| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Division of Abdominal Imaging and Intervention,
Massachusetts General Hospital, 55 Fruit St., White 270, Boston, MA
02114.
2 Department of Medicine, Division of Endocrinology, St. Elizabeth's Medical
Center, 736 Cambridge St., Boston, MA 02135.
3 Department of Pathology, Massachusetts General Hospital, Boston, MA
02114.
Received January 24, 2003;
accepted after revision June 2, 2003.
Address correspondence to M. A. Blake
(mblake2{at}partners.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. CT attenuation and size of nine adrenal nodules producing pheochromocytoma syndrome were measured on unenhanced CT in nine patients. For five patients who received IV contrast material, washout profiles were also calculated.
RESULTS. Two of the nine patients had adrenal lesions with attenuation values of less than 10 H; one had a pheochromocytoma with an attentuation of 9.0 H, and the other had a medullary hyperplasia with an attenuation of 1.8 H. These two nodules showed evidence of microscopic fat at histologic examination. No macroscopic fat was seen on the CT scans. The remaining seven patients had lesions with attenuation values exceeding 10 H (mean value, 25.6 H; range, 1.8–41 H). Mean diameter of the nine tumors (including the hyperplastic nodule) was 3.2 cm (range, 0.8–6.7 cm; SD, ± 2.3 cm). The two low-attenuation lesions also mimicked adenomas by displaying more than 60% contrast washout on 10-min-delayed contrast-enhanced scans, unlike the other three pheochromocytomas for which we had washout data.
CONCLUSION. On CT, pheochromocytomas may have attenuation values less than 10 H and also may display more than 60% washout of contrast agents on delayed scanning. Adrenal pheochromocytomas should be included with adenomas in the differential diagnosis both for masses with low attenuation on unenhanced CT and for lesions exhibiting a high percentage of contrast washout.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Characterization of adrenal masses, whether suspected or incidental, is thus desirable because involvement of the adrenal glands is frequently of critical importance in the staging of malignancies [1]. These circumstances have prompted numerous studies examining the characteristics of adrenal masses on unenhanced, dynamic, and delayed CT scans [3–12]. The seminal observation that adenomas frequently contain sufficient intracytoplasmic fat to produce lower attenuation values is a property that has been used to distinguish lipid-rich adenomas from indeterminate lesions [2]. These higherdensity indeterminate masses include some lipid-poor adenomas that fortunately appear to share the same contrast washout profile as lipid-rich adenomas. Many studies have confirmed the usefulness of unenhanced attenuation values in identifying lipid-rich adenomas and the value of contrast washout data for helping to differentiate lipid-rich and lipid-poor adenomas from nonadenomatous masses in the adrenal gland [3–12].
Pheochromocytoma is a rare catecholamine-secreting tumor derived from chromaffin cells. Pheochromocytomas are often well-defined masses with attenuation values similar to those of muscle tissue, measuring approximately 30–40 H [13]. A major impetus for diagnosing pheochromocytomas noninvasively is that any handling of these neoplasms can precipitate a hypertensive crisis [1]. Therefore, making the correct diagnosis before surgery is important. MRI has been shown to be somewhat useful for this purpose; lesions are frequently bright on T2-weighted sequences [14]. However, most lesions are detected during CT scanning, and adrenal mass characterization on CT relies primarily on attenuation parameters.
Studies to date have assumed that on CT, pheochromocytomas have characteristics similar to malignant (lipid-poor) lesions and that CT attenuation measurements are of little help in this differential diagnosis [3–11]. We questioned whether some pheochromocytomas have attenuation values low enough to be erroneously labeled adenomas if one applied the accepted clinical practice of considering all adrenal masses measuring under 10 H as adenomas (with allowance for other usually recognizable benign lesions, such as myelolipomas and cysts). We describe the attenuation characteristics of a series of histologically proven and resected pheochromocytomas on unenhanced CT.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Imaging was performed on helical CT scanners (HiLight Advantage or LightSpeed, General Electric Medical Systems, Milwaukee, WI). All examinations were performed using similar scanning parameters with a slice thickness of 5 mm, a pitch of 1:1, 140 kVp, and 200–300 mA. All scanners underwent daily calibration using water phantoms to ensure accurate attenuation values.
The diameter of each adrenal mass was measured on the slice with the largest surface area. The largest dimension was used to represent the diameter of the lesion. The region of interest (ROI) selected was an ovoid or circular area that was as large as possible without including adjacent retroperitoneal fat or inhomogeneous areas, such as necrosis. The average attenuation value of each ROI was measured twice by two independent observers who then calculated the mean values. Evidence of calcification, necrosis, or macroscopic fat was recorded.
