ELSEVIER
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology
journal homepage: www.elsevier.com/locate/ejrad
—
EUROPEAN JOURNAL OF RADIOLOGY
=
-
Differentiation of adrenal tumors in patients with hepatocellular carcinoma: Adrenal adenoma versus metastasis
CrossMark
Koichiro Yasaka a,*, Wataru Gonoia, Hiroyuki Akaiª, Masaki Katsuraª, Masaaki Akahaneª, Shigeru Kiryub, Kuni Ohtomoa
a Department of Radiology, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan
b Department of Radiology, The Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan
ARTICLE INFO
Article history: Received 21 November 2012 Received in revised form 6 February 2013 Accepted 9 February 2013
Keywords:
Adrenal gland Computed tomography Hepatocellular carcinoma Incidentaloma Metastasis
ABSTRACT
Objective: To investigate whether computed tomography (CT) attenuation test for differential diagnosis of adrenal nodule is applicable in patients with hepatocellular carcinoma (HCC) which shows similar image characteristics to adrenal adenoma.
Materials and methods: This retrospective study was approved by our institutional review board, and the requirement for informed consent from study patients was waived. Searching picture archiving and com- munication system, we identified 3678 patients with HCC who underwent upper abdominal unenhanced CT scans between April 2002 and March 2010, and 114 adrenal nodules (39 adenomas and 75 metastases) were included for analysis. Ten nodules were confirmed pathologically while 104 had imaging diagnosis (enlarged or emerged during the study period). Size, CT number, and the internal characteristics of the lesions were recorded.
Results: Mean CT numbers of adrenal adenomas were significantly lower than those of metastases (P<0.0001, t-test) on unenhanced CT. Thresholds of 17 and 33 Hounsfield units (HU) provided the fol- lowing sensitivity, specificity, and accuracy: 46.2%, 100%, and 81.6% at 17 HU, and 94.9%, 89.3%, and 91.2% at 33 HU, respectively. The area under receiver operating characteristic curve for the CT number test was 0.96. Metastases were significantly larger than adrenal adenoma (P=0.009, t-test). However, the accuracy of testing using mass size was 64.0% at most. All adenomas and metastases were depicted as homoge- neous masses with the exception of two metastases that presented as heterogeneous masses (necrotic or lipomatous).
Conclusion: Adrenal adenomas can be differentiated from HCC metastases using CT number on unen- hanced CT.
@ 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Hepatocellular carcinoma (HCC) is the sixth-most prevalent car- cinoma globally [1]. Extrahepatic metastasis is reported to occur in 13.5-42% of HCC patients, and the adrenal gland (12%) is the fourth- most common site for metastasis, following the lung (47%), lymph node (45%), and bone (37%) [2].
Adrenal incidentaloma, which is depicted at radiologic exami- nation conducted for indications other than adrenal disease, is not rare. The prevalence of adrenal incidentaloma at computed tomo- graphy (CT) is reportedly 0.4-5% [3,4]. If the patient has no history of malignancy, the tumor is most likely an adenoma [3]. Mansmann reported that among patients with adrenal incidentalomas and a
history of malignancy, 50-75% have metastasis [5]. As for the pri- mary focus, many kinds of malignant tumors metastasize to the adrenal glands, including lung carcinoma (35%), gastric carcinoma (14%), esophageal carcinoma (12%), and liver/bile duct carcinoma (10%) [6]. In patients with HCC, the choice of treatment depends on whether adrenal incidentaloma is a benign adrenal adenoma or a metastasis; thus, precise diagnosis is very important.
There have been some reports of the differential diagnosis of adrenal incidentalomas [7-12], but to our knowledge none have focused on patients with HCC. Imaging characteristics of adrenal metastasis from HCC may be similar to those of adrenal adenoma. HCC is sometimes depicted as a heterogeneous mass on unen- hanced CT with a low-attenuation area that represents fatty change [13], enhancement in the arterial phase, and washout in the delayed phase on dynamic contrast-enhanced CT [14]. These imaging char- acteristics are similar to those of adrenal adenomas, which have intracytoplasmic fat [7] and show early enhancement [15] and
* Corresponding author. Tel .: +81 3 5800 8666; fax: +81 3 5800 8935. E-mail address: koyasaka@gmail.com (K. Yasaka).
