Localization Evaluation of Primary Middle Ear Cholesteatoma With Fusion of Turbo Spin-Echo Diffusion-Weighted Imaging and High-Resolution Computed Tomography : Journal of Computer Assisted Tomography

Secondary Logo

Journal Logo

Neuroimaging: Head and Neck

Localization Evaluation of Primary Middle Ear Cholesteatoma With Fusion of Turbo Spin-Echo Diffusion-Weighted Imaging and High-Resolution Computed Tomography

Fan, Xiaoxue MD; Ding, Changwei PhD; Liu, Zhaoyu PhD

Author Information
Journal of Computer Assisted Tomography 47(1):p 144-150, 1/2 2023. | DOI: 10.1097/RCT.0000000000001389
  • Free

Abstract

A cholesteatoma has localized invasiveness and can induce conductive hearing impairment, facial paralysis, labyrinthine fistulae, cerebral abscesses, and sigmoid sinus thrombosis when the lesions expand to invade adjacent structures.1 At present, surgical resection is the only therapy for cholesteatomas. The choice of surgical approach is mainly based on the extent of involvement of the cholesteatoma. Transcanal endoscopic ear surgery (TEES) is a minimally invasive procedure that is used to resect a cholesteatoma, and it can protect the bone and mucosa of the mastoid process.2 When a cholesteatoma is limited to the middle ear cavity, it may be resected by TEES, but a cholesteatoma lesion invading the mastoid process region may be resected by microscopic ear surgery (MES) or microscopic ear surgery + TEES.3,4 Therefore, it is very important to accurately detect and precisely locate the cholesteatoma to guide the surgical approach.

High-resolution computed tomography (HRCT) can clearly display small middle ear structures that cannot be visualized directly.5–7 However, because HRCT cannot accurately distinguish a cholesteatoma from inflammatory tissue, its diagnostic sensitivity and specificity are both low.8,9 Cholesteatoma lesions have a high keratin content, which results in a high signal on diffusion-weighted imaging (DWI); therefore, DWI has been used as a common technique to detect cholesteatoma.10 Turbo spin-echo DWI (TSE-DWI) is a type of non–echo-planar imaging (non-EPI) DWI that has high spatial resolution and diagnostic efficiency and demonstrates few bone artifacts.10–12 However, it cannot show the fine structure of bones and poorly localized cholesteatomas.13 Although previous studies have shown that DWI-HRCT fusion images can display both cholesteatomas and a clear visualization of the adjacent anatomic structures,13–15 their sample sizes were small, with typically fewer than 14 cases.

The STAM cholesteatoma classification and staging system (difficult access sites [S], the tympanic cavity [T], the attic [A], and the mastoid [M]) was created in 2016 with the consensus of the experts of the European Academy of Otology and Neurotology and the Japanese Otological Society with the purposes of dividing the middle ear space to aid localization and staging and simplifying the description of cholesteatoma invasion.16 The application of fusion imaging to diagnose and localize cholesteatomas utilizing the STAM system has not yet been reported. Therefore, this study aimed to evaluate the application of fusion imaging with regard to efficacy of diagnosing cholesteatomas and localizing cholesteatomas based on STAM.

MATERIALS AND METHODS

Case Data

This was a prospective study approved by the appropriate ethics committee (2019PS069J), and informed consent was obtained from all patients. The patients suspected with a middle ear cholesteatoma were prospectively enrolled and treated in our hospital from May 2019 to April 2020. The inclusion criteria were as follows: (1) patients who did not receive any surgical treatment prior to examination; (2) patients who complained of hearing loss, long-term continuous or intermittent pustules, and tympanic membrane perforation (clinical symptoms suggestive of cholesteatoma); (3) patients who underwent HRCT, standard magnetic resonance imaging (MRI) including the TSE-DWI sequence, and those with TSE-DWI-CT fusion images provided by postprocessing; (4) patients who were treated surgically after examination; and (5) patients whose cholesteatoma lesions were precisely recorded in detail when found during the operation (if the pathological assessment revealed a cholesteatoma).

