Abdominal Imaging
CT sensitivities for large (≥3 cm) adrenal adenoma and cortical carcinoma
Sung Yoon Park, Byung Kwan Park, Jung Jae Park, Chan Kyo Kim
Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-dong, Kangnam-ku, Seoul 135-710, Korea
Abstract
Purpose: To retrospectively compare computed tomog- raphy (CT) sensitivities of large (≥3 cm) adenoma and cortical carcinoma.
Methods: Between January 2004 and November 2012, 43 non-oncologic patients with 43 adrenal masses [31 large adenomas, and 12 carcinomas] underwent unenhanced CT, early contrast-enhanced CT, and delayed contrast- enhanced CT scans prior to adrenalectomy. Three types of region-of-interest (ROI) were used on early contrast- enhanced CT images:aROI (large ROI) covering more than half of a mass and two small ROIs fitted to the highest (high ROI) or lowest (low ROI) attenuation area. These ROIs were also placed in the same area on the other CT images. Adenoma was diagnosed if a mass measured ≤10 HU on unenhanced CT image, or if it had ≥60% absolute percentage washout (APW)or ≥40% relative percentage washout (RPW).Carcinoma was diagnosed if a mass had <60% APW and <40% RPW. CT sensitivities for large adenoma and carcinoma were compared.
Results: CT sensitivities for large adenoma vs. carcinoma were 64.5% (20/31) vs. 100% (12/12) using a large ROI, 100% (31/31) vs. 50.0% (6/12) using a high ROI, and 51.6% (16/31) vs. 100% (12/12) using a low ROIs, respectively.
Conclusions: CT sensitivities for large adenoma and cortical carcinoma are influenced by the size or location of an ROI.A large ROI helps to minimize loss of CT sensitivity for large adenoma and to detect carcinoma.
Key words: Adrenal gland-Adenoma- Carcinoma-Computed tomography
Dedicated adrenal computed tomography (CT) protocol that consists generally of unenhanced, early contrast- enhanced, and delayed contrast-enhanced CT scans has been accepted as the imaging technique of choice for adenoma characterization because of high diagnostic accuracy [1-5]. Such CT scans have been increasingly performed to evaluate an adrenal mass that is inciden- tally detected on routine abdomen CT scan. As a result, the number of adrenal mass biopsies has been markedly reduced because of excellent CT sensitivity for adenoma, which accounts for almost all adrenal incidentalomas [6].
Previous studies, however, reported that dedicated adrenal CT was useful for characterizing mainly adeno- mas (small adenoma) with a mean size <3 cm [1-4]. They included only a small number of adenomas (large adenoma) measuring ≥3 cm and so did not focus on the characterization of large adenoma [1-4]. Differentiation of large adenoma and carcinoma has been rarely re- ported [7]. Small adenoma is histologically homogeneous because degenerative changes rarely occur [8, 9]. Con- versely, if the size of adenoma increases, the adenoma may become heterogeneous, since calcification, hemor- rhage, cystic change, or necrosis is likely to occur [9, 10]. Intuitively, a large adenoma with tissue heterogeneity poses a dilemma to radiologists concerning where to place a region-of-interest (ROI) or how to control the size of an ROI, which can subsequently affect CT diag- nostic performance in the characterization of large ade- noma.
Therefore, it can be hypothesized that that a large adenoma is difficult to differentiate from a carcinoma using adrenal CT protocol. It is still unclear what size or location of an ROI is optimal for CT attenuation mea- surement and how it influences the diagnostic perfor- mance of CT for large adenoma characterization.
The purpose of our study was to retrospectively compare CT sensitivities of large adenoma and corti- cal carcinoma according to the size or location of an ROI.
Materials and methods
This retrospective study was approved by our institu- tional review board and informed consent was waived.
