| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, New York University Medical Center, 560 First Ave., HW202, New York, NY 10016.
Received July 12, 2004;
accepted after revision September 25, 2004.
Address correspondence to G. M. Israel
(gary.israel{at}yale.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twenty-three angiomyolipomas, 24 hemorrhagic-proteinaceous cysts, and 23 solid renal masses (nonangiomyolipomas) were retrospectively evaluated at 1.5 T with chemical shift MRI using TEs of 2.1-2.7 msec (opposed phase) and 4.8-5.3 msec (in phase). Two independent observers reviewed the MR images for signal loss on opposed-phase images and for the presence or absence of india ink artifact. An angiomyolipoma was diagnosed if the india ink artifact was identified at the interface of the mass and the kidney or was present within the renal mass.
RESULTS. Twenty-three (100%) of the 23 angiomyolipomas showed india ink artifact within the mass or at its interface with the kidney, and 18 (78.3%) of the 23 angiomyolipomas showed signal loss on opposed-phase MR images. In 24 (100%) of the 24 hemorrhagic-proteinaceous cysts, india ink artifact was not present within the mass or at its interface with the kidney. No signal loss was seen in hemorrhagic-proteinaceous cysts on opposed-phase MR images. In one (4%) of the 23 solid (nonangiomyolipoma) renal masses, the india ink artifact was identified at the interface of the mass with the kidney, and in two (9%) of the 23 masses, loss of signal was identified on the opposed-phase MR images.
CONCLUSION. The presence of india artifact at a renal mass-kidney interface or within a renal mass is indicative of angiomyolipoma.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two types of edge artifacts occur on chemical shift MRI: the chemical shift artifact and the india ink artifact. The chemical shift artifact is due to positional misregistration of the fat signal resulting from the difference in precessional frequency of fat and water protons [6, 7] and manifests as alternating bands of bright and dark signal along the frequency-encoding direction at fat-water interfaces. It is dependent on the receiver bandwidth (the artifact is more pronounced at lower bandwidths) [8], field strength (the artifact is less pronounced at lower field strengths) [6, 7], and the shape and orientation of the fat-water interface [9]. Alternatively, the india ink artifact is due to the presence of fat and water protons within the same imaging voxel, resulting in signal loss. This artifact can be recognized on opposed-phase MR images as a characteristic sharp black line at fat-water (fat-muscle or fat-solid organ) interfaces [10]. It occurs along the entire border of fat-water interfaces and not only in the frequency-encoding direction because it is a result of fat and water proton phase cancellation in all directions [8].
Because most angiomyolipomas contain macroscopic fat, the india ink artifact will appear at the interface of the angiomyolipoma with the kidney or at the interface of the fatty and nonfatty portions of the mass. Other renal masses (hemorrhagic-proteinaceous cysts and renal cell carcinomas) do not contain macroscopic fat, and for that reason, the india ink artifact would appear at the interface of the renal mass with the perinephric fat when the mass is exophytic. In the case of a completely intrarenal mass (nonangiomyolipoma), an india ink artifact would not be seen. Consequently, differentiating an angiomyolipoma from other renal masses on the basis of the location and the presence of the india ink artifact may be possible. However, to our knowledge, no studies have been published to support this hypothesis. The purpose of this study was to determine the reliability of the location of the india ink artifact and signal loss on opposed-phase chemical shift MRI to differentiate angiomyolipomas from other solid renal masses and hemorrhagic-proteinaceous renal cysts.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both groups of patients were cross-referenced with a radiology information system to determine which patients had also undergone CT of the abdomen. Twenty-six of the 82 patients in the angiomyolipoma group and 21 of the 155 patients of the hemorrhagic cyst group had undergone both CT and MRI. All these examinations were retrieved from the archive except for those for which the CT archive discs could not be found—four patients in the angiomyolipoma group and two patients in the hemorrhagic cyst group. One additional patient from the angiomyolipoma group was excluded secondary to extensive respiratory motion on the CT examination, and two patients from each group were excluded because a chemical shift T1-weighted MRI sequence was not performed. This yielded a total of 19 patients (four men and 15 women; average age, 67.8 years; range, 40-93 years) in the angiomyolipoma group and 17 patients (eight men and nine women; average age, 66.2 years; range, 40-80 years) in the hemorrhagic cyst group.
If a patient had more than one hemorrhagic-proteinaceous cyst or angiomyolipoma, the largest lesions, up to a maximal of three lesions, were evaluated. Twenty-three lesions in the angiomyolipoma group (average size, 2.6 cm; range, 0.7-15 cm) and 24 lesions in the hemorrhagic-proteinaceous cyst group (average size, 2.8 cm; range, 1.0-7.0 cm) were evaluated.
