Original Research
Adrenocortical Carcinoma: MR Imaging Appearance with Current Techniques1
James F. Schlund, MD . Philip J. Kenney, MD . Elizabeth D. Brown, MD . Susan M. Ascher, MD Jeffrey J. Brown, MD . Richard C. Semelka, MD
The purpose of the study was to define the mag- netic resonance (MR) imaging appearance of adrenocortical carcinoma (ACC) with current MR techniques. Eight patients with ACC underwent high-field-strength MR imaging with the following sequences: axial T1-weighted gradient echo, fat- suppressed T1-weighted spin echo, fat-suppressed T2-weighted spin echo, and gadolinium-enhanced T1-weighted gradient echo and fat-suppressed T1- weighted spin echo. Postcontrast images were also acquired in the sagittal (six patients) and coronal (three patients) planes. Out-of-phase gra- dient-echo images were obtained in two patients. Lesion morphology, signal intensity features, and presence of metastatic disease were retrospec- tively evaluated. MR imaging features of ACC in- cluded internal hemorrhage (seven of eight pa- tients), central necrosis (seven of eight), and pe- ripheral enhancing nodules (seven of eight). Out-of-phase images in two of two patients dem- onstrated signal loss compared with in-phase im- ages, which may be a common feature of these tu- mors. Liver metastases were present in four pa- tients. Consistent MR features of ACC are identified.
Index terms: Adrenal gland, MR. 862.121415. 862.12143 · Adrenal gland. neoplasms, 862.324 . Contrast enhancement . Gadolinium
JMRI 1995: 5:171-174
Abbreviation: ACC = adrenocortical carcinoma.
1 From the Department of Radiology, Campus Box 7510. Old Clinic Bldg. University of North Carolina, Chapel Hill. NC 17599-7510 (J.F.S., E.D.B., R.C.S.): Department of Radiology. University of Alabama at Birmingham (P.J.K.): Department of Radiology, Georgetown University Medical Center. Washington, DC (S.M.A.); and Mallinckrodt Institute of Radiology. Wash- ington University, St Louis, Mo (J.J.B.). Received May 24, 1994: revision requested July 1; revision received July 14: accepted July 15. Address reprint requests to R.C.S.
· SMR. 1995
ADRENOCORTICAL carcinoma (ACC) is a rare ma- lignancy with an annual incidence of 0.0006%- 0.00017% (1). The majority of reports concerning the imaging of ACC are based on computed tomographic (CT) findings (2-7), with a relative paucity of reports describing the magnetic resonance (MR) appearance of these tumors (7-12). The use of MR imaging for characterizing adrenal masses as benign or malig- nant has been described, with T2-weighted images, serial contrast agent-enhanced gradient-echo im- ages, contrast-enhanced fat-suppressed T1-weighted spin-echo images, and out-of-phase images (11-15). Understanding the spectrum of appearance of ACC is essential because it may have features that over- lap with those of benign disease. We retrospectively examined MR images from patients with ACC from four university hospitals in which newer techniques of fast imaging, gadolinium enhancement, and out- of-phase imaging were used to demonstrate the MR appearance of these tumors.
. MATERIALS AND METHODS
MR imaging was performed in eight patients with ACC between December 1989 and November 1993. There were seven women and one man, aged 25-73 years (mean, 45.5 years). All cases of ACC were pathologically proved by means of surgical resection (n = 5), open surgical biopsy (n = 2), or needle aspira- tion cytology (n = 1).
MR imaging was performed with a 1.0-T (n = 2) or 1.5-T (n = 6) MR system (Impact and SP4000, re- spectively; Siemens Medical Systems, Iselin, NJ). Im- ages were acquired with the following sequences: fat- suppressed T1-weighted (TR msec/TE msec = 500/ 15) spin echo, fat-suppressed T2-weighted (2,300- 2,400/13, 80-90) spin echo, and T1-weighted (130- 140/4 [1.5 T], 140-150/6 [1.0 T]) spoiled gradient echo. Gadopentetate dimeglumine was administered at a dose of 0.1 mmol/kg, and spoiled gradient-echo and fat-suppressed spin-echo images were again ac- quired after contrast agent administration. T1- weighted sagittal (n = 6) and coronal (n = 3) images were also acquired after contrast agent administra- tion. Two patients underwent precontrast out-of- phase T1-weighted (150/6 [1.5 T]/ 120/4 [1.0 T]) spoiled gradient-echo imaging.
Lesions were characterized retrospectively by two investigators (R.C.S., J.F.S.). ACCs were evaluated
for size, presence of central necrosis, internal hemor- rhage. enhancement pattern, signal loss on out-of- phase images (two patients), and distant spread. Ne- crosis was described as central, large irregular areas of signal void on T1-weighted images after gadolin- ium enhancement, and internal hemorrhage was de- scribed as areas of high signal intensity on T2- weighted images and precontrast T1-weighted spoiled gradient-echo and fat-suppressed images. Signal loss on out-of-phase images was visually as- sessed.
