| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Faculty of Medicine, Kagoshima University, 8-35-1 Sakuragaoka, Kagoshima-shi, Kagoshima, 890-8520, Japan.
Received August 26, 2002;
accepted after revision November 4, 2002.
Address correspondence to F. Fujiyoshi.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We assessed 102 adrenal tumors in 88 patients (27 hyperfunctioning and 44 nonhyperfunctioning adenomas, 19 metastases, nine pheochromocytomas, and three other adrenal tumors) using chemical shift MR imaging. On the chemical shift imaging, signal intensity index, calculated as [(signal intensity on in-phase imaging – signal intensity on opposed-phase imaging) / (signal intensity on in-phase imaging)] x 100%, was compared with the adrenal-to-spleen ratio, adrenal-to-muscle ratio, and adrenal-to-liver ratio for signal change on opposed-phase fast low-angle shot MR imaging. The tissues in the spleen, paraspinal muscle, and liver were reference tissues.
RESULTS. The signal intensity index had several advantages over the other three parameters calculated. We found no overlap in indexes between adenomas and metastatic tumors. The accuracy in distinguishing adenomas from metastatic tumors was 100% if the cutoff value of the signal intensity index selected was 11.2–16.5%.
CONCLUSION. The signal intensity index is the most reliable evaluation method for differentiating adrenal adenomas from metastatic adrenal tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In 1992, Mitchell et al. [3] reported that chemical shift techniques and simple visual analyses could be used to help identify adrenal adenomas by detecting the lipid frequently present in these tumors. Chemical shift–induced signal intensity changes for the adrenal mass relative to the liver and muscle were calculated by comparing these ratios on conventional T1-weighted images and chemical shift images. Tsushima et al. [4] described the utility of chemical shift fast low-angle shot (FLASH) imaging for differentiating adrenal masses. When the signal intensity indexes of adrenal masses, calculated as [(signal intensity on in-phase imaging – signal intensity on opposed-phase imaging) / (signal intensity on in-phase imaging)] x 100%, were determined, all adenomas had signal intensity indexes that were greater than 5%, whereas the signal intensity indexes of metastatic tumors and pheochromocytomas were less than 5%. These researchers found that the degree of accuracy for differentiating adenomas from nonadenomas was 100%.
Although the value of chemical shift imaging for accurately diagnosing adrenal masses has been established in a number of quantitative [3, 4, 5, 6, 7, 8, 9, 10, 11] and qualitative studies [5, 7, 9], considerable differences in methodology have made direct comparisons difficult. These differences include magnetic field strength, gradient strength, TE, and the formulas used to calculate the percentage of relative signal intensity change when comparing opposed-phase images with in-phase gradient-echo images. Furthermore, researchers have used several reference tissues with which to compare signal intensities, including the spleen [6, 7, 8, 9, 10], liver [3, 5, 6, 7, 9, 11], and muscle [3, 5, 6, 9].
Our retrospective study was performed to determine the efficacy and limits of chemical shift imaging for the diagnosis of adrenal tumors. We used two previously reported evaluation methods [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17]. One method uses the percentage of chemical shift–induced signal intensity change for the adrenal mass relative to the liver, paraspinal muscle, and spleen, and the change in signal intensity is calculated by comparing in-phase and opposed-phase chemical shift images [3, 5, 6, 7, 8, 9, 10, 11]. The other method is the signal intensity index as proposed by Tsushima et al. [4].
These two evaluation methods were compared for reliability with a scatterplot and the receiver operating characteristic (ROC) analysis. One of our goals was to determine whether a quantitative threshold value existed that would allow high specificity for the diagnosis of adrenal adenomas and to compare this threshold value with those that have been reported previously. Finally, we evaluated chemical shift imaging as a tool with which to differentiate between hyperfunctioning and nonhyperfunctioning adenomas and between cortisol-producing and aldosterone-producing adeomas.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Diagnosis of the five cortisol-producing adenomas that resulted in Cushing's syndrome were confirmed by surgical removal. Nine of 22 aldosterone-producing adenomas were confirmed by pathologic examination after open or laparoscopic surgery. The other 13 aldosterone-producing adenomas were confirmed by clinical and hormonal examinations, visualization of an adrenal nodule on CT and a "hot" nodule on 131I-adosterol adrenocortical scintigrams, venous sampling (elevated blood levels of aldosterone in the ipsilateral inferior adrenal vein or the renal vein), and normalized levels of hormones after transcatheter embolization of the adenoma [18]. Forty-two of the 44 nonhyperfunctioning adenomas were diagnosed on the basis of stability in size or appearance on CT scans after a minimum of 6 months' observation (range, 6–33 months; mean, 14.5 months) and the absence of clinical and endocrinologic dysfunction. Diagnosis of the other two nonhyperfunctioning adenomas were confirmed by pathologic examination after open or laparoscopic surgery.
