Radiology
Dieter H. Szolar, MD Melvyn Korobkin, MD Pia Reittner, MD Andrea Berghold, PhD Thomas Bauernhofer, MD Harald Trummer, MD Helmut Schoellnast, MD Klaus W. Preidler, MD Hellmuth Samonigg, MD
Published online 10.1148/radiol.2342031876 Radiology 2005; 234:479-485
Abbreviation: ROI = region of interest
1 From Diagnostikum Graz-Südwest and Medical School, Karl Franzens University, Weblinger Gürtel 25, 8054 Graz, Austria (D.H.S., K.W.P.); Depart- ments of Radiology (P.R., H. Schoell- nast), Internal Medicine (T.B., H. Sa- monigg), and Urology (H.T.) and Institute for Medical Informatics, Sta- tistics and Documentation (A.B.), Karl Franzens University and University Hospital, Graz, Austria; and Depart- ment of Radiology, University of Michigan Medical Center, Ann Ar- bor, Mich (M.K.). Received Novem- ber 21, 2003; revision requested February 6, 2004;revision received March 27; accepted May 17. Address correspondence to D.H.S. (e-mail: dieter.szolar@diagnostikum-graz.at).
Authors stated no financial relation- ship to disclose.
Author contributions:
Guarantor of integrity of entire study, D.H.S .; study concepts, D.H.S .; study design, D.H.S., M.K .; literature re- search, D.H.S., P.R .; clinical studies, T.B., H.T., H. Samonigg; data acquisi- tion, D.H.S., P.R., K.W.P .; data analy- sis/interpretation, D.H.S., P.R., K.W.P., A.B .; statistical analysis, A.B .; manu- script preparation, definition of intel- lectual content, revision/review, and final version approval, D.H.S., M.K .; manuscript editing, H. Schoellnast · RSNA, 2005
Adrenocortical Carcinomas and Adrenal
Pheochromocytomas: Mass and Enhancement Loss Evaluation at Delayed Contrast-enhanced CT1
PURPOSE: To retrospectively measure the adrenal gland attenuation and the percentage loss of adrenal gland enhancement at delayed contrast medium-enhanced computed tomography (CT) in patients with adrenocortical carcinomas and pheochromocytomas and to compare these data with those in patients with adenomas and metastases.
MATERIALS AND METHODS: The study protocol was approved by the ethics committee, which waived informed consent. Eleven patients with proved adreno- cortical carcinoma, 17 with proved pheochromocytoma, 23 with adrenal adenoma, and 16 with metastasis to the adrenal gland underwent helical CT. Nonenhanced CT was followed by contrast-enhanced CT 1 minute and 10 minutes later. Attenu- ation and enhancement loss values were calculated.
RESULTS: The mean attenuation of adenomas (8 HU ± 18 [standard deviation]) was significantly lower than those of adrenocortical carcinomas (39 HU ± 14), pheochromocytomas (44 HU ± 11), and metastases (34 HU ± 11) on nonenhanced CT scans (P < . 001). Although the mean attenuation values for nonadenomas (ie, adrenocortical carcinomas, pheochromocytomas, and metastases) were signifi- cantly higher than the value for adenomas on the 1-minute contrast-enhanced CT scans (P < . 001), there was more overlap in attenuation between adenomas and nonadenomas on contrast-enhanced scans than on nonenhanced scans. On the 10-minute delayed contrast-enhanced scans, the mean attenuation of adenomas (32 HU ± 17) was significantly lower than the mean attenuations of carcinomas (72 HU ± 15), pheochromocytomas (83 HU ± 14), and metastases (66 HU ± 13) (P < .001). At optimal threshold values of 50% for absolute percentage of enhancement loss and 40% for relative percentage of enhancement loss at 10 minutes, both the sensitivity and the specificity for the diagnosis of adenoma were 100% when adenomas were compared with carcinomas, pheochromocytomas, and metastases.
