strong echoes being typically assigned a shade near the white end of the spectrum and weak echoes given a shade near the black end of the spectrum. In addition, the depth of the reflecting tissue can be calculated from the known total beam travel time and the average sound velocity in human tissue (1540 meters/second). Limitations of this modality are based on its operator-dependent nature, sensitivity of equipment to slow flow, variable visualization of midline abdominal organs (pancreas) or vasculature when obscured by overlying bowel gas, and inability of sound waves to penetrate gas or bone.

Ultrasonography has many common applications that involve imaging of the abdomen [liver, gallbladder (Fig. 1–6), pancreas, kidneys], pelvis (female reproductive organs), fetus (routine fetal surveys for detection of anomalies), vascular system (aneurysms, arteriovenous communications, deep-venous thrombosis), testicles (tumor, torsion, infection), pediatric brain (hemorrhage, congenital malformations), breast, and chest (size and location of pleural fluid collections). In addition, ultrasound-guided interventions have been used for facilitating lesion biopsy, abscess drainage, and radiofrequency ablation.

Doppler ultrasound is used primarily to evaluate vascular flow by detecting frequency shifts in the reflected beam, utilizing a principle termed the Doppler effect. This effect occurs when a sound emitter or reflector is moving relative to the receiver of sound. Objects moving toward the detector appear to have a higher frequency and shorter wavelength, while objects moving away from the detector appear to have a lower frequency and longer wavelength. If the ultrasound beam strikes a reflector (vessel wall, heart) moving toward it, the reflected sound will have a higher frequency than the original beam. The reflected sound will be of lower frequency than the original beam if the reflector is moving away when the beam strikes it. The Doppler shift is the frequency difference between the original beam frequency and the reflected beam frequency. These frequency differences are used to calculate the corresponding flow velocities, from which a Doppler waveform, or tracing, can be generated. This tracing depicts the relationship between velocity and time and is unique to the flow pattern within the vessel. Color flow Doppler assigns colors (blue and red) to structures according to their motion toward or away from transducers. This information can be superimposed on a gray-scale image.

Endoluminal sonography uses a high-frequency catheter-based transducer (9–20 MHz) to image structures beyond the lumen of the hollow viscus. It is accurate in local staging of cancer and in detecting small lesions that may not be visualized with other imaging modalities. Limitations for optimal evaluation include inability to precisely position the transducer within an area of interest that may restrict full entry.

Endoluminal sonography has been applied in many fields. GI applications of endoluminal sonography include quantification of the size and wall thickness of esophageal carcinoma or varices. Transrectal ultrasound is performed for evaluation of the prostate. Genitourinary (GU) applications include guidance of collagen injections, examination of the severity and length of ureteral strictures, diagnosis of upper tract neoplasms and urethral diverticula, identification of submucosal calculi, and visualization of crossing vessels prior to endopyelotomy. The uterus, adnexa, and routine fetal examinations can be conducted using a transvaginal probe in the presence of an empty bladder. Sonohysterography, an ultrasound-guided procedure, requires instillation of a sterile saline solution into the uterine cavity following cannulation for evaluation of endometrial masses or other abnormalities. Transesophageal echocardiography is used for evaluating cardiovascular abnormalities after placement of a probe into the esophagus. More recently, intravascular applications of sonography have been promising for quantitating the degree of arterial stenosis and for monitoring the therapeutic effects of angioplasty in both peripheral and coronary arteries. Intravascular ultrasound (IVUS) has been applied to modeling plaque morphology, blood flow, and the geometry of the vessel lumen.

Three-dimensional ultrasound (3D-US) has developed during the last decade with advancements in computer processing power and has rapidly achieved widespread use with numerous clinical applications, including obstetrics and vascular imaging. 3D-US is used to quantify the volume of organs and pathology. It has been used predominantly in obstetrics for studying normal embryonic and/or fetal development and for detecting specific congenital anomalies in a fetus at risk given a known family history.

MAGNETIC RESONANCE IMAGING

In 1952, Felix Bloch and Edward Purcell were awarded the Nobel Prize in physics for their independent discovery of the magnetic resonance phenomenon in 1946. Between 1950 and 1970, nuclear magnetic resonance (NMR) was developed and used for chemical and physical molecular analysis. In 1971, Raymond Damadian showed that NMR may have utility in cancer diagnosis, based on observed prolonged relaxation times in pathologic tissue. The first 2D proton NMR image of a water sample was generated in 1972 by Paul Lauterbur using a back-projection technique, similar to that used in CT. In 1975, Richard Ernst used phase and frequency encoding, as well as Fourier transform analysis, to form the basis of current MR imaging techniques. All of these experiments used defined, nonuniform magnetic fields, or linear variations of field strength, along all coordinate axes. The application of these nonuniform fields (magnetic field gradients) permitted discrimination of various signals from different spatial locations. In MR imaging, a pulsed radio-frequency (rf) beam is used in the presence of a strong main magnetic field to generate high-quality images of the body. These images can be acquired in virtually any plane, although sagittal, coronal, and axial images are commonly obtained.