Metastatic tumors were assessed by percutaneous needle biopsy (one tumor), pathologic examination after surgery (four tumors), and rapid enlargement of an existing adrenal mass on serial CT scans (10 tumors). Bilateral malignant lymphomas of two patients were confirmed by percutaneous needle biopsy of an adrenal mass of one side, bilateral abnormal deposits in adrenal masses on gallium scintigrams, and a reduction in size of the masses on follow-up CT after chemotherapy. All nine pheochromocytomas, one adrenocortical carcinoma, one ganglioneuroma, and one schwannoma were confirmed histologically after surgical removal.
MR Imaging Techniques
MR imaging was performed on a 1.5-T super-conductive unit (Magnetom Vision,
Siemens, Erlangen, Germany) with a phased array body coil. After localization
imaging, coronal T2-weighted MR imaging and chemical shift imaging were
performed while the patients held their breath after full expiration. The
coronal T2-weighted images were obtained with a half-Fourier acquisition
single-shot turbo spin-echo sequence with an echo space of 4.4 msec, an
effective TE of 64 msec, an infinite TR, a slice thickness of 6 mm, an
intersection gap of 0.6 mm, a field of view of 300 cm, and a matrix size of
160 x 256. Seventeen contiguous transverse MR images then were obtained
for localization of the adrenal lesions during a single 18-sec
breath-hold.
Chemical shift imaging was performed with FLASH to detect lipid. In-phase
and opposed-phase T1-weighted FLASH transverse images were obtained at the
level of the adrenal lesion. The spleen, liver, and paraspinal muscle were
imaged in the same slice as the adrenal lesion. Five sequential FLASH images
of the adrenal lesion were obtained with a single 19-sec breath-hold at a TR
of 133, opposed-phase and in-phase TEs of 2.2 and 4.4, a flip angle of
75°, a slice thickness of 6 mm, a field of view of 300
350 mm, a
matrix size of 134 x 256 using a rectangular field of view, an
intersection gap of 0.6 mm, and one signal acquisition. Because the
opposed-phase (TE, 2.2) images and in-phase (TE, 4.4) images must be obtained
with similar transmitter power and receiver gain, the images were obtained
without autotuning after radiofrequency tuning at a TE of 2.2.
Methods of Evaluation
Signal intensities of the adrenal lesions and reference tissues were
measured with an electronic cursor using circular regions of interest. The
regions of interest were set by a single investigator who was unaware of the
clinical data.
Reference tissues were measured as close as possible to the adrenal masses on the same slice. If splenic tissue was not present on the same slice, the closest available slice was used. The region of interest over the adrenal mass was chosen to cover most of the mass at the widest dimension and to avoid causing a partial volume artifact at the edge of the mass. Cystic, necrotic, hemorrhagic, or calcified components of the mass were excluded from the region of interest whenever possible. The regions of interest in the reference tissues were made as large as they could be without also including blood vessels in the liver and obvious fat striations in the paraspinal muscle whenever possible. The region of interest over the liver was selected from the posterior segment of the right lobe. The signal intensities of the adrenal lesions, spleen, paraspinal muscle, and liver were measured at the same positions on both opposed-phase and in-phase images. Four quantitative parameters of signal changes between opposed-phase and in-phase images were calculated from the measured signal intensities, as described in previous reports [3, 4, 5, 6, 7, 8, 9, 10, 11].
These parameters included the signal intensity index, adrenal-to-spleen ratio, adrenal-to-muscle ratio, and adrenal-to-liver ratio. The signal intensity index was calculated as [(signal intensity on in-phase imaging – signal intensity on opposed-phase imaging) / (signal intensity on in-phase imaging)] x 100%. The adrenal-to-spleen ratio was calculated as {[(adrenal signal intensity on opposed-phase imaging / spleen signal intensity on opposed-phase imaging) / (adrenal signal intensity on in-phase imaging / spleen signal intensity on in-phase imaging)] – 1} x 100%. The adrenal-to-muscle ratio was calculated as {[(adrenal signal intensity on opposed-phase imaging / muscle signal intensity on opposed-phase imaging) / (adrenal signal intensity on in-phase imaging / muscle signal intensity on in-phase imaging)] – 1} x 100%. The adrenal-to-liver ratio was calculated as {[(adrenal signal intensity on opposed-phase imaging / liver signal intensity on opposed-phase imaging) / (adrenal signal intensity on in-phase imaging / liver signal intensity on in-phase imaging)] – 1} x 100%.
