1450 AJR:194, June 2010

major clinical importance because, for ex-ample, an isolated ipsilateral adrenal me-tastasis in a patient with resectable prima-ry non–small cell lung cancer is treated as localized disease. In these patients, re-section of isolated adrenal metastasis has been shown to extend disease-free survival [10]. CT, MRI, and PET readily character-ize many benign adrenal lesions from their characteristic diagnostic imaging features, including lipid-poor adenomas that can be identified by washout techniques and PET. A small minority of adrenal masses, however, remain indeterminate after imaging. These may include metastases (however character-izable by follow-up), adrenocortical carcino-ma (ACC), lymphoma, and pheochromocy-toma. A variety of imaging techniques have been advocated in the literature to distin-guish between benign and malignant adrenal lesions, and these techniques, along with the sometimes challenging appearances of ACC, pheochromocytoma, and lymphoma on these techniques, are highlighted in this article. General morphologic imaging principles are also emphasized.

General Morphologic FeaturesThe structural features of most adrenal le-

sions are usually not specific enough to allow them to be characterized with imaging alone. On unenhanced CT, imaging findings that are

Adrenal Imaging

Michael A. Blake1

Carmel G. Cronin Giles W. Boland

Blake MA, Cronin CG, Boland GW

1All authors: Division of Abdominal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, 55 Fruit St., Boston, MA 02114. Address correspondence to M. A. Blake (mblake2@partners.org).

Genitour inar y Imaging • Review

AJR 2010; 194:1450–1460

0361–803X/10/1946–1450

© American Roentgen Ray Society

The widespread use of imaging has led to increased detection of adrenal lesions and has under-lined the importance of accurate

adrenal lesion characterization [1, 2]. In-deed, it is essential to characterize any adre-nal lesion in patients with a known cancer because many tumors may metastasize to the adrenal glands [3], and a metastasis might contraindicate a curative treatment of the pa-tient and affect survival [4]. However, inci-dental adrenal nodules in patients with no known malignancy or endocrine abnormali-ty are present in approximately 5% of all ab-dominal CT examinations [5, 6]. The inci-dence of an adrenal nodule further increases to 9–13% in patients being scanned for a known malignancy [5], but only 26–36% of such adrenal lesions are metastatic [7]. Fur-thermore, adrenal adenomas are seen more frequently in patients with certain inherited diseases, including multiple endocrine neo-plasia type 1 and the Carney complex. The incidence of adenomas also increases with patient age, and about 6% of patients more than 60 years old have an adrenal adenoma [8]. The incidence of metastasis increases to 71%, however, if the adrenal lesion is larger than 4 cm or significantly increases in size within 1 year [9].

Accurate adrenal characterization in pa-tients with a known primary carcinoma has

Keywords: adrenal cortical carcinoma imaging, adrenal CT, adrenal imaging, adrenal lymphoma imaging, adrenal MRI, adrenal PET/CT, pheochromocytoma imaging

DOI:10.2214/AJR.10.4547

Received March 2, 2010; accepted without revision March 6, 2010.

FOCU

S O

N:

OBJECTIVE. Adrenal nodules are frequently encountered on current high-resolution im-aging, and accurate characterization of such lesions is critical for appropriate patient care. Our article highlights how imaging techniques such as CT densitometry, CT washout char-acteristics, chemical shift MRI, PET, and PET/CT help characterize most adrenal lesions. We focus on these techniques as well as specifically, because of space constraints, the varied imaging appearances of adrenocortical carcinoma, pheochromocytoma, and lymphoma on these techniques.

CONCLUSION. The imaging characterization of adrenal lesions has continued to ad-vance over the past decade as new technologies have evolved. CT, MRI, PET, and PET/CT are now established clinical techniques capable of differentiating benign from malignant ad-renal lesions.

Blake et al.Adrenal Imaging

Genitourinary ImagingReview

AJR:194, June 2010 1451

Adrenal Imaging

more suspicious for malignancy include large lesion size, irregular margins, heterogeneous appearance, and growth in size. Lesions greater than 4 cm in diameter are more likely to be malignant, and if the patient has no oth-er history of malignancy, an ACC should be considered [11] (Fig. 1). Some myelolipomas also are large but are confidently recognized owing to the presence of macroscopic fat [12, 13] (Figs. 2 and 3). Although adenomas tend to have a smooth contour and large malignant

lesions to have an irregular border, there is a very large overlap; some adenomas also show irregularity, and adrenal multinodularity is usually associated with benignity [14]. Shape is therefore not usually helpful for diagnosis. However, stability of a lesion signifies benig-nity because it is highly unusual for untreated malignant lesions to remain stable for more than 6 months [2, 15].

In contrast, any adrenal lesion that signifi-cantly increases in size over time can usu-

ally be considered malignant; some pitfalls include certain benign lesions (adenomas and myelolipomas) that rarely can increase in size but usually do so very slowly unless due to elevated levels of adrenocorticotropic hormone (ACTH) as in patients with ectopic ACTH secretion. In addition, hemorrhage into the adrenal gland will cause abrupt adre-nal enlargement [13]. Many adrenal lesions, both benign and malignant, can be hetero-geneous in attenuation, particularly after the

A

C

Fig. 1—61-year-old woman who presented with left lower quadrant pain.A–C, Arterial phase (A), portal venous phase axial (B), and coronal (C) images show well-encapsulated large 13.5-cm mass lesion arising from left adrenal gland with internal calcifications but containing no focal fat. Pancreas and left kidney are displaced by mass, but there is no evidence of invasion into adjacent vascular structures. On resection this mass represented adrenocortical carcinoma.