HCC patients 3678 patients
HCC patients without Adrenal masses 3488 patients
HCC patients with Adrenal masses 190 patients (213 lesions)
Exclude 99 lesions
Benign 39 lesions
Malignant 75 lesions
“No unenhanced CT: 12 lesions
” slice thickness other than 5mm: 36 lesions
“tube voltage other than 120kVp: 1 lesion
· Stable for more than 6 months : 38 lesions · Pathology: 1 lesion
*Enlarged : 50 lesions
* after treatment: 8 lesions
·Emerged during the period: 16 lesions “Pathology: 9 lesions
* other malignancy: 11 lesions
* direct invasion from HCC: 6 lesions
“undiagnosed: 17 lesions
*myelolipoma (4 lesions), cyst (2 lesions), hemorrhage (2 lesions)
washout in the delayed phase [8-11] on contrast-enhanced CT. It is unclear whether adrenal adenomas can be differentiated from adrenal metastases from HCC or not using only unenhanced CT. Studies of the differential diagnoses of adrenal masses in patients with HCC should be performed.
The purpose of our study was to examine whether adrenal ade- nomas can be accurately differentiated from adrenal metastases in patients with HCC using unenhanced CT.
2. Materials and methods
This retrospective study was approved by our institutional review board, and the requirement for informed consent from study patients was waived.
2.1. Subjects
The study subjects were selected by searching the picture archiving and communication system. All consecutive patients who underwent abdominal CT scans and were diagnosed with HCC in our single tertiary-care center from April, 2002 to March, 2010 were included. Patients who did not have HCC in the liver during this period, but had a past history of HCC, were also included because they had a potential for metastasis from HCC. HCC was defined as a mass showing enhancement in the arterial phase and washout in the delayed phase on dynamic contrast-enhanced CT, according to a previous study [14].
We identified 3678 patients with HCC, and all CT images were reviewed with respect to the presence of an adrenal mass. There were 213 adrenal masses in 190 patients. Exclusion criteria for adrenal lesions were: did not undergo unenhanced CT (12 lesions); underwent CT with a slice thickness other than 5 mm (36 lesions); underwent CT with a tube voltage other than 120kVp (1 lesion); treated with transarterial embolization, radiofrequency ablation, chemotherapy, and radiation therapy (8 lesions); the presence of other malignancies (11 lesions); direct invasion from HCC in the liver (6 lesions); and no radiological diagnosis by the follow- up criteria described later nor pathological diagnosis (17 lesions). Myelolipoma (4 lesions), cysts (2 lesions), and hemorrhage (2 lesions) were also excluded according to previously reported crite- ria [8]. Of 213 adrenal nodules, 114 lesions (43 right, 71 left; mean
[SD] patient age 68.5 [8.0] years; 101 males, 13 females) were employed for analysis. A flow chart of the inclusion and exclusion criteria is shown in Fig. 1.
2.2. Referential criteria
Adrenal lesions were diagnosed pathologically or by imaging follow-up criteria. If an adrenal lesion did not increase in size for at least 6 months, it was diagnosed as an adrenal adenoma [9-11,16]. If an adrenal lesion was enlarged [9-11] or emerged during the study period [15], it was diagnosed as a metastasis.
2.3. CT scanning technique
Unenhanced CT images were obtained using multidetector-row CT scanners manufactured by GE Medical Systems, Toshiba, Hitachi, and Philips. Scanning parameters were as follows: slice thickness of 5 mm, slice interval of 5 mm, tube voltage of 120 kVp, and gantry rotation time of 250-1543 ms (mean, 710 ms). A total of 82 lesions were scanned with an automatic exposure control system, and 32 lesions were scanned with fixed milliampere seconds (mAs) (mean, 250 mAs; range, 75-428 mAs).