To identify a cohort of patients with an enhancing renal mass, we retrospectively searched the records in the MRI database for all cases in which "enhancing renal" was present in the final interpretation of the examination. This search was performed starting with records dated November 2003 and progressing back in time to identify 20 consecutive patients who also had pathologic correlation of the renal mass. This search yielded 12 men and eight women (average age, 63.4 years; range, 33-79 years). A total of 23 renal masses (average size, 2.8 cm; range, 0.8-6.8 cm) with pathologic correlation were available, including 15 renal cell carcinomas (clear cell subtype, n = 8; papillary subtype, n = 3; chromophobe subtype, n = 3; multilocular cystic renal cell carcinoma, n = 1), seven oncocytomas, and one metanephric adenoma. All examinations were retrieved from the archive and reloaded onto a PACS (Sienet, Siemens Medical Systems).
Establishing the Reference Standard
For the pathologically proven renal masses, the final pathology diagnosis
was the reference standard. For those renal masses without pathologic
correlation (angiomyolipoma and hemorrhagic-proteinaceous cyst groups), a
single radiologist reviewed the CT and MRI examinations to identify the renal
masses. Hounsfield unit values were obtained on CT using manually defined
circular or oval region-of-interest measurements performed in the portion of
the mass that visually had the lowest CT attenuation. If unenhanced CT images
were available, they were used for the Hounsfield unit values. If unenhanced
CT images were not available, the contrast-enhanced images were used.
An angiomyolipoma was diagnosed if a lesion measured -20 H or less [11]. The lesions that measured greater than -20 H on CT were considered hemorrhagic-proteinaceous cysts and were diagnosed as such at the initial clinical interpretation of the MRI examination. For that MRI examination, unenhanced T1-weighted, frequency-selective fat-suppressed T1-weighted, and gadolinium-enhanced frequency-selective fat-suppressed T1-weighted images were obtained. For all lesions, at least a portion of the mass was hyperintense with respect to the renal parenchyma on the unenhanced T1-weighted images, did not decrease in signal on the frequency-selective fat-suppressed T1-weighted images, and did not enhance after gadolinium administration using image subtraction. The combined results of the MRI and CT examinations were used as the reference standard.
Image Analysis
Two independent observers (with 3 and 10 years' experience in interpreting
abdominal MRI examinations) blinded to the CT, MRI, and pathologic results
reviewed the in- and opposed-phase MR images. Each observer evaluated the
opposed-phase MR images for the presence or absence of india ink artifact
[10] at the interface of the
renal mass and the kidney or within the renal mass. Each observer also
evaluated the in- and opposed-phase images for the presence or absence of
signal loss (subjectively) in the renal mass using the kidney as a reference.
Any discrepancy between the two radiologists was settled by consensus. A mass
was considered an angiomyolipoma if the india ink artifact was identified at
the interface of the mass and the kidney or was present within the renal
mass.
One investigator, who was not involved with image analysis, determined whether the angiomyolipomas were exophytic or intrarenal. An angiomyolipoma was considered exophytic if more than half of the mass was extrarenal. All other angiomyolipomas were considered intrarenal.
Comparison of Imaging Findings in the Misdiagnosed Lesions
After all lesions were evaluated, a side-by-side comparison of the in- and
opposed-phase MR images with the final results was performed for all
false-positive and false-negative masses to identify possible causes of
misdiagnoses.
Imaging Technique
All MRI examinations were performed on a 1.5-T system (Vision or Symphony,
Siemens Medical Systems) using a torso phased-array coil. All patients
underwent axial breath-hold dual-echo T1-weighted imaging (in phase and
opposed phase) using a 2D gradient-echo sequence. The TR was 160-200 msec and
the TE was 2.1-2.7 msec (opposed phase) and 4.8-5.3 msec (in phase). The
average slice thickness was 6.9 mm (range, 5-8 mm). Other imaging parameters
were as follows: matrix, 128 x 256; imaging bandwidth, 475-610 Hz;
transmitter bandwidth, 2 kHz; interslice gap, 0-2 mm; and rectangular field of
view, 200-263 x 320-375 mm.