B

1452 AJR:194, June 2010

Blake et al.

administration of IV contrast medium. How-ever, large areas of necrosis in a lesion usu-ally signify malignancy. Most adrenal cysts can be characterized from their simple mor-phology, although some can be complex and are occasionally confused with necrotic ma-lignancies [2, 15, 16].

CTUnenhanced CT adrenal densitometry

takes advantage of the fact that around 70% of adrenal adenomas contain significant in-tracellular lipid (mainly cholesterol, fatty ac-ids, and neutral fat), in contrast to almost all malignant lesions that do not [17–19]. The high intracellular lipid content lowers the un-enhanced CT density of most adenomas. Lee et al. [17] first reported that unenhanced CT densitometry could effectively differentiate many adrenal adenomas from nonadenoma-tous disease. They found that the mean at-tenuation of adenomas (−2.2 HU) was signif-icantly lower than that of nonadenomas (28.9 HU). By using a threshold of 0 HU, these le-sions could be then differentiated with sensi-tivity and specificity of 47% and 100%, re-spectively. Later, Korobkin et al. [18] showed that there is an inverse linear relationship be-tween fat concentration and attenuation on unenhanced CT images. Conversely, almost all nonadenomatous lesions are low in intra-cellular fat, and their CT attenuation is con-sequently higher. A meta-analysis [20] of published studies later found that if the CT attenuation threshold is raised to 10 HU, the test sensitivity becomes far higher (71%) while high specificity (98%) is maintained. In clinical practice therefore, 10 HU is the most widely used threshold value for the di-agnosis of a lipid-rich adrenal adenoma. Song et al. [6] have shown that in 973 con-secutive patients with 1,049 incidental adre-nal masses adenomas accounted for 75% of incidental masses, of which 78% were lipid-rich adenomas with native CT attenuation values of less than 10 HU.

There are several limitations to unen-hanced CT densitometry, however. Most standard CT scans are now obtained after IV contrast. Thus, unenhanced attenuation mea-surements cannot be made unless performed at dual-kilovoltage scanning, which pro-vides the potential for a virtual unenhanced CT from an IV contrast-enhanced CT study. Furthermore, up to 30% of adenomas are lip-id poor and have an attenuation value great-er than 10 HU on unenhanced CT scans, as do almost all malignant lesions [2, 15, 17,

19, 21–26]. Therefore, lesions with an unen-hanced CT attenuation greater than 10 HU require further evaluation to characterize them. A further caveat is that two indepen-dent studies by Hahn et al. [27] and Stadler et al. [28] reported that different single-de-tector and MDCT helical scanners produce slightly different but statistically significant unenhanced attenuation levels, which some-times could conceivably lead to erroneous categorization [29].

More recently, CT histogram analysis has been applied to both unenhanced CT and contrast-enhanced CT images [30–34]. Bae et al. [30] have advocated this method as be-ing more sensitive than the 10-HU threshold method for the diagnosis of adrenal adeno-ma. The technique involves placement of a region of interest (ROI) over approximately one half to two thirds of the adrenal surface area excluding areas of necrosis as is done for standard attenuation measurement. This ROI is further processed with a histogram analysis tool, which is now standard on most CT viewing workstations [30]. The individ-ual attenuation values of all the pixels in the ROI are then plotted against their frequency. The amount of lipid in the mass is propor-tional to the number of negative pixels (less than 0 HU) within it.

The study by Bae et al. [30] showed that 97% of adenomas contain negative pixels. Eighty-five percent have more than 5% neg-ative pixels, and 83% have more than 10% negative pixels. No metastases had negative pixels [34]. However, results of subsequent histogram analysis studies have varied, re-porting negative pixels in both adenomas and nonadenomas, including metastases, pheochromocytomas, and ACC [31, 32]. Ho

et al. [33] reported a much lower percent-age of lipid-poor adenomas (50%) exceeding 10% negative-attenuation pixels, compared with 100% of lipid-rich adenomas. Other authors report improved sensitivities of 85–91% while maintaining perfect specificity for adenomas [33, 34].

CT histogram analysis on unenhanced CT images can therefore be used as an adjunct to the CT attenuation values because the com-bination of CT attenuation value less than 10 HU or greater than 10% negative pixel content would correctly identify 91% of ad-enomas compared with 66% using CT atten-uation values alone [33, 34]. On contrast-en-hanced CT, a 10% or greater negative pixel threshold indicates an adenoma with sen-sitivity of 12% and specificity of 99%. Al-though histogram analysis specificity for the diagnosis of adenomas on contrast-enhanced CT scans was high using a 10% negative pix-el threshold, its low sensitivity limits its clin-ical usefulness [31, 32]. However, certainly it may occasionally obviate further imaging. The most practical clinical application of his-togram analysis appears to be as an adjunct to unenhanced CT, where it can improve the sensitivity to almost 90% while maintaining high specificity for adenomas. Nevertheless, this methodology is scanner- and technique-dependent, and most departments do not yet routinely use this analysis method [31].

CT WashoutSeveral authors [23, 26, 35] reported that

although the CT densitometry method was unable to characterize adrenal lesions during the dynamic phase of a contrast-enhanced CT examination, it could be used on de-layed (15 minutes to 1 hour) images. It was

Fig. 2—58 -year-old woman with history of adrenal mass. CT image shows 3.4-cm partially calcified right adrenal mass containing small focus of macroscopic fat (measuring −40 HU), consistent with myelolipoma.

AJR:194, June 2010 1453

Adrenal Imaging

A

C

E

Fig. 3—43-year-old man with some abdominal discomfort.A, CT image shows 12.3-cm well-defined heterogeneous mass arising from left adrenal that contains several areas of macroscopic fat.B–D, Fat manifests as T2 and T1 hyperintensity similar to extraadrenal fat on T2-weighted (B) and T1-weighted in-phase (C) images and shows loss of signal on fat-saturation (C) and out-of-phase (D) images. Mass contains areas of mild internal enhancement. Findings are consistent with large left adrenal myelolipoma.E, IV gadolinium-enhanced T1-weighted image shows that mass contains areas of mild internal enhancement.