2.4. Clinical and image analysis
Patient sex and age at the first detection of the adrenal mass were recorded. TNM staging of HCCs was performed for both the benign adrenal adenoma group and the metastasis group accord- ing to the seventh edition of the classification of malignant tumors published by the International Union Against Cancer [17]. Lymph node metastasis was defined as 2 cm or larger in its longest diame- ter, according to a previous report, to avoid lymphadenopathy due to liver cirrhosis [18].
The observation period of the adrenal adenoma and the adrenal metastasis were recorded. The growth rate (changes in size per year) of adrenal metastasis was also recorded.
To analyze the imaging features of adrenal lesions, a com- mercial viewer (Centricity Radiology RA1000; General Electric Company, Barrington, IL, USA) was used. Adrenal lesions when first detected were recorded in terms of CT number, longest diameter (on transaxial CT images), and internal characteristics
(homogeneous or heterogeneous). All image analyses were per- formed by the consensus of two radiologists (K.Y. and H.A., with 4 and 10 years of experience, respectively). When measuring CT number, a circular or ovoid region of interest (ROI) was placed centrally over the adrenal lesions, covering approximately two- thirds. The peripheral part of the lesion was carefully avoided to minimize a partial-volume effect. If present, calcified regions were carefully avoided. If the adrenal mass was heterogeneous, the ROI was placed on the low-attenuation area in addition to the normal ROI described above. In such cases, CT attenuation of the normal ROI was subjected to statistical analysis.
2.5. Statistical analysis
Diagnostic sensitivity, specificity, accuracy, positive predictive value, and negative predictive value for the lesion size test and CT number test were calculated.
To compare patient age, lesion size, and lesion CT number between the groups, an unpaired Student’s t-test was used. Com- parison of patient sex and lesion internal characteristics between groups was performed using Fisher’s exact test. These tests were performed using commercial statistical software (JMP 9.0.0; SAS Institute Inc., Cary, NC, USA). Whenever possible, results were expressed as the means ± standard deviation. P<0.05 was consid- ered to suggest statistically significant difference.
Binormal receiver operating characteristic (ROC) analysis was performed to calculate the area under the curve (AUC) for the CT attenuation test using statistical software (LABROC Analysis; The University of Chicago, Chicago, IL, USA).
3. Results
A total of 114 adrenal lesions were analyzed. By pathologi- cal analysis, one lesion was diagnosed as adrenal adenoma, and 9 as metastases from HCC. The remaining 104 lesions were diag- nosed by imaging follow-up. Thirty-eight lesions were stable in size for a variable period (mean, 33.9±22.1 months; range, 6-82 months) and diagnosed as adrenal adenomas. Among these lesions, 29 lesions (76%) were stable in size for more than a year. Fifty lesions increased in size from 17.1 ±6.8 mm to 36.4 ± 13.1 mm during the follow up period (mean 181 ± 127 days) (Fig. 2), and among them 45 lesions emerged during the study period and increased in size. The mean growth rate of 50 lesions was 53.9 ±38.1 mm/year. Six- teen lesions emerged during the study period (mean size when first detected, 19.7 ±5.8 mm; range, 11.5 - 29.6 mm). These enlarged and/or emerged lesions were diagnosed as metastases.
3.1. Clinical and imaging characteristics
The TNM stage of HCC for both groups is shown in Table 1. There were no significant differences in patient age or sex between the two groups (P=0.06 and 0.36, respectively); these parameters are shown in Table 2.
