| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Universitätsklinikum Carl Gustav
Carus an der Technischen Universität Dresden, Fetscherstr. 74, D-01307
Dresden, Germany.
2 Clinic for Ear, Nose and Throat, Universitätsklinikum Carl Gustav Carus
an der Technischen Universität Dresden, D-01307 Dresden, Germany.
Received May 13, 2002;
accepted after revision December 19, 2002.
Address correspondence to P. Aikele.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Twenty-two patients who had undergone resection of cholesteatoma were referred for MR imaging. MR imaging (1.5 T) was performed using a diffusion-weighted single-shot spin-echo echoplanar sequence, a proton density and T2-weighted double-echo turbo spin-echo sequence, and T1-weighted spin-echo sequences before and after IV injection of gadopentetate dimeglumine (0.1 mmol/kg of body weight). An experienced reviewer evaluated the diffusion-weighted MR images for the presence of a high-signal-intensity cholesteatoma. Imaging findings were correlated with findings from surgery in 17 patients and with findings from clinical follow-up examination in five patients.
RESULTS. Diffusion-weighted MR imaging combined with conventional MR imaging depicted 10 of 13 cholesteatomas (sensitivity, 77%). The three lesions that were missed were smaller than 5 mm. All the MR images of patients without cholesteatoma were correctly interpreted as showing negative findings for cholesteatoma (specificity, 100%). The positive predictive value and negative predictive value were 100% and 75%, respectively.
CONCLUSION. Diffusion-weighted MR imaging combined with conventional MR imaging can confirm residual or recurrent cholesteatoma in patients who have undergone middle ear surgery by showing a high-signal-intensity lesion. Because tumors smaller than 5 mm may be missed, a diffusion-weighted MR imaging study with negative findings does not exclude small residual or recurrent cholesteatoma.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
High-resolution CT [9] is the method of choice for imaging the middle ear. In patients with recurrent cholesteatoma, high-resolution CT can show the extent of tumor tissue in relation to the ossicles, the labyrinthine structures, the epitympanic space, and the mastoid process. Osseous destructions are also reliably depicted. High-resolution CT has a high negative predictive value if neither a soft-tissue mass nor bony destructions are shown [10]. However, if a soft-tissue mass in the cavity of the middle ear is seen on high-resolution CT, diagnosis of the mass is not possible because cholesteatoma, mucoid secretion, granulation tissue, fibrous tissue, and cholesterol granuloma cannot be differentiated from one another on high-resolution CT [11, 12].
MR imaging using standard T1- and T2-weighted sequences shows tissues better than high-resolution CT. Therefore, in patients with primary (genuine) cholesteatoma, MR imaging plays an important complementary role to high-resolution CT in the preoperative diagnostic workup. Although high-resolution CT provides excellent bony resolution for showing the anatomy, MR imaging provides specificity in characterizing potential soft-tissue abnormalities shown on high-resolution CT. However, MR imaging with standard T1- and T2-weighted pulse sequences frequently fails to allow differentiation of cholesteatoma from other soft tissues or mucoid secretions [13–15], particularly in patients who have undergone middle ear surgery (acquired cholesteatoma). Recent studies of diffusion-weighted MR imaging show that it is sensitive to cholesteatoma tissue [16–18]. One recent case report highlights the potential of diffusion-weighted MR imaging in differentiating cholesteatoma from granulation tissue in patients who have undergone mastoidectomy [19].
The aim of our prospective study was to determine the sensitivity and specificity of diffusion-weighted MR imaging combined with conventional MR imaging for the detection of residual or recurrent cholesteatoma in the petrous bone in patients who have undergone middle ear surgery.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
MR Imaging
MR imaging was performed at 1.5 T (Magnetom Vision, Siemens, Erlangen,
Germany) using a circularly polarized transmit-and-receive head coil. The same
protocol was used for all patients. Diffusion-weighted images were obtained
with a transverse diffusion-weighted single-shot spin-echo echoplanar pulse
sequence (TR/TE, 3900/100; slice thickness, 4 mm; matrix, 96 x 200;
field of view, 240 x 240 mm; b value, 1000 sec/mm).