Patients
A total of 424 consecutive patients with a histologically proven adrenal mass underwent dedicated adrenal CT scans to evaluate an adrenal mass prior to biopsy or adrenalectomy between January 2004 and November 2012. These patients were histologically categorized to adenoma in 144 and non-adenoma in 280. Of the 144 patients with adenoma, 113 were excluded due to small adenoma <3 cm (n = 97), a history of extra-adrenal malignancy (n = 14), and bilateral or multiple adrenal lesions (n = 2) (Fig. 1). Finally, 31 patients (mal- e:female = 16:15; mean 47.9 ± 10.3 years; range 28- 69 years)with 31 large (23 cm) adenomas(mean 4.2 cm; range 3.0-7.3 cm) were included. Adrenalectomy in these 31 patients with adenoma was performed because of large mass size (n = 23) and cushing syndrome (n = 8).
Of the 280 patients with non-adenomas, 17 were his- tologically confirmed with carcinoma and the remaining 263 were confirmed with other non-adenomas. Of the 17 patients with carcinoma, five were excluded because of distant metastasis (n = 3) or invasion to adjacent organs (n = 2) at initial CT images. Finally, a total of 12 carci- nomas (mean 6.0 cm; range 2.4-13.1 cm) in 12 patients
(male:female = 4:8; mean 55.3 ± 9.4 years; range 40- 67 years) were included to compare with large adenomas. Of the 12 carcinomas, nine were ≥3 cm (mean 7.1 cm; range 3.1-13.1 cm) and three were <3 cm (mean 2.7 cm; range 2.4-2.9 cm). The sizes of large adenoma and carci- noma were defined as the maximal diameter measured on the surgical specimen.
CT protocols
CT examination was performed using one of six multi- detector CT scanners (LightSpeed QX/I, LightSpeed Ultra16, or LightSpeed VCT, GE Medical System, Mil- waukee, WI; Brilliance 40, Philips Medical Systems, Cleveland, OH; Aquilion, Toshiba Medical Systems, Otawara, Japan; SOMATOM Definition Flash, Siemens Medical Solutions, Forchheim, Germany). Imaging parameters for unenhanced and contrast-enhanced CT examinations were 2.5-3.0 mm of slice thickness, 0.75- 1.13:1 of pitch, 120 kVp for single-energy CT and 140/ 80 kVp for dual-energy CT, and 150-370 mA.
A total of 120 mL of non-ionic iodine contrast material (Iopromide, Ultravist 370, Schering, Germany; Iomeprol, Iomeron, Ilsung, Seoul, South Korea; Iohexol 300, Omnipaque, Nycomed, New York, N.Y., USA) were injected intravenously using a power injector at a rate of 3.0 mL/s. All patients underwent unenhanced CT and two-phase contrast-enhanced CT scans, which were obtained at 1 min (early contrast-enhanced CT) and
Patients with adrenal CT protocol and histologic confirmation for adrenal lesions between Jan. 2004 and Nov. 2012 (n=424)
Adenomas (n=144 )
Nonadenomas (n=280)
< Exclusion >
(a) Small adenoma < 3cm (n=95)
< Exclusion >
(b) History of extra-adrenal malignancy ( n=14)
(a) Noncarcinomas (n=263)
(b) Carcinomas with invasion
(c) Less than 1cm (n=2)
or metastasis (n=5)
(d) Bilateral or multiple lesions (n=2)
Large adenomas (n=31)
Carcinomas (n=12)
15 min (delayed contrast-enhanced CT) after adminis- tration of contrast material [2]. The same slice thickness or collimation was used for all three CT acquisitions in a same patient.
Image analyses
Two radiologists with 11-year and 1-year experience, respectively, for genitourinary imaging interpretation reviewed dedicated adrenal CT images and reached a consensus. CT images were uploaded from a picture archiving and communication system (Centricity 2.0, General Electric Medical Systems, Milwaukee, WI). The radiologists were blinded to patients’ clinical and path- ologic information when CT images were evaluated.