60
· metastases A Adenomas
50
Changes in size (mm)
40
30
20
10
0
10
100
1000
10000
-10
Followup period (days)
CT number, size, and internal characteristics of the lesions are shown in Table 2. Mean CT numbers of adrenal adeno- mas (16.8 HU) were significantly lower than those of metastases (43.1 HU) (P<0.0001). The sensitivity, specificity, accuracy, posi- tive predictive value, and negative predictive value of CT number for differential diagnosis are shown in Table 3. A threshold of 33 Hounsfield units (HU) provided the greatest accuracy (91.2%). No malignancy was found under 17 HU. Scatter plots and ROC curves of CT number for differential diagnosis are shown in Figs. 3 and 4. The AUC for the CT number test was 0.96.
The mean size of adrenal adenomas (14.7 mm) was significantly smaller than that of metastases (18.5 mm) (P=0.009). However, as shown in Fig. 3, there was much overlap in size between the two groups. The accuracy of differentiation of metastasis from adenoma by size was up to 64.0% (Table 4).
All adenomas and metastases appeared as homogeneous masses (Fig. 5) with the exception of the two metastases, which pre- sented as heterogeneous masses (Fig. 6) with low-attenuation areas (4 HU and 29 HU), which represented either intracytoplasmic fat or necrosis.
| Adenomas (n=39) | Metastases (n=75) | |
|---|---|---|
| T0 | 4(10.3%) | 15(20.0%) |
| T1 | 17(43.6%) | 6(8.0%) |
| T2 | 14(35.9%) | 31 (41.3%) |
| T3a | 2(5.1%) | 7(9.3%) |
| T3b | 2(5.1%) | 14(18.7%) |
| T4 | 0 | 2(2.7%) |
| N0 | 39 (100%) | 58 (77.3%) |
| N1 | 0 | 17 (22.7%) |
| M0 | 39 (100%) | 0 |
| M1 | 0 | 75(100%) |
| Adenomas (n=39) | Metastases (n=75) | P value | |
|---|---|---|---|
| Age (year, [SD], range) | 66.6 [8.1](50-86) | 69.4 [7.8](42-92) | 0.06ª |
| Sex (male/female, [%male]) | 33/6 [84.6%] | 68/7 [90.7%] | 0.36b |
| Tumor CT number (HU, [SD], range) | 16.8 [12.9](-21.0 to 40.2) | 43.1 [9.4] (17.0-70.6) | <0.0001ª |
| Tumor size (mm, [SD], range) | 14.7 [3.3] (10.3-24.4) | 18.5[8.6](10.0-65.9) | 0.009ª |
| Homogeneous/heterogeneous [%homogenous] | 39/0 [100%] | 73/2 [97.3%] | 0.55b |
HU, Hounsfield units; SD, standard deviation.
a Student’s t-test.
b Fisher’s exact test.
| Threshold of CT number (HU) | Sensitivity (%) | Specificity (%) | Accuracy (%) | PPV (%) | NPV (%) |
|---|---|---|---|---|---|
| 10 | 25.6 (10/39) | 100 (75/75) | 74.6 (85/114) | 100 (10/10) | 72.1 (75/104) |
| 17 | 46.2 (18/39) | 100 (75/75) | 81.6 (93/114) | 100 (18/18) | 78.1 (75/96) |
| 33 | 94.9 (37/39) | 89.3 (67/75) | 91.2 (104/114) | 82.2 (37/45) | 97.1 (67/69) |
| 41 | 100 (39/39) | 66.7 (50/75) | 78.1 (89/114) | 60.9 (39/64) | 100 (50/50) |
HU, Hounsfield units; PPV, positive predictive value; NPV, negative predictive value. A threshold of 33 HU provides a high accuracy of 91.2%, and that of 17 HU provides a specificity of 100%. Sensitivity of 100% could be obtained using 41 HU as a threshold, but specificity was only 66.7%.
