Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Cross-sectional imaging plays a major role in the evaluation of adrenal gland disease. In many patients not only can adrenal gland pathology be identified, but also a specific diagnosis can be made. The appropriate selection and accurate interpretation of adrenal imaging and interventional studies are the subject of this chapter, which is divided into three sections. In the first section, the embryology, physiology, anatomy, and imaging of the adrenal gland are reviewed. Mass lesions of the adrenal gland are discussed in the second section. The third section reviews the approach to several common clinical problems in which adrenal imaging plays an integral role.
The embryology of the adrenal gland reflects its physiologic and anatomic separation into the cortex and the medulla. The adrenal cortex develops from the coelomic mesoderm in the fourth to sixth weeks of life as a cluster of cells between the root of the mesentery and the genital ridge. The adrenal medulla is of neural crest origin derived from the neuroectoderm. The development of the adrenal gland is independent from that of the kidney, and the ipsilateral adrenal gland is positioned in its normal anatomic location in more than 90% of patients with agenesis or malposition of the kidney.
Adrenal cortical tissue, which makes up approximately 90% of the adrenal gland by weight, synthesizes cholesterol-derived steroid hormones. Adrenal steroids contain either 19 or 21 carbon atoms. Steroids with 21 carbon atoms (C21 steroids) have either glucocorticoid or mineralocorticoid activity, whereas the C19 steroids have androgenic activity predominantly.
The major glucocorticoid produced by the adrenal gland is cortisol, which plays an important role in the regulation of protein, carbohydrate, lipid, and nucleic acid metabolism. In addition, cortisol has potent anti-inflammatory properties. Adrenal synthesis and secretion of cortisol are stimulated by adrenocorticotropic hormone (ACTH), also known as corticotropin , a peptide produced in basophilic cells of the anterior pituitary gland. The important factors that influence release of ACTH include the sleep/wake cycle, stress, plasma cortisol concentration, and corticotropin-releasing hormone (CRH), which is produced in the hypothalamus. Thus a negative-feedback servomechanism involving cortisol, ACTH, and CRH regulates adrenal secretion of glucocorticoids.
The renin-angiotensin system plays a pivotal role in the regulation of extracellular fluid, largely through its action on the adrenal mineralocorticoid, aldosterone. Renin is an enzyme produced and stored in the granules of the juxtaglomerular cells, which surround the afferent arterioles of the renal glomerulus. Renin is released in response to reduced renal perfusion as signaled by reduced afferent arteriole perfusion pressure, increased delivery of filtered sodium to the distal tubule, and increased sympathetic nerve stimulus. Renin acts on angiotensinogen to form angiotensin I, and the converting enzyme forms angiotensin II from angiotensin I. Angiotensin II is a potent stimulator of aldosterone production by the adrenal cortex. Increasing blood levels of aldosterone lead to sodium retention and an expansion of the extracellular fluid volume. In addition, aldosterone is an important regulator of potassium metabolism.
The major androgen secreted by the adrenal cortex is dehydroepiandrosterone (DHEA), which is the main precursor of the urinary 17-ketosteroids. The relatively weak adrenal androgens exert a greater effect after conversion in extra-adrenal tissues to the more potent androgen, testosterone. ACTH regulates the production of DHEA and other weak androgens by the adrenal cortex.
Physiologically, the adrenal medulla is best thought of as an endocrinologic homolog with the postganglionic sympathetic neuron. The medulla maintains high concentrations of catecholamines, of which 85% is epinephrine. In contrast to the regulation of adrenal cortical steroid secretion by hormones or enzymes, release of catecholamines into the bloodstream in response to systemic stress occurs due to stimulation by the preganglionic sympathetic nerves. The medulla is composed of chromaffin cells, so named because these cells stain brown with chromic acid salts, which oxidize intracellular catecholamines.