Statistical Analysis
Statistical analysis was performed using a commercially available software
program (StatView, SAS Institute, Cary, NC). We compared the signal intensity
index, adrenal-to-spleen ratio, adrenal-to-muscle ratio, and adrenal-to-liver
ratio of adenomas and nonadenomas, of adenomas and metastatic tumors, of
hyperfunctioning and nonhyperfunctioning adenomas, and of cortisol-producing
adenomas and aldosterone-producing adenomas with a two-tailed unpaired
Student's t test. Four malignant lymphomas were included in the group
of metastatic tumors we evaluated.
The most effective method for distinguishing adenomas from metastatic tumors was determined by scatterplotting and ROC curves, which were constructed from the data using the ROCKIT program [19]. The ROC curve is a plot of the true-positive fraction (sensitivity) against the false-positive fraction (1 – specificity) for each mean of assessment. The area under the curve values (Az) and the standard errors were calculated with the LABROC5 portion of the ROCKIT program. The CLABROC portion was used to calculate the statistical significance of the differences between the Az values of the discriminators. A p value of less than 0.05 was considered to be a statistically significant difference.
An optimal cutoff value to distinguish adenomas from metastatic tumors was calculated from the most effective method of assessment. It was defined as the value that produced the maximal sum of sensitivity and specificity in the identification of adenomas.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Figures 1A, 1B, 1C, 1D shows the scatterplots comparing the signal intensity index, adrenal-to-spleen ratio, adrenal-to-muscle ratio, and adrenal-to-liver ratio values for adenomas with those of metastatic tumors. Although the adrenal-to-spleen, adrenal-to-muscle, and adrenal-to-liver ratios of adenomas and metastatic tumors over-lapped considerably, no overlaps were evident in the signal intensity index of adenomas and metastatic tumors. The ROC curves of the four parameters for discriminating adenomas from metastatic tumors are presented in Figure 2. The Az values and standard errors for these parameters are presented in Table 3. The Az value of the signal intensity index was 1.0, which was the highest value among all evaluation parameters. In the chemical shift imaging, the Az value for the signal intensity index was significantly higher than the values for the adrenal-to-muscle ratio (p <0.05) and the adrenal-to-liver ratio (p< 0.05) but not that of the adrenal-to-spleen ratio (p = 0.92). All the adenomas had signal intensity indexes of 16.5% or greater (Figs. 3A, 3B, 4A, 4B, 5A, 5B), whereas all metastatic tumors had signal intensity indexes of 11.2% or less (Figs. 6A, 6B). If the value for the signal intensity index was taken from this range (11.2–16.5%), the accuracy for differentiating adenomas from metastatic tumors was 100%.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
New MR techniques have helped in the differentiation of adrenal adenomas from nonadenomas. Contrast-enhanced or dynamic-enhanced MR imaging has improved the ability of radiologists to differentiate adenomas from malignant tumors and pheochromocytomas [5, 7, 21, 22]. Chemical shift MR imaging takes advantage of the reduced signal on opposed-phase images to differentiate adenomas from malignant adrenal masses. Most adrenal adenomas contain lipid, whereas most malignant lesions of the adrenal gland do not [23]. Lipid and water protons precess at different frequencies when exposed to a given magnetic field, and the signals from these protons can be either additive or subtractive, depending on the TE chosen [3, 24, 25]. At 1.5 T, water protons precess at a frequency that is 225 Hz faster than the frequency of lipid protons. On conventional MR imaging, the differences in precession frequencies for lipid and water protons lead to a chemical shift artifact, which appears as a black-and-white band along the lipid–water interface. The signals from fat and water remain in-phase despite the presence of spatial misregistration.
In contrast, gradient-echo MR images do not contain 180° refocusing pulses, so that the different resonance frequencies of water and fat protons cause the phases of these two substances to cycle in opposed phase to each other as the TE increases [24, 25]. Thus, fat and water protons are either in-phase or opposed-phase, depending on the echo time. The periodicity of in-phase and opposed-phase imaging occurs at 2.2-msec intervals using 1.5-T imagers, so that the periodicity of opposed-phase cycling occurs at 2.2 msec, 6.6 msec, 11.0 msec, and so forth. This effect is especially helpful for identifying adenomas that contain relatively high cytoplasmic lipid concentration and consequently develop signal dropout with opposed-phase sequences. Masses with either fat or water components, but not both, in the same voxel will not display signal cancellation [24, 25].
