Вы находитесь на странице: 1из 9

Genitourinar y Imaging • Original Research

Friedrich-Rust et al.
Sonography of Adrenal Masses

Genitourinary Imaging
Original Research

Contrast-Enhanced
Sonography of Adrenal Masses:
Differentiation of Adenomas
and Nonadenomatous Lesions
Mireen Friedrich-Rust 1 OBJECTIVE. The aim of this proof-of-principle study was to evaluate contrast-enhanced
Guenther Schneider 2 sonography in the characterization of adrenal masses.
Rainer M. Bohle 3 Subjects AND METHODS. Thirty-five consecutively registered patients with new-
Eva Herrmann 4 ly detected adrenal masses underwent hormonal evaluation and duplex and Doppler sonogra-
Christoph Sarrazin1 phy followed by contrast-enhanced sonography and CT or MRI. The dynamics of contrast
enhancement were analyzed with time–intensity curves. CT and MRI were used as the refer-
Stefan Zeuzem1
ence methods for the diagnosis of adenoma and myelolipoma. Metastasis was diagnosed with
Joerg Bojunga1 fine-needle biopsy, and all other adrenal masses were diagnosed at adrenalectomy. Fisher’s
Friedrich-Rust M, Schneider G, Bohle RM, et al. exact test was used to evaluate the criteria for diagnosis of malignant adrenal masses.
RESULTS. Size greater than 4 cm and hypervascularization were found significantly
more often in malignant than in benign lesions (71% vs 21% for size; 57% vs 7% for hyper-
vascularization). At contrast-enhanced sonography, early arterial or arterial contrast en-
hancement and rapid washout were seen in all patients with primary or secondary malignant
lesions of the adrenal gland and in only 22% of patients with benign adrenal masses (p < 0.05).
All primary malignant lesions were confirmed at histologic examination. In 32 of 35 patients
(91%), findings at CT or MRI were congruent with those at contrast-enhanced sonography in
regard to characterization of adenoma versus nonadenomatous lesion (p < 0.001). In two of
the 35 cases, however, all imaging methods favored the diagnosis of nonadenomatous lesion,
but the histologic result after adrenalectomy was adrenal adenoma. The sensitivity and speci-
ficity of contrast-enhanced sonography in the diagnosis of malignant adrenal mass were
100% and 82%.
Keywords: adenoma, adrenal gland, contrast-enhanced CONCLUSION. Contrast-enhanced sonography can be used to differentiate adenomas
sonography, incidentaloma, nonadenomatous, sonography and nonadenomatous lesions with a sensitivity comparable with that of CT and MRI and may
be a cost-effective method for preselection of patients with adrenal masses.
DOI:10.2214/AJR.07.3565

S
Received December 18, 2007; accepted after revision onography can be performed with drawn solely from the size of the adrenal
June 28, 2008. high sensitivity and specificity for mass. Therefore, differentiation of benign and
1 the detection of adrenal masses malignant adrenal masses with sonography
Medizinische Klinik I, Klinikum der Johann Wolfgang
Goethe Universität, Theodor-Stern-Kai 7, 60590 [1–5]. One study [1] showed that has not been possible.
Frankfurt am Main, Germany. Address correspondence even in healthy subjects, the right and left ad- In only one pilot study [6] to our knowledge
to J. Bojunga (Joerg.Bojunga@kgu.de). renal glands can be visualized with transab- has contrast-enhanced Doppler sonography of
2
dominal B-mode sonography in 99% (79 of adrenal masses with the contrast agent SH U
Department of Radiology, Saarland University
Hospital, Homburg, Germany.
80) and 69% (55 of 80) of cases, respectively. 508A (Levovist, Bayer Schering Pharma) been
The overall sensitivity and specificity for the evaluated. The investigators found improved
3
Department of Pathology, Saarland University sonographic detection of adrenal masses inde- visualization of the vascularization of adrenal
Hospital, Homburg, Germany. pendent of size were 96% (48 of 50 cases) and masses during power Doppler sonography.
4 92% (92 of 100 cases). The size of the adrenal However, differentiation of benign and malig-
Faculty of Medicine, Internal Medicine–
Biomathematics, Saarland University, Homburg, mass did not have a significant effect on sensi- nant adrenal masses was not possible in that
Germany. tivity and specificity [5]. Similar results were study. To our knowledge, no study has been
found in another study [4] with 119 patients. conducted to examine adrenal glands and adre-
AJR 2008; 191:1852–1860
Although a study [3] showed that benign adre- nal masses with a contrast agent consisting of
0361–803X/08/1916–1852 nal masses are on average smaller than malig- a suspension of phospholipid-stabilized micro­
nant adrenal masses, no conclusion concern- bubbles filled with sulfur hexafluoride (Sono-
© American Roentgen Ray Society ing the benignity of an adrenal mass was Vue, Bracco), which has the advantage that