The right adrenal gland is suprarenal in location and is first imaged 1 to 2 cm cephalad to the upper pole of the right kidney. Its inferior extent can be seen anterior and medial to the upper pole. The right adrenal gland is posterior to the inferior vena cava, lateral to the right crus of the diaphragm, and medial to the right lobe of the liver. The left adrenal gland is located at or caudal to the level of the right adrenal gland. It is most often imaged anteromedially to the upper pole of the left kidney and frequently extends to the level of the left renal hilum. The left adrenal gland is lateral to the aorta and left crus of the diaphragm, and posterior to the pancreas and splenic vessels. The anatomic relationship of the right and left adrenal glands to the inferior vena cava and the splenic vein, respectively, is important because it may suggest an adrenal origin for a large upper-quadrant mass.
At birth, the adrenal glands are almost one third the size of the kidneys, whereas in adults they are about one thirtieth the size of the kidneys. The cephalocaudal length of the adrenal gland varies from 4 to 6 cm and the width varies from 2 to 3 cm. Because of this variation, these dimensions are used infrequently as criteria for the assessment of adrenal gland size. This variation also explains why endocrinologically hyperfunctioning and pathologically hyperplastic adrenal glands may appear normal in size at imaging and surgery ( Fig. 9-1 ). Depending on the level of transverse (axial) computed tomography (CT) images, the adrenal gland may have a variety of configurations, varying from oblique linear to an inverted Y, inverted V, or inverted T shape. The normal adrenal gland is composed of the adrenal body and two limbs, medial and lateral. The width of a normal adrenal limb, when measured perpendicular to the long axis on the transverse plane, has a range of 4 to 9 mm. In patients with congenital absence of a kidney or a pelvic kidney, the ipsilateral adrenal gland will have an elongated or flattened appearance ( Fig. 9-2 ).
Three arteries typically supply each adrenal gland: the inferior, middle, and superior adrenal arteries. The inferior adrenal artery most often is a branch of the proximal, ipsilateral renal artery and usually is the major artery to the adrenal gland. A single central adrenal vein drains each adrenal gland. From the right adrenal gland, three segmental veins join to form a short central vein that enters the posterior inferior vena cava. From the left adrenal gland, a long, single adrenal vein enters the superior aspect of the left renal vein opposite the gonadal vein.
Imaging of the enlarged adrenal gland or adrenal mass can be accomplished with a variety of modalities including ultrasonography (US), CT, magnetic resonance imaging (MRI), nuclear scintigraphy, and angiography. CT, with its capability to create multiplanar reformatted images, is the most readily available and consistently effective means of imaging the healthy and the abnormal adrenal gland. In the investigation of a large upper-quadrant mass, the multiplanar imaging capabilities of MRI and US may also be valuable when the relationship of a mass to the kidney or to other retroperitoneal organs or vessels needs to be determined. Angiography and nuclear scintigraphy have limited and specific roles in the evaluation of adrenal gland abnormalities.
The CT scan can image the normal adrenal gland in nearly all patients. Most adrenal masses can be detected when 5 mm or thinner slices are acquired. Oral contrast administration is not usually necessary in the evaluation of the adrenal gland. Intravenous contrast administration with enhancement of the kidney, liver, or pancreas may be helpful to distinguish an adrenal mass from these organs. In addition, intravenous contrast enhancement may be a valuable adjunct for characterizing an adrenal mass. Following unenhanced and intravenous contrast-enhanced CT, rescanning through the adrenals 15 minutes after the start of contrast material injection can be used to assess contrast material washout for indeterminate adrenal masses. This is discussed later in the chapter. CT guidance can also be used for percutaneous needle biopsy of an adrenal mass. Biopsy of an adrenal mass can be accomplished with the patient in a prone position, but often this necessitates steep-needle angulation to avoid transgressing the pleural space in the posterior sulcus. The transhepatic approach, which avoids the pleural space and the need for cephalocaudal needle angulation, is an effective method for biopsy of a right adrenal mass. A useful alternative to the prone position is the ipsilateral decubitus position, particularly when a left adrenal mass is to be sampled. This position elevates the dependent diaphragm and reduces respiratory excursion of the ipsilateral thorax and upper abdomen, thereby reducing the risk of pleural entry by the biopsy needle.