CONCLUSION: The enhancement loss in adrenocortical carcinomas and pheochro- mocytomas is similar to that in adrenal metastases but significantly less than that in adrenal adenomas. The percentage change in contrast material washout is a useful adjunct to absolute CT attenuation values in differentiating adrenal adenomas from adrenocortical carcinomas and pheochromocytomas. · RSNA, 2005
Extensive research has been focused on strategies to differentiate benign from malignant adrenal lesions by using noninvasive imaging methods (1-23). Several studies have ad- dressed the value of delayed contrast medium-enhanced computed tomography (CT) in distinguishing adrenal adenomas from adrenal nonadenomas (9-14). The results of several studies have demonstrated that adrenal adenomas have significantly lower attenuation
and significantly greater loss of enhance- ment than nonadenomas on delayed contrast-enhanced CT scans (9-14). However, detailed information on atten- uation and percentage change in contrast material washout in nonadenomas has been based primarily on data regarding metastases to the adrenal glands (hereaf- ter referred to as adrenal metastases).
To our knowledge, very few data on either the attenuation or the percentage change in contrast material washout in adrenocortical carcinomas and pheo- chromocytomas are available. Therefore, the purpose of this study was to measure retrospectively the adrenal attenuation and the percentage loss of adrenal gland enhancement (ie, percentage change in washout) in patients with adrenocortical carcinomas and pheochromocytomas and to compare these data with those in patients with adenomas and metastases.
MATERIALS AND METHODS
Study Population
The study protocol was approved by and conducted in accordance with the recommendations of the ethics commit- tee at Karl Franzens University. The eth- ics committee waived informed consent. Between 1996 and 2002, 211 patients with 248 adrenal masses were consecu- tively examined with helical CT and con- sidered for entry into our study. There were a total of 126 male and 85 female patients ranging in age from 12 to 84 years (mean age, 64 years). The presence of 173 masses in 136 patients was proved at biopsy (40 masses), surgery (29 masses), or imaging (repeated CT) and clinical fol- low-up (104 masses). The diagnosis of ad- enoma was proved on the basis of a stable lesion size or an attenuation value of less than 10 HU at nonenhanced follow-up CT for at least 12 months (mean, 13 months; range, 12-26 months).
Excluded from the study were 70 pa- tients in whom neither follow-up CT nor cytologic or histologic analysis was pos- sible, two patients with adrenal cysts, and one patient each with adrenal hema- toma, neuroblastoma, and ganglioneu- roma. Also excluded were 69 patients who underwent imaging protocols other than those described herein (eg, contrast- enhanced CT 30 and 90 seconds and de- layed contrast-enhanced CT 3 and 30 minutes after the start of the contrast material administration).
The remaining 67 patients (31 men and 36 women aged 22-83 years; mean age, 60 years) fulfilled the criteria for in-
clusion. Eleven patients (seven men and four women aged 22-58 years; mean age, 42 years) had adrenocortical carcinomas, and 17 (five men and 12 women aged 38-64 years; mean age, 49 years) had pheochromocytomas. Ten of the 17 pa- tients with pheochromocytoma were ex- amined because of abnormal plasma cat- echolamine levels and abnormal 24-hour urine vanillymandelic acid and meta- nephrine levels. In the remaining seven patients, who had no endocrine dysfunc- tion, the masses were found incidentally during CT performed for other reasons. In four of the 11 patients with adrenocor- tical carcinoma, metastatic spread to the liver, lungs, and bone was evident. Thirty-nine patients-23 (nine men and 14 women aged 28-74 years; mean age, 55 years) with 24 adrenal adenomas and 16 (10 men and six women aged 36-83 years; mean age, 64 years) with 21 adre- nal metastases-provided reference data (11). The primary cancers from which the adrenal metastases stemmed were small lung cell carcinoma in two patients, non- small lung cell carcinoma in 10 patients, breast carcinoma in one patient, renal cell carcinoma in two patients, and colo- rectal carcinoma in one patient.
A total of 73 masses-11 adrenocorti- cal carcinomas, 17 pheochromocytomas, 24 adenomas, and 21 metastases-in 67 patients were investigated with an iden- tical imaging protocol. Diagnoses for the final study group of 73 masses were es- tablished when histologic proof was ob- tained at surgery or percutaneous biopsy (n = 28), when stability or change in the size of the mass was observed at a subse- quent CT examination after at least 6 months (n = 28), or when an attenuation value of 10 HU or less was measured on nonenhanced CT scans (n = 17).