Lesion attenuation values were measured at three different areas using three types of an ROI.A ROI (large ROI) was increased to cover more than half of a lesion, while the other ROIs were decreased to cover less than half of a lesion. These small ROIs were fitted to cover the highest (high ROI) or lowest (low ROI) attenuated area on early contrast-enhanced CT images (Figs. 2, 3). These ROIs were also placed in the same area on unenhanced and delayed contrast-enhanced CT images. Lesion attenuation values were measured three times using each ROI. A representative lesion attenuation value was averaged from these three measurements. Calcification, blood vessel, fat tissue, hemorrhage, and cystic or ne-
crotic areas were avoided to place an ROI. Peripheral area of a lesion was also avoided to prevent partial vol- ume artifacts.
An absolute percentage washout (APW) was calcu- lated as (CTEE - CTDE) × 100/(CTEE - CTUE), and a relative percentage washout (RPW) was calculated as (CTEE - CTDE) × 100/CTEE, where CTUE was a lesion attenuation value at unenhanced CT; CTEE was a lesion attenuation value at early contrast-enhanced CT; and CTDE was a lesion attenuation value at delayed contrast- enhanced CT. Adenoma was diagnosed if a mass mea- sured ≤10 HU at unenhanced CT or if a mass had APW ≥60% or RPW ≥40%. Carcinoma was diagnosed if a mass measured > 10 HU at unenhanced CT and if the lesion had APW <60% and RPW <40%. Lesion attenuation values, APW, and RPW were compared between large adenoma and carcinoma. CT sensitivities for large adenoma and carcinoma were compared to determine whether or not there was difference according to the ROI size or location.
Large adenoma and carcinoma were qualitatively compared in terms of lobulated tumor margin, fat tissue, cystic or necrotic change, and calcification. Lobulated tumor margin was defined when the tumor surface was not smooth. Fat tissue was defined when there was a low attenuated area measuring ← 20 HU. Cystic or necro- tic change was defined when there was an unenhanced area with an attenuation difference <10 HU before and
after administration of contrast material. Calcification was defined when a cortical bone-like hyper dense focus was detected on unenhanced CT.
Reference standard
All 31 large adenomas and 12 carcinomas were histo- logically confirmed by means of the adrenalectomy. Pathologic diagnosis of large adenoma or carcinoma was based on the modified Weiss scoring system [11, 12]. A score ≥3 was considered indicative of carcinoma.
Statistical analyses
CT sensitivities using (a) lesion attenuation value mea- sured on unenhanced CT, CT sensitivities using (b) APW and RPW, and overall CT sensitivities (a + b) were obtained to compare between large adenoma and carci- noma. Specificity, positive predictive value, negative predictive value, overall accuracy for adenoma by dif- ferent ROI methods was also investigated.
In terms of lesion attenuation value, APW, and RPW, large adenoma, and carcinoma was compared using non- parametric (Mann-Whitney U test) tests. Unpaired t test was applied to compare the sizes of large adenoma and carcinoma because the size data were in the Gaussian distribution.
high, and low ROIs, the lesion attenuation values are mea- sured 34.2, 36.6, and 33.4 HU on unenhanced CT, respec- tively. Using large, high, and low ROIs, the APW and RPW were 53.4% and 31.4%, 66.9% and 41.0%, and 33.4 HU, 20.9%, respectively.
Fisher exact test was used to compare the frequency of qualitative CT features including tumor margin, fat tissue, cystic or necrotic change, and calcification be- tween large adenoma and carcinoma.
Statistical analysis was performed with SPSS 19.0 for windows (SPSS Inc., Chicago, IL, USA). A P value of <0.05 was considered statistically significant and was shown up to 1/1000 rounded up from 1/10000.
Results
Of 31 large adenomas and 12 carcinomas, CT images showed lobulated margin in 10 (32.3%) and three (25.0%) (P = 0.727), fat tissue in seven (22.6%) and 0 (0%) (P = 0.163), cystic or necrotic change in five (16.1%) and two (16.7%) (P = 1.000), and calcification in two (6.5%) and five (38.5%) (P = 0.012). Calcifi- cation was more frequent in carcinomas than large adenomas. However, there was no difference between large adenoma and carcinoma in terms of tumor margin, fat tissue, and cystic or necrotic change sta- tistically (P > 0.05).
