CT and Angiography in Adrenocortical Carcinoma

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: CT and Angiography in Adrenocortical Carcinoma F. Kolmannskog, A. Kolbenstvedt & I. B. Brekke To cite this article: F. Kolmannskog, A. Kolbenstvedt & I. B. Brekke (1992) CT and Angiography in Adrenocortical Carcinoma, Acta Radiologica, 33:1, To link to this article: Published online: 04 Jan Submit your article to this journal Article views: 60 Full Terms & Conditions of access and use can be found at

2 Acra Radiulugica 33 (1992) Fasc. I FROM THE DEPARTMENT OF RADIOLOGY AND SURGICAL DEPARTMENT B, THE NATIONAL HOSPITAL, OSLO, NORWAY. CT AND ANGIOGRAPHY IN ADRENOCORTICAL CARCINOMA F. KOLMANNSKOG, A. KOLBENSTVEDT and I.B. BREKKE Abstract CT and angiography were performed in 15 patients with adrenocortical carcinoma. The tumors had a mean diameter of II cm (range 4-20 cm). At CT, the largest tumors were ill-defined, and in these, the organ of tumor origin could not be established. Angiographically the correct organ of tumor origin was established in all but one patient. It is concluded that CT is excellent in showing the extent of an adrenal tumor, but is often unable to predict the organ of origin in large tumors. Angiography is still of great value in the preoperative work-up in patients with large adrenocortical carcinomas for correct identification of tumor origin and for vascular mapping. Key wiurds: Adrenal gland, neoplasms; -, CT; -, angiography. Adrenocortical carcinoma is a rare tumor. During a 15- year period (197C194) only 99 patients were recorded at The Cancer Registry of Norway (population 4.25 million) (S. 0. Toresen, personal communication 199). CT is the primary and probably the best imaging modality for localizing adrenal lesions (2, 5, 6,, 15). Specific CTfindings in adrenocortical carcinomas (3,s) and differentiation between malignant and benign adrenal tumors using CT have been reported (1 3). Since adrenal carcinomas are usually large when diagnosed, it may be difficult to decide the organ of tumor origin at CT (5, 17). Angiography is valuable for demonstration of the vascular supply, and hence in defining from which organ the tumor arises (7,, 17). The aim of the present study was to report on our experience with adrenocortical carcinomas using CT and angiography, and to assess if angiography is still necessary in the preoperative evaluation of these patients. Material and Methods Fifteen patients with histologically verified adrenocortical carcinoma were examined with CT and angiography. Pa- tient characteristics, survey of their main symptoms, and operative procedures are summarized in Table 1. Five patients had hormonally active tumors (Cushing syndrome), and 7 patients were referred for CT examination due to uncharacteristic abdominal symptoms or weight loss. The remaining 3 patients were examined due to paraparesis in one, search for a primary tumor in another with cerebral metastases, and a palpable abdominal mass in a third. Of the 7 patients with uncharacteristic abdominal symptoms, 3 were primarily examined with ultrasonography (US) due to suspected gallstone disease. US revealed a retroperitoneal mass, which prompted further investigation with CT and angiography. Five patients are alive (Nos 6, 10, 1 I, 13, 15) 1 to 7 years (mean 4 years) after the operation. Eight patients died l to 1 months (mean 7 months) following surgery, and one patient with cerebral metastases died 3 months after the diagnosis was made. One patient was lost to follow-up. The CT examinations were performed at our department in 11 patients using a second generation scanner (DeltaScan 50 FS, Ohio Nuclear) with contiguous 13-mm collimated slices in 5 patients and a third generation scanner (General Electric 900) with contiguous 10-mm thick slices in 6 patients. Four patients had CT examinationsat other hospitals before referral to our institution. Hence, different scanning techniques were used. All patients, however, had scans starting at the level of the diaphragm to below the kidneys and all examinations were performed after peroral contrast medium administration. Seven patients were examined both before and after intravenous contrast medium injection, 4 only had precontrast scans, and 4 only postcontrast examinations. On each scan, the size, tumor calcifications, contrast enhancement, shape, and outline of the tumor were analyzed. An attempt was made to reevaluate whether or Accepted for publication 21 August