Imaging Protocol
Imaging examinations were performed with a Somatom Plus 4 Volume Zoom (Siemens, Erlangen, Germany) or Light- Speed QX/i (GE Medical Systems, Mil- waukee, Wis) CT scanner. All patients gave written informed consent to un- dergo CT. The protocol for the single- section helical CT examinations performed in 51 patients included volumetric data acquisition through the upper part of the abdomen with 5-mm collimation, a 7.5 mm/sec table feed, and 4-mm increments before and after the intravenous bolus injection of contrast material (iopromide 300, Ultravist, Schering Pharmaceuticals, Berlin, Germany; or iohexol 300, Om- nipaque, Nycomed-Amersham, London,
England). The scanning time for one rev- olution of the x-ray tube was 0.75 sec- ond.
For the multisection helical CT exam- inations, the parameters varied because of the different detector array systems used. With use of the Volume Zoom scanner, which was used to examine nine patients, the protocol consisted of 2.5-mm collimated sections acquired with four detectors (4 × 2.5-mm collimation), a 12.5-mm section thickness with a pitch of 1.25, and 3.0-mm reconstruction in- tervals. With use of the LightSpeed QX/i scanner, which was used to examine seven patients, the parameters were 120 kVp, 200-240 mA, 4 × 2.5-mm collima- tion, a 0.8-second rotation time, a pitch of 3 (in thin-section mode), and 3.0-mm reconstruction intervals. To optimize the reproducibility of the starting measure- ments, each scan was obtained while the patient was in full inspiration.
First, a nonenhanced scan was ob- tained through the adrenal glands. An 18- or 20-gauge intravenous catheter (Angiocath; Becton Dickinson, Franklin Lakes, NJ) was then placed in an antecu- bital vein and tested by rapidly infusing 10 mL of saline by hand. Subsequently, 120 mL of nonionic contrast material (io- promide 300 or iohexol 300) was admin- istered at 2.5 mL/sec by using a power injector (MCT Plus; Medrad, Pittsburgh, Pa). In all patients, the second and third scans were obtained, respectively, 1 minute after and 10 minutes after the start of the contrast material injection. The imaging protocol was preprogrammed so that these scans were obtained by using the same parameters that were used to obtain the first nonenhanced scan. The scans (Figs 1, 2) were obtained with standard soft-tissue settings (window width, 400 HU; window level, 40 HU).
Image and Data Analyses
CT scans were interpreted and attenu- ation values were measured retrospec- tively by using a commercially available CT-magnetic resonance (MR) imaging workstation (Sienet Magic View 1100; Siemens). These tasks were performed by two radiologists who were experienced in performing CT of the adrenal glands (D.H.S. with 11 years experience; P.R. with 7 years experience). These radiolo- gists had no knowledge of the clinical, histologic, or follow-up findings, and they worked independently without con- sultation with one another.
For all adrenal masses detected at CT (regardless of the type of scan [nonen-
a.
b.
c.
a.
b.
c.
hanced, contrast-enhanced, or delayed contrast-enhanced] obtained), the atten- uation was measured by using circular region-of-interest (ROI) cursors placed over the area of disease. The ROI circle was made as large as possible, and lesion edges were avoided to preclude partial volume effects. Cystic, necrotic, and hemorrhagic components, as well as cal- cifications of the adrenal masses, were excluded if they were present. Necrosis was defined as a region (within the mass) with an attenuation value similar to that of water (<20 HU) at 1-minute contrast- enhanced CT. Calcification was defined as a region with an attenuation value greater than 120 HU at nonenhanced CT.
Attenuation values were recorded and
averaged for final data analysis. To deter- mine the sizes of the masses, a distance cursor was used to measure the diameter in the transverse plane. The section that encompassed the largest square area of the adrenal mass was used for analysis. The measurements obtained by the two radiologists were averaged.
For all masses, the absolute and rela- tive percentage losses of enhancement (ie, washout) were determined. The abso- lute percentage loss of enhancement was calculated as follows: [1 - (AD - AN)/ (AE - AN)] . 100, where Ap is the attenu- ation at delayed (10-minute) contrast- enhanced CT; AE; the attenuation at con- trast-enhanced (1-minute) CT; and AN; the attenuation at nonenhanced CT. The
percentage loss of enhancement relative to the amount of initial enhancement (ie, relative loss of enhancement) was cal- culated as follows: [(AE - AD)/AE] - 100.