B

D

1454 AJR:194, June 2010

Blake et al.

found that adenomas enhance rapidly and also show a rapid loss of contrast medium—a phenomenon termed “contrast washout.” Malignant lesions also enhance rapidly but usually show a slower washout of contrast medium due to leaky capillaries.

It was further discovered that the ratio of attenuation values on the washout-delayed scan when compared with the initial dynam-ic contrast-enhanced study could help accu-rately characterize adrenal lesions [23]. Al-though these findings were first reported on MRI, the technique was considered insuffi-ciently reliable to be used in clinical prac-tice [36]. Several studies, however, then de-scribed a more clinically robust CT method using percentage washout threshold values (Table 1) with high accuracies for the diag-nosis of both lipid-rich and lipid-poor adre-nal adenomas [2, 19, 26, 37, 38].

If a 15-minute delay after contrast admin-istration protocol is used, an absolute per-centage washout of 60% or higher has re-ported sensitivity of 86–88% and specificity of 92–96% for the diagnosis of an adenoma [19, 22]. In similar fashion, if a 10-minute delayed protocol is used, sensitivity of 100% and specificity of 98% have been obtained for a threshold absolute percentage washout val-ue of 52% [38] (Fig. 4). If a 10-minute delay protocol is used, a relative percentage wash-out of 38–40% or higher has had reported sensitivity and specificity of 98% and 100% for the detection of adenoma [38]. After 15 minutes, if a relative percentage washout of 40% or higher is achieved, sensitivity is 96% and specificity is 100% for the diagno-sis of an adenoma [19]. Blake et al. [38] also highlighted that an unenhanced CT attenua-tion value of 0 HU or lower should supersede the contrast washout characteristics and that noncalcified, nonhemorrhagic adrenal le-sions with a native density of 43 HU or more should be considered indeterminate and sus-picious for malignancy irrespective of their contrast washout characteristics [38].

Pheochromocytomas, although rare, may mimic both adenomas and malignant masses on both CT densitometry and washout [38], and some authors strongly advise consider-

ing a pheochromocytoma if the contrast-en-hanced CT value is very high, for example, more than 150 HU [26]. However, in general, a combination of unenhanced CT, contrast enhancement, and washout characteristics correctly discriminates nearly all adrenal ad-enomas from malignant lesions.

Adrenal PET/CTPET and PET/CT also have been shown to

be valuable in the differentiation of adrenal masses [39]. Adrenal glands are larger than the spatial resolution of PET (about 5 mm), and, although rich in vascularity and metabolical-ly active, they are usually not visible on PET alone. Combined PET/CT can assign adre-nal tracer uptake to the location of the adrenal gland when present. Bagheri et al. [40] found only 5% (2/40) of normal adrenal glands could be visualized with PET alone, whereas 68% (27/40) were identified with integrated PET/CT. With regard to standardized uptake value (SUV), the average maximum and mean SUVs are recorded in Table 2 [40]. Adrenal FDG up-take is considered to be of malignant origin when intensity is higher than hepatic uptake.

Blake et al. [41] found that of 32 benign ad-renal masses, 30 showed 18F-FDG activity on visual analysis of PET/CT that was less than that of the liver (specificity, 94%). However, the maximum SUV of normal adrenal glands ranged from 0.95 to 2.46, and given that nor-mal liver tissue has an average mean SUV be-tween 1.5 and 2.0, physiologic adrenal uptake might in some cases be in the range of malig-nant lesions. Caoili et al. [42] found signifi-cant differences only in SUV ratios (adrenal/liver) but not in absolute SUVs in differentiat-ing adrenal adenomas and metastases.

Metser et al. [43] found using a maximum SUV of 3.1 for differentiating malignant from benign adrenal lesions had sensitivity and specificity of 98.5% and 92% for this

semiquantitative threshold. When they com-bined the SUV threshold with attenuation analysis from unenhanced CT (< 10 HU pos-itive for adenoma), they found sensitivity of 100% and specificity of 98%. Furthermore, they found that FDG uptake was not signifi-cantly different between lipid-rich and lipid-poor adenomas—sensitivity and specificity were 98.5% and 92% for differentiating lip-id-poor adenomas and 98.5% and 93% for all adenomas. Boland et al. [39] found that PET/CT had sensitivity, specificity, positive pre-dictive value, negative predictive value, and accuracy of 99%, 100%, 100%, 93%, and 99% for the detection of benign lesions and 100%, 99%, 93%, and 100%, respectively, for the detection of malignancy.

In the study of Metser et al. [43], PET alone (maximum SUV, 3.1) achieved sensi-tivity, specificity, and accuracy of 99% (67 of 68 nodules), 92% (98 of 107), and 94% (165 of 175), respectively, compared with in-tegrated PET/CT with values of 100% (68 of 68 nodules), 98% (105 of 107), and 99% (173 of 175), respectively. Also described were added advantages of small adrenal lesions detected on CT, which were correctly diag-nosed on the PET component by evaluating tracer uptake [43].

Five percent of adrenal abnormalities in-terpreted as positive at PET are false-posi-tive for malignancy. This is secondary to inflammatory lesions (sarcoidosis, tubercu-losis [44]), adrenal endothelial cysts, some adrenal adenomas, and periadrenal abnor-mality [45], and adrenal cortical hyperplasia may all mimic metastases because they have been associated with increased FDG uptake. Metastasis detection at PET/CT, however, depends on the primary tumor, metastasis size, and differentiation. PET scanners have a spatial resolution of about 5 mm. Hemor-rhage and necrosis are also known to cause

TABLE 2: Average Maximum and Mean Standardized Uptake Value (SUV) [40]

Adrenal Gland Average Maximum SUV Average Mean SUV

Right 0.90–0.15 0.83–0.17

Left 1.10–0.15 0.946–0.15

Note—Data are range of indicated value.