3 46 F. KOLMANNSKOG ET AL. Table 1 Patient characteristics, main symptoms, and operative findings in 15 patients with adrenocortical carcinoma Pat. No./ Hormonally Main symptom Surgery AgeISex active Operative procedure Tumor Metastasis weight (g) F Cushing Adrenalectomy l5/m Cushing Adrenalectomy, nephrectomy 3500 Liver 3157lF Weight loss Adrenalectomy, nephrectomy 0 Lung 41561F Paraparesis Adrenalectomy 650 Lung, liver 5132lF Cushing Adrenalectomy 30 Liver 6l341F Cushing Adrenalectomy 25 1/661M Weight loss Adrenalectomy 3300 Liver, lymph nodes /54/M Abdominal pain* Adrenalectomy 350 9/49/F Abdominal pain* Adrenalectomy, nephrectomy /41lM Abdominal pain Adrenalectomy, splenectomy, 1500 pancreatic resection 11 /66/F Abdominal pain* Adrenalectomy 66 /71 / M Cerebral symptoms Not performed due to cerebral metastases M None** Adrenalectomy /62/M Cushing Inoperable tumor F Abdominal pain Adrenalectomy 600 *US performed due to suspicion of gallstones disclosed retroperitoneal mass. **Palpable abdominal mass discovered at routine clinical check-up not the adrenal origin of the tumors could be predicted by CT. Abdominal aortography was performed in all 15 patients, and all had one or more selective catheterizations of adrenal or other abdominal arteries (renal artery with inferior adrenal artery 14, celiac artery 13, middle adrenal artery 5, inferior phrenic artery 4, splenic artery 1, and superior mesenteric artery 1). On the angiograms we noted the degree of vascularization and which arteries supplied the tumor. The presence of abnormal vessels, arteriovenous shunting, and tumor blush were recorded. From the angiograms we also gave our opinion about which organ the tumor originated from. All CT scans and angiograms were reviewed retrospectively and compared with the medical and surgical reports. Results Findings at CT and angiography are shown in Table 2. Tumor calcifications were evident in 6 patients. On noncontrast scans 4 tumors with a mean diameter of 7.5 cm (range 4- cm) were homogeneous and 7 tumors with a mean diameter of 13 cm (range -20 cm) were inhomogeneous. On postcontrast scans 3 tumors with a mean diameter of.5 cm (range 6- cm) were homogeneous and with a mean diameter of 13 cm (range 4-20 cm) showed an inhomogeneous appearance. The largest tumors were illdefined and in these the organ of tumor origin could not be established at CT. Two large adrenocortical carcinomas (Nos 2 and 7) were misinterpreted as liver tumors. In one patient we could not decide if the tumor originated from the liver or the adrenal (Fig. 1). In another patient we could not decide if the tumor originated from the kidney or the adrenal (Fig. 2), and in a 3rd patient the liver, kidney and adrenal were the alternative organs of tumor origin. In 6 patients with leftsided tumors, the tumor origin could not be decided in 3, and in one of these the tumor was diagnosed as a renal neoplasm. The remaining 7 adrenocortical carcinomas were correctly diagnosed as adrenal tumors, the largest being 10 cm in diameter (Fig. 3). At angiography 9 tumors were richly to moderately vascularized and 6 poorly vascularized. Tumor stain was shown in 4 but arteriovenous shunting with early venous filling in none. Caudal displacement of the kidney was shown in 10 patients with the largest tumors. Using angiography one tumor with a diameter of 20 cm (No. 7) was misinterpreted as originating from the liver, the remaining 14 were all correctly diagnosed as adrenal tumors. Liver metastases in 4 patients were shown at both modalities, but retroperitoneal lymph node metastases, found at operation in one patient, were not depicted by any met hod. Discussion Adrenocortical carcinomas often remain symptomless until advanced malignancy, a large mass, or hormonal excess. Hormone-producing tumors often do not produce early signs and symptoms since malignant cells are inefficient in the synthesis of steroids (11). Endocrine activity is reported to occur in 33% to 5% of patients with adrenocortical carcinomas (11, 14). Only 5 of our 15 patients (33%) presented with hormonally active tumors. In our series as in