1852 AJR:191, December 2008


Sonography of Adrenal Masses

contrast enhancement and washout can be fol- that benign and malignant adrenal masses can CT and MRI examinations are performed
lowed for up to 5 minutes. be differentiated at CT with a sensitivity of only to rule out malignancy. Therefore, pre-
With advances in diagnostic imaging, dif- 85–100% and a specificity of 100–95%. If an selection of patients with suspicious adrenal
ferentiation of malignant and benign adrenal adrenal mass remains indeterminate after CT, masses at screening sonography would be
masses has improved substantially. CT and MRI or adrenal biopsy is performed [13–15]. favorable. If this screening can be performed
MRI can be used to characterize benign adre- CT and MRI, however, are expensive ex- with high sensitivity, the number of patients
nal masses in most patients with adrenal aminations compared with sonography. CT who need additional diagnostic imaging with
masses. Whereas in the era before highly spe- is associated with substantial radiation expo- CT or MRI would be markedly reduced, and
cialized CT and MRI 40–57% of patients with sure, and CT and MRI are commonly per- malignant lesions would not be missed. We
adrenal adenomas underwent biopsy for diag- formed in addition to screening sonography. conducted a proof-of-principle study to eval-
nosis, only 12% of these patients undergo bi- Furthermore, adenomas are much more com- uate contrast-enhanced sonography in the
opsy at present [7]. Studies [8–12] have shown mon than nonadenomatous lesions, and many characterization of adrenal masses.

TABLE 1: Characteristics and Clinical Findings in 35 Patients with Adrenal Masses


Benign or Malignant
Diagnosis Correct
Length Pattern on
No. of of Contrast- Histologic Finding Contrast-
Adrenal Masses Enhanced Diagnosis at Congruence on Surgical Enhanced
Diagnosis Masses (mm) Sonography CT or MRI of Findingsa Specimen Sonography MRI or CT
Adenoma 8 16–43 4 Adenoma Congruent Yes Reference
method
Adenoma 11 18–104 3 Adenoma Congruent Yes Reference
method
Adrenal hyperplasia 1 27 3 Adrenal hyperplasia Congruent Yes Reference
method
Adenoma 1 35 2 Nonadenomatous Congruent Adenoma No No
(metastasis, neuroendocrine
carcinoma)
Adenoma 1 30 1 Nonadenomatous Congruent Adenoma No No
(pheochromocytoma)
Adenoma 1 22 3 Nonadenomatous Incongruent Adenoma Yes No
(metastasis)
Myelolipoma 1 55 3 Nonadenomatous Incongruent Myelolipoma Yes Yes
(myelolipoma)
Angiomyolipoma 1 80 2 Nonadenomatous Congruent No Reference
(angiomyolipoma) method
Hemangioma 1 114 1 Nonadenomatous Congruent Hemangioma No No
(adrenocortical carcinoma)
Pheochromocytoma 1 66 1 Nonadenomatous Congruent Pheochromocytoma No Yes
(pheochromocytoma)
Posttreatment 1 36 4 Nonadenomatous Incongruent Yes Reference
metastasis (posttreatment metastasis) method
Metastasis 2 19–78 1 Nonadenomatous Congruent Yes Reference
(metastasis) method
Metastasis 1 37 2 Nonadenomatous Congruent Yes Reference
(metastasis) method
Cortisol-producing 1 44 1 Nonadenomatous Congruent Cortisol-producing Yes Yes
adrenocortical pheochromocytoma) adrenocortical
carcinoma carcinoma
Adrenocortical 1 79 1 Nonadenomatous Congruent Adrenocortical Yes Yes
carcinoma (metastasis) carcinoma
Adrenocortical 1 190 1 Nonadenomatous (adrenal Congruent Adrenocortical Yes Yes
carcinoma carcinoma) carcinoma
T-cell lymphoma 1 93 1 Nonadenomatous Congruent T-cell lymphoma Yes Yes
aCongruence between contrast-enhanced sonographic and CT or MRI findings for adenoma and nonadenomatous lesions.

AJR:191, December 2008 1853


Friedrich-Rust et al.