Hydrogen nuclei are favored for MR imaging. Once the patient is in place in an MR scanner, the randomly oriented hydrogen nuclei align with the static magnetic field. To detect a signal, a perturbing rf pulse is transiently applied to the patient, resulting in a net change in alignment of these nuclei. When the rf pulse is turned off, the spins return to their equilibrium state by dissipating energy to the surrounding molecules. The rate of energy loss is mediated by the intrinsic relaxation properties of the tissue, designated as the longitudinal (T₁) and transverse (T₂) relaxation times. T₁represents the restoration of the longitudinal magnetization along the axis of the main magnetic field; T₂ represents the decay time of the magnetization in the transverse plane. Although a detailed explanation is beyond the scope of this chapter, substances (e.g., fluid) that have a long T₁ will appear dark on T₁-weighted images, whereas those with short T₁ (fat) will display high signal intensity. On T₂-weighted images, a long T₂ substance (fluid) will appear bright. Advantages of MR imaging include its superb contrast resolution, high spatial resolution, and lack of ionizing radiation.

The most commonly used, clinically approved contrast for MR imaging is a paramagnetic agent (atoms with unpaired electrons in their outer shells) containing gadolinium, termed gadolinium dimeglumine (or Gd-DTPA), a T₁-shortening agent. Tissue relaxation results from interactions between the unpaired electron of gadolinium and tissue hydrogen protons, which significantly decrease the T₁ of the blood relative to the surrounding tissues. Adverse reactions to this agent are far less frequent than those seen with iodinated compounds, with common reactions including nausea, vomiting, headache, paresthesias, or dizziness.

MR imaging is contraindicated for patients with metal implants or foreign bodies, such as intracranial aneurysm clips, intraorbital metallic foci, cardiac pacemakers, or specific types of cardiac valves. In these instances, these objects may be dislodged or damaged by the magnetic field. MR imaging may also be contraindicated for claustrophobic or uncooperative patients who may not respond to conscious sedation protocols.

Technical advances in gradient hardware, resulting in faster and stronger gradients, have permitted subsecond image scan times. Newer pulse sequences have been developed that currently augment the conventional MR pulse sequences (spin echo and gradient echo), increasing the sensitivity of clinical studies to disease detection. These rapid imaging techniques offer major advantages over conventional MR imaging, including decreased image acquisition times, minimized patient discomfort, and increased ability to image physiologic processes in the body.

Fast spin echo, fast gradient echo, diffusion imaging, perfusion imaging, and echo planar imaging are examples of fast imaging techniques that can be performed on clinical scanners. Diffusion-weighted imaging is exquisitely sensitive to the microscopic molecular motion of water, demonstrating areas of limited (restricted) intracellular diffusion following an acute ischemic event. This sequence is routinely utilized in clinical neuroimaging protocols, but is nonspecific for pathology; diffusion changes similar to that seen with acute ischemia can be observed with infection and some tumors. In conjunction with the results of diffusion-weighted imaging, areas of the brain at risk for further ischemia may be identified. Perfusion-weighted MR imaging, a less frequently used technique, provides information about the blood supply to a particular area of the brain following rapid bolus injection of Gd-DTPA. Echo planar imaging (EPI), introduced by Mansfield and Pykett in 1978, allows the collection of all data required for image reconstruction to occur within a fraction of a second, after a single rf pulse. This technology has resulted in significant clinical and scientific advances, such as in stroke evaluation and functional brain imaging, respectively. Functional MR imaging studies of the human brain using EPI techniques have allowed physiologic investigations of the functional organization of the brain.

Three-dimensional contrast-enhanced magnetic resonance angiography (MR angiography) is used for noninvasive assessment of many vascular abnormalities, including aneurysms, dissection, vessel anomalies, and coarctation. It has evolved from the use of fast scanning techniques on high gradient strength units, in combination with Gd-DTPA. Using this technique, volumetric acquisitions can be performed in a single breath-hold. In many cases, gadolinium-enhanced MR angiography has supplemented traditional noncontrast MR angiography (time-of-flight or phase contrast) techniques. Noncontrast MR angiography methods are partially hampered by longer acquisition times and motion artifacts. Single breath-hold, contrast-enhanced MR angiography techniques avoid many of these problems. Improvements in contrast resolution are achieved, regardless of the plane of acquisition. This has allowed reductions in the number of image sections needed to display a large vascular territory and in the overall imaging acquisition times. Multiphase dynamic imaging is usually performed after intravenous gadolinium administration, with the arteries best seen during the early phase and veins during the later phases.