Statistical Analyses
Statistical analyses were performed with commercially available software (StatView; Abacus Concepts, Berkeley, Calif). Primary statistical analysis of the pooled data was performed with the paired Student t test to determine mean differences in objective ROI measure- ments of attenuation between the differ- ent helical CT scans (nonenhanced, de- layed contrast-enhanced, and contrast- enhanced scans). Repeated measures
| Adrenal Mass | Diameter (cm) |
|---|---|
| Adenoma (n = 24) Adrenocortical carcinoma (n = 11) | 2.2 ± 0.8 (1.3-3.5) |
| 9.8 ± 2.5 (4.5-16.2) | |
| Pheochromocytoma (n = 17) | 5.1 ± 1.9 (4.7-10.8) |
| Metastasis (n = 21) | 4.5 ± 2.2 (1.6-10.4) |
Note .- Diameters are given as means ± stan- dard deviations, with ranges in parentheses. The mean diameter of adrenocortical carcino- mas was significantly larger than the mean diameters of adenomas, pheochromocyto- mas, and metastases (P < . 001).
| Adrenal mass | Nonenhanced CT | Contrast-enhanced CT (at 1 min) | Delayed Contrast- enhanced CT (at 10 min) |
|---|---|---|---|
| Adenoma (n = 24) | 8 ± 18 (-21 to 27) | 60 ± 27 (30-84) | 32 ± 17 (15-52) |
| Adrenocortical carcinoma (n = 11) | 39 ± 14 (23-52) | 83 ± 29 (51-108) | 72 ± 15 (40-93) |
| Pheochromocytoma (n = 17) | 44 ± 11 (28-60) | 94 ± 20 (72-131) | 83 ± 14 (60-123) |
| Metastasis (n = 21) | 34 ± 11 (17-55) | 81 ± 30 (49-110) | 66 ± 13 (50-84) |
Note .- Data are mean attenuation values (in Hounsfield units) ± standard deviations, with ranges in parentheses. Adenomas were significantly lower in attenuation than adrenocortical carcinomas, pheochromocytomas, and metastases at nonenhanced, contrast-enhanced, and delayed contrast- enhanced CT (P < . 001). There were no significant differences in attenuation between adreno- cortical carcinomas, pheochromocytomas, and metastases.
analysis of variance also was used. With this method, the mean values for more than one tumor group can be compared when measurements are obtained at more than one point in time. The method also allows one to control differ- ences among patients, differences among lesions within individual patients, and associations between patient characteris- tics and helical CT sequences. The mean size and mean attenuation for the tumor groups at nonenhanced and contrast-en- hanced CT were calculated and analyzed with analysis of variance. P < . 01 was considered to indicate statistical signifi- cance. For the post hoc tests, Bonferroni correction was applied to the results to correct for the multiple comparisons made.
RESULTS
Mass Size
The mean diameters and the ranges of diameters of the different adrenal masses are listed in Table 1. Adrenocortical car- cinomas were significantly larger than adrenal adenomas, pheochromocyto- mas, and metastases (P < . 001). Of the 11 adrenocortical carcinomas, only three (27%) were smaller than 6.0 cm in diam- eter. The mean diameter of the pheo- chromocytomas (5.1 cm ± 1.9 [standard deviation]) was significantly greater than that of the adenomas (2.2 cm ± 0.8, P < .001) but similar to that of the metastases (4.5 cm ± 2.2). At a threshold diameter of 2.5 cm, the sensitivity and specificity for the diagnosis of adrenal adenoma were 75% (18 of 24 masses) and 88% (21 of 24 masses), respectively.
Nonenhanced CT
At nonenhanced CT, the 11 adrenocor- tical carcinomas had a mean attenuation of 39 HU ± 14, as compared with a mean attenuation of 44 HU + 11 for the 17 pheochromocytomas. The 24 adenomas had a mean attenuation of 8 HU ± 18, as compared with a mean attenuation of 34 HU + 11 for the 21 adrenal metastases (Table 2). The mean attenuation of the adenomas was significantly lower than the mean attenuations of the adrenocor- tical carcinomas, pheochromocytomas, and metastases (P < . 001) (Fig 3). There were no significant differences in attenua- tion between the adrenocortical carcino- mas, pheochromocytomas, and metastases (P = . 24). None of the 17 pheochromocy- tomas, as compared with one (9%) of the 11 adrenocortical carcinomas and two (10%) of the 21 metastases, had an atten- uation value lower than 28 HU. On the other hand, none of the 24 adenomas had an attenuation value greater than 27 HU at nonenhanced CT. The sensitivity and spec- ificity for the diagnosis of adenoma at a threshold attenuation value of 11 HU were 79% (19 of 24 masses) and 100% (24 of 24 masses), respectively. The positive predic- tive value was 100%, and the negative pre- dictive value was 65%.