Subjects and Methods seconds after injection), arterial phase (20–40 portal venous (50–60 seconds after injection), and
Patients seconds after injection), and parenchymal (late) equilibrium (> 5 minutes after injection) phases
Between April 2006 and June 2007, 35 con­ phase (> 40 seconds–5 minutes after injection). of contrast enhancement. The field of view for all
secutively registered patients (14 men, 21 women; Specific vascularization patterns such as central, sequences was 350–450 mm 2, and the matrix size
mean age, 59 years; range, 36–84 years) sent to peripheral, and chaotic contrast enhancement was 107–186 × 256. All sequences were perform­
our endocrinology referral center for further diag­ were assessed. ed with a breath-hold. The contrast agent was
nostic evaluation of incidentally detected adrenal The video clip was digitally recorded, and a administered with a power injector at a constant
masses were enrolled in this study. The adrenal region of interest was placed in the adrenal mass injection rate of 2.5 mL/s. The dose of gado­
masses were detected incidentally at sonography and in the reference organ (spleen or liver) and pentetate dimeglumine (Magnevist, Schering)
or CT. Baseline characteristics of the patients are analyzed over time with time–intensity curves. was 0.1 mmol/kg body weight, corresponding to
shown in Table 1. For analysis of hormonal Contrast enhancement was compared with pre­ 0.2 mL/kg body weight of a 0.5 mol/L formulation
activity, all patients underwent detailed laboratory injection values and expressed in percentage of injected in a volume based on body weight and
testing (see later). baseline value. The perfusion patterns were followed by a 20-mL saline flush.
All patients underwent sonography of the compared with the CT and MRI diagnoses and Criteria for the diagnosis of adenoma on MRI
adrenal mass that included duplex and Doppler judged congruent or noncongruent with CT or were homogeneous signal intensity on un­
sonography followed by contrast-enhanced sonog­ MRI findings for differentiation of adenoma and enhanced T2-weighted and T1-weighted images,
raphy. In addition, all patients underwent CT, nonadenomatous lesions. homo­geneous enhancement on dynamic images
MRI, or both. Patients without clear signs of a after contrast injection (0.1 mmol/kg body weight
benign adrenal mass on MRI or CT underwent CT and MRI of gadopentetate dimeglumine), and loss of signal
fine-needle aspiration biopsy or adrenalectomy If the incidental diagnosis of adrenal mass was intensity of the lesion on out-of phase images
and histologic evaluation of the specimen. The made at CT performed for other purposes, such as (evaluated by visual assessment of signal intensity
study was performed in accordance with the ethi­ abdominal pain, staging of primary malignant loss on opposed-phase images compared with in-
cal guidelines of the Declaration of Helsinki. tumor, and hormonal dysfunction, the CT scans phase images) [18]. As an internal standard, the
Informed consent in compliance with legislation were reviewed for quality criteria by a radiologist signal intensity of the adrenal mass was compared
was obtained from each patient. with 20 years of experience. Only a triple-phase with the signal intensity of the spleen.
helical CT examination with delayed contrast
Sonographic Examination enhancement (iomeprol 400 mg I/mL, Iomeron Hormonal Evaluation
All patients were examined in the supine 400, Bracco) was accepted for analysis. Percentage In addition to routine laboratory testing, a specific
position with B-mode Doppler sonography with a of washout was calculated with the following evaluation to detect the hormonal activity of an
3.5-MHz transducer (EUB-8500, Hitachi). The formula: (1 – delayed enhanced attenuation value / adrenal mass was performed on all patients after
sonographic examinations were performed by two initial enhanced attenuation value) × 100 [16]. If overnight fasting. The hormonal evaluation included
experienced examiners blinded to the results of clear criteria for a benign adrenal mass were base levels of corticotropin, cortisol, dehydro­
MRI and CT. All patients had fasted overnight. present (unenhanced attenuation ≤ 10 HU and epiandrosterone sulfate, aldosterone, renin activity,
The adrenal mass was examined for size, CT contrast washout ≥ 50% 10 minutes after plasma-free normetanephrine, plasma-free meta­
echogenicity, echotexture, margins, and perfusion contrast administration) [16, 17], no additional nephrine, and chromogranin A. In addition, an over­
pattern. After B-mode sonography, conventional MRI was performed. night dexamethasone suppression test with 2 mg of
power Doppler imaging was performed. Vascu­ If the primary diagnosis of adrenal mass was dexamethasone was performed to rule out Cushing’s
larization of the adrenal mass was classified as found during a sonographic examination, low- syndrome. A plasma aldosterone concen­tration to
hypovascular, isovascular, or hypervascular in image-quality CT, or CT without clear benign plasma renin activity ratio of less than 20 together
comparison with the vascularization of the liver or criteria, MRI was performed at our clinic. MRI with an aldosterone concentration within the normal
spleen. The presence of specific vascular patterns, was performed on a 1.5-T unit (Vision, Siemens range and a cortisol concentration of less than 3 µg/
such as afferent blood vessels and irregular tumor Medical Solutions). The MRI examinations dL after an overnight suppression test were con­
vessels was recorded. included unenhanced T1-weighted and T2- sidered normal. The laboratory results were judged
Contrast-enhanced dynamic sonography was weighted sequences, chemical shift imaging (in- by an endo­crinologist with 14 years of experience.
performed with contrast-specific continuous-mode and out-of-phase imaging), and dynamic T1-
software with a 3.5-MHz transducer (EUB- 8500) weighted contrast-enhanced imaging (gadobenate Final Reference Diagnosis
and a low mechanical index (0.11). Aqueous sus­ dimeglumine, MultiHance, Bracco). The reference standard was determined by an
pension of phospholipid-stabilized micro­bubbles MRI was performed with T2-weighted turbo experienced radiologist and an experienced endo­
filled with sulfur hexafluoride (SonoVue) was spin-echo sequences (TR/TE, 4,000/108; flip crinologist working together. The final reference
used as the sonographic contrast agent. The angle, 150°; echo-train length, 29), a T2-weighted diagnosis was defined by combining all available
contrast agent was injected IV in a 4.8-mL bolus RARE sequence (HASTE, Siemens Medical information from imaging (CT and MRI) and
through a 20-gauge cannula into an antecubital Solutions) (infinite/74; flip angle, 180°), and T1- additional information from the hormonal evalu­
vein and was followed by a 10-mL saline flush. weighted gradient-recalled echo in-phase (160/ ation and histologic examination. Postsurgical
The adrenal mass was scanned for 5 minutes with 4.7; flip angle, 70°) and out-of phase (160/2.6; histo­logic verification of adrenal masses after
the liver (right adrenal mass) or spleen (left flip angle, 70°) images. For dynamic contrast- adrenalectomy was available in 10 cases. Metastatic
adrenal mass) as an in vivo reference. The first enhanced imaging, a T1-weighted FLASH 2D lesions (n = 3) were diagnosed with fine-needle
contrast enhancement was observed and sequence (174.9/4.1; flip angle, 80°) was used biopsy. In the other 22 cases, the diagnosis was
categorized in the early arterial phase (< 20 during the arterial (20–25 seconds after injection), confirmed with CT (n = 10) or MRI (n = 12).