The mean sizes of large adenomas and carcinomas were 4.2 ± 1.1 cm (3.0-7.3 cm) and 6.0 ± 3.3 cm (2.4- 13.1 cm), respectively (P = 0.085) (Fig. 4). There was broad overlap between large adenoma and carcinoma in terms of the size of a lesion.
14
12
Size (cm)
10
8
6
0 00 0
4
2
Large adenoma
Carcinoma
On unenhanced CT, the mean attenuation values of large adenomas and carcinomas using large, high, and low ROIs were 19.9 ± 13.4 and 32.7 ± 6.3 HU, 22.3 ± 16.5 and 33.2 ±6.0 HU, and 19.2 ±15.0 and 29.8 ± 10.3 HU, respectively (Table 1). The mean lesion attenuation values of carcinomas were higher than those of large adenomas, when a large or low ROI was used (P = 0.003-0.035). However, there was no difference between large adenoma and carcinoma when a high ROI was used (P = 0.051).
Using a large ROI, the mean APW and RPW of large adenoma vs. carcinomas were 45.5% ± 46.7% and 38.6% ± 34.2% vs. 6.3% ± 61.8% and 7.8% ± 23.1%; using a high ROI, 75.2% ± 11.3% and 61.4% ± 9.3% vs.
48.8% ± 20.5% and 30.8% ± 14.8%; using a low ROI, - 55.5% ± 195% and -3.7% ± 95.8% vs. - 207.1% ± 477.4% and -27.8% ± 48.9%, respectively (Table 2) (Figs. 2, 3). The mean APW and RPW of carcinomas were lower than those of large adenomas, when a large or high ROI was used (P = 0.000-0.008). However, the mean APW and RPW of carcinoma were not different from those of large adenoma when a low ROI was used (P = 0.055- 0.066).
Of 31 large adenomas, 20, 31, and 16 had ≥60% APW or ≥40% RPW using large, high, and low ROIs, respec- tively. Of 12 carcinomas, six had ≥60% APW or ≥40% RPW using a high ROI alone. Overall CT sensitivities for carcinoma using large, high, and low ROIs were 100% (12/ 12), 50.0% (6/12), and 100% (12/12), respectively (Fig. 5).
Sensitivity, specificity, positive predictive value, and negative predictive value, for characterizing large adeno- mas using a large ROI were 64.5% (20/31), 100% (12/12), 100% (31/31), and 52.2% (12/23); those using a high ROI were 100% (31/31), 50.0% (6/12). 83.8% (31/37), and 100% (6/6); those using a low ROI were 51.6% (16/31), 100% (12/ 12), 100% (16/16), and 44.4% (12/27), respectively (Ta- ble 3). Overall accuracies for diagnosing large adenoma using a large ROI, high ROI, and small ROI were 74.4% (32/43), 86.0% (37/43), and 65.1% (28/43), respectively. Even though a high ROI provided the highest accuracy for large adenoma, it misdiagnosed carcinoma in six patients (50%) as adenoma. A large ROI provided higher accuracy for large adenoma than a low ROI but also correctly diagnosed all patients with carcinoma.