| Threshold of mass size (mm) | Sensitivity (%) | Specificity (%) | Accuracy (%) | PPV (%) | NPV (%) |
|---|---|---|---|---|---|
| 12 | 80.0 (60/75) | 20.5 (8/39) | 59.6 (68/114) | 65.9 (60/91) | 34.8 (8/23) |
| 15 | 62.7 (47/75) | 66.7 (26/39) | 64.0 (73/114) | 78.3 (47/60) | 48.1 (26/54) |
| 20 | 29.3 (22/75) | 92.3 (36/39) | 50.9 (58/114) | 88.0 (22/25) | 40.4 (36/89) |
| 25 | 14.7 (11/75) | 100 (39/39) | 43.9 (50/114) | 100 (11/11) | 37.9 (39/103) |
PPV, positive predictive value; NPV, negative predictive value. A threshold of 15 mm provided an accuracy of up to 64.0%.
4. Discussion
We examined the imaging characteristics of adrenal inciden- taloma on unenhanced CT in patients with HCC. Since some HCCs have imaging characteristics similar to those of adrenal adenoma, whether benign adrenal adenoma could be differentiated from adrenal metastasis from HCC was unclear. To the best of our knowl- edge, this study is the first to report the characteristics of adrenal metastasis from HCC on unenhanced CT. Our data suggest that adrenal adenoma can be differentiated from metastasis in patients with HCC by means of a threshold CT number.
a 70
60
Size (mm)
50
40
30
20
10
Adenoma
Metastasis
CT number (Hounsfield Unit)
80
70
60
50
40
=
30
20
10
0
:
-10
-20
:
-30
Adenoma
Metastasis
To-date, three imaging methods for the diagnosis of adrenal inci- dentalomas has been developed [8]. The first is based on the fact that some adrenal adenomas contain intracytoplasmic fat, while most malignant lesions do not. On unenhanced CT, lipid-rich ade- nomas are depicted as low-attenuation masses, while malignant lesions are not. According to Boland et al., a threshold of 10 HU or less corresponds to a sensitivity of 71% and a specificity of 98% for the diagnosis of adrenal adenoma [12]. Chemical shift MR test is also based on the detection of intracytoplasmic fat. The sec- ond imaging method is based on the fact that adrenal adenoma exhibits a higher washout ratio than that of malignant lesions in dynamic contrast-enhanced studies. The third method is based on the metabolic activity of masses and radioisotopes such as 131I-6- beta-iodomethyl-19-norcholesterol (NP59), MIBG, and fluorine-18 fluorodeoxyglucose are used [8].
It has been reported that CT attenuation test is useful for dif- ferentiating benign adrenal nodules from malignant with high specificity, but no study has focused on adrenal nodules in patients with HCC or tried to differentiate them with using only unenhanced
1
0.9
0.8
0.7
sensitivity
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1-specificity
CT (not using enhanced CT). If it were possible, many patients who cannot use contrast medium would reap the benefit. As HCCs have imaging characteristics similar to those of adrenal adenoma, it is difficult to differentiate benign adrenal nodules from malignant by dynamic enhanced CT as reported quite recently [19], however we successfully evidenced in the present study that they are clearly dif- ferentiable just using non-enhanced CT. We showed that adrenal
metastases in patients with HCC have a higher CT number than do adrenal adenomas, and metastases could be excluded with a specificity of 100% using 17 HU as the threshold. Using 33 HU as the threshold, we could obtain a rather high sensitivity (94.9%) and specificity (89.3%), and the highest accuracy (91.2%).
There were two heterogeneous metastases with low- attenuation areas representing necrosis or intracytoplasmic fat. Nakamura et al. reported that three of four adrenal metastases from HCC were heterogeneous and contained necrotic regions [20]. We should consider the possibility of metastasis when a mass is heterogeneous despite the fact that a partial hypoattenuating area representing intracytoplasmic fat was observed.
Some HCCs show fatty changes and heterogeneous characteris- tics [13]. This is more frequent in well-differentiated HCC. Lesions that have the potential to metastasize to extrahepatic organs tend to be low-differentiated histological types; therefore, metastatic lesions are less likely to appear as low-attenuation masses.