Contrast-enhanced CT
One minute after the start of the con- trast material injection, the mean atten- uation was 83 HU + 29 for the adreno- cortical carcinomas and 94 HU ± 20 for the pheochromocytomas (P = . 29). The mean attenuation was 60 HU ± 27 for the adenomas and 81 HU + 30 for the metastases (Table 2). There were no sig- nificant differences in attenuation be- tween the adrenocortical carcinomas,
pheochromocytomas, and metastases (P = . 31) (Fig 3). Although the mean at- tenuation values for all of the nonadeno- mas (ie, adrenocortical carcinomas, pheochromocytomas, and metastases) were significantly greater than that for the adenomas at 1 minute (P < . 001), there was much more overlap in attenu- ation between the adenomas and non- adenomas on the contrast-enhanced CT scans than on the nonenhanced images.
Contrast Material Washout at Delayed Contrast-enhanced CT
Ten minutes after the start of the con- trast material injection, the adrenocorti- cal carcinomas had a mean attenuation of 72 HU ± 15 (range, 40-93 HU), as compared with a mean attenuation of 83 HU ± 14 (range, 60-123 HU) for the pheochromocytomas. The adenomas had a mean attenuation of 32 HU ± 17 (range, 15-52 HU), as compared with a mean attenuation of 66 HU ± 13 (range, 50-84 HU) for the adrenal metastases (Table 2, Fig 4). The mean attenuation of the ade- nomas was significantly lower than those of the adrenocortical carcinomas, pheo- chromocytomas, and metastases (P < .001). There were no significant differ- ences in attenuation between the adreno- cortical carcinomas, pheochromocyto- mas, and metastases (P = . 25) at delayed contrast-enhanced CT. At a threshold at- tenuation value of 52 HU, the sensitivity and specificity for the diagnosis of ade- noma were 92% (22 of 24 masses) and 96% (23 of 24 masses), respectively. The positive predictive value was 96%, and the negative predictive value was 91%.
On the 10-minute scans, the mean ab- solute percentage loss of enhancement was 34% ± 9 for the adrenocortical car-
Radiology
100
90-
80
Attenuation (HU)
70-
60
50
40
30-
20
10-
0
unenhanced scan
60-sec scan
10-min scan
120
110
100
Percentage Loss of Enhancement
90
80
70
60
50
40
30
20
10
0
absolute
relative
cinomas and 22% ± 12 for the pheochro- mocytomas, as compared with 62% ± 17 for the adenomas (P < . 001) and 31% ± 16 for the metastases (Table 3). The per- centage loss of enhancement relative to the amount of initial enhancement was 13% + 12 for the adrenocortical carcino- mas and 14% ± 7 for the pheochromo- cytomas, as compared with 108% ± 87 for the adenomas (P < . 001) and 19% ± 11 for the metastases (P = . 35). Both the absolute loss of enhancement and the relative loss of enhancement were signif- icantly greater for adenomas than for ad- renocortical carcinomas, pheochromocy- tomas, and metastases (P < . 001). At threshold values of 50% for absolute en- hancement loss and 40% for relative en- hancement loss, both the sensitivity and the specificity for the diagnosis of ade- noma were 100% (24 of 24 adenomas), as compared with the sensitivity and specific- ity for the diagnosis of the 11 adrenocorti- cal carcinomas and the 17 pheochromocy- tomas (P < . 001).
There were no calcifications in the pheo- chromocytomas or adrenocortical carcino- mas. Unlike most of the adenomas, seven of the 11 adrenocortical carcinomas, three of the 21 adrenal metastases, and 10 of the 17 adrenal pheochromocytomas had re- gions of necrosis. As mentioned earlier (see Materials and Methods), care was taken to exclude these regions from the attenuation measurements. In any case, there were no obvious differences in attenuation charac- teristics among the adrenal metastases.