Equipment and Scan Protocol

The scan range was from the superior border of the petrosum to the inferior border of the mastoid process. The signals were collected using a Philips Ingenia 3.0 T superconductive MRI scanner (Philips Electronics, Inc, Amsterdam, the Netherlands) and 32-channel phased array coils (Invivo Corporation, Gainesville, Fla). The following parameters were used: (1) cross-sectional T1-weighted: repetition time (TR): 3000 ms, echo time (TE): 80 ms, matrix: 308 × 192, slice thickness: 2 mm, and slice interval: 1 mm; (2) cross-sectional T2-weighted: TR: 3000 ms, TE: 80 ms, matrix: 308 × 192, slice thickness: 2 mm, and slice interval: 1 mm; (3) cross-sectional EPI-DWI: b = 0, 1,000 s/mm2, TR: 10,202.25 ms, TE: 89.16 ms, matrix: 256 × 256, slice thickness: 1 mm, and slice interval: 1.1 mm; and (4) cross-sectional TSE-DWI: b = 0, 1000 s/mm2, TR: 3000 ms, TE: 72 ms, matrix: 118 × 87, slice thickness: 1.5 mm, and slice interval: 1 mm. The images were acquired with a Philips iCT 256CT scanner (Philips Healthcare, Cleveland, Ohio) with the following parameters: focal spot size: 1 × 1.2 mm, collimation: 20 × 0.625, ear HRCT scan mode: spiral scan, pitch: 0.25, matrix: 768 × 768, peak tube voltage: 120 kV, 200 mA/s, slice thickness for reconstruction: 1 mm, interval: 0.5 mm, rotation time: 0.4 seconds, filter function: Y-Sharp (YE), window width: 4000, and window level: 700.

Turbo Spin-Echo DWI CT Image Fusion

The original HRCT images and TSE-DWI images were uploaded to a 3-dimensional reconstruction postprocessing workstation (Philips, Amsterdam, the Netherlands) for image processing. High-resolution computed tomography images were reconstructed with the field of view and slice thickness of TSE-DWI images as references, such that the reconstructed CT images had the same field of view and slice thickness as the TSE-DWI images. These 3 sets of images—the original CT images, the TSE-DWI and the TSE-DWI-CT fusion images—were saved back into our PACS system under the same study identification number, thereby ensuring that all 3 series would be available to review together upon opening a given study folder. According to the anatomy of the internal acoustic meatus and semicircular canal, the reconstructed CT images and TSE-DWI images were fused and trimmed to form TSE-DWI-CT fusion images. The 3-dimensional reconstruction software fused the CT and TSE-DWI images with the TSE-DWI image data displayed in color. Anecdotally, we note that the software displayed the higher intensity part of the spectrum of DWI values as visible colors, for example, when comparing Figures 3C and D. The image fusion process was performed by head and neck radiologists. The time to create the fused images was approximately 3 minutes.

Image Analysis

Cholesteatomas were defined according to different methods as follows: (1) HRCT images: a soft-tissue density shadow with bone destruction of the scutum, tegmen tympani, or auditory ossicle17; (2) TSE-DWI images: a middle ear region brighter than the cerebellar parenchyma at a DWI signal of b = 1000 s/mm210; and (3) TSE-DWI-CT fusion images: a focus of hyperintensity in the middle ear region and an increase in the visualization of its anatomic position on the fusion image (Fig. 1). All the images were evaluated by 2 experienced head and neck radiologists, and both were blinded to the clinical data and surgical results of the patients. Based on the previously mentioned definitions, the 2 experts agreed on the diagnosis of the presence or absence of a middle ear cholesteatoma. In case of any discrepancy, a diagnosis was arrived at by mutual consensus. The experts evaluated the same type of images in all groups but were blinded to the results of the other images.

F1
FIGURE 1:
A–C, The fusion process that creates the TSE-DWI-CT fusion image. Axial (A) HRCT image showed an irregular soft-tissue density shadow in the mastoid process region and adjacent bone destruction (white arrow). The TSE-DWI (B) image indicated that the left middle ear region had significantly high signals (white arrow), with a clear border. C, Color fused image: the TSE-DWI image and HRCT image were fused by head and neck radiologists using a Philips postprocessing workstation to form a TSE-DWI-CT fusion image for the accurate presentation of anatomic information; the fusion image showed cholesteatoma lesions (white arrow) with a clear border located in the attic and tympanic antrum. Images can be viewed in color online. Figure 1 can be viewed online in color at www.jcat.org.