When IV contrast–enhanced and delayed scans were obtained, the resulting average attenuation values of each pheochromocytoma were recorded in similar fashion. In these cases, scans were first obtained before the administration of IV contrast material. Using a power injector, 120 mL of iodinated (300 mg I/mL) contrast material (Oxilan [ioxilan], Cook Imaging, Bloomington, IN) was delivered through an antecubital vein at a rate of 2.5 mL/sec with a 75-sec delay. Delayed scans were then obtained after a 10-min delay without moving the patient from the CT gantry. The relative percentage of washout was calculated as (Hounsfield units on enhanced CT scan) – (Hounsfield units on delayed scan) / (Hounsfield units on enhanced CT scan). The result of this formula was multiplied by 100 to produce a percentage. Absolute washout was calculated as (Hounsfield units on enhanced CT scan) – (Hounsfield units on delayed scan) / (Hounsfield units on enhanced CT scan – Hounsfield units on unenhanced scan). The result of this formula was also multiplied by 100 to produce a percentage [4, 5]. The pathologist who analyzed the descriptions of the gross pathologic specimens and microscopic slides was unaware of the imaging results.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
None of the patients had documented metastatic disease, which is the only reliable criterion for the diagnosis of malignant pheochromocytoma. Neither tumor size, mitotic rate, nor vascular or capsular invasion is a sufficient discriminating feature with which to distinguish benign from malignant tumors [19]. Five masses were on the right and four, on the left. CT attenuation measurements from uniformly solid components resulted in a mean value of 25.6 H (range, 1.8–41 H). The mean and standard deviation of the diameter of the nine tumors was 3.2 ± 2.3 cm (range, 0.8–6.7 cm). Five tumors were uniform in attenuation, three were heterogeneous, and one was grossly hemorrhagic, with a large portion that measured 100 H.
Two (22%) of the nine patients had adrenal glands with attenuation values below 10 H; one measured 1.8 ± 24 H and the other 9.0 ± 22 H, (Figs. 1A, 1B, 1C, 1D and 2A, 2B). ROIs taken from within the gallbladder and uniform portions of the liver and renal cortex all had SDs similar to those of the adrenal masses in the series ranging between 20 and 29 H. On CT, we saw masses of low attenuation with relatively uniform density (Figs. 1A and 2A). One of these masses was a histologically classic pheochromocytoma with foci of tumor cells containing abundant fatty cytoplasm (Fig. 2A, 2B). Although the other mass was clinically and surgically considered to be a pheochromocytoma, our extensive histologic analysis found the mass to be medullary hyperplasia with intermingled fat-containing cortical cells (Fig. 1A, 1B, 1C, 1D) The medullary cells did not contain fat; however, the medulla was enlarged by both an increase in the number of benign medullary cells and an intermingling of fat-containing cortical cells, which resulted in the medulla displaying low attenuation on CT.
|
|
|
|
|
|
One patient with an attenuation measurement at the high end of the range (38 H) had extensive hemorrhage (100 H) within and surrounding the adrenal gland. Findings in pathologic specimens proved that the mass contained classic features of a pheochromocytoma. Two of the heterogeneous masses showed radiologic evidence of central necrosis, and one displayed well-defined cystic degeneration. None of the studied masses showed evidence of calcification.
Contrast washout profiles were available for five of the nine patients. The range of relative washout percentage was 15.5–83.3%, with absolute washout percentages ranging from 35.9% to 69.2%. Both lesions with unenhanced attenuation values of less than 10 H displayed over 60% relative and absolute washout, a washout profile similar to that of typical adenomas [6]. No adverse reactions resulting from IV contrast administration, and no record of an adrenergic crisis was documented in any of the five patients who received IV contrast material.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The occurrence of pheochromocytoma has been reported in 0.05–0.2% of hypertensive individuals. Patients may be completely asymptomatic, with up to 10% of cases being silent [20], and the masses in such patients tend to be larger than hyperfunctioning tumors [21]. A retrospective study that covered 1950–1979 revealed that in 50% of cases, the diagnosis was made at autopsy [22]. This record, although certainly not a reflection of the current rate of detection on imaging, is a powerful testament to the notorious difficulty of clinically diagnosing pheochromocytomas. Two major changes in clinical practice have aided in the diagnosis of pheochromocytomas: Imaging studies such as CT and MRI have dramatically improved our ability to detect adrenal masses, and excellent serum and urine biochemical tests are now available that can reveal elevated levels of catecholamines. Nevertheless, pheochromocytomas in a significant number of patients will not be recognized at the time of their imaging studies.