TABLE 1: Absolute Percentage Washout and Relative Percentage Washout Formulas

Parameter Formula

Absolute percentage washout 100 x (contrast-enhanced CT HU − delayed CT HU) / (CT HU − unenhanced CT HU)

Relative percentage washout 100 x (contrast-enhanced CT HU − delayed CT HU) / (CT HU)

Note—If unenhanced CT has been obtained, absolute percentage washout can be calculated by the formula shown. If no unenhanced CT is available, then relative percentage washout can be calculated by the formula shown. Contrast-enhanced scan performed at 60–75 seconds.

AJR:194, June 2010 1455

Adrenal Imaging

false-negative results in FDG PET [42, 46, 47]. Metastases from primary carcinomas that are non-FDG avid have been found to be false-negative on PET, including carci-noid (neuroendocrine tumors) and pulmo-

nary carcinoma of the bronchioloalveolar type [46–48]. FDG PET cannot differenti-ate among malignant lesions, for example, among metastases, ACC, malignant pheo-chromocytoma, and lymphoma.

ACC, Pheochromocytoma, and Lymphoma on Adrenal PET/CTACC

Becherer et al. [49] showed sensitivity and specificity of 100% and 95%, respectively,

A

C

E

Fig. 4—45-year-old woman with incidental adrenal mass.A–F, Right adrenal gland shows 1.1-cm lesion measuring 8 HU on unenhanced (A and B), 40 HU on dynamic phase (C and D), and 18 HU on 10-minute delayed phase images (E and F). These density and washout properties are consistent with adrenal adenoma.

B

D

F

1456 AJR:194, June 2010

Blake et al.

(n = 10) for ACC with FDG PET. They also found that PET detected additional lesions in 30% of cases. Leboulleux et al. [50] report-ed sensitivities of 90% and 93% for the de-tection of distinct lesions and the diagnosis of metastatic organs, respectively. With the re-cent use of 11C-metomidate (METO), a mark-er of 11beta-hydroxylase, Zettinig et al. [51] found that METO PET had sensitivity and specificity of 89% and 96% for the identifi-cation of tumors of adrenocortical origin (n = 73). METO can distinguish adrenal metasta-ses and pheochromocytomas from adrenocor-tical tumors (adenomas and ACC); however, it cannot differentiate adenomas from ACC. FETO, (R)-1-(1-phenylethyl)-1H-imidazole-5-carboxylic acid 2-[18F] fluoroethylester, has a longer half-life, allows longer imaging pro-tocols, and is currently under investigation. However, presently FDG PET/CT remains the functional study of choice because it also may show additional sites of metastatic disease.

PheochromocytomaThe imaging technique of choice in local-

izing pheochromocytomas has been metaiodo-benzylguanidine (MIBG), which has high sen-sitivity (95–100%) and specificity (100%) [52]. Pheochromocytoma, whether benign or malig-nant, might accumulate FDG. It was reported, however, that among malignant pheochromo-cytomas, the percentage with increased up-take is higher than among benign pheochro-mocytomas [46]. Shulkin et al. [52] compared FDG PET with MIBG and found sensitivity of 83% for the detection of benign pheochromo-

cytomas, whereas FDG PET had sensitivity of 58%. For malignant pheochromocytomas, MIBG had sensitivity of 88%, whereas FDG PET had sensitivity of 82%. Although MIBG had better sensitivity, all of the MIBG-nega-tive lesions showed avid FDG uptake. These authors concluded that most pheochromocy-tomas accumulate FDG, uptake is found in a greater percentage of malignant than benign pheochromocytomas, and FDG PET is es-pecially useful in defining the distribution of those pheochromocytomas that fail to concen-trate MIBG. Newer PET-specific tracers for the sympathetic system, 11C-hydroxyephed-rine (HED) and 18F-dihydroxyphenylalanine (F-DOPA), are under investigation and have shown promising results [1].

LymphomaPET/CT has been found to be valuable in

distinguishing nonfunctioning adrenal neo-plasm or hyperplasia from lymphomatous in-volvement (Fig. 5). The degree of FDG avidi-ty in adrenal lymphoma tends to parallel that in other involved areas so that the resolution of adrenal gland uptake after treatment often follows that of uptake in other regions [43]. It should be remembered, however, that certain subtypes of lymphoma (for example, margin-al zone and peripheral T-cell) and low-grade lymphomas may not show reliable FDG tu-mor uptake on the PET component.

MRI and Developing MR TechniquesMRI, with its inherent tissue character-

izing ability, offers utility in the assessment

of adrenal disorders. The normal adrenal is of low to intermediate signal on T1- and T2-weighted imaging. Adrenal adenomas usu-ally show relatively uniform enhancement on immediate gadolinium-enhanced images [53]. Small, rounded foci of altered signal in-tensity may be seen within an adenoma ow-ing to cystic changes, hemorrhage, or varia-tion in vascularity. Chemical shift imaging (CSI) is the mainstay of MR evaluation of solid adrenal lesions.

Visual analysis of CSI compared with splenic intensity is the most common mode of adrenal lesion analysis and has been re-ported to be as effective as quantitative as-sessments (defined at the end of this sec-tion) [54]. MRI identifies intracellular lipid because of the different resonant frequen-cies of fat and water protons in a given voxel [15, 54–59]. Fat protons precess at a lower frequency than water protons, and they thus cancel each other out during out-of-phase breath-hold gradient-echo MRI [59–62]. This phenomenon results in loss of signal in-tensity on out-of-phase imaging when com-pared with in-phase images (Fig. 6.).