All CT examinations were performed on helical scanners (HiSpeed Advantage [n = 6] or CTi [n = 20], GE Healthcare; or Volume Zoom [n = 6] or Sensation 16 [n = 4], Siemens Medical Systems). For the angiomyolipoma group, 14 examinations were performed without IV contrast material, and five examinations were performed with IV contrast material. The slice collimation varied (7 mm, n = 6; 5 mm, n = 9; 4 mm, n = 2; 3 mm, n = 2). For the hemorrhagic cyst group, 11 examinations were performed without IV contrast material, and six examinations were performed with IV contrast material. The slice collimation varied (7 mm, n = 5; 5 mm, n = 7; 4 mm, n = 1; 3 mm, n = 2; 1.25 mm, n = 2).
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The 24 masses in the hemorrhagic-proteinaceous cyst group had an average CT attenuation of 39.3 H (range, 19-80 H). In 24 (100%) of the 24 masses, india ink artifact was not identified at the interface of the mass with the kidney or within the mass (Figs. 2A, and 2B). Loss of signal on the opposed-phase MR image was not identified in any of these masses.
|
|
In 20 (87%) of the 23 pathologically proven renal masses, the india ink artifact was not identified at the interface of the mass with the kidney or within the mass, nor was there loss of signal on the opposed-phase MR images. In two (9%) of the 23 masses, both of which were exophytic and measured 1.9 and 3.2 cm, respectively, the india ink artifact was not identified at the interface of the mass with the kidney or within the mass, but loss of signal was identified on the opposed-phase MR images (Figs. 3A, and 3B). Both of these masses represent the clear cell subtype of renal cell carcinoma. In one (4%) of the 23 masses, loss of signal was not identified on the opposed-phase MR images, but the india ink artifact was identified at the interface of the mass with the kidney, which is indicative of an angiomyolipoma (Fig. 4). At pathology, this mass represented a renal cell carcinoma (chromophobe subtype).
|
|
|
Comparison of Imaging Findings in the Misdiagnosed Lesions
One 8-mm pathologically proven renal cell carcinoma was misdiagnosed as an
angiomyolipoma (Fig. 4). On
reviewing the images, we noted that there was an apparent loss of signal at
the interface of the mass and the kidney that had a superficial resemblance to
the india ink artifact. This finding is likely artifactual and secondary to a
combination of volume averaging (slice thickness, 8 mm; interslice gap, 0) and
"Gibbs" artifact
[8] (secondary to the use of a
128 phase matrix).
A consensus interpretation was necessary in a total of 11 masses. In six angiomyolipomas and three renal cell carcinomas, observers initially disagreed about whether the mass lost signal on the opposed-phase images. In one angiomyolipoma and one renal cell carcinoma, the observers initially disagreed about the location of the india ink artifact, but both of these masses were correctly characterized at the consensus interpretation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When compared with in-phase images, opposed-phase MR images showed 18 (78.3%) of 23 angiomyolipomas, zero (0%) of 24 hemorrhagic-proteinaceous cysts, and two (9%) of 23 enhancing renal masses as having lost signal. Although this may be characteristic of angiomyolipomas, it is not diagnostic because the clear cell subtype of renal cell carcinoma may also lose signal on opposed-phase images [12]. In this study, both enhancing renal masses (nonangiomyolipomas) that lost signal on opposed-phase images represented the clear cell subtype of renal cell carcinomas. Their differentiation from angiomyolipomas should not be difficult because clear cell carcinomas are almost never hyperintense on in-phase gradient-echo MRI unless they have undergone hemorrhage, and in this setting, blood products will not lose signal on opposed-phase MRI.
Characterizing small angiomyolipomas or angiomyolipomas containing small amounts of macroscopic fat can be problematic secondary to partial volume artifacts and limitations in spatial resolution. In this study, an 8-mm renal cell carcinoma was misdiagnosed as an angiomyolipoma because observers thought that india ink artifact was present at the interface of the mass with the kidney. In this case, the apparent loss of signal at the interface of the mass and the kidney was likely artifactual and secondary to a combination of volume averaging (slice thickness, 8 mm) and Gibbs artifact [8]. However, a 7-mm angiomyolipoma in this series was correctly characterized by identifying the india ink artifact at its interface with the kidney.
Small angiomyolipomas or angiomyolipomas that contain only a small amount of macroscopic fat can be difficult to characterize. A limitation of using opposed-phase chemical shift MRI to characterize these angiomyolipomas is the potential difficulty in differentiating loss of signal within a mass from the india ink artifact surrounding the mass or the fatty portion of the mass. In these cases, frequency-selective fat-suppressed T1-weighted imaging may prove to be more sensitive in characterizing these types of angiomyolipomas, and further studies are necessary.