The middle ear and mastoid space were divided into 4 sites (5 positions) based on the STAM system as follows: difficult access sites (S), the tympanic cavity (T), the attic (A), and the mastoid (M). S included the supratubal recess (S1) and the sinus tympani (S2). The cholesteatoma was wholly divided into 2 regions according to the boundary of the lateral posterior semicircular canal, that is, the middle ear region, which was defined by the lesions in front of the posterior limb of the lateral semicircular canal and the mastoid process region, which was defined by the lesions that exceeded the posterior limb of the lateral semicircular canal. Therefore, the lateral posterior semicircular canal served as a boundary.17 If a cholesteatoma was determined to be present by the 2 experienced head and neck radiologists, the lesions were localized according to the STAM system and the overall 2-region localization described previously.

Statistical Analysis

The data were analyzed with SPSS 17.0 (SPSS, Inc, Chicago, Ill) and MedCalc version 15.2.2 (MedCalc, Ostend, Belgium) statistical software. The measurement data are presented as the mean ± SD, and the measurement parameters were analyzed using independent sample t tests. Pathological diagnosis was used as the criterion standard to diagnose cholesteatomas. Accordingly, the accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the HRCT images, TSE-DWI images, and TSE-DWI-CT fusion images were calculated. The area under the receiver operating characteristic curve (AUC) of these images was compared using the receiver operating characteristic (ROC) analysis. Regarding the localization of the lesions, agreement between the HRCT image, TSE-DWI image, or TSE-DWI-CT fusion image and surgery was statistically analyzed using Cohen κ statistic, with intraoperative localization as the criterion standard. The agreement analysis with the κ value was based on the following standard: κ ≥ 0.75, good agreement; κ < 0.40, poor agreement; and 0.40 ≤ κ < 0.75, moderate agreement. A P value less than 0.05 was considered statistically significant.

RESULTS

General Data

Eighty-six patients (mean age, 45.58 ± 16.79; range, 9–78 years) were ultimately included in the qualitative assessment (Fig. 2). All the 86 patients who were clinically suspected with cholesteatoma received a surgical treatment and pathological examination. The mean time from MRI examination to surgery was 4.03 ± 2.15 days (maximum, 10 days). The enrolled patients were divided into a cholesteatoma group (n = 50) and a noncholesteatoma group (n = 36), according to the postoperative pathology results. The general data of the patients in this study are shown in Table 1.

F2
FIGURE 2:
Flow chart of the patients enrolled.
TABLE 1 - General Data of Patients (N = 86)
Item n % Cholesteatoma Noncholesteatoma P
Sex 0.018
 Male 44 51.2% 31 (36.0%) 13 (15.1%)
 Female 42 48.8% 19 (22.1%) 23 (26.7%)
Age 44.20 ± 17.67 47.50 ± 15.79 0.374
Side 0.121
 Left 37 43.0% 18 (20.9%) 19 (22.1%)
 Right 49 57.0% 32 (37.2%) 17 (19.8%)
Clinical symptoms
 Perforation 60 69.8% 35 (40.7%) 25 (29.1%) 0.956
 Otopyorrhea 75 87.2% 43 (50.0%) 32 (37.2%) 0.692

Qualitative Assessment

In terms of diagnosing a cholesteatoma, the AUC of the TSE-DWI-CT fusion images was identical to that of the TSE-DWI images (AUCTSE-DWI-CT = 0.924 vs AUCTSE-DWI = 0.924), but it was significantly higher than that of the HRCT images (AUCTSE-DWI-CT = 0.924 vs AUCHRCT = 0.767; P = 0.0005; Table 2). The sensitivity and specificity of the TSE-DWI-CT fusion images were 96.0% and 88.9%, respectively (Fig. 3). In the TSE-DWI-CT fusion images, there were 4 false-positive cases with a pathological diagnosis of mastoid abscess (n = 2) and cholesterol crystal (n = 2) and 2 false-negative cases with a pathological diagnosis of massive inflammatory cell infiltration and granulation obscuring small cholesteatoma foci.