MRI of the adrenal gland should include T1- and T2-weighted images. The normal adrenal gland has low to intermediate signal on both T1- and T2-weighted images. Chemical-shift imaging (in phase and out of phase) is the mainstay of MRI of the adrenal glands and can be performed with T1-weighted, gradient-echo pulse sequences. Echo times for chemical-shift imaging vary by magnetic field strength. At 1.5 T, the echo time for in-phase imaging is 4.6 ms and for out-of-phase imaging is 2.3 ms, whereas at 3 T, the echo time for in-phase and out-of-phase imaging is 2.3 and 1.1 ms, respectively. Although surface coils have been used, the body coil provides a larger field of view and makes the evaluation less cumbersome. The transverse and coronal imaging planes are most often used to evaluate the adrenal gland. With respect to demonstrating the healthy adrenal gland and masses exceeding 1 to 2 cm in diameter, MRI is comparable with CT. Multiplanar imaging may be useful in the evaluation of the origin and extension of a large upper-quadrant mass.
Normal adrenal tissue is usually slightly lower in signal intensity than that of normal liver and renal cortex on T1-weighted images. On T2-weighted images, the healthy adrenal gland may be difficult to distinguish from adjacent retroperitoneal fat.
US can be used to evaluate the normal adrenal gland, but it is variably successful because of the small size of the adrenal gland and its frequent obscuration by retroperitoneal fat. One method to image the right adrenal gland uses a lateral or anterior approach with the patient in the supine or left lateral decubitus position. Scanning through the ninth to tenth intercostal spaces is performed. An imaginary line connecting the center of the right kidney and the inferior vena cava should pass through the right adrenal gland in either the sagittal plane or the transverse plane. The left adrenal gland is imaged on a coronal plane by scanning in the posterior axillary line with the patient in the right lateral decubitus position. An imaginary line drawn through the spleen or the left kidney to the aorta should intersect the left adrenal gland. Normally, the adrenal gland appears as a hypoechoic triangular or semilunar structure, but often only echogenic retroperitoneal fat is seen. Ultrasound plays a greater role in the examination of the patient with a known or suspected adrenal mass lesion. It is particularly valuable when there is a need to distinguish an adrenal mass from one originating from the upper pole of the kidney, the liver, or the pancreas. In some patients an adrenal mass can be characterized accurately sonographically, such as when the fat content of an adrenal myelolipoma is recognized ( Fig. 9-3 ).
Imaging of the adrenal glands with radiopharmaceuticals provides functional information that is unavailable from other imaging modalities. Radioactive iodine I 131 -labeled 6-beta-iodomethyl-19-norcholesterol (NP-59) is a cholesterol analog that accumulates in adrenocortical tissues. Adrenal uptake of NP-59 is affected by circulating levels of ACTH and is inversely related to the size of the body cholesterol pool. Exogenous corticosteroid administration suppresses pituitary secretion of ACTH, decreasing baseline adrenal uptake of this radiotracer. In the presence of a known hyperfunctioning adrenal lesion, administration of a potent corticosteroid, such as dexamethasone, before an NP-59 scan permits assessment of the autonomy of the adrenal lesion. Furthermore, the distribution of adrenal radioactivity after dexamethasone suppression distinguishes between lesions that are unilateral (i.e., neoplasm) from those that are bilateral (i.e., hyperplasia). This radiopharmaceutical has been shown to be useful in the evaluation of patients with biochemical evidence of adrenal hyperfunction.