1854 AJR:191, December 2008


Sonography of Adrenal Masses

Statistical Analysis mass was found for malignant or benign le- Power Doppler sonography revealed hy-
Statistical analysis was performed with SPSS sions. Malignant lesions, however, signifi- pervascularization or an afferent blood ves-
software (version 12.0, SPSS). Clinical and labora­ cantly more often had an inhomogeneous sel in six cases. Hypervascularization of the
tory characteristics of patients were expressed as echotexture (71% vs 18%, p < 0.05) and a adrenal mass or an afferent blood vessel was
mean ± SD, median, and range. Fisher’s exact test mixed hyperechoic and hypoechoic pattern found significantly more often in malignant
was used to evaluate the criteria for diagnosis of (43% vs 4%, p < 0.05). The mean size of be- than in benign lesions (57% vs 7%, p < 0.05).
malignant adrenal masses. A value of p < 0.05 was nign lesions was 30 × 39 mm (median, 25 × Nevertheless, no sufficient differentiation
judged to be statistically significant. 34 mm; range, 10–114 mm), and the mean between benign and malignant adrenal
size of malignant lesions was 60 × 77 mm masses was possible at B-mode and power
Results (median, 37 × 78 mm; range, 18–190 mm). A Doppler sonography.
B-Mode Sonography and Power diameter greater than 4 cm was found signifi-
Doppler Sonography cantly more often for malignant than for be- Contrast-Enhanced Sonography
Thirty-five patients with incidentally de- nign lesions (71% vs 21%, p < 0.05) and had Four patterns of sonographic contrast en-
tected adrenal masses were included in the a sensitivity of 71% (95% CI, 0.35–0.92) and hancement were found (Fig. 1). Nine adrenal
study. At B-mode sonography no specific a specificity of 79% (95% CI, 0.60–0.90) in masses exhibited early arterial contrast en-
shape (round, oval, polycyclic) of adrenal the detection of malignant adrenal masses. hancement (< 20 seconds) with an intensity

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10
12 24 36 48 60 72 84 96 108 120 12 24 36 48 60 72 84 96 108 120
Seconds Seconds

2
1

2
1
3
A B

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10
12 24 36 48 60 72 84 96 108 120 12 24 36 48 60 72 84 96 108 120
Seconds Seconds

2
1
2

1
C D
Fig. 1—Sonographic contrast enhancement patterns.
A, Graph of pattern 1 shows early arterial (< 20 seconds) contrast enhancement with intensity peak between 11 and 43 seconds followed by rapid washout. Regions of
interest (ROIs) for yellow (ROI 2) and red (ROI 1) lines were placed in adrenal mass.
B, Graph of pattern 2 shows arterial contrast enhancement beginning in arterial phase (21–40 seconds) with intensity peak after 43–86 seconds. ROI for yellow line was
placed in spleen, ROI of red line in adrenal mass.
C, Graph of pattern 3 shows late and little contrast enhancement in late phase (> 40 seconds) with intensity peak after 60–110 seconds. ROI for green line was placed in
liver, ROI of red line in adrenal mass.
D, Graph of pattern 4 shows no recordable contrast enhancement. ROI for yellow line was placed in liver, ROI of red line in adrenal mass.

AJR:191, December 2008 1855


Friedrich-Rust et al.