Diffusion-sensitizing gradients were applied sequentially along the three
orthogonal planes, and trace-weighted images of the diffusion tensor in the
three orthogonal directions were generated with software provided by the
manufacturer. Apparent diffusion coefficient maps and exponential
diffusion-weighted imaging were not used. Proton density and T2-weighted
images were acquired using a transverse double-echo turbo spin-echo sequence
(TR/first-echo TE, second-echo TE, 2735/17, 102; slice thickness, 5 mm;
matrix, 260 x 512; field of view, 190 x 250 mm). T1-weighted
images were obtained using a transverse plane and transverse and coronal
contrast-enhanced spin-echo sequence (TR/TE, 600/14; slice thickness, 3 mm;
matrix, 512 x 512; field of view, 230 x 230 mm) before and after
contrast administration. Gadopentetate dimeglumine (Magnevist, Schering,
Berlin, Germany) was injected at a dose of 0.1 mmol/kg of body weight.
Imaging Evaluation
A radiologist with long-standing experience in head and neck imaging
prospectively evaluated all the MR images without knowing the results from
surgery or follow-up examinations. The criteria for the diagnosis
cholesteatoma on MR imaging were an area of markedly hyperintense signal
compared with that of brain tissue on diffusion-weighted MR imaging and a
soft-tissue mass seen on at least one of the conventional MR sequences (Figs.
1A,
1B,
1C). The reviewer classified MR
images as showing either positive or negative findings for cholesteatoma.
|
|
|
Histopathology was the gold standard in the 17 patients who underwent surgery after MR imaging. The five patients in the nonsurgical group had unremarkable findings at follow-up examination 11–19 months after MR imaging and were categorized as not having residual or recurrent cholesteatoma. Because the surgical technique was fractionated tissue resection, the tumor could not be measured intraoperatively; therefore, tumor size was determined using the MR images. The greatest tumor dimension on a single slice was measured.
Statistical Evaluation
The positive predictive value, negative predictive value, sensitivity, and
specificity of MR imaging in the diagnosis of cholesteatoma were
calculated.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
Clinical findings for the 10 patients with true-positive results on MR imaging included perforated eardrum with and without moist tympanum, retraction of the eardrum with and without moist eardrum, polypous granulation tissue with secretion from the external auditory canal, granulation tissue at the eardrum, and facial nerve palsy with an intact but arid eardrum in one patient each. In the remaining three patients with true-positive results on MR imaging, two had an intact eardrum and clinically unremarkable findings, whereas the eardrum was not visible because of a tight meatus in the remaining patient.
MR imaging failed to show the cholesteatoma as an area of markedly high signal intensity relative to brain tissue on diffusion-weighted imaging in three cases of surgically proven cholesteatoma. The greatest diameter of all three lesions was less than 5 mm. Clinical findings in these three patients were retraction of the eardrum in two patients and eardrum destruction with visible tumor tissue and secretion in one patient.
In the four patients with surgical correlation for true-negative MR imaging results, one patient presented with an obturating polyp in the external auditory meatus, one with an intact but moist eardrum, and one with a perforated eardrum with secretion. The eardrum could not be evaluated in the fourth patient because of a tight meatus.
By the time of MR imaging, five of the 22 patients had clinically unremarkable findings and were not referred for surgery. Findings from follow-up examinations were also unremarkable. In these patients, diffusion-weighted MR imaging showed low signal intensity and was correctly interpreted as showing negative findings for cholesteatoma. There was no case in which diffusion-weighted imaging displayed a markedly hyperintense signal without a soft-tissue mass seen on at least one of the other conventional sequences (i.e., no false-positive results). Sensitivity and specificity of conventional MR imaging with diffusion-weighted MR imaging were 77% and 100%, respectively. Positive predictive value and negative predictive value were 100% and 75%, respectively.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Both cholesteatomas and epidermoid cysts are cystic lesions lined with keratin-producing squamous epithelium and filled with desquamation debris. In epidermoid cysts, two mechanisms have been discussed to explain high signal intensity on diffusion-weighted imaging—that is, restricted molecular diffusion and the T2 shine-through effect. Restricted diffusion occurs in patients with cytotoxic damage from ischemia, inflammation, trauma, or tumor. The T2 shine-through effect is caused by the bright signal of epidermoid cysts on standard T2-weighted images.
The diffusion-weighted MR imaging technique allows water molecules to be observed moving within tissue structures. On diffusion-weighted imaging, the signal intensity of mobile water molecules is lower than that of immobile water molecules [20]. For example, cerebrospinal fluid has low signal intensity on diffusion-weighted imaging because of its abundant mobile water molecules. Tumor tissue usually also has high water content, but the diffusion of water molecules between the extra- and intracellular compartments is impaired. Recent studies have shown that the mean apparent diffusion coefficient of epidermoid tumors is lower than that of cerebrospinal fluid, slightly higher than that of white or gray matter [21, 22], or similar to that of gray matter [23]. Schaefer et al. [23] and Annet et al. [22] concluded that both restricted diffusion and the T2 shine-through effect are involved in the bright signal of epidermoid tumors on diffusion-weighted imaging. Chen et al. [21] concluded that the hyperintensity of epidermoid tumors on diffusion-weighted imaging is not caused by diffusion restriction but by the T2 shine-through effect.