Despite their typically unilateral and benign presentation, pheochromocytomas can be bilateral (10%) or malignant (10%) [23]. The hereditary percentage was formerly considered to be 10% also, but Neumann et al. [24] have recently shown that approximately 25% of patients with sporadic pheochromocytoma and no family history of the disease have germ-line mutations in one of four susceptible genes for pheochromocytoma.
Eight of the patients in our study had adrenal pheochromocytomas, but one had functioning adrenal medullary hyperplasia, which other researchers have suggested is the precursor of pheochromocytoma [18, 19]. These studies suggest that diffuse medullary hyperplasia may be the initial pathologic change in the adrenal gland that leads to the development of nodular hyperplasia and adrenal medullary tumor. Marked uptake of iobenguane iodine-131 by adrenal medullary hyperplastic nodules on scans performed to detect pheochromocytomas has been reported [25]. However, to our knowledge, the CT appearance of a nodule caused by medullary hyperplasia has not been previously reported.
Several gross pathologic and microscopic patterns may be seen in pheochromocytomas, many of which we encountered in our series. Small neoplasms tend to be solid, whereas large lesions are often cystic or hemorrhagic [26]. Nonsecreting pheochromocytomas tend to be larger than secreting ones [27]. Cystic degeneration may be so marked that only a thin rim of identifiable cells may remain to disclose the true nature of the lesion. Gross features of pheochromocytomas described in the radiology literature are cystic regions [26], calcifications [26], fibrosis [26], necrosis [1], and internal hemorrhage [1].
Ramsay et al. [28] were the first to report lipid degeneration within a pheochromocytoma. These researchers described a case of bilateral pheochromocytomas with massive accumulation of lipid in the cytoplasm of clear cells in the tumors. The researchers found that on electron microscopy, intracellular cytoplasmic processes were surrounded by electron-lucent lipid. Light microscopy also revealed these changes with positive results on oil red O staining of the lipid. In our series, two lesions showed low attenuation on CT because of fat in either medullary cells or intermingled cortical cells; the presence of fat was confirmed histologically using H and E staining (Figs. 1A, 1B, 1C, 1D and 2A, 2B).
The use of CT attenuation values has proved valuable for differentiating an adenoma from other masses; the intracellular fat of adenomas causes CT attenuation values to drop below 10 H. A metaanalysis of several studies set the optimal threshold at 10 H. Labeling lesions with attenuation values of less than 10 H as adenomas has been reported to have a sensitivity of 71% and a specificity of 98% [3]. Proposed algorithms have presumed that lesions with attenuation values of less than 10 H represent benign adenomas or benign myelolipomas or cysts. Our results and the work of Ramsay et al. [28] indicate that pheochromocytomas can contain sufficient intracellular fat to display attenuation values of less than 10 H. Because the diagnosis of adenomas on MRI also depends on their intracellular fat content [29], these findings are intriguing; they suggest that pheochromocytomas can be misclassified as adenomas on opposed-phase MRI. Unfortunately, neither of the two patients with low-attenuation lesions in our study had MRIs available, so we could not test this hypothesis.
Studies of the washout profiles of adrenal masses have shown that such profiles can aid in the successful differentiation of lipid-rich and lipid-poor adenomas from metastases [4–6]. Research has also outlined reliable methods of differentiating adenomas not just on the basis of their lipid content but also on the basis of their washout characterization on delayed CT [4–9]. By comparing the drop in attenuation values from dynamic CT scans to values on the 10-min-delayed scans, Peña et al. [5] showed that adenomas have a relative percentage of washout that exceeds 50%, regardless of the fat content of the mass. The percentage of washout for malignant lesions is less than 50%; these researchers calculated a sensitivity of 98% and specificity of 100% for this technique. In a study using a related method, enhancement washout values at 15 min were compared with the enhanced attenuation values; this method achieved a sensitivity and a specificity of 98% and 92%, respectively [6]. Estimation of percentage washout values is therefore useful for adrenal mass characterization. However, in five patients in our study, pheochromocytomas displayed different and variable washout patterns after 10 min and therefore could have been confused with either adenomas or metastases. Indeed, although five of six pheochromocytomas studied by Szolar and Kammerhuber [9] showed washout curves similar to those of metastases, one displayed washout similar to that of adenomas, as did one of the two pheochromocytomas in the study of Caoili et al. [6].