peak between 11 and 43 seconds followed by tween primary and secondary malignant tu- metastatic at MRI but found to be adenoma at
rapid washout (pattern 1). Three adrenal mors of the adrenal gland was possible at histologic examination). With patterns 3 and 4
masses exhibited arterial contrast enhance- contrast-enhanced sonography. Eight of the as the criteria for benign adrenal mass and
ment beginning in the arterial phase (21–40 10 resected lesions were found to be nonade- patterns 1 and 2 as the criteria for malignant
seconds) with an intensity peak after 43–86 nomatous at sonography with the use of con- lesion, the diagnosis of malignant lesion was
seconds (pattern 2). Fourteen adrenal masses trast enhancement patterns 1 and 2 for the made with contrast-enhanced sonography
exhibited late and little contrast enhance- diagnosis of nonadenomatous lesions. with a sensitivity of 100% (95% CI, 60–100%),
ment in the late phase (> 40 seconds) with an Among the nine adrenal masses with no specificity of 82% (64–93%), positive predic-
intensity peak after 60–110 seconds (pattern contrast enhancement on sonography (pat- tive value of 58% (32–81%), and negative
3). Nine adrenal masses exhibited no record- tern 4), all but one had CT or MRI evidence predictive value of 100% (83–100%).
able contrast enhancement (pattern 4). of adrenal adenoma, and one was a metastat- In 91.4% (32 of 35) of examined patients,
ic lesion after successful chemotherapy and the findings at CT or MRI and contrast-en-
Contrast-Enhanced Sonographic Findings in in accordance with the sonographic findings hanced sonography were congruent concern-
Comparison with CT, MRI, and Histologic did not exhibit perfusion at MRI. This lesion ing the characterization of adenoma versus
Findings was therefore judged to be nonadenomatous nonadenomatous lesion. In two of these cas-
All nine adrenal masses with early arterial and nonmalignant. es, however, findings with all imaging meth-
peak enhancement with rapid washout on Among the 14 adrenal masses with slow ods suggested a nonadenomatous lesion, but
sonograms (pattern 1) exhibited the criteria and late contrast enhancement at sonography histologic examination after adrenalectomy
for nonadenomatous lesions on CT or MRI. (pattern 3), 11 were diagnosed as adrenal ade- revealed adrenal adenoma. One of the cases
Six of these nine adrenal masses exhibited nomas on the basis of CT or MRI findings. of divergent findings at CT or MRI and con-
criteria of malignancy on CT and MRI, and One mass was misdiagnosed as metastasis at trast-enhanced sonography was metastasis
three were suspected of being pheochromocy- CT, but the histologic finding was adenoma after chemotherapy without remaining vas-
toma according to MRI findings. Fine-needle (Table 1). One mass was diagnosed as adrenal cularization. Perfusion was seen neither on
biopsy revealed metastasis of lung cancer in hyperplasia at MRI, and one as myelolipoma MRI nor on contrast-enhanced sonography,
two of the nine lesions, and the patients with at MRI and histologic examination. All three but the lesion could not be differentiated from
the other seven adrenal masses were trans- adrenal masses with arterial contrast enhance- adenoma at sonography. In the two other cas-
ferred to surgery. Examination of the histo- ment on sonography (pattern 2) beginning in es of divergent findings, the correct diagnosis
logic specimens confirmed the presence of the arterial phase had the CT or MRI criteria was made at CT or MRI in one case (angio-
malignancy in four cases (two nonfunctional of a nonadenomatous lesion (one myelolipoma, myelolipoma) and with contrast-enhanced
adrenocortical carcinomas, one cortisol-pro- one metastatic lesion, and one lesion deemed sonography in the other case (adenoma).
ducing adrenocortical carcinoma, and one T-
cell lymphoma). The other three lesions were
benign (one pheochromocytoma, one adre-
nal adenoma, and one partially thrombosed
cavernous hemangioma). Among the three
lesions suspected of being pheochromocyto-
ma at MRI, the histologic finding was pheo-
chromocytoma in only one case; the other
two lesions were carcinoma and adenoma.
The adenoma exhibited central hemorrhage
and calcification. All patients tolerated sur-
gery without complications. A B
In addition to the nine lesions with sono-
graphic pattern 1, three other adrenal masses
were resected. Two of these lesions (22 × 15
mm and 30 × 26 mm) had a suspicious ap-
pearance on CT or MRI, and one lesion was
large (55 × 50 mm) and caused local symp-
toms. The former two adrenal masses exhib-
ited sonographic contrast enhancement pat-
terns 3 and 2, respectively, and the latter,
pattern 3. The former two masses were his-
tologically adrenal adenoma, and the latter
had MRI evidence of myelolipoma. C D
Among the 10 cases managed surgically, Fig. 2—Comparison of sonographic findings in adenoma and nonadenomatous lesions.
the correct diagnosis concerning the histo- A and B, 57-year-old woman with histologically proven adrenocortical carcinoma. Images show B-mode (A)
and power Doppler (B) sonographic findings.
pathologic entity was made with CT and C and D, 43-year-old man with adrenal adenoma. Images show B-mode (C) and power Doppler (D) sonographic
MRI in three cases. No differentiation be- findings.