MR imaging has traditionally been used for neurologic indications, including brain tumors (Fig. 1–7), acute ischemia, infection, and congenital abnormalities. More recently, MR imaging has been used for a number of non-neurologic indications, namely, spine, musculoskeletal, cardiac, hepatic, biliary, pancreatic, adrenal, renal, breast, and female pelvis applications. Spine MR studies are useful for evaluating degenerative changes, disk herniation, infection, metastatic disease, or congenital abnormalities. Common musculoskeletal applications involve the knee, shoulder, and hip. The primary indication for the knee is the assessment of the menisci and ligaments following internal derangement. Rotator cuff tear is the most typical shoulder indication. Cardiac studies are performed to identify congenital anomalies and complex malformations (malpositioning of great vessels and/or cardiac chambers). In the abdomen, hepatic MR imaging studies are often used to diagnose atypical presentations of liver lesions, metastatic diseases, or hepatocellular carcinoma. Adrenal studies are performed primarily to distinguish adrenal adenomas from metastatic disease. Atypical renal masses, found incidentally on ultrasonography or CT, can often be better characterized on MR imaging. In addition, renal MR imaging is used to establish the presence and extent of tumor thrombus in cases of renal cell carcinoma for tumor staging purposes. Breast MR imaging is predominantly utilized for detection and evaluation of ruptured implants, and studies of the use of MR to detect and characterize heterogeneous cancers in the breast are promising. Finally, oncologic applications in the female pelvis have included the diagnosis and characterization of cervical and endometrial carcinomas, as well as adnexal lesions.

Magnetic resonance cholangiopancreatography (MRCP) is used to evaluate choledocholithiasis, retained gallstones, pancreatobiliary neoplasms, strictures, primary sclerosing cholangitis, and chronic pancreatitis. This noncontrast technique relies on the relatively stationary nature of the bile (compared with blood) to depict the predominantly fluid-filled pancreatic ducts and biliary tree. Rapid, heavily T₂-weighted, breath-hold sequences are utilized, resulting in high signal intensity ductal structures. In patients who have failed endoscopic retrograde cholangiopancreatography (ERCP) or who are unable to tolerate this procedure, MRCP has become a suitable alternative. MRCP is particularly useful in postoperative patients, patients with biliary system anomalies, and as a screening tool in patients with an otherwise low probability of a biliary abnormality. ERCP is generally reserved for therapeutic purposes, such as stent placement, stone extraction, or stricture dilatation.

Molecular MR imaging is a functional imaging strategy that probes biologic processes at a cellular or molecular level using targeted MR contrast agents. Applications include imaging gene expression and visualization of surface receptors.

NUCLEAR MEDICINE

Nuclear medicine studies are performed by administering a radiopharmaceutical to the patient and subsequently recording its distribution in the body over a defined period of time. In general, these studies are very sensitive, but relatively nonspecific, in the detection of pathophysiology. Correlation with pertinent clinical history, physical findings, laboratory data, and other diagnostic imaging procedures is, therefore, essential to rendering an accurate interpretation and for maximizing clinical benefits to both the patient and ordering physician.

Radiopharmaceuticals are radioactive compounds that typically do not elicit a physiologic response when administered for diagnostic or therapeutic purposes. Selective uptake of these compounds by various organs forms the basis for nuclear imaging. In addition, the radionuclide must be normally involved in the physiologic metabolism of the organ to be successfully imaged. These radiopharmaceuticals typically consist of two components: (1) the main component, which is the compound distributed to various organs by a number of physiologic mechanisms, and (2) the radionuclide tagged to this component, which emits gamma rays, permitting detection of the compound in the body. Mechanisms by which these radiopharmaceuticals localize in organ systems include active transport, phagocytosis, capillary blockade, cell sequestration, compartmental localization, exchange diffusion, chemisorption, antigen–antibody reactions, receptor binding, and metabolic trapping. Physiologic function can then be assessed, but spatial resolution is relatively poor.

Most nuclear medicine studies are performed with gamma cameras, which provide planar (two-dimensional) images. A gamma camera converts photons emitted by the ingested or injected radionuclide into a light pulse. This pulse is then converted into a voltage signal, which is used to produce an image of the radionuclide distribution. Gamma cameras may be analog or digital. An analog signal, used throughout the analog camera, is inherently noisy and can encompass an infinite array of values. A digital signal has a discrete number of values. Single photon emission computed tomography (SPECT) is a tomographic technique that utilizes a rotating gamma camera system. One advantage over two-dimensional planar images is the improved image contrast achieved by focusing on a thin slice of tissue, while eliminating the overlying and underlying confounding activity that may obscure lesions of interest.