Bilbey et al. [6] suggested that the adrenal-to-spleen ratio was the best quantitative means for distinguishing adrenal adenomas from other adrenal masses on chemical shift images, leading many investigators to adopt the spleen as a reference organ [7, 8, 9, 10]. In our study, the adrenal-to-spleen ratio had the second largest Az value (0.984) on ROC analysis, and this value was larger than those for adrenal-to-muscle ratio and adrenal-to-liver ratio; this finding is consistent with the findings reported by Bilbey et al. [6]. Fatty infiltration of the liver and of the adipose tissue in the paraspinal muscle can result in signal intensity loss on opposed-phase images and could mask a lipid-containing adrenal mass. Unlike the liver and muscle, the spleen is not subject to fatty infiltration [6]. Although the adrenal-to-spleen ratio has been an excellent discriminator (except in patients with conditions involving abnormal iron deposits), the problem of considerable overlap between adenomas and metastatic tumors remains (14.9% [13/87] of tumors in our study).
Tsushima et al. [4] were the first to propose the signal intensity index. They reported that adrenal lesions with signal intensity indexes of more than 5% were adenomas and those with values of less than 5% were metastatic tumors and pheochromocytomas. The signal intensity index was associated with an accuracy of 100% and did not require measuring reference organs, eliminating the problem of signal intensity variability produced by the reference tissues.
In spite of the apparent advantages of this technique, some reports [9, 11] have indicated that the signal intensity index is less reliable than the method used by Mitchell et al. [3]. These researchers calculated the percentage of signal loss for adrenal lesions on opposed-phase T1-weighted MR images, using the liver as the reference tissue. Mayo-Smith et al. [9] also evaluated the signal intensity index together with the adrenal-to-spleen ratio, adrenal-to-muscle ratio, and adrenal-to-liver ratio. Their Az value for the signal intensity index was less than that for the adrenal-to-spleen ratio value; this result differed from our finding that the signal intensity index showed the greatest Az value of all discriminators. Reinig et al. [11] found the signal intensity index less reliable than the other ratios.
In our study, the signal intensity index was calculated with the same FLASH imaging sequence used by Tsushima et al. [4], and it proved to be an excellent discriminator for characterizing adenomas and metastatic tumors. Our accuracy in distinguishing between the lesions was 100% when the cutoff value for the signal intensity index was 11.2–16.5%. If the cutoff value of 5% had been used (as reported by Tsushima et al.), five of 19 metastatic tumors would have been misdisagnosed as cortical adenomas in our series. These five tumors included one lung carcinoma, one hepatocellular carcinoma, and three malignant lymphomas.
Inconsistencies between the Tsushima et al. [4] study and ours in establishing threshold values to achieve 100% accuracy may be related to the lipid content of metastatic tumors and to different imaging parameters such as TR, TE, or flip angle (TR, 100; in-phase TE, 11; opposed-phase TE, 13; flip angle, 20° were the parameters used by Tsushima et al.). The lowest signal intensity index of an adenoma in our series was 16.5%. Although the signal intensity indexes for all adenomas were dispersed in our study, they were all in the positive range and indicated intratumor lipid. No adenomas in our series had signal intensity indexes of less than 16.5%, whereas some adenomas in the study by Tsushima et al. had signal intensity indexes of less than 10%. In the study by Tsushima et al., the mean signal intensity indexes for adenomas and metastatic tumors were 17.0 ± 8.2% and –6.8 ± 9.2%, respectively. In our study, the mean signal intensity indexes for adenomas and metastatic tumors were 57.49 ± 15.33% and 0.22 ± 6.88%, respectively.
Although our values for the signal intensity index differed from those of Tsushima et al. [4], it seems to be possible to differentiate adenomas from metastatic tumors accurately using the signal intensity index. Reinig et al. [11, 26] implied that loss of relative signal intensity on opposed-phase MR imaging could be due to small amounts of histologic lipid that have been reported as being found in some metastases and adrenal cortical adenocarcinomas, although Reinig et al. found no pathologic evidence of lipid in their four metastatic lesions. Mitchell et al. [3] indicated that four of eight metastases they studied quantitatively showed reductions in the percentage of signal intensity loss relative to the signal intensity in the liver on opposed-phase MR imaging. In the study by Korobkin et al. [5], four of 15 nonadenomas had a 2–12% decrease in the signal intensity ratio relative to the the signal intensity of the liver. Tsushima et al. [4] reported that three of the 19 nonadenomas (16%) had a positive signal intensity index of 1–5%. In our study, eight (42.1%) of 19 metastatic tumors had positive signal intensity indexes, which ranged from 1% to 11.21%.