1856 AJR:191, December 2008


Sonography of Adrenal Masses

The timing of contrast enhancement for with sensitivity comparable with that of CT malignant adrenal masses is essential for re-
contrast-enhanced sonography and that for and MRI in differentiating adenoma and ducing the number of expensive and invasive
CT or MRI were not directly comparable nonadenomatous lesions. Contrast-enhanced imaging studies performed. A high sensitiv-
owing to the different pharmacokinetics of sonography, however, is not reliable for dif- ity for the diagnosis of malignancy is needed
the contrast agents. The interpretations there- ferentiating the various histopathologic non- to ensure that a primary or secondary malig-
fore were made for each method separately adenomatous lesions from one another. nant lesion of the adrenal gland is not missed.
(Figs. 2–4). The incidence of incidentaloma found in Conversely, specificity for the diagnosis of
autopsy and CT studies is 4–6% [19–21]. malignancy is less critical, because the worst
Hormonal Evaluation Noninvasive differentiation of benign and consequence of a false-positive diagnosis is
All but two adrenal masses were nonfunc-
tional. One patient had clinical and laboratory
signs of Cushing’s syndrome. The dexametha-
sone overnight suppression test showed a
pathologic cortisol concentration of 24 µg/dL
(normal range < 3 µg/dL) at 8:00 am after a
dose of 2 mg of dexamethasone at 12 mid-
night. MRI showed a suspicious adrenal mass
with the differential diagnosis of pheochro-
mocytoma. Contrast-enhanced sonog­raphy
showed pattern 1 vascularization. Histologic
examination revealed cortisol-producing ad-
renocortical carcinoma.
Another patient had clinical and laborato-
ry signs of pheochromocytoma with a plasma
metanephrine concentration of 2,531 pg/mL
(normal range < 90 pg/mL) and plasma A B
normetanephrine concentration of 466 pg/
mL (normal range < 200 pg/mL). The MRI
findings were congruent with the diagnosis
of pheochromocytoma. Contrast-enhanced
sonography showed pattern 1 vasculariza-
tion. Histologic examination revealed benign
pheochromocytoma.

Discussion
This proof-of-principle study showed for
the first time, to our knowledge, that con-
trast-enhanced sonography can be performed

Fig. 3—Contrast-enhanced sonographic images


of patients in Figure 2. Adrenocortical carcinoma
(A, C, E, G) in 57-year-old woman exhibits rapid C D
early arterial enhancement with rapid washout.
No adjacent liver tissue is present. Center of
adrenocortical carcinoma is necrotic, and therefore
no perfusion is present in center. No contrast
enhancement is evident in adrenal adenoma (B, D, F,
H) in 43-year-old man, but arterial and portal venous
enhancement is present in adjacent liver.
A and B, Contrast-enhanced sonographic images
obtained in early arterial phase (16 seconds after
contrast injection).
C and D, Contrast-enhanced sonographic images
obtained in arterial phase, 22 seconds after contrast
injection.
E and F, Contrast-enhanced sonographic images
obtained in parenchymal phase, 48 seconds after
contrast injection.
G and H, Contrast-enhanced sonographic images
obtained in late phase, 70 seconds after contrast
injection.
(Fig. 3 continues on next page)
E F

AJR:191, December 2008 1857


Friedrich-Rust et al.

Fig. 3 (continued)—Contrast-enhanced sonographic


images of patients in Figure 2. Adrenocortical
carcinoma (A, C, E, G) exhibits rapid early arterial
enhancement with rapid washout. No adjacent liver
tissue is present. Center of adrenocortical carcinoma
is necrotic, and therefore no perfusion is present
in center. No contrast enhancement is evident in
adrenal adenoma (B, D, F, H), but arterial and portal
venous enhancement is present in adjacent liver.
G and H, Contrast-enhanced sonographic images
obtained in late phase, 70 seconds after contrast
injection.

G H

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10
12 24 36 48 60 72 84 96 108 120
Seconds

2
1
3

A B

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10
12 24 36 48 60 72 84 96 108 120
Seconds

1
3
2

C D
Fig. 4—Region of interest (ROI) location and time–intensity curves for patients in Figures 2 and 3. ROIs are placed before contrast enhancement begins; therefore
background is very dark. Insets (A and C) show contrast enhancement for orientation of ROI placement.
A and B, 57-year-old woman with histologically proven adrenocortical carcinoma. All three ROIs are in different regions of adrenal mass. ROI 3 (green) in necrotic area
exhibits no contrast enhancement. Other two ROIs (yellow and red) show early arterial contrast enhancement and rapid washout (pattern 1).
C and D, 43-year-old man with adrenal adenoma. ROI 1 (red) is in liver; ROIs 2 (yellow) and 3 (green) are in adrenal mass. No contrast enhancement is present in adrenal
mass compared with liver (pattern 4).