TABLE 2 - Comparison of HRCT, TSE-DWI, and TSE-DWI-CT Fusion With Pathology in the Diagnosis of Cholesteatoma
Accuracy Sensitivity Specificity PPV NPV AUC
HRCT 77.9% 84.0% 69.4% 97.7% 75.8% 0.767
TSE-DWI 93.0% 96.0% 88.9% 92.3% 94.1% 0.924*
TSE-DWI-CT 93.0% 96.0% 88.9% 92.3% 94.1% 0.924*
*Indicates AUC (TSE-DWI-CT vs HRCT) and AUC (TSE-DWI vs HRCT), P < 0.05.
NPV, negative predictive value; PPV, positive predictive value.

F3
FIGURE 3:
A–D, Magnetic resonance images with pathologically diagnosed cholesteatoma. The axial T1WI (A) image showed low signals as shown by the white arrow. In the T2WI (B), the image showed high signals with an unclear border (as shown by the white arrow). The axial TSE-DWI image (b = 1000 s/mm2) showed limited diffusion and apparently high signals (as shown by the white arrow) with a clear border. D, Color fused image: TSE-DWI-CT fusion image showed focus of hyperintense images (as shown by the white arrow) that were located in the mesotympanum. Images can be viewed in color online. Figure 3 can be viewed online in color at www.jcat.org.

Localization Assessment

Among the patients who were pathologically diagnosed with a cholesteatoma (n = 50), 2 were excluded because they were not found to have positive lesions on the TSE-DWI images or TSE-DWI-CT fusion images. The remaining 48 patients clearly demonstrated positive lesions on the HRCT images, TSE-DWI images, and TSE-DWI-CT fusion images.

According to the STAM system, there was good agreement between the TSE-DWI-CT fusion images and intraoperative localization in the A, T, and M regions (κ = 0.934, 0.789, and 0.808 > 0.75, respectively). Furthermore, the diagnostic accuracy of the TSE-DWI-CT fusion images for these regions was higher than 90%. For the S area, the diagnostic accuracy of the TSE-DWI-CT positioning was more than 80%. Compared with the HRCT and TSE-DWI images, the TSE-DWI-CT fusion images demonstrated higher accuracy in all the regions of the STAM system for localization of cholesteatoma lesions (Table 3).

TABLE 3 - Comparison of HRCT, TSE-DWI, and TSE-DWI-CT Fusion Imaging and Surgery Based on the STAM System
HRCT TSE-DWI TSE-DWI-CT
κ Accuracy P κ Accuracy P κ Accuracy P
A 0.592 83.3% <0.001 0.453 83.3% 0.002 0.934 97.9% <0.001
T 0.730 89.6% <0.001 0.576 81.3% <0.001 0.789 91.7% <0.001
M 0.488 66.7% 0.001 0.684 85.4% <0.001 0.808 91.7% <0.001
S1 0.628 81.3% <0.001 0.269 64.6% 0.006 0.661 83.3% <0.001
S2 0.559 79.2% <0.001 0.226 66.7% 0.013 0.689 85.4% <0.001
κ ≥ 0.75 indicates good agreement; κ < 0.40 indicates poor agreement; 0.40 < κ ≤0.75 indicates moderate agreement.

There was good agreement in the overall 2-region localization of cholesteatoma lesions between the TSE-DWI-CT fusion images and surgery (κ = 0.808; Fig. 4). Compared with the HRCT images and TSE-DWI images, the TSE-DWI-CT fusion images demonstrated higher agreement and accuracy with surgery (Table 4).

F4
FIGURE 4:
A–C, Surgically confirmed cholesteatoma lesions located in the attic. The HRCT (A) image showed an irregular soft-tissue density shadow in the mastoid process region and adjacent bone destruction (white arrow). The TSE-DWI (B) image indicated that the right middle ear region had significantly high signals (white arrow), with a clear border. C, Color fused image: the TSE-DWI-CT fusion image for the accurate presentation of anatomic information; the fusion image showed the focus of cholesteatoma lesions (white arrow) that were located in the attic. Images can be viewed in color online. Figure 4 can be viewed online in color at www.jcat.org.
TABLE 4 - Comparison of Agreement in the 2-Region Localization of Cholesteatoma Lesions Between HRCT Imaging, TSE-DWI Imaging, or TSE-DWI-CT Fusion Imaging and Surgery
Agreement Disagreement Accuracy κ P
HRCT 39 9 81.3% 0.488 0.001
TSE-DWI 41 7 85.4% 0.684 <0.001
TSE-DWI-CT 44 4 91.7% 0.808 <0.001
Limited to middle ear region and invading mastoid process region. κ ≥ 0.75, good agreement; κ < 0.40, poor agreement; 0.40 < κ ≤ 0.75, moderate agreement.