DISCUSSION
Adrenal masses are common: They are found at an estimated 9% of postmortem examinations (1,20). Even in oncologic set- tings, approximately 70% of adrenal masses are benign (24-26). However, the precise characterization of adrenal masses is important for therapeutic guidance. Adrenocortical carcinomas and pheochro- mocytomas of the adrenal glands are rare, and both occur with equal frequency in women and men (15,16). Adrenocortical carcinomas account for only 0.02% of ma- lignant tumors, and the true incidence of pheochromocytomas is unknown (15,16). More than 50% of adrenocortical carcino- mas are endocrine related and caused by extensive cortisol production; Cushing syndrome is the most frequently seen con- dition at clinical presentation (15,16). Oc- casionally, patients present with symp- toms of virilization or Conn syndrome.
The use of CT and MR imaging has resulted in more frequent and incidental detections of adrenal masses. More re- cent research has been focused on differ- ent approaches to characterizing adrenal masses on cross-sectional images, includ- ing assessing lesion size, describing mor- phologic criteria, measuring attenuation or signal intensity, calculating the per- centage loss of enhancement, and deter- mining the lipid content (1-23). How- ever, to our knowledge, there are no
literature reports of comparisons be- tween the attenuation and percentage loss of enhancement values for adreno- cortical carcinomas and pheochromocy- tomas and those for the more frequently encountered adrenal adenomas and adre- nal metastases.
Two main CT features used to distin- guish adenomas from nonadenomas have been identified: the intracellular lipid content and the enhancement loss in adrenal masses. Adenomas often con- tain abundant intracytoplasmatic fat and thus have low attenuation on nonen- hanced CT scans and signal loss on op- posed phase MR images. Conversely, nonadenomas contain little or no intra- cytoplasmatic fat and thus have higher attenuation at nonenhanced CT and no signal loss at opposed phase MR imaging (4,5,8,18,23). Korobkin et al (23) observed that the intratumoral lipid content of re- sected adrenal adenomas correlated with lower attenuation at nonenhanced CT. Al- though an attenuation value of less than 10 HU at nonenhanced CT is considered sufficient for characterizing an adrenal mass as an adenoma, it has been esti- mated that up to 30% of adenomas do not contain large amounts of lipid and may remain indeterminate and indistin- guishable from nonadenomas on nonen- hanced CT scans or chemical shift MR images.
Quantitative evaluation of adrenal mass enhancement on delayed contrast-
Radiology
enhanced CT scans has been shown to be highly accurate for differentiating adeno- mas from nonadenomas (9-14). Adeno- mas exhibit rapid enhancement and con- trast medium washout; nonadenomas also enhance vigorously but have prolonged contrast material washout. Korobkin et al (9) reported that adrenal adenomas have a much earlier and more rapid contrast material washout than do nonadenomas. In their study, the mean percentage loss of enhancement for adrenal adenomas was 51% at 5 minutes and 70% at 15 minutes after the start of the contrast material administration, as compared with 8% and 20%, respectively, for non- adenomas. Caoili et al (13,14) reported a sensitivity of 98% and a specificity of 92% for the diagnosis of adenoma with use of 15-minute delayed contrast-en- hanced CT scans obtained in 166 pa- tients with adrenal masses.
Although a few adrenocortical carcino- mas and pheochromocytomas have been described in previously published related series, to our knowledge, little detailed information on the behavior of these masses as compared with the behavior of adenomas has been provided. For in- stance, Szolar and Kammerhuber (11) in- cluded six pheochromocytomas and six adrenocortical carcinomas in their non- adenoma group. All of the pheochromo- cytomas had delayed enhancement and percentage of enhancement loss values in the range of nonadenoma values, but data on the individual carcinomas were not reported.
The results of the present study indi- cate that at nonenhanced CT, both adre- nocortical carcinomas and pheochromo- cytomas have mean attenuation values similar to those of metastases but signif- icantly higher than those of adenomas. On the 1-minute contrast-enhanced CT scans, however, there was considerable overlap in attenuation between the ade- nomas and all three types of nonadeno- mas, and, hence, it was not possible to identify a threshold attenuation value that could be used to reliably differenti- ate the different masses. On the 10- minute delayed contrast-enhanced scans, both the mean attenuation and the mean percentage loss of enhancement for ade- nomas differed significantly from those observed for pheochromocytomas, adre- nocortical carcinomas, and metastases.