4 CT AND ANGIOGRAPHY IN ADRENOCORTICAL CARCINOMA 47 Table 2 Findings at CT and angiography in 15 patients with adrenocortical carcinoma Pat. CT findings Angiographic findings No. Tumor Calcifi- Inhomogeneous Homogeneous Ill- Tumor Hypo- Moderately Tumor size cation Pre- Post- Pre- Post- defined origin vascular origin (cm) contrast contrast Liver Liver? Kidney? Pancreas? Kidney? Kidney? Liver Pancreas? Kidney? Kidney Liver? Liver a b C Fig. 1. Adrenocortical carcinoma (No. 15). a) CT during i.v. contrast medium injection shows a large inhomogeneous tumor. It is not possible to decide if the tumor originates from the liver or the adrenal. b) Selective arteriography of the right inferior phrenic artery demonstrates rich arterial blood supply from superior adrenal vessels, indicating that the tumor is of adrenal origin. c) Capillary phase showing tumor blush. several others, Cushing syndrome is the most common endocrine manifestation of adrenocortical carcinoma (6, 10, 11, 14). Carcinoma of the adrenal cortex causing primary hyperaldosteronism is rare (16). In the present series all 5 patients with Cushing syndrome had large tumors. This is in disagreement with others, e.g. FISHMAN et al. (6) found that endocrinopathy was present in all 5 adult patients with tumors 6 cm or less in diameter, but in only 11 of 25 with lesions larger than 6 cm. In a review of 307 cases of adrenocortical carcinoma, GUERRERO (9) found carcinomas smaller than 5 cm in 5 patients only, and in 4 the tumors were hormonally functioning. Symptoms causing the patients to seek medical attention usually are abdominal complaints or an abdominal mass. Indications for CT examinations in 7 of our patients were uncharacteristic abdominal symptoms or weight loss. Incidental discovery of asymptomatic adrenal gland tumors has been reported with a frequency of 0.7% of CT examinations of the upper part of the abdomen (1). Five of approximately 000 patients evaluated by abdominal CT

5 4 F. KOLMANNSKOG ET AL. a b Fig. 2. Adrenocortical carcinoma (No. 6). a) Precontrast CT shows a large homogeneous tumor arising either from the upper renal pole or the adrenal. b) Selective renal arteriography demonstrates moderate vascularization of an adrenal tumor supplied by the inferior adrenal artery. a b Fig. 3. Adrenocortical carcinoma (No. 14). a) Postcontrast CT shows an inhomogeneous sharply outlined left adrenal tumor. Adjacent scans demonstrated that this tumor was clearly separated from the pancreas and the kidney. b) Selective arteriography of the middle adrenal artery demonstrates a poorly vascularized adrenal tumor. were incidentally found to have adrenocortical carcinoma (I), measuring by CT from 6 to 15 cm, with a mean of 10 cm. The 10 nonfunctioning adrenocortical carcinomas discovered at CT in our series had a diameter of 4 to 20 cm with a mean of 10 cm. When an adrenal lesion is found incidentally at CT, it is more likely to be a benign nonfunctioning adenoma or an adrenal metastasis than a primary adrenocortical carcinoma (1, 6). The distinction, however, is important, since patients with adrenocortical carcinoma have a median survival of only 6 months if left untreated, but are potentially curable by surgery (1 1). There exists no single criterion to help resolve the differential diagnosis between benign and malignant adrenal lesions using either US or CT (10, 15). The CT criteria used in the distinction between different adrenal tumors are size, calcification, inhomogeneity, and outline. A large inhomogeneous adrenal tumor with calcification and irregular margins is probably an adrenocortical carcinoma. Adrenal me-

6 CT AND ANGIOGRAPHY IN ADRENOCORTICAL CARCINOMA 49 tastases and pheochromocytomas may, however, show the same CT features (, 13). Fine needle aspiration with cytologic evaluation of aspirated material is of little value in distinguishing benign from malignant lesions because of the difficulty in the cytologic definition of malignant adrenal disease (1). Few papers have reported on the role of angiography compared to CT in the diagnostic work-up in adrenocortical carcinoma. DUNNICK et al. (3) performed angiography and CT in 3 patients and correlated the findings. Their conclusion was that the tumor was primarily supplied by adrenal vessels in all 3 patients, and in one patient, the demonstration of arterial supply by adrenal arteries was helpful in predicting an adrenal rather than renal origin of a tumor. TISNADO et al. (17) studied 4 large rightsided adrenal carcinomas by CT and angiography. In 3 of these, the adrenal origin of the tumor could not be ascertained using CT, while arteriography was diagnostic in all, by demonstrating that the tumor was supplied by adrenal arteries. Many authors