DISCUSSION

This study has the following advantages: the presence of a cholesteatoma was confirmed in all patients alongside their pathological results, and this was the largest sample of DWI-CT fusion images and intraoperative controls to date. The cholesteatoma lesions with significantly high signals in the TSE-DWI images overlapped in the corresponding HRCT images, and the anatomy of the middle ear was clearly displayed in the fusion images.

With intraoperative positioning as the criterion standard, the TSE-DWI-CT fusion images had a high accuracy in the STAM system. Furthermore, it had high diagnostic accuracy in the overall 2-region localization of cholesteatoma lesions, which are of guiding value for the selection of operative plans. Turbo spin-echo DWI uses radio frequency refocusing pulses and is less sensitive to susceptibility artifacts,18 which permits rapid multiplanar imaging in artifact-prone regions, such as the posterior fossa and inferior frontal and temporal lobes.19 Therefore, for the diagnosis of a cholesteatoma in the middle ear, non-EPI sequences are thought to be superior to standard EPI sequences.20 Previous research has shown that non–EPI-DWI images have high diagnostic accuracy for intraoperative 2-region localization, which may help in selecting clinical operative procedures.3 In our study, the accuracy of TSE-DWI images for 2-region positioning was 85.4%, which was slightly lower than the finding (98.0%) previously reported by Migirov et al.3 Despite this, this value relatively limits the consideration of clinical operations. In our study, the HRCT images demonstrated moderate agreement with surgery 2-region localization. This may be because cholesteatoma lesions cannot be precisely distinguished from inflammatory tissues in HRCT images when both cholesteatoma lesions and infections are present4,18; hence, the extent of the cholesteatoma spread cannot be assessed accurately. Therefore, HRCT images cannot accurately guide the selection of clinical operative plans.

The STAM system is used to simplify description of a cholesteatoma's location.16 Based on the STAM system, the TSE-DWI-CT fusion images in this study had a high diagnostic accuracy rate with A, T, and M region localization, which is consistent with the conclusions obtained by previous studies.13,14 This study concluded that the accuracy of the TSE-DWI-CT fusion images for the S1 area was 83.3%, which is nearly identical to that reported by Locketz et al.13 In the current study, the diagnosis of the S2 area (85.4%) was lower than that reported by Felici et al,14 who obtained a rate of 94% in the diagnosis of the S2 area. Such a difference may be related to how the patients were selected; while Felici et al.14 excluded patients who had middle ear infections, patients with inflammation of the middle ear in this study were included. This may have led to a reduction in the diagnostic accuracy of this study. Based on the middle ear division in the STAM system, the TSE-DWI-CT fusion images have a higher diagnostic accuracy rate for S region localization, thereby providing a greater potential to assist clinical surgery decisions.