· RESULTS
The ACCs were solitary tumors in all patients and ranged in diameter from 3.5 to 14 cm, with an aver- age maximum diameter of 10.7 cm. Tumors arose from the right adrenal gland in four patients and from the left adrenal gland in the other four. All le- sions, except the single 3.5-cm-diameter tumor, demonstrated internal hemorrhage, which was seen as ill-defined areas of high signal intensity on T1- weighted images. Internal hemorrhage was predomi- nantly central. Large areas of central necrosis were noted in all seven large ACCs, and uniform central low signal intensity was seen in the 3.5-cm tumor. After gadolinium administration, seven of the eight lesions demonstrated marked enhancement, with the presence of periphery-based nodules (Fig 1 ). This appearance was accentuated by the presence of central necrosis and internal hemorrhage, which did not enhance (Fig 2). Out-of-phase images in two of two patients with 8- and 9-cm-diameter cancers, re- spectively, demonstrated areas of signal loss com- pared with in-phase images. In one of these tumors, the region of signal loss measured approximately 15% of the total tumor volume (Fig 3). Loss of signal was more clearly depicted on out-of-phase images than on fat-suppressed images in both patients.
On sagittal images in six of six patients, the extrarenal origin of the tumor was well shown in pro- file, with preservation of an intact renal cortex (Fig 4). However, coronal images failed to demonstrate the extrarenal origin of the tumor in all three pa- tients in whom this sequence was used.
Metastatic or invasive disease was identified on MR images of six of the eight patients. In four of the eight, including the patient with the 3.5-cm tumor, liver metastases were noted. These were best demon- strated on fat-suppressed T2-weighted spin-echo im-
ages as ill-defined, focal high-intensity lesions. Ir- regular rim enhancement was apparent on images obtained immediately after contrast agent adminis- tration. One patient had direct invasion of the inferior vena cava. The sixth patient with tumor spread had multiple pulmonary nodules compatible with pulmo- nary metastases. The pulmonary lesions were noted on the proton-density and T2-weighted spin-echo im- ages used for evaluation of liver metastases. The lung metastases appeared as circumscribed nodules of moderate signal intensity in the areas of signal void representing aerated lung.
· DISCUSSION
CT features of ACC include a diameter greater than 5 cm, central areas of low attenuation, calcifications, and hepatic, nodal, or venous spread (2,3,7). Consis- tent MR features in our series were central necrosis (seven of eight patients), peripheral nodules of en- hancement (seven of eight), and internal hemorrhage (seven of eight). The greater sensitivity of MR imaging for detecting hemorrhage (16,17) and the presence or absence of intravenous contrast enhancement (16- 19) may explain why these features have not been previously emphasized with CT. Hemorrhage is fre- quently identified in pathologic specimens (20,21).
9
a.
b.
c.
Regarding MR imaging techniques, T2-weighted images were important for the evaluation of liver me- tastases (present in four of eight patients). Precon- trast fat-suppressed T1-weighted images were useful for demonstrating high-intensity blood, and postcon- trast T1-weighted images demonstrated central hypovascularity and peripheral nodularity. The sag- ittal plane was useful for demonstrating the extra- renal origin of the tumors. We found that the coronal plane less useful for this purpose because the tissue interface was observed en face. The presence of an uninterrupted renal cortex on either immediately postcontrast spoiled gradient-echo or fat-suppressed T1-weighted spin-echo images is also helpful for separating adrenal from renal tumors (11,12).
In both patients in whom out-of-phase images were obtained, regions of signal loss were identified relative to in-phase images. This is of note because
signal loss on out-of-phase images has been consid- ered a reliable indication that an adrenal mass is a benign adenoma (13). The underlying basis for this observation is that both normal adrenal tissue and benign adenomas contain cortisol and related fatty hormone precursors. However, the series of Mitchell et al (13) did not include any ACCs. Since ACCs fre- quently are hormonally active (1,3), and hence con- tain some form of fat, it is not surprising that these lesions exhibited signal loss on out-of-phase images. The signal loss was heterogeneous, however; uniform loss of signal may be consistent with a benign pro- cess. The degree of signal loss in small ACCs is un- known. Out-of-phase images were superior to fat- suppressed images because functional adrenal tis- sue contains both fat and water tissue. On out-of- phase images, signal loss is maximal when fat and water exist in equal proportions; on fat-suppressed
a.
b.
c.
d.
a.
b.
images signal loss is maximal when fat approximates 100% of the tissue.
MR imaging documented extra-adrenal spread of tumor in six of eight patients. Although CT is prob- ably superior for the evaluation of pulmonary metas- tases, their presence was well shown on the proton- density and fat-suppressed T2-weighted spin-echo images of one patient. The advantages of MR imaging may be in the demonstration of liver metastases and invasion of the inferior vena cava. Because of the multiplanar capability of MR imaging, direct inva- sion of the liver by tumor is well shown on either sagittal or coronal MR images. Sagittal images dem- onstrated the tumor interface with liver and kidneys because these interfaces were shown in profile. Di- rect extension of tumor into the inferior vena cava was documented in one case by the presence of a filling defect invading the bright column of enhanced blood in the vessel on postcontrast gradient-echo im- ages. Other bright blood or dark blood techniques could be used to demonstrate venous involvement.