One limitation of using attenuation and enhancement washout calculations to diagnose adrenal masses is that at least one half to two thirds of the adrenal mass must have homogeneous attenuation. Differences in attenuation can arise from cystic changes, necrosis, hemorrhage, or calcification. Such regions may contain abnormal capillary networks that can enhance slowly, thereby slowing enhancement washout. Thus, diagnosis using CT and calculation of enhancement washout values can be complicated by various types of abnormal degeneration [5]. We took ROIs from uniformly solid portions of each mass, but various forms of degeneration described earlier may have contributed to the differing washout profiles of the pheochromocytomas studied. Both low-attenuation lesions had washout profiles similar to those of adenomas. If future investigations of pheochromocytomas find this pattern of attenuation and washout characteristics to be consistent, CT evaluation of adrenal lesions will require additional caution.
Another potential cause of diagnostic confusion is the presence of macroscopic fat in adrenal lesions. Macroscopic fat (<–30 H) within an adrenal mass is characteristic of a benign myelolipoma [30]. However, fat within a large soft-tissue mass is not specific for myelolipoma. Ramsay et al. [28] found that pheochromocytomas can contain extensive lipid degeneration that at pathologic examination can be mistaken for myelolipomas and thus should also be considered in the radiologic differential diagnosis. Lipid-containing adenomas have also been reported in rare association with pheochromocytomas [31]. Adenomas and Cushing's syndrome have also been reported to be associated with myelolipoma. Adrenal cortical hyperplasia, which may also contain fat, has been described in conjunction with adrenocorticotropic-hormone-producing pheochromocytomas [32]. However, the precise cause and pathogenesis of lipid degeneration in pheochromocytomas remain unclear. Chase [33] found a paraganglioma with vacuolated cells that stained positively for fat. This finding suggests a similar phenomenon in these closely related extraadrenal tumors. The case of medullary hyperplasia in this series displayed multiple fat-containing cortical cells inter-mingled with hyperplastic medullary cells within an enlarged medulla.
In clinical practice, many unsuspected adrenal masses are found on CT and may be present on as many as 5% of CT scans [34]. Up to 10% of pheochromocytomas are silent [20] and before the era of cross-sectional imaging, the diagnosis in 50% of cases was made at autopsy [22]. The prior cutoff value of 10 H cannot be used to absolutely differentiate benign from malignant lesions or adenomas from nonadenomas because the attenuation of some pheochromocytomas in our study fell below this threshold value. Radiologists should also note that there are no reliable radiologic or histologic criteria by which to distinguish benign from malignant pheochromocytomas other than the documentation of metastatic disease. Even small, apparently benign pheochromocytomas are clinically and biologically unpredictable and should be removed. Hence, knowledge of the CT characteristics of pheochromocytomas is necessary to fully assess patients with adrenal masses.
Our study has limitations. It is an observational review and comprises a relatively small number of cases. However, to our knowledge, it is the largest reported collection of pathologically proven pheochromocytomas in patients who underwent unenhanced CT. All patients had functioning benign adrenal catecholamine-secreting lesions, with no silent tumors or extraadrenal locations. Of course, patients presenting clinically may be different from those whose pheochromocytoma is discovered incidentally. We find it interesting that no mention of pheochromocytoma was made on the referring request form of five of our patients, although all nine patients were found, after full assessment, to have a clinically active pheochromocytoma. Thus, sometimes in a busy radiology practice, one may discover a functioning pheochromocytoma that could not have been anticipated from the clinical information provided to the radiologist.
Contrast washout data are available for just over half the cases in our study. Partial volume averaging is an issue in all attenuation measurements. The ROI was placed in the center of the lesion with respect to both the transverse and the vertical axes, but the smaller the nodule is, the greater the concern over partial volume averaging. The two lesions with low attenuation were two of the smallest, but their SD measurements were similar to those of other masses. One patient included in this series had a histologically proven case of medullary hyperplasia. We included this case in our series for practical purposes because radiologists may encounter this entity during the imaging evaluation of a suspected pheochromocytoma.
We believe it is important to alert physicians to the spectrum of densities observed in pheochromocytomas, given the increasing use of CT and the history of clinical underdiagnosis of this condition. The variable nature of pheochromocytomas certainly merits the inclusion of this entity in the differential diagnosis of fatty, cystic, or calcified adrenal masses. To our knowledge, ours is the first study to report that a pheochromocytoma may mimic a lipid-rich adenoma on unenhanced CT. Furthermore, the contrast washout profiles of pheochromocytomas may also mimic those of adenomas. Future research— especially prospective studies—with a greater number of cases is neccessary to validate these findings. More controlled studies evaluating the washout characteristics of these tumors would be interesting, although the current accepted practice is to avoid the use of iodinated contrast material in patients with a known pheochromocytoma [35, 36].