Korobkin et al. [18, 58] found an inverse linear relationship between the percentage of lipid-rich cells and the relative change in MR signal intensity on CSI. When there are almost equal concentrations of fat and water protons in most voxels, as is seen with lipid-rich ade-nomas, there will be almost complete signal in-tensity loss on out-of-phase images. Converse-ly, with a low lipid-to-water proton ratio, as may be seen in lipid-poor adrenal adenomas,

A

Fig. 5—73-year-old man with extranodal marginal zone–type B-cell lymphoma now with transformation to diffuse large B cell lymphoma of adrenals.A and B, There is nodular thickening measuring approximately 2.3 cm in thickness involving entire left adrenal gland with some preservation of its shape on CT image (A) with corresponding abnormal 18F-FDG uptake on overlaid PET/CT image (B).

B

AJR:194, June 2010 1457

Adrenal Imaging

the signal intensity is essentially unchanged on out-of-phase images; such lesions remain inde-terminate by CSI methods. Important caveats to be aware of, however, are ACC, pheochro-mocytoma, and clear cell renal cell cancer me-tastasis, all of which may sometimes show sig-nal loss on out-of-phase images.

The sensitivity and specificity of CSI for the differentiation of incidental adrenal lesion are similar to those of unenhanced CT densitom-etry, at 81–100% and 94–100%, respectively [57, 59, 60]. Studies have shown that for lip-id-rich adenomas, there is effectively no dif-ference between CT and MRI, but CSI might be superior when evaluating lipid-poor ade-nomas [60]. However, one study [61] showed that CSI might be useful only when the unen-hanced CT attenuation is less than 30 HU.

In terms of quantitative analysis, calcu-lation of the adrenal-to-spleen CSI ratio or the signal intensity index allows quantifica-tion of the chemical phenomenon (Table 3). A CSI ratio of less than 0.71 or a signal inten-sity index of more than 16.5% both indicate a lipid-rich adenoma [62].

Diffusion MRI for Adrenal LesionsPublished reports of diffusion analysis

have so far been disappointing for adrenal assessment. Tsushima et al. [63] in a review of 43 adrenal tumors found no difference in apparent diffusion coefficient (ADC) values between adenomas and metastatic tumors. Miller et al. [64] found similar results of ADC values not being useful in distinguish-ing benign from malignant adrenal lesions (n = 160 lesions).

MR Adrenal SpectroscopyFaria et al. [65] showed that using thresh-

old values of 1.20 for the choline–creatine

ratio (92% sensitivity, 96% specificity; p < 0.01), 0.38 for the choline–lipid ratio (92% sensitivity, 90% specificity; p < 0.01), and 2.10 for the lipid–creatine ratio (45% sensi-tivity, 100% specificity) enabled adenomas and pheochromocytomas to be distinguished from carcinomas and metastases. A 4.0–4.3 ppm/creatine ratio greater than 1.50 enabled distinction of pheochromocytomas and car-cinomas from adenomas and metastases (87% sensitivity, 98% specificity; p < 0.01). The best distinction was obtained by com-paring choline–creatine and 4.0–4.3 ppm/creatine ratios. Further larger studies will be needed to validate these promising spectro-scopic findings.

ACC, Pheochromocytoma, and Lymphoma on Adrenal MRI

ACCs are seen as heterogeneous on both T1- and T2-weighted imaging because of hemorrhage and necrosis [66]. Necrotic ar-eas can have high signal intensity on T2-weighted imaging, and blood products can result in areas of high signal intensity within the lesion on T1-weighted imaging. ACC can contain foci of intracytoplasmic lipid, which results in a loss of signal intensity on out-of-phase images—a feature that could cause them to be erroneously diagnosed as benign adenomas (even at pathology [67]), although their usually large size and associated clin-ical findings alert the interpreting radiolo-gist to this pitfall. The diagnosis of ACC

also may be suggested on MRI or CT by its known propensity to spread via venous tu-mor thrombus.

Lymphomatous lesions show washout characteristics similar to those of other pri-mary or secondary adrenal malignancies. Lymphoma shows heterogeneous low sig-nal intensity on T1-weighted images and heterogeneous high signal intensity on T2-weighted images [68], with progressive en-hancement after administration of contrast material. Lymphoma can sometimes main-tain the adreniform shape of the gland and so in its early stage could mimic adrenal hy-pertrophy or hyperplasia in contour.

Up to 70% of pheochromocytomas show relatively high signal intensity on T2-weight-ed images—a feature classically known as the light bulb sign, which was originally thought to be characteristic of pheochromocytoma [56, 69]. Currently, however, that description is not considered accurate, with at least 30% of pheochromocytomas showing moderate or low T2-weighted signal intensity and appear-ing similar to other adrenal diseases [15, 70]. Pheochromocytomas do not usually have sig-nificant cytoplasmic lipid and generally main-tain their signal intensity on out-of-phase im-ages, although exceptions are possible because fatty degeneration is sometimes known to oc-cur [70–72]. Most lesions exhibit intense en-hancement after contrast injection [73] (Fig. 7). At MR spectroscopy, pheochromocytomas were found to have a unique MR spectral sig-

TABLE 3: Quantitative MRI Adrenal Chemical Shift Imaging Analysis

Parameter Formula

Adrenal-to-spleen CSI ratio Lesion-to spleen SIIP / lesion-to-spleen SIOP

Adrenal signal intensity index 100 x [(SIIP − SIOP) / SIIP]

Note—SIIP and SIOP are the signal intensities measured on in-phase and out-of-phase images, respectively.

A

Fig. 6—53-year-old man with Cushing syndrome.A and B, MRI shows left adrenal nodule on in-phase image (A), with marked loss of signal relative to splenic signal on out-of-phase image (B) consistent with lipid-rich adenoma.