In our study, the TSE-DWI-CT fusion images had high accuracy, sensitivity, and specificity for diagnosing cholesteatomas, which was consistent with the results of previous literature reports.13,15 The study by Locketz et al13 showed that DWI-HRCT fusion images increased the diagnostic sensitivity of DWI for cholesteatomas (0.88 vs 0.75); however, our study results showed that the TSE-DWI-CT fusion images had the same sensitivity and specificity as the TSE-DWI images. This is because the TSE-DWI-CT fusion images were generated by the overlapping TSE-DWI images. Moreover, the TSE-DWI images in this study had high diagnostic sensitivity and specificity for cholesteatomas, which is similar to the results of previous studies21,22; however, the sensitivity was slightly higher in the present study.21,22 This may be due to the use of a 3.0 T MRI scanner for image acquisition and a higher signal-noise ratio. Despite the high sensitivity and specificity for diagnosing cholesteatomas according to the intraoperative results, the TSE-DWI-CT fusion images failed to reduce the number of false-positive and false-negative cases. There were 2 false-negative cases, with a pathological diagnosis of massive inflammatory cell infiltration and granulation. This may be because massive inflammatory tissues enclosed few cholesteatoma lesions, and on TSE-DWI imaging, this was presented as massive inflammatory low-signal lesions enclosing small cholesteatoma high-signal lesions. This phenomenon weakened the overall lesion signals on the TSE-DWI-CT fusion images. However, when the false-negative case was a retraction pocket without keratin, delayed gadolinium-enhanced T1-weighted images may have helped in differentiating the peripheral enhancement of inflammatory stroma and central nonenhanced cholesteatoma area.23 There were 4 false-positive cases, with a pathological diagnosis of a mastoid abscess (n = 2) and cholesterol crystal (n = 2). The mastoid abscesses were highly limited in terms of diffusion and were displayed as significantly high-signal shadows on DWI imaging. This presented as lesions with significantly high signals on the TSE-DWI-CT fusion images. Current studies have suggested that the apparent diffusion coefficient (ADC) value can be used to distinguish between cholesteatomas and abscesses.24,25 Furthermore, the ADC value of mastoid abscesses is lower than that of cholesteatomas, and false-positive cases can be excluded. Furthermore, there were 2 false-positive cases, with a pathological diagnosis of a cholesterol crystal. Cholesterol crystals show high signals on T1-weighted imaging (T1WI), while cholesteatoma lesions present as low or moderate signals; this can be used to distinguish cholesterol granulomas.26 Furthermore, this suggests that combining the T1WI sequence and ADC value can reduce the false-positive rate.

This study has some limitations. First, manual trimming was required at the last link in the TSE-DWI-CT image fusion process; therefore, positioning deviations were unavoidable. Second, it is necessary to further add T1WI sequences to assist with diagnosis in future studies as use of the T1WI sequence is beneficial for excluding cholesterol granulomas.26 Finally, only patients with a primary cholesteatoma were included in this study, and patients with postoperative recurrence were not included. In the future, fusion images can be added to the study of secondary surgical positioning of patients with suspected postoperative cholesteatoma recurrences.

Taken together, the results show that the preoperative completion of TSE-DWI-CT fusion imaging has a high diagnostic value in the qualification and localization of cholesteatomas. Turbo spin-echo–DWI–CT fusion imaging is helpful for clinicians in the performance of preoperative evaluations, identification of the best operation plans, selection of the surgical approach, and prevention of a potential change in the surgical procedure during the operation.