In a study of 52 adenomas and 24 non- adenomas (mainly metastases), Korobkin et al (9) reported that 10 minutes after contrast material administration, the op- timal thresholds for absolute and relative enhancement loss were 50% and 40%,
| Adrenal Mass | Absolute Loss of Enhancement (%) | Relative Loss of Enhancement (%) |
|---|---|---|
| Adenoma (n = 24) | 62 ± 17 | 108 ± 87 |
| Adrenocortical carcinoma (n = 11) | 34 ± 9 | 13 ± 12 |
| Pheochromocytoma (n = 17) | 22 ± 12 | 14 ± 7 |
| Metastasis (n = 21) | 31 ± 16 | 19 ± 11 |
Note .- Values are given as means ± standard deviations. Absolute and relative losses of enhance- ment were significantly greater for adenomas than for adrenocortical carcinomas, pheochromo- cytomas, and metastases (P < . 001). There were no significant differences in contrast material washout between adrenocortical carcinomas, pheochromocytomas, and metastases.
respectively. Applying these same thresh- old values to the data in our series yielded a sensitivity of 100% and a specificity of 100% for the diagnosis of adenoma when we compared the 24 adenomas with the 11 cortical carcinomas and the 17 pheo- chromocytomas. Caoili et al (14) re- ported that one patient with adrenocor- tical carcinoma and one patient with pheochromocytoma had a percentage of enhancement loss greater than the opti- mal threshold value of 60% with a 15- minute enhancement delay, so the high accuracy observed in our series certainly may not apply to all of these uncommon neoplasms. Our findings do suggest, however, that enhancement loss calcula- tions at 10-minute delayed CT can be helpful in differentiating adrenal adeno- mas from adrenal carcinomas, pheochro- mocytomas, and metastases.
In patients with a solitary adrenal mass and no evidence of either a hyperfunc- tioning endocrine disorder or an extra- adrenal primary neoplasm, the differen- tial diagnosis usually includes adenoma and adrenocortical carcinoma. Most solid adrenal masses larger than 6 cm in diam- eter are resected because a substantial proportion of them are found to be ma- lignant and most adrenocortical carcino- mas are larger than 6 cm when they are first detected clinically. Although only a small fraction of adrenocortical carcino- mas are smaller than 6 cm and only a small fraction of adenomas are larger than 4 cm at presentation, it is the group of adrenal masses that have diameters be- tween 4 and 6 cm that is so challenging to characterize at imaging. In our study, three of 11 adrenocortical carcinomas were smaller than 6 cm. The nonen- hanced CT attenuation, delayed contrast- enhanced CT attenuation, and percent- age of delayed enhancement loss values for these three masses were all in the range of nonadenoma values; thus, these lesions would not have been incorrectly categorized as adenomas.
The present study may have had sev- eral limitations. First, the diagnosis of ad- renal adenomas was not histologically verified in most cases. However, the find- ings of multiple studies (6-14) have con- firmed that the presence of adrenal ade- noma can be proved on the basis of an attenuation value of less than 10 HU at nonenhanced CT and a stable mass size during a follow-up period of at least 6 months. Second, although our series in- cluded a substantial number of adrenal masses, the numbers of adrenocortical carcinomas and pheochromocytomas still may have been too small for an ad- equate statistical work-up. Third, we did not standardize the rate of administra- tion or the injected volume of contrast material with consideration of the car- diac output and weight of each patient. Had we done so, we might have obtained different results, but these approaches are not used in clinical practice. Therefore, we believe that the results we obtained are more likely to reflect actual daily clin- ical routine.
In summary, adrenocortical carcino- mas and pheochromocytomas have en- hancement loss that is similar to that of adrenal metastases and significantly less than that of adrenal adenomas. The per- centage change in contrast material washout is a useful adjunct to absolute CT attenuation values in differentiating adrenal adenomas from adrenocortical carcinomas and pheochromocytomas.
References
1. Dunnick NR, Korobkin M. Adrenal radi- ology: distinguishing benign from malig- nant adrenal masses. AJR Am J Roentge- nol 1996; 167:861-867.
2. Lee MJ, Hahn PF, Papanicolaou N, et al. Benign and malignant adrenal masses: CT distinction with attenuation coefficients, size, and observer analysis. Radiology 1991; 179:415-418.
3. Korobkin M, Brodeur FJ, Yutzy GG, et al. Differentiation of adrenal adenomas from nonadenomas using CT attenuation val-
ues. AJR Am J Roentgenol 1996; 166:531- 536.
4. Mitchell DG, Crovello M, Matteucci T, Petersen RO, Miettinen MM. Benign ad- renocortical masses: diagnosis with chemical shift MR imaging. Radiology 1992; 185:345-351.