B

1458 AJR:194, June 2010

Blake et al.

A

C

Fig. 7—63-year-old woman with neurofibromatosis type 1 and history of resected left adrenal pheochromocytoma. Patient underwent left adrenalectomy.A–D, On T1-weighted images (A and B), 18-mm right adrenal nodule is heterogeneous in signal; however, on T2-weighted image (C), lesion is predominantly T2 hyperintense. This lesion shows no signal decrease on out-of phase image (B) and heterogeneous marked enhancement (D), features concerning for pheochromocytoma.

B

D

nature, showing 6.8-ppm resonance that is not seen in adenomas and is attributed to the pres-ence of catecholamines and catecholamine metabolites [74].

ConclusionThe imaging characterization of adrenal

lesions has continued to advance over the past decade as new technologies have contin-ued to evolve. CT, MRI, PET, and PET/CT are now established clinical techniques capa-ble of differentiating benign from malignant adrenal lesions. Developments in adrenal im-aging, such as spectroscopy, also have shown promise as potentially useful adrenal appli-cations. It is important to be knowledgeable about the imaging appearances of pathology that can affect the adrenal gland. The dimin-utive size of the adrenal gland belies its piv-otal importance in medicine, and imaging

now plays a critical role in adrenal pathology detection and characterization.

References 1. Boland GW, Blake MA, Hahn PF, Mayo-Smith

WW. Incidental adrenal lesions: principles, tech-

niques, and algorithms for imaging characteriza-

tion. Radiology 2008; 249:756–775

2. Dunnick NR, Korobkin M. Imaging of adrenal

incidentalomas: current status. AJR 2002;

179:559–568

3. Lam KY, Lo CY. Metastatic tumours of the adre-

nal glands: a 30-year experience in a teaching

hospital. Clin Endocrinol (Oxf) 2002; 56:95–101

4. Mitchell IC, Nwariaku FE. Adrenal masses in the

cancer patient: surveillance or excision. Oncolo-

gist 2007; 12:168–174

5. Bovio S, Cataldi A, Reimondo G, et al. Prevalence

of adrenal incidentaloma in a contemporary com-

puterized tomography series. J Endocrinol Invest

2006; 29:298–302

6. Song JH, Chaudhry FS, Mayo-Smith WW. The

incidental adrenal mass on CT: prevalence of ad-

renal disease in 1,049 consecutive adrenal masses

in patients with no known malignancy. AJR 2008;

190:1163–1168

7. Oliver TW Jr, Bernardino ME, Miller JI, Mansour

K, Greene D, Davis WA. Isolated adrenal masses

in nonsmall-cell bronchogenic carcinoma. Radi-

ology 1984; 153:217–218

8. Libe R, Bertherat J. Molecular genetics of adreno-

cortical tumours: from familial to sporadic dis-

eases. Eur J Endocrinol 2005; 153:477–487

9. Frilling A, Tecklenborg K, Weber F, et al. Impor-

tance of adrenal incidentaloma in patients with a his-

tory of malignancy. Surgery 2004; 136:1289–1296

10. Kocijancic I, Vidmar K, Zwitter M, Snoj M. The

significance of adrenal metastases from lung car-

cinoma. Eur J Surg Oncol 2003; 29:87–88

11. Szolar DH, Melvyn Korobkin M, Reittner P, et al.

AJR:194, June 2010 1459

Adrenal Imaging

Adrenocortical carcinomas and adrenal pheo-

chromocytomas: mass and enhancement loss

evaluation at delayed contrast-enhanced CT. Ra-

diology 2005; 234:479–485

12. Young WF Jr. The incidentally discovered adre-

nal mass. N Engl J Med 2007; 356:601–610

13. Russell C, Goodacre BW, vanSonnenberg E, Ori-

huela E. Spontaneous rupture of adrenal myeloli-

poma: spiral CT appearance. Abdom Imaging

2000; 25:431–434

14. Benitah N, Yeh BM, Qayyum A. Minor morpho-

logic abnormalities of adrenal glands at CT: prog-

nostic importance in patients with lung cancer.

Radiology 2005; 235:517–522

15. Mayo-Smith WW, Boland GW, Noto RB, et al.

State-of-the-art adrenal imaging. RadioGraphics

2001; 21:995–1012

16. Rozenblit A, Morehouse HT, Amis SE Jr. Cystic

adrenal lesions: CT features. Radiology 1996;

201:541–548

17. Lee MJ, Hahn PF, Papanicolau N, et al. Benign

and malignant adrenal masses: CT distinction

with attenuation coefficients, size, and observer

analysis. Radiology 1991; 179:415–418

18. Korobkin M, Giordano TJ, Brodeur FJ, et al. Ad-

renal adenomas: relationship between histologic

lipid and CT and MR findings. Radiology 1996;

200:743–747

19. Caoili EM, Korobkin M, Francis IR, et al. Adre-

nal masses: characterization with combined un-

enhanced and delayed enhanced CT. Radiology

2002; 222:629–633

20. Boland GW, Lee MJ, Gazelle GS, et al. Charac-

terization of adrenal masses using unenhanced

CT: an analysis of the CT literature. AJR 1998;

171:201–204

21. Korobkin M, Brodeur FJ, Yutzy GG, et al. Dif-

ferentiation of adrenal adenomas from nonade-

nomas using CT attenuation values. AJR 1996;

166:531–536

22. Korobkin M, Brodeur FJ, Francis IR, Quint LE,

Dunnick NR, Londy F. CT time–attenuation

washout curves of adrenal adenomas and nonade-

nomas. AJR 1998; 170:747–752

23. Szolar DH, Kammerhuber FH. Adrenal adenomas

and nonadenomas: assessment of washout at de-

layed contrast-enhanced CT. Radiology 1998;