Discussion
Our results showed that unenhanced CT or washout CT sensitivity for large adenoma varied widely according to
| Types of ROI | Unenhanced CT attenuation values | P value | |
|---|---|---|---|
| Large adenoma | Carcinoma | ||
| Large ROI | 19.9 ± 13.4 HU(-13.5-39.3 HU) | 32.7 ± 6.3 HU(23.3-42.9 HU) | 0.003 |
| High ROI | 22.3 ± 16.5 HU(-26.2-45.9 HU) | 33.2 ± 6.0 HU(25.6-46.1 HU) | 0.051 |
| Low ROI | 19.2 ± 15.0 HU(-17.0-40.8 HU) | 29.8 ± 10.3 HU(11.6-47.4 HU) | 0.035 |
Data are mean ± standard deviation (range)
ROI, region-of-interest; HU, Hounsfield unit
| Types of ROI | CT washout | Large adenoma | Carcinoma | P value |
|---|---|---|---|---|
| Large ROI | APW (%) | 45.5 ± 46.7 (-95.7-97.4) | 6.3 ± 61.8 (-164.3-53.4) | 0.008 |
| RPW (%) | 38.6 ± 34.2 (-52.5-91.9) | 7.8 ± 23.1 (-43.4-32.5) | 0.003 | |
| High ROI | APW (%) | 75.2 ± 11.3 (49.7-104.2) | 48.8 ± 20.5(1.2-67.8) | <0.001 |
| RPW (%) | 61.4 ± 9.3(38.9-85.8) | 30.8 ± 14.8 (0.6-48.7) | <0.001 | |
| Low ROI | APW (%) | -55.5 ± 194.8 (-707.7-97.5) | -207.1 ± 426.5 (-1255.2-48.5) | 0.055 |
| RPW (%) | -3.7 ± 95.8 (-305.8-101.7) | -27.8 ± 48.9 (-129-19.7) | 0.066 |
Data are mean ± standard deviation (range)
ROI, region-of-interest; HU, Hounsfield unit; APW, absolute percentage washout; RPW, relative percentage washout
| CT performance | Types of ROI | ||
|---|---|---|---|
| Large ROI | High ROI | Low ROI | |
| Sensitivity | 64.5% (20/31)[46.9%-79.0%]* | 100% (31/31)[86.9%-100%] | 51.6% (16/31)[34.8%-68.0%] |
| Specificity | 100% (12/12)[71.8%-100%] | 50.0% (6/12)[25.4%-74.6%) | 100% (12/12)[71.8%-100%] |
| Positive predictive value | 100% (31/31)[86.9%-100%] | 83.8% (31/37)[68.4%-92.7%] | 100% (16/16)[77.3%-100%] |
| Negative predictive value | 52.2% (12/23)[33.0%-70.8%] | 100% (6/6)[55.7%-100%] | 44.4% (12/27)[27.6%-62.7%] |
| Overall accuracy | 74.4% (32/43)[59.6%-85.2%] | 86.0% (37/43)[72.4%-93.8%] | 65.1% (28/43)[50.1%-77.6%] |
* Data are 95% confidence intervals
ROI, region-of-interest
ROI size or location. Large adenoma and carcinoma were overlapped in terms of qualitative or quantitative CT imaging features.
Almost all adrenal masses measuring ≤4 cm or less are considered benign conditions although the size of an adrenal mass is not a definitive indicator of benignity [13, 14]. Thus, incidental adrenal masses >4 cm are of con- cern to clinicians because of an increasing likelihood for carcinoma in non-oncologic patients [15-17]. This size threshold achieved a high sensitivity (93%) for carcinoma but the specificity (42%) was poor [18]. Several studies reported that the mean sizes of most adenomas were 2.1- 2.4 cm and that only a small number of adenomas were measured around 4.1-6.0 cm [1-4]. Our study, however, showed that the size of a lesion was much overlapped between large adenoma and carcinoma and besides, that 25% (3/12) carcinomas were measured <3 cm. There- fore, using the size criterion alone without CT evaluation may lead to unnecessary surgery to patients with large adenoma and delayed cancer detection to patients with small carcinoma, as well.