Slapa et al. [27] reported a large series involving adrenal tumors and chemical shift imaging using FLASH. These researchers found higher values for the signal intensity index of malignant tumors that were attributed to cortical carcinomas and metastatic tumors of renal cell carcinoma. The frequency of positive values for the signal intensity index from chemical shift images of metastatic adrenal tumors may be higher than we anticipated. However, few investigations have used chemical shift imaging to compare a large series of metastatic tumors and adenomas. Some metastatic adrenal tumors in our study had positive signal intensity indexes, suggesting lipid content; however, the amount of lipid seemed to be small in comparison with the amount present in adenomas. This finding suggests that 100% accuracy in discriminating between metastatic tumors and adenomas may be achievable.
Incomplete differentiation of adenomas from metastatic tumors on chemical shift imaging in previous studies may be attributable to the dispersion of lipid content in adenomas and higher levels of intratumor lipid in metastatic tumors. Furthermore, Leroy-Willig et al. [28] suggested that signal cancellation with opposed-phase MR imaging is also related to T1 values and that the signal intensity of some malignant tumors could overlap that of benign lesions on short-TR opposed-phase images. These researchers' studies of lipid quantification with MR imaging found that malignant lesions contained less than 6% lipid.
Several authors have disputed the hypothesis by Mitchell et al. [3] that hyperfunctioning adenomas may have lower concentrations of lipid than other benign cortical masses [4, 29, 30, 31]. In support of Mitchell et al., Slapa et al. [27], reported that signal intensity indexes for hyperfunctioning adenomas and nonhyperfunctioning adenomas differed so significantly that nonhyperfunctioning adenomas could be confidently differentiated from hyperfunctioning adenomas when the signal intensity index exceeded 33%. We found no significant differences between hyperfunctioning and nonhyperfunctioning adenomas on chemical shift imaging.
One interesting result that we found was the significant differences for all evaluation parameters between cortisol-producing and aldosterone-producing adenomas in chemical shift imaging. The signal intensity index values were significantly higher in aldosterone-producing adenomas than in cortisol-producing adenomas. The adrenal cortex usually contains abundant intracytoplasmic lipid, mainly composed of cholesterol, fatty acids, and neutral fat [32]. Histologically, cortical cells with abundant lipid appear to be "clear cells," whereas those with fewer lipids appear as "compact cells" [30]. Adenomas may contain varying proportions of clear and compact cells. The adenomas in patients with Cushing's syndrome are generally composed of enlarged lipid-laden clear cells scattered among more predominant compact cells. The cells in primary aldosteronism are mostly large lipid-laden clear cells similar to those seen in the zona fasciculata [30, 33, 34]. Therefore, the amount of cytoplasmic lipid may be greater in adenomas of primary aldosteronism than in those of Cushing's syndrome. Ichiyanagi et al. [29] and Korobkin et al. [30] have reported that CT attenuation was not significantly different for cortisol-producing, aldosterone-producing, and nonfunctioning adenomas, whereas there was a negative correlation between clear cell–compact cell ratios and CT attenuation. These results suggest that chemical shift imaging can be used to detect pathologic differences on the basis of the differences in lipid content in the aldosterone-producing and cortisol-producing adenomas.
Because small adenomas are more susceptible to partial volume effects with surrounding fat that can reduce signal on opposed-phase images, the lesions with diameters smaller than 1 cm were excluded from our assessment, and the regions of interest over the lesions were correctly chosen to avoid the partial volume artifacts at the edge of the lesions. We believe that we eliminated the partial volume effects as much as possible. The distinction between aldosterone-producing adenomas and cortisol-producing adenomas has not been easy to make, and the diagnosis is usually based on clinical findings and biochemical and hormonal data. However, this distinction between aldosterone-producing adenomas and cortisol-producing adenomas is less important than discriminating between hyperfunctioning and nonhyperfunctioning adenomas in patients with equivocal abnormal biochemical values.