Functional lesions of the adrenal medulla can be imaged using I 131 - or I 123 -labeled meta-iodobenzyl-guanidine (MIBG) and In 111 -octreotide. Radioiodine-labeled MIBG is a radiopharmaceutical that bears a structural similarity to norepinephrine. In 111 -octreotide is a somatostatin analog taken up by some neuroendocrine tumors, including most pheochromocytomas. Both of these radionuclides have been effectively used in the assessment of patients with suspected extra-adrenal or recurrent pheochromocytomas. They also may be used to evaluate sites of metastases in patients with malignant pheochromocytoma and to follow patients with multiple endocrine neoplasia (MEN) syndromes. Radioiodine-labeled MIBG and In 111 -octreotide are somewhat complementary in that virtually all pheochromocytomas show uptake of at least one of these agents ( Fig. 9-4 ).
Fluorine-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) and PET-CT scanning have shown great promise in detecting adrenal malignancies. PET-CT has high sensitivity and specificity for lesion detection and classification of lesions as benign or malignant ( Fig. 9-5 ). False-positive diagnosis of malignancy at PET-CT is uncommon (approximately 5% of cases) and may be secondary to inflammatory/granulomatous processes (including sarcoidosis and tuberculosis), adrenal cortical hyperplasia, or abnormalities adjacent to the adrenal gland. False-negative diagnosis may occur secondary to hemorrhage or necrosis within an adrenal mass.
There are several specific indications for angiography in the evaluation of the adrenal gland. Patients may undergo adrenal arteriography as part of a comprehensive arteriographic examination to determine the organ of origin of a large abdominal mass if this remains unclear despite adequate evaluation by cross-sectional imaging. Adrenal venography and venous sampling are reserved for patients with primary aldosteronism or, more rarely, Cushing syndrome; in these patients adenoma must be distinguished from adrenal hyperplasia. Total-body venous sampling can be used effectively when the site of recurrent or persistent pheochromocytoma is sought. Complications of adrenal venography include extravasation, venous thrombosis, adrenal infarction, and minor retroperitoneal hemorrhage.
Adrenal masses are identified in up to 5% of patients who undergo abdominal CT examination. Most of these masses are adenomas, benign tumors of the adrenal cortex. A majority of adenomas are nonfunctional and are discovered as an incidental finding. The majority of nonhyperfunctioning adenomas consist of large cells containing abundant cytoplasmic lipid. Functional adrenal adenomas may be the cause of Cushing syndrome, primary hyperaldosteronism, virilization, or feminization.
Size, contour, attenuation values on precontrast and postcontrast imaging, and growth are features that are used frequently to characterize adrenal masses by cross-sectional imaging ( Table 9-1 ). There are two key properties of adrenal adenomas that can be exploited with imaging for diagnosis. First, most adenomas have increased intracellular lipid content. This results in near-zero attenuation values on unenhanced CT. Attenuation values of less than 10 Hounsfield units (HU) are specific for lipid-rich adrenal adenoma, with 80% sensitivity and 95% specificity in differentiating adenomas from adrenal metastases. The majority of masses with measured attenuation less than 10 HU are lipid-rich adenomas ( Fig. 9-6 ). The second feature important for imaging diagnosis of adrenal adenomas results from the fact that adenomas, even the lipid-poor group, have more rapid contrast material dissipation, or washout, than metastatic tumors ( Box 9-1 ). Combining the findings of attenuation 10 HU or less on unenhanced CT and an absolute contrast material washout value of 60% or greater on a 15-minute-delayed CT scan, 98% of adenomas can be distinguished from nonadenomas ( Fig. 9-7 ). If unenhanced images were not obtained, relative washout can be calculated based on attenuation values of the adrenal mass on dynamic and 15-minute-delayed images. Relative washout of 40% or more supports the diagnosis of adenoma. Adrenal washout calculators are readily available on the Internet. In addition, most adrenal adenomas are 3 cm or smaller in diameter at the time they are discovered, and it is rare for these tumors to measure over 6 cm when initially detected. The contour of these tumors is usually well defined and smooth. The typical appearance is a solid and homogeneous mass with an attenuation that is lower than that of adjacent muscle. Calcification and central necrosis are unusual in adenomas ( Fig. 9-8 ). An important feature of an adrenal adenoma is that this tumor is stable in size at serial imaging for intervals of up to 2 years.