Our study was limited by the relatively small number of tumors in our series, despite acquisition of cases from several institutions. A large prospective series, including benign and malignant adrenal masses, would be required to determine the specific- ity and diagnostic accuracy of the above-described MR imaging criteria for the diagnosis of ACC. In par- ticular, out-of-phase images of ACC and adrenal ad- enoma must be compared in a larger series to deter- mine if reliable differences in the pattern of signal loss can be shown.
In conclusion, in our series of eight ACCs, MR im- aging proved to be excellent for demonstrating char- acteristics of this rare tumor.
References
1. Copeland PM. The incidentally discovered adrenal mass. Ann Intern Med 1983: 98:940-945.
2. Dunnick NR. Heaston D, Halvorsen R, et al. CT appear- ance of adrenal cortical carcinoma. J Comput Assist Tomogr 1982; 6:978-982.
3. Fishman EK, Deutch BM, Hartman DS, Goldman SM, Zerhouni EA. Siegelman SS. Primary adrenocortical car- cinoma: CT evaluation with clinical correlation. AJR 1987: 148:531-535.
4. Lee MJ, Hahn PF, Papanicolaou N, et al. Benign and ma- lignant adrenal masses: CT distinction with attenuation
coefficients, size, and observer analysis. Radiology 1991; 179:415-418.
5. Berland LL, Koslin DB, Kenney PJ. Stanley RJ, Lee JY. Differentiation between small benign and malignant adre- nal masses with dynamic incremented CT. AJR 1988: 51:95-101.
6. Hussain S. Belldegrun A. Seltzer SE, et al. Differentia- tion of malignant from benign adrenal masses: predictive indices on computed tomography. AJR 1985; 144:61-65.
7. Francis IR, Gross MD, Shapiro B, Lorobkin M. Quint LE. Integrated imaging of adrenal disease. Radiology 1992: 184:1-13.
8. Smith SM, Patel SK. Turner DA, Matalon TAS. Magnetic resonance imaging of adrenal cortical carcinoma. Urol Radiol 1989; 11:1-6.
9. Dunnick NR. Adrenal carcinoma. Radiol Clin N Am 1994: 32:99-148.
10. DeAtkine AB. Dunnick NR. The adrenal glands. Semin Oncol 1991: 18:131-139.
11. Chang A. Blazer HS, Lee JKT, Heiken JP. Adrenal gland: MR imaging. Radiology 1987: 163:123-128.
12. Krestin GP. Steinbrich W, Friedmann GI. Adrenal masses: evaluation with fast gradient-echo MR imaging and Gd-DTPA-enhanced dynamic studies. Raidology 1989: 171:675-680.
13. Mitchell DG. Crovello M, Matteucci T. Petersen RO, Miet- tinen MM. Benign adrenocortical masses: diagnosis with chemical shift MR imaging. Radiology 1992: 185:345-351.
14. Tsushima Y. Ishizaka H, Matsumoto M. Adrenal masses: differentiation with chemical shift, fast low-angle shot MR imaging. Radiology 1993: 186:705-709.
15. Reinig JW. MR imaging differentiation of adrenal masses: has the time finally come? Radiology 1992: 185:339-340.
16. Semelka RC, Shoenut JP, Magro CM, Kroeker MA, MacMahon R. Greenberg HM. Renal lesions: controlled comparison between CT and 1.5-T MR imaging with non- enhanced and gadolinium-enhanced fat-suppressed spin- echo and breath-hold FLASH techniques. Radiology 1992; 182:425-430.
17. 7. Semelka RC, Shoenut JP, Lawrence PH. et al. Evaluation of adrenal masses with gadolinium enhancement and fat- suppressed MR imaging. JMRI 1993; 3:337-343.
18. Semelka RC. Shoenut JP, Magro CM, Kroeker MA. MacMahon R. Greenberg HM. Renal cancer staging: comparison of contrast-enhanced CT and gadolinium-en- hanced fat-suppressed spin-echo and gradient-echo MR imaging. JMRI 1993; 3:597-602.
19. Semelka RC, Shoenut JP, Kroeker MA, et al. Bile duct disease: prospective comparison of ERCP, CT and fat sup- pression MRI. Gastrointest Radiol 1992; 17:347-352.
20. Warner NE, Strauss FH II. The adrenal gland. In: Silver- berg SG, ed. Principles and practice of surgical pathology. New York, NY: Churchill Livingstone, 1990; 1957-1979.
21. Grahn JG, Gould VE. The adrenal glands. In: Kissane JM. ed. Anderson’s pathology. Vol 2. St Louis, Mo: Mosby, 1990; 1602-1603.