On unenhanced CT images, homogeneous attenua- tion has been reported in 87% of adenomas [19]. According to the size or location of a ROI, however, the lesion attenuation value on unenhanced CT may vary even within the same adenoma because the amount of
intra-cytoplasmic lipid is different in each tumor cell and because tumor cells are not evenly distributed. Subse- quently, it may influence unenhanced CT sensitivity for lipid-rich adenoma. Our study showed that unenhanced CT sensitivities for large adenoma ranged from 19.4% (6/ 31) to 29.0%(9/31) according to the size or location of a ROI. Therefore, the number of lipid-rich adenomas was identified from 6 to 9 alone among 31 large adenomas. Large adenoma had a smaller number of lipid-rich ade- noma than did small adenomas for which unenhanced CT sensitivity is approximately 70% [20]. A small ROI that fits to the lowest attenuated area should be used to maximize unenhanced CT sensitivity for lipid-rich ade- noma because no carcinomas were measured ≤10 HU on unenhanced CT regardless of ROI size or location.
On post-contrast CT images, homogenous attenua- tion has been reported only in 58% of adenomas [19]. For this reason, greater care should be taken to determine the location of an ROI or to control the size of an ROI for measurement of lesion attenuation values on post-con- trast CT than unenhanced CT. The CT sensitivities for large adenoma were widely ranged in our study. Both large and low ROIs achieved 100% (12/12) CT sensitivity for carcinoma, while those did 64.5% (20/31) and 51.6% (16/31) CT sensitivity for large adenoma. In contrast, a high ROI alone achieved 100% CT sensitivity for large
100
90
80
CT Sensitivity (%)
70
60
Large ROI
50
High ROI
Low ROI
40
30
20
10
0
Large adenoma
Carcinoma
S. Y. Park et al .: CT sensitivities for large adrenal adenoma and cortical carcinoma
adenoma, while achieving 50% (6/12) CT sensitivity for carcinoma. If a carcinoma is misdiagnosed as an ade- noma using a high ROI, the lesion may progress at 6 moths or later on a follow-up CT examination. This is why adenoma is known to be stable for at least 6 months, and thus regular follow-up is a management of choice for adenoma. Therefore, a large or low ROI should be used not to miss any carcinoma, even though some large adenomas are surgically treated. However, our study showed that a large ROI was preferable to a low ROI because of slightly better CT sensitivity for large adenoma.
Szolar et al. [7] reported that both sensitivity and specificity for adenoma were 100% compared with carci- nomas at dedicated adrenal CT images. In their study, however, the mean size of adenomas was 2.2 ± 0.8 cm, and thus most of the adenomas were likely to be <3 cm. Most of these adenomas were not histologically con- firmed. Moreover, they included 36.4% (4/11) advanced carcinomas that already had metastasized to other organs at initial CT scan. Quantitative assessment using lesion attenuation value or percentage loss of enhancement is meaningless in these adrenal incidentaloma because ade- noma does not metastasize. As a result, carcinoma with metastasis should have been excluded from their data analysis, as our study in which five patients were excluded from 17 patients with carcinoma due to invasion to adja- cent organs or distant metastasis at initial CT scan. To measure lesion attenuation value, many studies adopted a large ROI enough to cover more than half of an adrenal mass [1-5]. However, these studies did not provide any evidence of why a large ROI was chosen instead of a small ROI. Therefore, it remains unclear why a large ROI should be used for lesion attenuation measurement. When a small cancer focus exists in the adenoma, a large ROI covering more than half of a lesion can increase the probability for non-adenoma than a small ROI covering less than half of a lesion. In contrast, using a small ROI may make a false negative or positive diagnosis, when it is placed in the highest or lowest area on post-contrast CT images. Therefore, the lesion attenuation value that is measured by a large ROI may be considered a represen- tative value for an adrenal incidentaloma.
Previous studies reported that CT examination showed high sensitivities for adenoma [1-4]. However, the majority of the adrenal masses included in these studies were measured <3 cm. Some investigators re- ported high agreement between training resident and genitourinary specialist when they independently re- viewed dedicated adrenal CT images of small adenomas [21, 22]. In contrast to excellent CT diagnostic perfor- mance for small adenoma, CT sensitivity for large ade- noma was relatively poor in our study because increasing the size of adenoma decreases CT sensitivity.