The exact cause of increased signal of cholesteatoma on diffusion-weighted MR imaging is still unknown. However, cholesterol-containing detritus of cholesteatoma also has high signal intensity on standard T2-weighted MR images, which suggests the T2 shine-through effect contributes to hyperintensity on diffusion-weighted imaging. In addition, reduced diffusion in cholesteatoma compared with granulation tissue, fibrous tissue, or mucoid secretion is assumed to contribute to high signal intensity of cholesteatoma on diffusion-weighted imaging [16].
In the present study, diffusion-weighted MR imaging depicted 10 cholesteatomas, of which three were between 5 and 10 mm. Consistent with the findings of a previous study [16], our study showed that diffusion-weighted imaging failed to depict three tumors with a diameter smaller than 5 mm. Presumably, high lesion contrast of cholesteatoma on diffusion-weighted imaging does not compensate for limitations of spatial resolution in tumors of that size. However, the sensitivity for residual or recurrent cholesteatomas that are larger than 10 mm in our study was 100%. Of the limited number of reports dealing with MR imaging of patients who have undergone surgery of cholesteatoma, no data are available about the sensitivity depending on tumor size. The overall sensitivity of MR imaging with conventional pulse sequences after previous cholesteatoma surgery ranges between 57% and 79%, but sensitivity data for diffusion-weighted imaging are lacking [13, 15, 24]. In the same clinical setting, MR imaging without the diffusion-weighted sequence provided correct radiosurgical correlation in only 50–70% of cases. The lowest percentage of false-negative findings quoted in the recent MR imaging studies without diffusion-weighted imaging is 10% [15, 24].
Specificity is another parameter that determines the strength of an imaging test. For example, high-resolution CT of the middle ear has excellent spatial resolution, and even small soft-tissue structures can accurately be delineated against bony structures and air-filled cavities (high sensitivity). However, high-resolution CT fails to allow cholesteatoma to be differentiated from other soft-tissue structures, which is a significant limitation for postoperative imaging. Thus, the method has poor specificity due to a considerable number of false-positive cases. However, if high-resolution CT shows a free air cavity of the mastoid process, the negative predictive value is 100% [10]. As for MR imaging without diffusion-weighted imaging, the lowest percentage of false-positive cases in patients who have undergone middle ear surgery for cholesteatoma was 13% [15, 24]. Recent studies of MR imaging without diffusion-weighted imaging after previous surgery showed a specificity of 63–71% and a positive predictive value of 50–78% [13, 15, 24]. In the same clinical setting, high-resolution CT provided a specificity of 48–54% and a positive predictive value of 29–41% [10, 12, 25]. Thus, conventional MR imaging combined with diffusion-weighted MR imaging has superior specificity (100%) and positive predictive value (100%) in depicting cholesteatoma recurrence. In the present study, no false-positive case was recorded. The lack of a soft-tissue mass on conventional pulse sequences in combination with a hypointense signal on diffusion-weighted imaging provided 100% specificity along with a negative predictive value of 75%.
Although specificity and positive predictive value were 100% in our study, diffusion-weighted imaging may potentially cause false-positive results. High signal intensity in the area of the petrous bone is seen not only with cholesteatomas. Susceptibility artifacts at the air–bone border at the base of the skull, chordomas, cholesterol granulomas, and abscesses may also appear markedly hyperintense on diffusion-weighted imaging [26]. Susceptibility artifacts at the base of the skull can be minimized when a scan is angled parallel to the base of the skull. In doubtful cases of hyperintense areas, performing the diffusion-weighted imaging sequence with two different scan angles can be helpful. Susceptibility artifacts appear in a typical location at the air–bone border, are commonly bilateral, frequently show a streaky aspect, and lack a soft-tissue mass in the same location on standard MR imaging sequences. Chordoma can be differentiated from cholesteatoma by its typical location in the clivus region and its marked contrast enhancement as compared with the faint signal changes seen with cholesteatoma after injection of a contrast agent. Cerebral abscess has been described as displaying high signal intensity on diffusion-weighted imaging [26], but no single case of mastoid abscess imaged on diffusion-weighted imaging is, to our knowledge, available in the literature. A recent article reported cases of cholesterol granuloma with high signal intensity on diffusion-weighted imaging. However, differentiation from cholesteatoma was possible by means of the conventional pulse sequences [27]. Therefore, conventional T1-weighted, T2-weighted, proton density–weighted, and contrast-enhanced T1-weighted images should always be obtained when MR imaging with a diffusion-weighted sequence is performed.