REFERENCES

1. Xie S, Xiang Y, Wang X, et al. Acquired cholesteatoma epithelial hyperproliferation: roles of cell proliferation signal pathways. Laryngoscope. 2016;126:1923–1930.
2. Anschuetz L, Presutti L, Marchioni D, et al. Discovering middle ear anatomy by transcanal endoscopic ear surgery: a dissection manual. J Vis Exp. 2018;131:56390.
3. Migirov L, Wolf M, Greenberg G, et al. Non-EPI DW MRI in planning the surgical approach to primary and recurrent cholesteatoma. Otol Neurotol. 2014;35:121–125.
4. Migirov L, Greenberg G, Eyal A, et al. Imaging prior to endoscopic ear surgery: clinical note. Isr Med Assoc J. 2014;16:191–193.
5. Aljehani M, Alhussini R. The correlation between preoperative findings of high-resolution computed tomography (HRCT) and intraoperative findings of chronic otitis media (COM). Clin Med Insights Ear Nose Throat. 2019;12:1179550619870471.
6. Rogha M, Hashemi SM, Mokhtarinejad F, et al. Comparison of preoperative temporal bone CT with intraoperative findings in patients with cholesteatoma. Iran J Otorhinolaryngol. 2014;26:7–12.
7. Ng JH, Zhang EZ, Soon SR, et al. Pre-operative high resolution computed tomography scans for cholesteatoma: has anything changed. Am J Otolaryngol. 2014;35:508–513.
8. Songu M, Altay C, Onal K, et al. Correlation of computed tomography, echo-planar diffusion-weighted magnetic resonance imaging and surgical outcomes in middle ear cholesteatoma. Acta Otolaryngol. 2015;135:776–780.
9. Más-Estellés F, Mateos-Fernández M, Carrascosa-Bisquert B, et al. Contemporary non–echo-planar diffusion-weighted imaging of middle ear cholesteatomas. Radiographics. 2012;32:1197–1213.
10. Lingam RK, Bassett P. A Meta-analysis on the diagnostic performance of non-echoplanar diffusion-weighted imaging in detecting middle ear cholesteatoma: 10 years on. Otol Neurotol. 2017;38:521–528.
11. Van Egmond SL, Stegeman I, Grolman W, et al. A systematic review of non–echo planar diffusion-weighted magnetic resonance imaging for detection of primary and postoperative cholesteatoma. Otolaryngol Head Neck Surg. 2016;154:233–240.
12. Wiesmueller M, Wuest W, May MS, et al. Comparison of readout-segmented echo-planar imaging and single-shot TSE DWI for cholesteatoma diagnostics. Am J Neuroradiol. 2021;42:1305–1312.
13. Locketz GD, Li PM, Fischbein NJ, et al. Fusion of computed tomography and PROPELLER diffusion-weighted magnetic resonance imaging for the detection and localization of middle ear cholesteatoma. JAMA Otolaryngol Head Neck Surg. 2016;142:947–953.
14. Felici F, Scemama U, Bendahan D, et al. Improved assessment of middle ear recurrent cholesteatomas using a fusion of conventional CT and non-EPI-DWI MRI. AJNR Am J Neuroradiol. 2019;40:1546–1551.
15. Yamashita K, Hiwatashi A, Togao O, et al. High-resolution three-dimensional diffusion-weighted MRI/CT image data fusion for cholesteatoma surgical planning: a feasibility study. Eur Arch Otorhinolaryngol. 2015;272:3821–3824.
16. Yung M, Tono T, Olszewska E, et al. EAONO/JOS joint consensus statements on the definitions, classification and staging of middle ear cholesteatoma. J Int Adv Otolaryngol. 2017;13:1–8.
17. Khemani S, Lingam RK, Kalan A, et al. The value of non–echo planar HASTE diffusion-weighted MR imaging in the detection, localisation and prediction of extent of postoperative cholesteatoma. Clin Otolaryngol. 2011;36:306–312.
18. Dundar Y, Akcan FA, Dilli A, et al. Does diffusion-weighted MR imaging change the follow-up strategy in cases with residual cholesteatoma. J Int Adv Otolaryngol. 2015;11:58–62.
19. De Foer B, Vercruysse JP, Pilet B, et al. 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. 2006;27:1480–1482.
20. Mikayama R, Yabuuchi H, Sonoda S, et al. Comparison of intravoxel incoherent motion diffusion-weighted imaging between turbo spin-echo and echo-planar imaging of the head and neck. Eur Radiol. 2018;28:316–324.
21. Laske RD, Roth TN, Barath K, et al. The role of non–echoplanar diffusion-weighted magnetic resonance imaging in diagnosis of primary cholesteatoma and cholesteatoma recidivism as an adjunct to clinical evaluation. Ann Otol Rhinol Laryngol. 2018;127:919–925.
22. Dubrulle F, Souillard R, Chechin D, et al. Diffusion-weighted MR imaging sequence in the detection of postoperative recurrent cholesteatoma. Radiology. 2006;238:604–610.
23. De Foer B, Vercruysse JP, Bernaerts A, et al. Middle ear cholesteatoma: non–echo-planar diffusion-weighted MR imaging versus delayed gadolinium-enhanced T1-weighted MR imaging—value in detection. Radiology. 2010;255:866–872.
24. Karandikar A, Loke SC, Goh J, et al. Evaluation of cholesteatoma: our experience with DW Propeller imaging. Acta Radiol. 2015;56:1108–1112.
25. Fan X, Liu Z, Ding C, et al. The value of turbo spin-echo diffusion-weighted imaging apparent diffusion coefficient in the diagnosis of temporal bone cholesteatoma. Clin Radiol. 2019;74:977.e1–e7.
26. Fukuda A, Morita S, Harada T, et al. Value of T1-weighted magnetic resonance imaging in cholesteatoma detection. Otol Neurotol. 2017;38:1440–1444.
Keywords:

cholesteatoma; diffusion-weighted imaging; fusion image; localization; turbo spin echo

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.