When an adrenal nodule is large, it is more likely to be malig- nant. In a previous report, 70% of the adrenal masses >4 cm were malignant [8]. In our study, all lesions >25 mm were metastases. However, in the range 10-25 mm, there was much overlap in size between adrenal adenomas and metastases. The accuracy of testing using mass size was 64.0% at most. Thus, tumor size did not seem to be as useful as CT number for differential diagnosis.
There are some limitations to our study. First, we used more than one CT scanner; however, differences among scanners report- edly have little effect on the CT values of adrenal masses [21]. Second, not all adrenal nodules were pathologically confirmed in the present study. However, the proportion of pathologically confirmed nodules in our study (10/114) was higher than previ- ous reports (5/66, by Lee et al. [16]; 19/323 by Sangwaiya et al. [22]). Third, because enlarged adrenal nodules without histologi- cal proofs were diagnosed as metastases using imaging follow-up criteria [9-11,16], there is possibility that some enlarged adrenal adenomas were overdiagnosed as malignant. However, 45 of 50 nodules in our study not only increased in size but also emerged during the study period. Barzon et al. reports that mass enlargement of benign adrenal incidentaloma is generally limited to 10-20 mm increase in diameter over a period of 1-3 years [23]. The growth rate of 50 enlarged adrenal nodules (mean 53.9±38.1 mm/year) was high enough compared with that of benign adrenal nod- ules reported by Barzon et al .; thus we believe those enlarged nodules in the present study could be assumed to be malignant. Finally, we preliminarily ruled out the possibility of metasta- sis from lesions other than HCC, excluding patients with other known malignancies; however, not all metastases were patholog- ically confirmed and may have included some primary adrenal neoplasms. In a previous study, 50-75% of adrenal nodules were metastases in patients with a malignant tumor [5], while the pro- portion of adrenocortical carcinomas was reported to be as low as 4% [24].
5. Conclusion
In conclusion, although metastatic hepatocellular carcinoma to the adrenal demonstrates somewhat similar imaging charac- teristics as adrenal adenomas, we could differentiate them using Hounsfield unit measurements on unenhanced CT. A threshold of 33 HU achieved the highest accuracy of 91.2%. Using a threshold of 17 HU, metastasis could be excluded with a specificity of 100%.
Conflict of interest
There is no acknowledgement of grants or conflict of interest with regard to this manuscript.
References
[1] El-Serag HB. Hepatocellular carcinoma: an epidemiologic view. Journal of Clin- ical Gastroenterology 2002;35(5 Suppl. (2)):S72-8.
[2] Uka K, Aikata H, Takaki S, Shirakawa H, Jeong SC, Yamashina K, et al. Clinical features and prognosis of patients with extrahepatic metastases from hepatocellular carcinoma. World Journal of Gastroenterology: WJG 2007;13(3):414-20.
[3] Song JH, Chaudhry FS, Mayo-Smith WW. The incidental adrenal mass on CT: prevalence of adrenal disease in 1049 consecutive adrenal masses in patients with no known malignancy. American Journal of Roentgenology 2008;190(5):1163-8.
[4] Brunt LM, Moley JF. Adrenal incidentaloma. World Journal of Surgery 2001;25(7):905-13.
[5] Mansmann G. The clinically inapparent adrenal mass: update in diagnosis and management. Endocrine Reviews 2004;25(2):309-40.
[6] Lam KY, Lo CY. Metastatic tumours of the adrenal glands: a 30-year experience in a teaching hospital. Clinical Endocrinology 2002;56(1):95-101.
[7] Korobkin M, Brodeur FJ, Yutzy GG, Francis IR, Quint LE, Dunnick NR, et al. Differ- entiation of adrenal adenomas from nonadenomas using CT attenuation values. American Journal of Roentgenology 1996;166(3):531-6.
[8] Boland GW, Blake MA, Hahn PF, Mayo-Smith WW. Incidental adrenal lesions: principles, techniques, and algorithms for imaging characterization. Radiology 2008;249(3):756-75.