There are case reports of renal cell carcinomas that contain fat [13-15]. Theoretically, this would raise uncertainty regarding the proper diagnosis of a fat-containing renal mass. However, these renal cell carcinomas are usually large and contain only a small amount of fat that may represent engulfed renal sinus, perirenal fat, or even an engulfed coexistent angiomyolipoma. Nevertheless, it is rare for a solid renal mass that contains fat not to represent an angiomyolipoma [14].
Opposed-phase chemical shift MRI has some advantages over frequency-selective fat-suppression techniques. At times, frequency-selective fat-suppressed sequences may have nonuniform fat suppression, and in this instance, the loss of signal within an angiomyolipoma may not be evident. In addition, at low field strengths, frequency-selective fat-suppressed sequences cannot be performed during a comfortable breath-hold. Opposed-phase chemical shift MRI can be performed at low field strengths without adding to the examination time and is not dependent on field homogeneity; therefore, an angiomyolipoma can still theoretically be diagnosed with confidence.
We recognize limitations to this study including its retrospective nature, which introduces a case-selection bias. Also, the imaging protocols were not standardized across all the cases, and in some cases, the CT slice collimation was 7 mm. In cases of angiomyolipomas that contain a small amount of fat, a thinner slice collimation may be necessary for accurate characterization. Nevertheless, all angiomyolipomas in this series were correctly characterized. Next, six of the CT examinations in the hemorrhagic cyst group were performed with IV contrast material, and it is possible that a small amount of fat within a renal mass could have been masked by the adjacent enhancing renal parenchyma. Finally, pathologic proof was not available for all the cases; however, they were characterized using clinically accepted and widely used techniques.
In conclusion, although a comparison of T1-weighted and frequency-selective fat-suppressed T1-weighted images has been traditionally used to diagnose angiomyolipomas of the kidney, chemical shift MRI can also accurately characterize angiomyolipomas. On opposed-phase chemical shift MRI, the presence of the india ink artifact at the interface of a renal mass with the renal parenchyma or within the renal mass is indicative of an angiomyolipoma. Therefore, the presence and location of the india ink artifact on opposed-phase MR images should be evaluated carefully when characterizing a renal mass. If the findings on opposed-phase and frequency-selective fat-suppressed T1-weighted images are discrepant, more careful scrutiny or possibly repeating sequences may be necessary.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. R. Oliva, J. N. Glickman, K. H. Zou, S. Y. Teo, K. J. Mortele, M. S. Rocha, and S. G. Silverman Renal Cell Carcinoma: T1 and T2 Signal Intensity Characteristics of Papillary and Clear Cell Types Correlated with Pathology Am. J. Roentgenol., June 1, 2009; 192(6): 1524 - 1530. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Taouli, R. K. Thakur, L. Mannelli, J. S. Babb, S. Kim, E. M. Hecht, V. S. Lee, and G. M. Israel Renal Lesions: Characterization with Diffusion-weighted Imaging versus Contrast-enhanced MR Imaging Radiology, May 1, 2009; 251(2): 398 - 407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Silverman, J. R. Leyendecker, and E. S. Amis Jr What Is the Current Role of CT Urography and MR Urography in the Evaluation of the Urinary Tract? Radiology, February 1, 2009; 250(2): 309 - 323. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Silverman, G. M. Israel, B. R. Herts, and J. P. Richie Management of the Incidental Renal Mass Radiology, October 1, 2008; 249(1): 16 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Pedrosa, M. R. Sun, M. Spencer, E. M. Genega, A. F. Olumi, W. C. Dewolf, and N. M. Rofsky MR Imaging of Renal Masses: Correlation with Findings at Surgery and Pathologic Analysis1 RadioGraphics, July 1, 2008; 28(4): 985 - 1003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kemmling, I. Noelte, L. Gerigk, S. Singer, C. Groden, and J. Scharf A Diagnostic Pitfall for Intracranial Aneurysms in Time-of-Flight MR Angiography: Small Intracranial Lipomas Am. J. Roentgenol., January 1, 2008; 190(1): W62 - W67. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Macari, R. Yeretsian, and J. Babb Assessment of Low Signal Adjacent to the Falciform Ligament on Contrast-Enhanced MRI Am. J. Roentgenol., December 1, 2007; 189(6): 1443 - 1448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Silverman, K. J. Mortele, K. Tuncali, M. Jinzaki, and E. S. Cibas Hyperattenuating Renal Masses: Etiologies, Pathogenesis, and Imaging Evaluation RadioGraphics, July 1, 2007; 27(4): 1131 - 1143. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