1858 AJR:191, December 2008


Sonography of Adrenal Masses

that an additional or invasive examination is cellent sensitivity, specificity, and positive masses are needed for confirmation of our
necessary to establish the correct diagnosis. predictive value in the diagnosis of benign preliminary data.
In our proof-of-principle study, contrast-en- adrenal mass and are currently accepted as
hanced sonography fulfilled these criteria reference methods in the evaluation of non- References
with a sensitivity of 100% and a specificity functional adrenal masses smaller than 4 cm 1. Dietrich CF, Wehrmann T, Hoffmann C, Herr­
of 82% in the diagnosis of malignant lesions in diameter [16, 17, 24–29]. Therefore, CT or mann G, Caspary WF, Seifert H. Detection of the
of the adrenal gland. MRI was used as the reference standard when adrenal glands by endoscopic or transabdominal
As reported in previous studies [3] and signs of adenoma were overt, as they were in ultrasound. Endoscopy 1997; 29:859–864
found in our study, the average size of benign the 22 patients with nonfunctional adrenal 2. Suzuki Y, Sasagawa, Suzuki H, Izumi T, Kaneko
adrenal masses is smaller than that of malig- masses. These patients were not referred for H, Nakada T. The role of ultrasonography in the
nant adrenal masses (30 × 39 mm vs 60 × 77 surgery. Imaging (after 6, 12, and 24 months) detection of adrenal masses: comparison with
mm). In a study involving 887 patients with and laboratory (annually for 4 years) follow- computed tomography and magnetic resonance
adrenal incidentaloma, a diameter greater up was suggested [17, 24]. Three patients imaging. Int Urol Nephrol 2001; 32:303–306
than 4 cm was found to have 90% sensitivity with metastasis of primary lung cancer did 3. Lumachi F, Borsato S, Brandes AA, et al. Fine-
and 24% specificity in the detection of adre- not undergo surgery because of the palliative needle aspiration cytology of adrenal masses in
nocortical carcinoma [22]. In our study, 11 treatment concept. The diagnosis was con- noncancer patients: clinicoradiologic and histo-
adrenal masses were larger than 4 cm in di- firmed with fine-needle biopsy. logic correlations in functioning and nonfunction-
ameter, and three of these lesions were adre- In accordance with present recommenda- ing tumors. Cancer 2001; 93:323–329
nocortical carcinoma, for a sensitivity of tions [17], all nine patients with suspicious 4. Liao JT, Huang TH, Wu BY. Ultrasonic evalua-
100% (95% CI, 38–100%) and specificity of findings at imaging were referred for sur- tion of adrenal masses. Hunan Yi Ke Da Xue Xue
75% (58–87%). The prevalence of adreno- gery. Malignancy was found in four patients, Bao 2001; 26:453–454
cortical carcinoma was 8.6%, higher than and pheochromocytoma and myelolipoma 5. Trojan J, Schwarz W, Sarrazin C, Thalhammer A,
the 4.7% reported in the literature [21]. A were confirmed in two patients. The misdi- Vogl TJ, Dietrich CF. Role of ultrasonography in
size greater than 4 cm was found to have a agnosis of adenoma for a nonadenomatous the detection of small adrenal masses. Ultraschall
sensitivity of 71% and a specificity of 79% in lesion at CT or MRI can be explained by the Med 2002; 23:96–100
the detection of malignant adrenal masses. presence of hemorrhage and calcification in 6. Slonina J, Nienartowicz E, Agrawal AK, Malcze-
Therefore, in accordance with previously the large adenomas. wska J, Moron K. The usefulness of contrast-en-
published data [22, 23], no sufficient conclu- An algorithm for the evaluation of adrenal hanced sonography in the differential diagnostic
sion concerning the benignity of the adrenal incidentaloma starts with a history interview, of adrenal tumors [in Polish]. Endokrynol Pol
mass could be drawn solely from the size of physical examination, and hormonal testing 2006; 57:230–236
the mass. [17]. The next step, if the findings are nor- 7. Paulsen SD, Nghiem HV, Korobkin M, Caoili
According to present recommendations mal, is imaging for phenotype. At this point, EM, Higgins EJ. Changing role of imaging-guid-
[17], all patients received a hormonal evalua- in light of the results of our study, contrast- ed percutaneous biopsy of adrenal masses: evalu-
tion (after the history interview and physical enhanced sonography can be used as the first ation of 50 adrenal biopsies. AJR 2004; 182:1033–
examination) as the first diagnostic proce- and most cost-effective imaging method for 1037
dure. In this study, two adrenal masses (6%) differentiating benign and suspicious ap- 8. Korobkin M, Brodeur FJ, Yutzy GG, et al. Dif-
were functional (Cushing’s syndrome and pearances. Because the sensitivity for the ferentiation of adrenal adenomas from nonade-
pheochromocytoma). Both patients were re- diagnosis of a malignant lesion in this study nomas using CT attenuation values. AJR 1996;
ferred for surgery, and the diagnosis was was 100%, no malignant lesions were missed. 166:531–536
confirmed. According to the literature [23], Because adenomas are common and malig- 9. van Erkel AR, van Gils AP, Lequin M, et al. CT
approximately 85% of adrenal incidentalo- nancy of incidentally diagnosed adrenal and MR distinction of adenomas and nonade-
mas are nonfunctional, and 15% are func- masses is rare, contrast-enhanced sonogra- nomas of the adrenal gland. J Comput Assist To-
tional. The slightly lower number of func- phy may be a useful tool for excluding most mogr 1994; 18:432–438
tional adenomas in our study might have adenomas and selecting patients who need 10. Dunnick NR, Korobkin M. Imaging of adrenal inci-
been due to the small number of patients ex- additional diagnostic evaluation. dentalomas: current status. AJR 2002; 179:559–568
amined in this pilot study. Another explana- Because the specificity of MRI and CT is 11. Szolar DH, Kammerhuber F. Quantitative CT
tion may be that an increasing number of in- higher than that of contrast-enhanced sonog- evaluation of adrenal gland masses: a step for-
cidentalomas are detected as the result of raphy, the next step for all suspicious lesions ward in the differentiation between adenomas and
improvements in sonographic systems and should be a second imaging study (CT or nonadenomas? Radiology 1997; 202:517–521
the inclusion of adrenal gland examinations MRI). However, the number of CT and MRI 12. Boland GW, Lee MJ, Gazelle GS, Halpern EF,
during screening sonography. examinations needed can be markedly re- McNicholas MM, Mueller PR. Characterization
According to current recommendations duced if preselection is performed with con- of adrenal masses using unenhanced CT: an anal-
[17], all patients with negative results at hor- trast-enhanced sonography. Therefore, con- ysis of the CT literature. AJR 1998; 171:201–204
monal testing underwent high-quality CT or siderable cost reduction can be attained with 13. Outwater EK, Siegelman ES, Radecki PD, Piccoli
underwent MRI at our clinic to differentiate the expanded algorithm. On the basis of the CW, Mitchell DG. Distinction between benign
benign and suspicious lesions that necessi- results of this proof-of-principle study, more and malignant adrenal masses: value of T1-
tated further evaluation. Contrast-enhanced extensive studies of contrast-enhanced sono­ weighted chemical-shift MR imaging. AJR 1995;
CT and gadolinium-enhanced MRI have ex- graphy for the characterization of adrenal 165:579–583