[9] Pena CS, Boland GW, Hahn PF, Lee MJ, Mueller PR. Characterization of inde- terminate (lipid-poor) adrenal masses: use of washout characteristics at contrast-enhanced CT. Radiology 2000;217(3):798-802.
[10] Caoili EM, Korobkin M, Francis IR, Cohan RH, Platt JF, Dunnick NR, et al. Adrenal masses: characterization with combined unenhanced and delayed enhanced CT. Radiology 2002;222(3):629-33.
[11] Blake MA, Kalra MK, Sweeney AT, Lucey BC, Maher MM, Sahani DV, et al. Dis- tinguishing benign from malignant adrenal masses: multi-detector row CT protocol with 10-min delay. Radiology 2006;238(2):578-85.
[12] Boland GW, Lee MJ, Gazelle GS, Halpern EF, McNicholas MM, Mueller PR. Char- acterization of adrenal masses using unenhanced CT: an analysis of the CT literature. American Journal of Roentgenology 1998;171(1):201-4.
[13] Kutami R, Nakashima Y, Nakashima O, Shiota K, Kojiro M. Pathomorphologic study on the mechanism of fatty change in small hepatocellular carcinoma of humans. Journal of Hepatology 2000;33(2):282-9.
[14] El-Serag HB, Marrero JA, Rudolph L, Reddy KR. Diagnosis and treatment of hepatocellular carcinoma. Gastroenterology 2008; 134(6):1752-63.
[15] Foti G, Faccioli N, Manfredi R, Mantovani W, Mucelli RP. Evaluation of rela- tive wash-in ratio of adrenal lesions at early biphasic CT. American Journal of Roentgenology 2010; 194(6):1484-91.
[16] Lee MJ, Hahn PF, Papanicolaou N, Egglin TK, Saini S, Mueller PR, et al. Benign and malignant adrenal masses: CT distinction with attenuation coefficients, size, and observer analysis. Radiology 1991;179(2):415-8.
[17] Sobin LH, Gospodarowicz MK, Wittekind C. International union against cancer. TNM classification of malignant tumours. 7th ed. Chichester, West Sussex, UK; Hoboken, NJ: Wiley-Blackwell; 2010.
[18] Lewandowski RJ, Mulcahy MF, Kulik LM, Riaz A, Ryu RK, Baker TB, et al. Chemoembolization for hepatocellular carcinoma: comprehensive imag- ing and survival analysis in a 172-patient cohort. Radiology 2010;255(3): 955-65.
[19] Choi YA, Kim CK, Park BK, Kim B. Evaluation of adrenal metastases from renal cell carcinoma and hepatocellular carcinoma: use of delayed contrast- enhanced CT. Radiology 2013;266(2):514-20.
[20] Nakamura K, Sato Y, Nakata H. Computed tomography of adrenal metastases in hepatocellular carcinoma. Report of four cases. Acta Radiologica (Stockholm, Sweden: 1987) 1989;30(5):550-2.
[21] Hahn PF, Blake MA, Boland GW. Adrenal lesions: attenuation measurement differences between CT scanners. Radiology 2006;240(2):458-63.
[22] Sangwaiya MJ, Boland GW, Cronin CG, Blake MA, Halpern EF, Hahn PF. Inci- dental adrenal lesions: accuracy of characterization with contrast-enhanced washout multidetector CT - 10-min delayed imaging protocol revisited in a large patient cohort. Radiology 2010;256(2):504-10.
[23] Barzon L, Sonino N, Fallo F, Palu G, Boscaro M. Prevalence and natural his- tory of adrenal incidentalomas. European Journal of Endocrinology/European Federation of Endocrine Societies 2003; 149(4):273-85.
[24] Siren JE, Haapiainen RK, Huikuri KT, Sivula AH. Incidentalomas of the adrenal gland: 36 operated patients and review of literature. World Journal of Surgery 1993;17(5):634-9.