Our study had some limitations. First, CT images were evaluated in consensus by two radiologists. Gener-
ally, consensus interpretation is considered a more sub- jective analysis than independent interpretation which can show high correlation of inter-reader agreement in making a diagnosis of adenoma or non-adenoma. How- ever, if independent analysis had been conducted, we should have analyzed a lot of quantitative data that were obtained from three ROIs. If a lesion attenuation value had been measured by two independent readers using three ROIs, six lesion attenuation values would have been obtained at each lesion. Subsequently, a dilemma would have been posed about which one should be chosen as a representative value. Consensus evaluation for evaluating adrenal masses can be adopted in cases where experienced radiologists are involved in image analysis [5, 23-25]. Second, CT findings were correlated with pathologic re- port alone. When we began the retrospective study, we did not have surgical specimens but some representative slides in most cases. Thus, a lesion-by-lesion correlation between CT and pathologic examination was impossible. Third, the number of carcinomas was relatively small. Of 17 patients with histologically proven carcinoma, five were excluded because of tumor invasion or distant metastasis at initial CT scan. Fourth, CT diagnostic performance except sensitivity was over-estimated be- cause the other non-adenomas except carcinoma were excluded. Fifth, our study was retrospective. Selection bias was possible because majority of adenoma is not surgically treated but just followed up.
In conclusion, overall CT sensitivity for large ade- noma varies widely according to the size or location of an ROI and it is much lower than that of small adenoma (<3 cm) that were previously reported studies [1-5, 20- 22, 25, 26]. Moreover, We cannot rely on qualitative CT imaging features in order to differentiate large adenoma from carcinoma because of broad overlap of lesion size, cystic or necrotic change, and tumor margin between these two lesions. Although a large ROI does not provide better CT sensitivity for large adenoma than a small ROI fitted to the highest attenuated area alone, it can achieve 100% sensitivity for carcinoma. Also, a large ROI can provide slightly better CT sensitivity for large adenoma than a small ROI fitted to the lowest attenuation area alone. Therefore, a ROI covering more than half of a lesion should be used for evaluating large adrenal masses (≥3 cm), which minimizes loss of CT sensitivity for large adenoma and aids in detection of carcinoma.
Conflict of interest. None of authors have conflict or interest or financial support to disclose.
References
1. Caoili EM, Korobkin M, Francis IR, Cohan RH, Dunnick NR (2000) Delayed enhanced CT of lipid-poor adrenal adenomas. AJR Am J Roentgenol 175:1411-1415
2. Caoili EM, Korobkin M, Francis IR, et al. (2002) Adrenal masses: characterization with combined unenhanced and delayed enhanced CT. Radiology 222:629-633
S. Y. Park et al .: CT sensitivities for large adrenal adenoma and cortical carcinoma
3. Korobkin M, Brodeur FJ, Francis IR, et al. (1998) CT time- attenuation washout curves of adrenal adenomas and nonadeno- mas. AJR Am J Roentgenol 170:747-752
4. Pena CS, Boland GW, Hahn PF, Lee MJ, Mueller PR (2000) Characterization of indeterminate (lipid-poor) adrenal masses: use of washout characteristics at contrast-enhanced CT. Radiology 217:798-802
5. Park BK, Kim CK, Kim B, Lee JH (2007) Comparison of delayed enhanced CT and chemical shift MR for evaluating hyperattenu- ating incidental adrenal masses. Radiology 243:760-765