207:369–375

24. Pena CS, Boland GW, Hahn PF, Lee MJ, Mueller

PR. Characterization of indeterminate (lipid-

poor) adrenal masses: use of washout characteris-

tics at contrast-enhanced CT. Radiology 2000;

217:798–802

25. Johnson PT, Horton KM, Fishman EK. Adrenal

imaging with MDCT: nonneoplastic disease. AJR

2009; 193:1128–1135

26. Johnson PT, Horton KM, Fishman EK. Adrenal

mass imaging with multidetector CT: pathologic

conditions, pearls, and pitfalls. RadioGraphics

2009; 29:1333–1351

27. Hahn PF, Blake MA. Boland GW. Adrenal lesions:

attenuation measurement differences between CT

scanners. Radiology 2006; 240:458–463

28. Stadler A, Schima W, Prager G, et al. CT density

measurements for characterization of adrenal tu-

mors ex vivo: variability among three CT scan-

ners. AJR 2004; 182:671–675

29. Johnson PT, Horton KM, Fishman EK. Adrenal

imaging with multidetector CT: evidence-based

protocol optimization and interpretative practice.

RadioGraphics 2009; 29:1319–1331

30. Bae KT, Fuangtharnthip P, Prasad SR, Joe BN,

Heiken JP. Adrenal masses: CT characterization

with histogram analysis method. Radiology 2003;

228:735–742

31. Remer EM, Motta-Ramirez GA, Shepardson LB,

Hamrahian AH, Herts BR. CT histogram analysis

in pathologically proven adrenal masses. AJR

2006; 187:191–196

32. Jhaveri KS, Wong F, Ghai S, Haider MA. Com-

parison of CT histogram analysis and chemical

shift MRI in the characterization of indeterminate

adrenal nodules. AJR 2006; 187:1303–1308

33. Ho LM, Paulson EK, Brady MJ, Wong TZ, Schin-

dera ST. Lipid-poor adenomas on unenhanced

CT: does histogram analysis increase sensitivity

compared with a mean attenuation threshold?

AJR 2008; 191:234–238

34. Halefoglu AM, Bas N, Yasar A, Basak M. Differ-

entiation of adrenal adenomas from nonadenomas

using CT histogram analysis method: a prospec-

tive study. Eur J Radiol 2010; 73:643–651

35. Boland GW, Hahn PF, Pena C, Mueller PR. Adre-

nal masses: characterization with delayed contrast-

enhanced CT. Radiology 1997; 202:693–696

36. Krestin GP, Steinbrich W, Friedmann G. Adrenal

masses: evaluation with fast gradient-echo MR

imaging and Gd-DTPA-enhanced dynamic stud-

ies. Radiology 1989; 171:675–680

37. Park BK, Kim CK, Kim B, Lee JH. Comparison

of delayed enhanced CT and chemical shift MR

for evaluating hyperattenuating incidental adrenal

masses. Radiology 2007; 243:760–765

38. Blake MA, Kalra MK, Sweeney AT, et al. Distin-

guishing benign from malignant adrenal masses:

multi-detector row CT protocol with 10-minute

delay. Radiology 2005; 238:578–585

39. Boland GW, Blake MA, Holalkere NS, Hahn PF.

PET/CT for the characterization of adrenal mass-

es in patients with cancer: qualitative versus quan-

titative accuracy in 150 consecutive patients. AJR

2009; 192:956–962

40. Bagheri B, Maurer AH, Cone L, Doss M, Adler L.

Characterization of the normal adrenal gland with 18F-FDG PET/CT. J Nucl Med 2004; 45:1340–1343

41. Blake MA, Slattery JM, Kalra MK, et al. Adrenal

lesions: characterization with fused PET/CT im-

age in patients with proved or suspected malig-

nancy—initial experience. Radiology 2006;

238:970–977

42. Caoili EM, Korobkin M, Brown RK, Mackie G,

Shulkin BL. Differentiating adrenal adenomas

from nonadenomas using (18)F-FDG PET/CT:

quantitative and qualitative evaluation. Acad Ra-

diol 2007; 14:468–475

43. Metser U, Miller E, Lerman H, Lievshitz G, Avi-

tal S, Even-Sapir E. 18F-FDG PET/CT in the eval-

uation of adrenal masses. J Nucl Med 2006;

47:32–37

44. Bakheet SM, Powe J, Ezzat A, Rostom A. F-18-

FDG uptake in tuberculosis. Clin Nucl Med 1998;

23:739–742

45. Boerner AR, Voth E, Theissen P, Wienhard K,

Wagner R, Schicha H. Glucose metabolism of the

thyroid in Graves’ disease measured by F-18-fluo-

ro-deoxyglucose positron emission tomography.

Thyroid 1998; 8:765–772

46. Yun M, Kim W, Alnafisi N, Lacorte L, Jang S,

Alavi A. 18F-FDG PET in characterizing adrenal

lesions detected on CT or MRI. J Nucl Med 2001;

42:1795–1799

47. Jana S, Zhang T, Milstein DM, Isasi CR, Blaufox

MD. FDG-PET and CT characterization of adre-

nal lesions in cancer patients. Eur J Nucl Med Mol

Imaging 2006; 33:29–35

48. Erasmus JJ, Patz EF Jr, McAdams HP, et al. Eval-

uation of adrenal masses in patients with broncho-

genic carcinoma using 18F-fluorodeoxyglucose

positron emission tomography. AJR 1997;