Adenoma | Metastasis | |
---|---|---|
Size at presentation | Often < 3 cm | Variable |
Change in size over time | No | Yes |
Measured CT attenuation | ≤10 HU (unenhanced CT) | >10 HU (unenhanced CT) |
Contrast washout | ≥60% absolute, ≥40% relative |
<60% absolute, <40% relative |
Signal intensity * | Isointense-to-hypointense to liver | Hyperintense to liver |
Out-of-phase image † | Decreased | Same or increased |
* On a conventional T2-weighted image.
† Out-of-phase image refers to the signal intensity of the mass on a T1-weighted gradient-echo out-of-phase image compared with an in-phase image.
Absolute percentage washout = ( E − D )/( E − U ) × 100
Relative percentage washout = (E − D)/E × 100
where:
E =CT attenuation of adrenal mass during the portal venous phase of contrast enhancement
D =CT attenuation of adrenal mass on a 15-minute-delayed scan
U =CT attenuation of adrenal mass on an unenhanced scan
An absolute percentage washout ≥ 60% or relative washout ≥ 40% is required for diagnosis of adenoma.
CT, Computed tomography.
Adenomas larger than 1 to 1.5 cm in diameter can be evaluated with MRI. Although the signal intensity of this tumor on T1- and T2-weighted images may be useful in characterization, unfortunately it is not always specific. Nonfunctional adenomas are usually isointense with normal adrenal tissue on T1- and T2-weighted images. These tumors are isointense to the liver on T1-weighted images and isointense-to-slightly hyperintense to the liver on T2-weighted images. A number of techniques have been investigated to distinguish adenomas from other adrenal masses on high-field MRI, including calculated T2 relaxation time, dynamic contrast-enhanced imaging, and chemical-shift gradient-echo imaging.
Chemical-shift imaging has become the standard MRI method for characterizing adrenal adenomas. The chemical-shift family of pulse sequences is based on the difference in resonance frequency between the two major constituents of the hydrogen resonance spectrum: water and triglyceride protons. This difference translates to a frequency shift of 224 Hz on a 1.5-T system and 445 Hz on a 3-T system. At 1.5 T protons in fat and water cycle in phase and out of phase approximately every 2.2 ms, and at 3 T this cycle occurs approximately every 1.1 ms. If the echo is sampled when water and fat are in phase, the fat and water combine to generate a larger signal than an echo collected when they are out of phase. With gradient-echo pulse sequences and short echo times, the signal intensity of an adrenal mass can be evaluated using in-phase and out-of-phase pulse sequences. Adrenocortical adenomas consist of large, lipid-laden cells similar to those of the zona fasciculata, and they are unlike most adrenal metastases and pheochromocytomas, although some malignant tumors may contain cytoplasmic lipid (e.g., hepatocellular carcinoma, renal cell carcinoma, liposarcoma, well-differentiated adrenocortical carcinoma). Postulating that the cancellation of signal in masses containing water and fat protons in the same voxel (i.e., adenomas) would result in a relative decrease in signal intensity on opposed-phase images, chemical-shift imaging can characterize adrenal masses with a sensitivity of 80% to 90% and a specificity of 95% to 100% ( Fig. 9-9 ). By contrast, other adrenal masses (most metastases and pheochromocytomas) and reference tissues, such as muscle and spleen, do not contain fat and should not demonstrate significant change in signal intensity between in-phase and opposed-phase sequences ( Fig. 9-9B ). False-negative results may be caused by unusually small amounts of fat in adenomas; false-positive results could be explained by the presence of small amounts of fat reported rarely in masses other than adenomas.