Our study has a number of limitations. Only nine of 22 patients presented without cholesteatoma. Four were surgically proven, and five were clinically unremarkable at follow-up examinations over a minimum of 11 months. In strict statistical terms, the uneven number of normal and pathologic cases and lack of surgical proof in five cases weaken specificity data. As for sensitivity, three of the 13 tumors were 2 cm or larger. One can assume that soft-tissue masses of that size can also be diagnosed as cholesteatoma recurrence in patients who have undergone middle ear surgery using other techniques, such as MR imaging without diffusion-weighted imaging. Thus, additional studies are needed to define the role of the diffusion-weighted sequence in a greater number of lesions smaller than 10 mm. Another potential limitation is the somewhat inconsistent patient recruitment. Regardless of the results of MR imaging, 17 of the 22 patients were prospectively scheduled for surgery. In the remaining five cases, the likelihood of tumor recurrence was extremely low, and MR imaging confirmed the lack of tumor tissue. Thus, the role of diffusion-weighted imaging in a clinical scenario of suspected tumor recurrence can be estimated only on the basis of 17 surgical cases of which histopathology confirmed the clinical diagnosis in 13. Additional studies with diffusion-weighted imaging should focus on cases in which clinical examinations are inconclusive and exploratory surgery is planned. If such studies provide evidence that conventional MR imaging with a diffusion-weighted sequence can reliably separate surgical candidates from patients without tumor recurrence, the method deserves a place in the clinical setting.
In conclusion, conventional MR imaging with diffusion-weighted imaging is recommended in patients with a suspicion of cholesteatoma recurrence or residual tumor tissue who have undergone middle ear surgery. High signal intensity on a diffusion-weighted sequence and a soft-tissue mass seen on conventional MR imaging sequences have a positive predictive value of 100% for cholesteatoma. Because tumors smaller than 5 mm may be missed (negative predictive value, 75%), a lack of high signal intensity on diffusion-weighted imaging does not exclude tumor tissue, and second-look operations may be performed.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Lehmann, G. Saliou, C. Brochart, C. Page, B. Deschepper, J.N. Vallee, and H. Deramond 3T MR Imaging of Postoperative Recurrent Middle Ear Cholesteatomas: Value of Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction Diffusion-Weighted MR Imaging AJNR Am. J. Neuroradiol., February 1, 2009; 30(2): 423 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Venail, A. Bonafe, V. Poirrier, M. Mondain, and A. Uziel Comparison of Echo-Planar Diffusion-Weighted Imaging and Delayed Postcontrast T1-Weighted MR Imaging for the Detection of Residual Cholesteatoma AJNR Am. J. Neuroradiol., August 1, 2008; 29(7): 1363 - 1368. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Chung, M. D. Murphey, C. S. Specht, R. Cube, and J. Smirniotopoulos From the Archives of the AFIP Pediatric Orbit Tumors and Tumorlike Lesions: Osseous Lesions of the Orbit1 RadioGraphics, July 1, 2008; 28(4): 1193 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. De Foer, J.-P. Vercruysse, B. Pilet, J. Michiels, R. Vertriest, M. Pouillon, T. Somers, J.W. Casselman, and E. Offeciers Single-Shot, Turbo Spin-Echo, Diffusion-Weighted Imaging versus Spin-Echo-Planar, Diffusion-Weighted Imaging in the Detection of Acquired Middle Ear Cholesteatoma AJNR Am. J. Neuroradiol., August 1, 2006; 27(7): 1480 - 1482. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Dubrulle, R. Souillard, D. Chechin, F. M. Vaneecloo, A. Desaulty, and C. Vincent Diffusion-weighted MR Imaging Sequence in the Detection of Postoperative Recurrent Cholesteatoma Radiology, November 22, 2005; (2005) 2381041649. [Abstract] [Full Text]
|
|
|
|
 |
|
 E. C. Kavanagh, D. M. Fenton, D. Griesdale, and D. A. Graeb MRI of Acquired Cholesteatoma Presenting as a Temporal Lobe Mass Am. J. Roentgenol., September 1, 2005; 185(3): 788 - 789. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