168:1357–1360

49. Becherer A, Vierhapper H, Pötzi C, et al. FDG-

PET in adrenocortical carcinoma. Cancer Biother

Radiopharm 2001; 16:289–295

50. Leboulleux S, Dromain C, Bonniaud G, et al. Di-

agnostic and prognostic value of 18-fluorodeoxyg-

lucose positron emission tomography in adreno-

cortical carcinoma: a prospective comparison

with computed tomography. J Clin Endocrinol

Metab 2006; 91:920–925

51. Zettinig G, Mitterhauser M, Wadsak W, et al.

Positron emission tomography imaging of adrenal

masses: (18)F-fluorodeoxyglucose and the 11beta-

hydroxylase tracer (11)C-metomidate. Eur J Nucl

Med Mol Imaging 2004; 31:1224–1230

52. Shulkin BL, Thompson NW, Shapiro B, Francis

IR, Sisson JC. Pheochromocytomas: imaging

with 2-[fluorine-18] fluoro-2-deoxy-D-glucose

PET. Radiology 1999; 212:35–41

53. Semelka RC, Shoenut JP, Lawrence PH, et al.

Evaluation of adrenal masses with gadolinium en-

hancement and fat-suppressed MR imaging. J

Magn Reson Imaging 1993; 3:337–343

54. Mayo-Smith WW, Lee MJ, McNicholas MM,

Hahn PF, Boland GW, Saini S. Characterization

1460 AJR:194, June 2010

Blake et al.

of adrenal masses (15 cm) by use of chemical shift

MR imaging: observer performance versus quan-

titative measures. AJR 1995; 165:91–95

55. Mitchell DG, Crovello M, Matteucci T, Peterson

RO, Miettinen MM. Benign adrenocortical mass-

es: diagnosis with chemical shift MR imaging.

Radiology 1992; 185:345–351

56. Reinig JW, Stutley JE, Leonhurdt CM, Spicer

KM, Margolis M, Caldwell CB. Differentiation of

adrenal masses with MR imaging: comparison of

techniques. Radiology 1994; 192:41–46

57. Outwater EK, Siegelman ES, Radecki PD, Piccoli

CW, Mitchell DG. Distinction between benign

and malignant adrenal masses: value of T1-

weighted chemical shift MR imaging. AJR 1995;

165:579–583

58. Korobkin M, Lombardi TJ, Aisen AM, et al.

Characterization of adrenal masses with chemical

shift and gadolinium-enhanced MR imaging. Ra-

diology 1995; 197:411–418

59. Israel GM, Korobkin M, Wang C, Hecht EN,

Krinsky GA. Comparison of unenhanced CT and

chemical shift MRI in evaluating lipid-rich adre-

nal adenomas. AJR 2004; 183:215–219

60. Haider MA, Ghai S, Jhaveri K, Lockwood G.

Chemical shift MR imaging of hyperattenuating

(< 10 HU) adrenal masses: does it still have a role?

Radiology 2004; 231:711–716

61. Outwater EK, Siegelman ES, Huang AB, Birn-

baum BA. Adrenal masses: correlation between

CT attenuation value and chemical shift ratio at

MR imaging with in-phase and opposed-phase

sequences. Radiology 1996; 200:749–752

62. Fujiyoshi F, Nakajo M, Kukukura Y, Tsuchimochi

S. Characterization of adrenal tumors by chemi-

cal shift fast-low angle shot MR imaging: com-

parison of four methods of quantitative evalua-

tion. AJR 2003; 180:1649–1657

63. Tsushima Y, Takahashi-Taketomi A, Endo K. Di-

agnostic utility of diffusion-weighted MR imag-

ing and apparent diffusion coefficient value for

the diagnosis of adrenal tumors. J Magn Reson

Imaging 2009; 29:112–117

64. Miller FH, Wang Y, McCarthy RJ, et al. Utility of

diffusion-weighted MRI in characterization of adre-

nal lesions. AJR 2010; 194:459; [web]W179–W185

65. Faria JF, Goldman SM, Szejnfeld J, et al. Adrenal

masses: characterization with in vivo proton MR

spectroscopy–initial experience. Radiology 2007;

245:788–797

66. Schlund JF, Kenney PJ, Brown ED, et al. Adreno-

cortical carcinoma: MR imaging appearance with

current techniques. J Magn Reson Imaging 1995;

5:171–174

67. Klibanski A, Stephen AE, Greene MF, Blake

MA, Wu CL, Case records of the Massachusetts

General Hospital: case 36–2006—a 35-year-old

pregnant woman with new hypertension. N Engl J

Med 2006; 355:2237–2245

68. Lee FT Jr, Thornbury JR, Grist TM, et al. MR

imaging of adrenal lymphoma. Abdom Imaging

1993; 18:95–96

69. Miyajima A, Nakashima J, Baba S, Tachibana M,

Nakamura K, Murai M. Clinical experience with

incidentally discovered pheochromocytoma. J

Urol 1997; 157:1566–1568

70. Blake MA, Kalra MK, Maher MM, et al. Pheo-

chromocytoma: an imaging chameleon. Radio-

Graphics 2004; 24[suppl 1]:S87–S89

71. Ramsay JA, Asa SL, van Nostrand AW, Hassaram

ST, de Harven EP. Lipid degeneration in pheo-

chromocytomas mimicking adrenal cortical tu-

mors. Am J Surg Pathol 1987; 11:480–486

72. Blake MA. Krishnamoorthy SK, Boland GW, et

al. Low-density pheochromocytoma on CT: a

mimicker of adrenal adenoma. AJR 2003;

181:1663–1668

73. Remer EM, Miller FH. Imaging of pheochromocy-

toma. In: Blake MA, Boland G, eds. Adrenal imag-

ing. New York, NY: Humana Press, 2009:109–124

74. Kim S, Salibi N, Hardie AD, et al. Characteriza-

tion of adrenal pheochromocytoma using respira-

tory-triggered proton MR spectroscopy: initial

experience. AJR 2009; 192:450–454