Whereas most adrenal adenomas contain microscopic cytoplasmic lipid, adenomas may occasionally contain macroscopic fat. This has been described in adenomas with areas of lipomatous metaplasia that appear as oval or circular areas of fat density within an adrenal adenoma ( Fig. 9-10 ).
If surgery is contemplated in a patient with a history of a primary malignancy, particularly lung, breast, kidney, or melanoma, metastasis must be excluded when an adrenal mass is discovered because this would indicate stage IV disease. Adrenal metastasis occurs in 10% to as many as 30% of patients early in the course of non–small cell lung cancer, and the autopsy incidence is as high as 40%. Similarly, as many as 30% of patients with breast cancer will have adrenal metastasis at the time of radiologic staging. Despite these statistics, the prevalence of adrenal metastasis in a patient with an adrenal mass and history of cancer ranges from 33% to 75%. Therefore even when bilateral adrenal masses are discovered, one should not assume they are metastatic in origin ( Box 9-2 and Fig. 9-11 ).
Although no specific radiologic signs of adrenal metastasis have been identified, the suspicion of malignancy should be raised when several observations are made. Change in size on serial imaging studies may be a more valuable parameter than absolute size on any given study. Although malignant adrenal tumors grow more rapidly, and therefore generally are larger at detection, metastases to the adrenal gland can be as small as 1 cm or as large as 10 cm when discovered. If an adrenal mass grows during a 4- to 6-month period of observation, or if it clearly decreases in size during systemic treatment of the primary malignancy, it is reasonable to assume that it is a metastatic lesion ( Fig. 9-12 ). When small, a metastasis can have a radiologic appearance similar to that of an adrenal adenoma (i.e., a well-circumscribed mass with homogeneous attenuation). However, several features of small tumors (<5 cm) suggest malignancy, including an irregular or poorly defined margin, moderately-to-markedly inhomogeneous attenuation, and an enhancing thick rim or nodular margin. On rare occasions, a metastasis can occur in the adrenal gland adjacent to or within a benign lesion such as adenoma. When two tumors of different histology are present in the same adrenal gland, it is known as a collision tumor ( Fig. 9-13 ). Invasion of local viscera or bone suggests malignancy. Finally, metastasis to the adrenal glands is often accompanied by evidence of widespread metastases to other organs, particularly liver, retroperitoneal nodes, or lung.
When the cause of an adrenal mass must be determined in a patient with a known primary malignancy, percutaneous needle biopsy is a safe and accurate alternative to additional or follow-up imaging. Because this invasive procedure is not without risk, it is important to establish before biopsy is performed that determining the nature of the adrenal mass will have a significant impact on therapy or prognosis. The positive and negative predictive value of image-guided needle biopsy in patients with a history of lung cancer is 100% and 90%, respectively. However, the negative predictive value is only approximately 80% when the adrenal biopsy specimen is nondiagnostic; false-negative adrenal biopsy results have been reported. Therefore in patients with a history of cancer in whom the pretest probability of adrenal metastasis is high, repeated aspiration or biopsy of an adrenal mass should be considered when the pathology report is nondiagnostic.
In the patient with no known history of malignancy, it is unusual for an incidentally discovered adrenal mass to be the initial presentation of a distant primary. This occurs less than 1% of the time. In these patients a role may exist for MRI or serial imaging studies to assess for change in size. On MRI, adrenal metastases tend to be of higher signal intensity than liver parenchyma on T2-weighted images. On T1-weighted images, the signal intensity of adrenal metastases is typically lower than that of liver and retroperitoneal fat. Metastases also tend to have a different dynamic enhancement pattern than that of adrenal adenomas, and the same signal intensity on opposed-phase images compared with in-phase gradient-echo T1-weighted images ( Fig. 9-14 ).
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here