did not state if the adrenal gland was established as the organ of tumor origin at CT (1, 6,, 13). Others mentioned that large adrenal masses often were difficult to differentiate from large tumors arising in adjacent structures, and in these, angiography provided an advantage over CT, by demonstration of the arterial supply of the tumors (4, 5). However, they did not state if they had performed angiography. In our series, comprising 15 patients, the organ of tumor origin was correctly predicted at CT in the 7 patients with the smallest tumors (mean diameter 7.7 cm, range 4-10 cm). At angiography, tumor origin was correctly estimated in 14 patients. One very large tumor (No. 7) with multiple liver metastases was falsely interpreted as a liver tumor. This adrenal tumor also had large parasitic blood supply from hepatic arteries. Our patients were not examined by US or MR imaging. HAMPER et al. (10) stated that US usually can separate between an adrenal mass and a tumor originating in the liver or upper renal pole by characteristic displacement of retroperitoneal fat planes. Nonfunctioning adenomas may be distinguished from carcinomas by MR imaging (2, lo), but we have not found any report in the literature where the value of MR imaging in diagnosing adrenocortical carcinomas has been evaluated. We conclude that CT should be the primary imaging modality in the evaluation of patients with suspected adrenal lesions. Using CT, the exact localization and characterization of an abdominal mass is usually disclosed, and further imaging methods are not necessary. However, large adrenal masses are often impossible to differentiate from large tumors arising in adjacent organs. In these patients, angiography should be performed, firstly to determine the adrenal origin of the mass, and secondly for preoperative vascular mapping. Request for reprints: Dr. Finn Kolmannskog, Department of Radiology, The National Hospital, N-0027 Oslo I, Norway. REFERENCES 1. BELLDEGRUN A., HUSSAIN S., SELTZER S. E., LOUGHLIN K. R., GITTES R. F. & RICHIE J. P.: Incidentally discovered mass of the adrenal gland. Surg. Gynecol. Obstet. 163 (196) DUNNICK N. R.: Adrenal imaging. Current status. AJR 154 (1990), DUNNICK N. R., HEASTON D., HALVORSEN R., MOORE A. V. & KOROBKIN M.: CT appearance of adrenal cortical carcinoma. J. Comput. Assist. Tomogr. 6 (192), EGHRARI M., MCLOUGHLIN M. J., ROSEN J. E. et al.: The role of computed tomography in assessment of tumoral pathology of the adrenal glands. J. Comput. Assist. Tomogr. 4 (190), EL-SHERIEF M. A,, HEMMINGSSON A. & LORELIUS L. E.: Computed tomography and angiography in the evaluation of adrenal diseases. Acta Radiol. Diagnosis 23 (192), FISHMAN E. K., DEUTCH B. M., HARTMAN D. S., GOLDMAN S. M., ZERHOUNI E. A. & SIEGELMAN S. S.: Primary adrenocortical carcinoma. CT evaluation with clinical correlation. AJR 14 (197), FRITZSCHE P., ANDERSEN C. & CAHILL P.: Vascular specificity in differentiating adrenal carcinoma from renal cell carcinoma. Radiology 5 (1977), GLAZER G. M., FRANCIS J. R. & QUINT L. E.: Imaging of the adrenal glands. Invest. Radiol. 23 (19), GUERRERO L. A.: Diagnostic and therapeutic approach to incidental adrenal mass. Urology 26 (195), HAMPER U. M., FISHMAN E. K., HARTMAN D. S., ROBERTS J. L. & SANDERS R. C.: Primary adrenocortical carcinoma. Sonographic evaluation with clinical and pathologic correlation in 26 patients. AJR 14 (197), 915. I. HENLEY D. J., VAN HEERDEN J. A,, GRANT C. S., CARNEY J. A. & CARPENTER P. C.: Adrenal cortical carcinoma. A continuing challenge. Surgery 94 (193) HOEVELS J. & EKELUND, L.: Angiographic findings in adrenal masses. Acta Radiol. Diagnosis 20 (1979), HUSSAIN S., BELLDEGRUN A., SELTZER S. E., RICHIE J. P., GITTES R. F. & ABRAMS H. L.: Differentiation of malignant from benign adrenal masses. Predictive indices on computed tomography. AJR 144 (195), NADER S., HICKEY R. C., SELLIN R. V. & SAMAAN. A.: Adrenal cortical carcinoma. A study of 77 cases. Cancer 52 (193), PAIVANSALO M., LAHDE S., MERIKANTO J. & KALLIONEN M.: Computed tomography in primary and secondary adrenal tumours. Acta Radiol. 29 (19), SLEE P. H. T. J., SCHABERG A. & VAN BRUMMELEN P.: Carcinoma of the adrenal cortex causing primary hyperaldosteronism. Cancer 51 (193), TISNADO J., CHO S. R., WALSH J. W., BEACHLEY M. C. & GOLDSCHMIDT R. A.: Computed tomography versus angiography in the diagnosis of large right adrenal carcinoma. J. Comput. Tomogr. (194), 27.