AJR:191, December 2008 1859


Friedrich-Rust et al.

14. Mayo-Smith WW, Lee MJ, McNicholas MM, puterized tomography series. J Endocrinol Invest 25. Fassnacht M, Kenn W, Allolio B. Adrenal tumors:
Hahn PF, Boland GW, Saini S. Characterization 2006; 29:298–302 how to establish malignancy? J Endocrinol Invest
of adrenal masses (< 5 cm) by use of chemical 20. Kloos RT, Gross MD, Francis IR, Korobkin M, 2004; 27:387–399
shift MR imaging: observer performance versus Shapiro B. Incidentally discovered adrenal mass- 26. Grumbach MM, Biller BM, Braunstein GD, et al.
quantitative measures. AJR 1995; 165:91–95 es. Endocr Rev 1995; 16:460–484 Management of the clinically inapparent adrenal
15. Mitchell DG, Crovello M, Matteucci T, Petersen 21. Young WF Jr. Management approaches to adrenal mass (“incidentaloma”). Ann Intern Med 2003;
RO, Miettinen MM. Benign adrenocortical mass- incidentalomas: a view from Rochester, Minne- 138:424–429
es: diagnosis with chemical shift MR imaging. sota. Endocrinol Metab Clin North Am 2000; 27. Hamrahian AH, Ioachimescu AG, Remer EM, et
Radiology 1992; 185:345–351 29:159–185 al. Clinical utility of noncontrast computed to-
16. Pena CS, Boland GW, Hahn PF, Lee MJ, Mueller PR. 22. Angeli A, Osella G, Ali A, Terzolo M. Adrenal mography attenuation value (Hounsfield units) to
Characterization of indeterminate (lipid-poor) adre- incidentaloma: an overview of clinical and epide- differentiate adrenal adenomas/hyperplasias from
nal masses: use of washout characteristics at contrast- miological data from the National Italian Study nonadenomas: Cleveland Clinic experience. J
enhanced CT. Radiology 2000; 217:798–802 Group. Horm Res 1997; 47:279–283 Clin Endocrinol Metab 2005; 90:871–877
17. Young WF Jr. Clinical practice: the incidentally 23. Mantero F, Terzolo M, Arnaldi G, et al. A survey 28. Korobkin M, Brodeur FJ, Francis IR, Quint LE,
discovered adrenal mass. N Engl J Med 2007; on adrenal incidentaloma in Italy. Study Group Dunnick NR, Londy F. CT time–attenuation
356:601–610 on Adrenal Tumors of the Italian Society of En- washout curves of adrenal adenomas and nonade-
18. Korobkin M, Lombardi TJ, Aisen AM, et al. docrinology. J Clin Endocrinol Metab 2000; nomas. AJR 1998; 170:747–752
Characterization of adrenal masses with chemical 85:637–644 29. Szolar DH, Korobkin M, Reittner P, et al. Adreno-
shift and gadolinium-enhanced MR imaging. Ra- 24. Herrera MF, Grant CS, van Heerden JA, Sheedy cortical carcinomas and adrenal pheochromocy-
diology 1995; 197:411–418 PF, Ilstrup DM. Incidentally discovered adrenal tomas: mass and enhancement loss evaluation at
19. Bovio S, Cataldi A, Reimondo G, et al. Prevalence tumors: an institutional perspective. Surgery 1991; delayed contrast-enhanced CT. Radiology 2005;
of adrenal incidentaloma in a contemporary com- 110:1014–1021 234:479–485

1860 AJR:191, December 2008