6. Paulsen SD, Nghiem HV, Korobkin M, Caoili EM, Higgins EJ (2004) Changing role of imaging-guided percutaneous biopsy of adrenal masses: evaluation of 50 adrenal biopsies. AJR Am J Roentgenol 182:1033-1037
7. Szolar DH, Korobkin M, Reittner P, et al. (2005) Adrenocortical carcinomas and adrenal pheochromocytomas: mass and enhance- ment loss evaluation at delayed contrast-enhanced CT. Radiology 234:479-485
8. Park BK, Kim B, Ko K, Jeong SY, Kwon GY (2006) Adrenal masses falsely diagnosed as adenomas on unenhanced and delayed contrast-enhanced computed tomography: pathological correla- tion. Eur Radiol 16:642-647
9. Johnson PT, Horton KM, Fishman EK (2009) Adrenal mass imaging with multidetector CT: pathologic conditions, pearls, and pitfalls. Radiographics 29:1333-1351
10. Newhouse JH, Heffess CS, Wagner BJ, et al. (1999) Large degen- erated adrenal adenomas: radiologic-pathologic correlation. Radi- ology 210:385-391
11. Aubert S, Wacrenier A, Leroy X, et al. (2002) Weiss system revis- ited: a clinicopathologic and immunohistochemical study of 49 adrenocortical tumors. Am J Surg Pathol 26:1612-1619
12. Lau SK, Weiss LM (2009) The Weiss system for evaluating adre- nocortical neoplasms: 25 years later. Hum Pathol 40:757-768
13. NIH (2002) NIH state-of-the-science statement on management of the clinically inapparent adrenal mass (“incidentaloma”). NIH Consens State Sci Statements 19:1-25
14. Kapoor A, Morris T, Rebello R (2011) Guidelines for the man- agement of the incidentally discovered adrenal mass. Can Urol Assoc J 5:241-247
15. Angeli A, Osella G, Ali A, Terzolo M (1997) Adrenal incidenta- loma: an overview of clinical and epidemiological data from the National Italian Study Group. Horm Res 47:279-283
16. Terzolo M, Ali A, Osella G, Mazza E (1997) Prevalence of adrenal carcinoma among incidentally discovered adrenal masses. A ret- rospective study from 1989 to 1994. Gruppo Piemontese Inci- dentalomi Surrenalici. Arch Surg 132:914-919
17. Mantero F, Arnaldi G (2000) Management approaches to adrenal incidentalomas. A view from Ancona, Italy. Endocrinol Metab Clin North Am 29:107-125, ix
18. Mantero F, Terzolo M, Arnaldi G, et al. (2000) A survey on adrenal incidentaloma in Italy. Study Group on Adrenal Tumors of the Italian Society of Endocrinology. J Clin Endocrinol Metab 85:637-644
19. Ctvrtlik F, Herman M, Student V, Ticha V, Minarik J (2009) Differential diagnosis of incidentally detected adrenal masses re- vealed on routine abdominal CT. Eur J Radiol 69:243-252
20. Boland GW, Lee MJ, Gazelle GS, et al. (1998) Characterization of adrenal masses using unenhanced CT: an analysis of the CT liter- ature. AJR Am J Roentgenol 171:201-204
21. Choi YA, Kim CK, Park BK, Kim B (2013) Evaluation of adrenal metastases from renal cell carcinoma and hepatocellular carcinoma: use of delayed contrast-enhanced CT. Radiology 266:514-520
22. Seo JM, Park BK, Park SY, Kim CK (2014) Characterization of lipid-poor adrenal adenoma: chemical shift MRI and washout CT. AJR Am J Roentgenol 202(5):1043-1050
23. Kumar R, Xiu Y, Yu JQ, et al. (2004) 18F-FDG PET in evaluation of adrenal lesions in patients with lung cancer. J Nucl Med 45:2058-2062
24. Boland GW, Blake MA, Holalkere NS, Hahn PF (2009) PET/CT for the characterization of adrenal masses in patients with cancer: qualitative versus quantitative accuracy in 150 consecutive patients. AJR Am J Roentgenol 192:956-962
25. Kim YK, Park BK, Kim CK, Park SY (2013) Adenoma charac- terization: adrenal protocol with dual-energy CT. Radiology 267:155-163
26. Korobkin M, Brodeur FJ, Francis IR, et al. (1996) Delayed en- hanced CT for differentiation of benign from malignant adrenal masses. Radiology 200:737-742