To our knowledge, no other researchers have found that pheochromocytomas have attenuation sufficiently low to be mistaken for adenomas. We have shown that pheochromocytomas can have attenuation values of less than 10 H, thereby expanding the differential diagnosis of such adrenal masses. This observation has important clinical implications.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. F. Faria, S. M. Goldman, J. Szejnfeld, H. Melo, C. Kater, P. Kenney, M. P. Huayllas, G. Demarchi, V. V. Francisco, C. Andreoni, et al. Adrenal Masses: Characterization with in Vivo Proton MR Spectroscopy Initial Experience Radiology, December 1, 2007; 245(3): 788 - 797. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Elaini, S. K. Shetty, V. M. Chapman, D. V. Sahani, G. W. Boland, A. T. Sweeney, M. M. Maher, J. T. Slattery, P. R. Mueller, and M. A. Blake Improved Detection and Characterization of Adrenal Disease with PET-CT RadioGraphics, May 1, 2007; 27(3): 755 - 767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Bessell-Browne and M. E. O'Malley CT of Pheochromocytoma and Paraganglioma: Risk of Adverse Events with IV Administration of Nonionic Contrast Material Am. J. Roentgenol., April 1, 2007; 188(4): 970 - 974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Yoon, E. M. Remer, and B. R. Herts Incidental pheochromocytoma mimicking adrenal adenoma because of rapid contrast enhancement loss. Am. J. Roentgenol., November 1, 2006; 187(5): 1309 - 1311. [Full Text] [PDF]
|
|
|
|
|
|
F. S. Chew and K. P. Banks Imaging of Pheochromocytoma and Incidental Adrenal Lesions: Self-Assessment Module Am. J. Roentgenol., September 1, 2006; 187(3_Supplement): S467 - S469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Happel, G. Heinz-Peer, D. H. Szolar, and M. T. Korobkin Enhancement Characteristics of Pheochromocytomas Radiology, January 1, 2006; 238(1): 373 - 374. [Full Text] [PDF]
|
|
|
|
|
|
M. A. Blake, M. K. Kalra, A. T. Sweeney, B. C. Lucey, M. M. Maher, D. V. Sahani, E. F. Halpern, P. R. Mueller, P. F. Hahn, and G. W. Boland Distinguishing Benign from Malignant Adrenal Masses: Multi-Detector Row CT Protocol with 10-Minute Delay Radiology, December 21, 2005; (2005) 2382041514. [Abstract] [Full Text]
|
|
|
|
|
|
G. A. Motta-Ramirez, E. M. Remer, B. R. Herts, I. S. Gill, and A. H. Hamrahian Comparison of CT Findings in Symptomatic and Incidentally Discovered Pheochromocytomas Am. J. Roentgenol., September 1, 2005; 185(3): 684 - 688. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Kalra, M. A. Blake, G. W. Boland, and P. F. Hahn CT Features of Adrenal Pheochromocytomas: Attenuation Value and Loss of Contrast Enhancement Radiology, September 1, 2005; 236(3): 1112 - 1113. [Full Text] [PDF]
|
|
|
|
|
|
K. M. Elsayes, V. R. Narra, J. R. Leyendecker, I. R. Francis, J. S. Lewis Jr., and J. J. Brown MRI of Adrenal and Extraadrenal Pheochromocytoma Am. J. Roentgenol., March 1, 2005; 184(3): 860 - 867. [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Hamrahian, A. G. Ioachimescu, E. M. Remer, G. Motta-Ramirez, H. Bogabathina, H. S. Levin, S. Reddy, I. S. Gill, A. Siperstein, and E. L. Bravo Clinical Utility of Noncontrast Computed Tomography Attenuation Value (Hounsfield Units) to Differentiate Adrenal Adenomas/Hyperplasias from Nonadenomas: Cleveland Clinic Experience J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 871 - 877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Blake, M. K. Kalra, M. M. Maher, D. V. Sahani, A. T. Sweeney, P. R. Mueller, P. F. Hahn, and G. W. Boland Pheochromocytoma: An Imaging Chameleon RadioGraphics, October 1, 2004; 24(suppl_1): S87 - S99. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W.M. van Laarhoven, J. J. Willemsen, H. A. Ross, L. V.A.M. Beex, J. W.M. Lenders, and F. C.G.J. Sweep Pitfall in HPLC Assay for Urinary Metanephrines: An Unusual Type of Interference Caused by Methenamine Intake Clin. Chem., June 1, 2004; 50(6): 1097 - 1099. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