Our study is limited by several factors including the small number of nonhyperfunctioning adenomas with histopathologic confirmation. These lesions are difficult to confirm because they are seldom resected.
Using regions of interest, we measured signal intensities of adrenal tumors twice with breath-holding; therefore, misregistration between in-phase and opposed-phase MR images could not be avoided. In a recent study, double-echo chemical shift FLASH imaging (in-phase TE, 5.2; opposed-phase TE, 2.7) was used to examine adrenal adenomas with the same MR imaging system as the one we used in our study [35]. Results of that study indicate that the double-echo sequence performed during a single breath-hold may be helpful in differentiating adrenal adenomas from other adrenal tumors without section misregistration. In fact, no overlaps were found in the signal intensity index between 16 adenomas and nine metastatic tumors in that series. Although theoretically the ideal TEs (2.2 and 4.4) seem to be preferable for precise evaluation, the difference between the double-echo sequence and the two ideal echoes was negligible according to the findings of the study.
Although it is not possible to identify whether the lipid or water is the dominant signal in the opposed-phase MR images, adrenal adenomas may contain less than 30% lipid [23, 28, 35], and the observed signal cancellation is thought to be due to lipid. According to the phantom study performed by Namimoto et al. [35], which took the T1 value of the normal adrenal gland into account, the signal intensity index reached a maximum value at 53% lipid fraction. These facts suggest a positive correlation between the lipid content of adrenal adenomas and signal intensity indexes.
In conclusion, the efficacy of chemical shift imaging for diagnosis of adrenal tumors has been established, but a consensus on the best method for evaluating adrenal lesions has yet to be reached despite a large number of studies on this subject. In our investigation, the signal intensity index was the most reliable evaluation method and the second most reliable was the adrenal-to-spleen ratio. We found no overlaps in the signal intensity index of cortical adenomas and metastatic tumors in our series.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. W. L. Boland, M. A. Blake, P. F. Hahn, and W. W. Mayo-Smith Incidental Adrenal Lesions: Principles, Techniques, and Algorithms for Imaging Characterization Radiology, December 1, 2008; 249(3): 756 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. T. Schindera, B. J. Soher, D. M. Delong, B. M. Dale, and E. M. Merkle Effect of Echo Time Pair Selection on Quantitative Analysis for Adrenal Tumor Characterization with In-Phase and Opposed-Phase MR Imaging: Initial Experience Radiology, July 1, 2008; 248(1): 140 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. K. Park, C. K. Kim, B. Kim, and J. H. Lee Comparison of Delayed Enhanced CT and Chemical Shift MR for Evaluating Hyperattenuating Incidental Adrenal Masses Radiology, June 1, 2007; 243(3): 760 - 765. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chong, K. S. Lee, H. Y. Kim, Y. K. Kim, B.-T. Kim, M. J. Chung, C. A Yi, and G. Y. Kwon Integrated PET-CT for the Characterization of Adrenal Gland Lesions in Cancer Patients: Diagnostic Efficacy and Interpretation Pitfalls RadioGraphics, November 1, 2006; 26(6): 1811 - 1824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Savci, Z. Yazici, N. Sahin, S. Akgoz, and E. Tuncel Value of Chemical Shift Subtraction MRI in Characterization of Adrenal Masses Am. J. Roentgenol., January 1, 2006; 186(1): 130 - 135. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Hussain, T. L. Chenevert, F. J. Londy, V. Gulani, S. D. Swanson, B. J. McKenna, H. D. Appelman, S. Adusumilli, J. K. Greenson, and H. S. Conjeevaram Hepatic Fat Fraction: MR Imaging for Quantitative Measurement and Display--Early Experience Radiology, December 1, 2005; 237(3): 1048 - 1055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Israel, M. Korobkin, C. Wang, E. N. Hecht, and G. A. Krinsky Comparison of Unenhanced CT and Chemical Shift MRI in Evaluating Lipid-Rich Adrenal Adenomas Am. J. Roentgenol., July 1, 2004; 183(1): 215 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Mansmann, J. Lau, E. Balk, M. Rothberg, Y. Miyachi, and S. R. Bornstein The Clinically Inapparent Adrenal Mass: Update in Diagnosis and Management Endocr. Rev., April 1, 2004; 25(2): 309 - 340. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and
adrenal-to-spleen (
), adrenal-to-muscle (
) and adrenal-to-liver
(•) ratios. Perfect discrimination is represented by 1.0. Signal intensity
index has largest area under curve (Az) and adrenal-to-spleen ratio
has second largest Az.