5. Korobkin M, Lombardi TJ, Aisen AM, et al. Characterization of adrenal masses with chemical shift and gadolinium-en- hanced MR imaging. Radiology 1995; 197:411-418.
6. Korobkin M, Brodeur FJ, Francis IR, Quint LE, Dunnick NR, Goodsitt M. Delayed en- hanced CT for differentiation of benign from malignant adrenal masses. Radiol- ogy 1996; 200:737-742.
7. Krestin GP, Friedmann G, Fishbach R, Neufang KF, Allolio B. Evaluation of ad- renal masses in oncologic patients: dy- namic contrast-enhanced MR vs CT. J Comput Assist Tomogr 1991; 15:104- 110.
8. Outwater EK, Siegelmann ES, Huang AB, Birnbaum BA. Adrenal masses: correla- tion between CT attenuation value and chemical shift ratio at MR imaging with in-phase and opposed-phase sequences. Radiology 1996; 200:749-752.
9. Korobkin M, Brodeur FJ, Francis IR, Quint LE, Dunnick NR, Londy F. CT time-atten- uation washout curves of adrenal adeno- mas and nonadenomas. AJR Am J Roent- genol 1998; 170:747-752.
10. Szolar DH, Kammerhuber FH. Quantita- tive CT evaluation of adrenal gland mass- es: a step forward in the differentiation between adenomas and nonadenomas. Radiology 1997; 202:517-521.
11. Szolar DH, Kammerhuber FH. Adrenal ad- enomas and nonadenomas: assessment of washout at delayed contrast-enhanced CT. Radiology 1998; 207:369-375.
12. Pena CS, Boland GW, Hahn PF, Lee MJ, Mueller PR. Characterization of indetermi- nate lipid-poor adrenal masses: use of washout characteristics at contrast-en- hanced CT. Radiology 2000; 217:798-802.
13. Caoili EM, Korobkin M, Francis IR, Co- han RH, Dunnick NR. Delayed enhanced CT of lipid-poor adrenal adenomas. AJR Am J Roentgenol 2000; 175:1411-1415.
14. Caoili EM, Korobkin M, Francis IR, et al. Adrenal masses: characterization with combined unenhanced and delayed en- hanced CT. Radiology 2002; 222:629-633.
15. Dunnick NR. Adrenal carcinoma. Radiol Clin North Am 1994; 32:99-108.
16. Francis IR, Korobkin M. Pheochromocy- toma. Radiol Clin North Am 1996; 34: 1101-1112.
17. Korobkin M. CT characterization of adre- nal masses: the time has come. Radiology 2000; 217:629-632.
18. Bilbey JH, McLoughlin RF, Kurkjian PS, et al. MR imaging of adrenal masses: value of chemical-shift imaging for dis- tinguishing adenomas from other tu-
mors. AJR Am J Roentgenol 1995; 164: 637-642.
19. Reinig JW, Stutley JE, Leonhardt CM, Spicer KM, Margolis M, Caldwell CB. Dif- ferentiation of adrenal masses with MR imaging: comparison of techniques. Ra- diology 1994; 192:41-46.
20. Gross MD, Saphiro B, Francis IR, et al. Scintigraphic evaluation of clinically si- lent adrenal masses. J Nucl Med 1994; 35:1145-1152.
21. Mayo-Smith WW, Boland GW, Noto RB, Lee MJ. State-of-the-art adrenal imaging. RadioGraphics 2001; 21:995-1012.
22. Glazer HS, Weyman PJ, Sagal SS, Levitt RG, McClennan BL. Adrenal masses: incident discovery on computed tomography. AJR Am J Roentgenol 1982; 139:81-85.
23. Korobkin M, Giordano TJ, Brodeur FJ, et al. Adrenal adenomas: relationship be- tween histologic lipid and CT and MR findings. Radiology 1996; 200:743-747.
24. Korobkin M, Francis IR. Adrenal imaging. Semin Ultrasound CT MR 1995; 16:317- 330.
25. Oliver TW Jr, Bernardino ME, Miller JI, Mansour K, Greene D, Davis WA. Isolated adrenal masses in nonsmall-cell broncho- genic carcinoma. Radiology 1984; 153: 217-218.
26. Pagani JJ. Normal adrenal glands in small cell lung carcinoma: CT guided biopsy. AJR Am J Roentgenol 1983; 140:949-951.