Positron emission tomography is a molecular imaging modality that uses a positron emitter, 18F-fluorodeoxyglucose (18F-FDG), to create tomographic images by detecting gamma rays. Gamma rays are produced when the emitted positrons interact with electrons. This technique permits metabolic alterations in tumor cells to be detected and is used for diagnosing and staging. Advantages of both SPECT and PET include the ability to map the distribution of the radiopharmaceutical in three dimensions, with the added possibility of quantifying uptake, and the ability to generate cinematic displays of the organ imaged.

Common nuclear medicine procedures include (1) cardiac studies for evaluation of myocardial perfusion and/or ventricular function; (2) skeletal studies for detection of early bone metastases (Fig. 1–8), skeletal trauma, osteomyelitis, and primary bone neoplasms; (3) renograms and renal scans for assessment of renal function and morphologic defects (e.g., pyelonephritis); (4) ventilation–perfusion studies for identification of suspected pulmonary emboli; and (5) PET studies for tumor diagnosis and staging (e.g., lung, colorectal, breast, lymphoma, melanoma), as well as for evaluation of neurodegenerative disorders (dementia), brain tumor recurrence, and myocardial viability. A number of less commonly performed nuclear medicine studies include (1) diagnostic thyroid studies for evaluation of nodules and following iodine-131 therapy for hyperthyroidism and thyroid cancer, (2) hepatobiliary studies for determination of acute cholecystitis and bile duct patency, (3) brain imaging for assessment of brain death and dementia, (4) white blood cell studies for detection of infectious or inflammatory processes, (5) gastrointestinal bleeding studies for detection and localization of small bleeds, (6) lymphoscintigraphy to identify sentinel lymph nodes for surgery, and (7) parathyroid scans to identify adenomas and hyperplasia.

Relatively new developments in PET technology include the use of combined PET/CT imaging devices, synthesis of new tracers for targeting the biologic properties of cancer cells, in vivo imaging of cellular processes, and high-resolution PET instruments (microPET). Several advantages to using combined molecular and anatomic imaging (PET/CT scanners) include (1) the performance of physiologic and anatomic scanning in one examination; (2) limited patient motion; permitting accurate fusion of biologic and anatomic images; (3) greater ability to assign physiologic abnormalities to anatomical structures; and (4) improved diagnostic accuracy in the evaluation of hard-to-image regions of the body (e.g., head and neck), given the more precise assignment of metabolic alterations to anatomic landmarks. New molecular imaging probes may target and characterize the genetics and metabolic properties of in vivo tumor cells, allow monitoring of treatment responses, and yield prognostic information. MicroPET technology permits imaging of gene expression and assessment of cellular processes, such as metabolism.

Dawn SK, Gotway MB, Webb WR. Multi-detector row spiral computed tomography in the diagnosis of thoracic diseases.Respir Care. 2001;46:912–921. [PMID: 11513764]

Di Salle F, Formisano E, Linden DE, et al. Exploring brain function with magnetic resonance imaging. Eur J Radiol.1999;30:84–94.

Eisenberg RL. Radiology: An Illustrated History. St. Louis: Mosby–Year Book; 1992.

Elster AD, Burdette JH. Questions and Answers in Magnetic Resonance Imaging. 2nd ed. St. Louis: Mosby; 2001.

James JJ, Davies AG, Cowen AR, O'Conner PJ. Developments in digital radiography: an equipment update. Eur Radiol.2001;11:2616–2626. [PMID: 11734969]

Mettler FA, Guiberteau MJ. Essentials of Nuclear Medicine Imaging. 4th ed. Philadelphia: Saunders; 1998.

Prince MR, Grist TM, Debatin JF. 3DContrast MR Angiography. 2nd ed. Berlin: Springer-Verlag; 1999.

Thrall JH. Directions in radiology for the next millennium. AJR 1998;171:1459–1462. [PMID: 9843271]

Webb WR, Brant WE, Helms CA. Fundamentals of Body CT. 2nd ed. Philadelphia: Saunders; 1998.

Williamson MR. Essentials of Ultrasound. Philadelphia: Saunders; 1996.

Winston CB, Schwartz LH. Advances in magnetic resonance imaging: applications in body imaging. Cancer Invest.1998;16:413–420. [PMID: 9679533]

Zagoria RJ, Tung GA. Genitourinary Radiology: The Requisites. St. Louis, Mosby, 1997.