RPS 808 - Temporal bone imaging

Author Block: S. Palizzi, A. Romano, F. Dellepiane, A. Bozzao; Rome/IT
Purpose or Learning Objective: Cholesteatoma is a serious pathological condition that requires surgical excision as definitive treatment, with the least invasive technique. CT is the first level examination for the anatomical evaluation of the middle ear; however, it does not allow the differentiation of cholesteatoma margins from surrounding inflammatory tissue. On the contrary, RM-DWI has high specificity, with a reduced spatial resolution. The purpose of the study is to evaluate whether CT-MRI fusion images can increase the accuracy in the diagnosis and localization of cholesteatoma, with the aim of improving pre-operative planning
Methods or Background: 83 patients with cholesteatoma were included in our study in the time frame between October 2019 and May 2021; 20 underwent CT-MRI examination for pre-surgical planning. MR and CT images were co-registered using a dedicated console (BrainLab, Feldkirchen, Germany). The images with hypersignal in DWI-SE compatible with cholesteatoma were used to construct a three-dimensional ""cholesteatoma"", to be displayed on multiplanar and 3D co-registered CT images.
Results or Findings: A statistically significant difference was found between the volume of the cholesteatoma tissue and the surrounding inflammatory tissue (mean cholesteatoma volume: 0.03cm3; mean inflammatory tissue volume: 1.8cm3; p <0.05). In 80% of cases, the surgical approach to the patient was modified according to the information obtained from CT-MRI fusion images. No tissue residue was detected in the group of 20 patients surgically treated using the information from CT-MRI fusion. In the control group (63 patients), residual cholesteatoma was present in 6 cases (9.5%).
Conclusion: CT and MRI images are complementary to each other. In selected cases, they should be co-registered in the pre-operative evaluation of cholesteatoma in order to optimize surgical time and aim at radical surgical resection.
Limitations: No limitations identified.
Ethics committee approval: Not applicable.
Funding for this study: No funding was provided for this study.

Author Block: A. L. Schönenberger¹, T. Flohr², B. Schmidt², S. Winklhofer¹, H. Alkadhi¹; ¹Zurich/CH, ²Forchheim/DE
Purpose or Learning Objective: Assessing the image quality, artefact severity and radiation dose after cochlear implantation of the temporal bone in clinical photon-counting CT (PCCT) depending on the imaging acquisition and postprocessing.
Methods or Background: PCCT was performed on a human cadaveric temporal bone after surgical cochlear implantation using the QuantumPlus mode at three different CTDIvol levels (15.3, 9.7, 5.8 mGy). Image reconstructions were performed using four different reconstruction kernels (Hr72, 84, 96, 98) and additional four different Quantum Iterative Reconstruction levels (QIR level 1-4) resulting in a total of 60 different imaging data sets. Quantitative (ROI for noise and metal artefacts) and qualitative image evaluation (5-point Likert scale for anatomical and electrode delineation). Image noise quantification using the standard deviation of attenuation in the surrounding fluid.
Results or Findings: The 15.3mGy CTDIvol demonstrated a superior anatomical delineation and reduced quantitative artefacts compared to the two lower radiation levels (mean 3.8 / 2.9 / 1.6; 406 / 471 / 506HU; each p>0.05). Image noise showed no significant differences. Image artefacts decreased with higher iterative levels (no QIR = 512HU, QIR level 4 =379HU, p>0.05). Hr72 showed the best anatomical delineation (median 3.8 / 3.2-1.8) and lowest image noise (68 vs. 127-320HU) compared to Hr84, 96, and 98 (each p>0.05). Hr72 and 84 demonstrated the lowest artefact levels compared to Hr96 and 98 (388 and 357HU / 515 and 582HU, each p>0.05). Electrode delineation showed no significant difference with perfect visibility in all radiation levels.
Conclusion: In clinical PCCT of the temporal bone after cochlear implantation, a sufficiently good electrode delineation was seen in all radiation levels. Optimal anatomical delineation and artefact reduction on higher radiation levels and lower kernels (Hr72) and quantitative artefacts reduction by higher iterative levels.
Limitations: Single subject.
Ethics committee approval: Approved.
Funding for this study: Not applicable.

Author Block: I. Sel, Y. Karagöz, E. Ateş, D. Ö. Ö. Aksoy, A. S. Mahmutoglu; Istanbul/TR
Purpose or Learning Objective: Incomplete partition type II (IP II) is one of the most frequently encountered congenital inner ear malformations in cochlear implant candidates. It is characterized by modiolar apical defects and mild vestibular enlargement. Anomalous child will develop sensorineural hearing loss (SNHL) worsening progressively with ageing supposedly with minor traumas. Radiological diagnosis of malformation could be difficult with equivocal changes in tiny labyrinthine structures, especially with inexperienced viewers.
Methods or Background: Our study was a retrospective case-control study including 34 SNHL patients with IP type II (66 ears) and 24 patients (48 ears) with normal audiographic findings who underwent temporal CT imaging. A normal group was examined tomographically for inflammatory conditions or trauma not involving labyrinthine structures. Measurements of defined inner ear structures were performed by two observers separately; observer 1 is a senior neuroradiologist and observer 2 is a novice radiologist with 1 year of experience in neuroradiology. The parameters measured in patient and control groups were compared and interobserver reliability was calculated.
Results or Findings: Lateral interscalar notch angle, lateral scalar height, medial interscalar notch depth, medial scalar height, vestibular aqueduct (VA) width at the middle, VA width at operculum level, vestibular height and width were found to be significantly larger while lateral interscalar notch depth and LSSC bone island surface area were significantly smaller in the patient group (p<0.05). These findings especially the lateral cochlear measurements are valuably proving structural changes secondary to apical modiolar deficiency.
Conclusion: CT measurements in SNHL could help to diagnose IP type II anomaly especially in patients with equivocal audiological and radiological findings.
Limitations: No limitations identified.
Ethics committee approval: Yes.
Funding for this study: There was no need for any funding.

Author Block: M. B. Eser, B. Atalay, M. T. Kalcioglu; Istanbul/TR
Purpose or Learning Objective: This study aimed to investigate whether there is a relationship between cochlear nerve myelination and hearing loss in children with congenital non-syndromic sensorineural profound hearing loss.
Methods or Background: Patients who underwent cochlear implant surgery in our university hospital were retrospectively evaluated, and 33 patients (19 boys, 14 girls) with congenital non-syndromic sensorineural profound hearing loss (>100db) were included. Current knowledge has shown that the myelinated cochlear nerve can be measured with MRI T2-weighted sequences. So that, the experienced observer measured cochlear nerve circumference (CNC) and surface area (CNSA) from FIESTA-C images taken in the sagittal oblique plane of these patients. Pearson correlation was done to understand the relationship between age and CNC and CNSA, and with t-test was done to compare means of 24 months under-over children of CNC and CNSA.
Results or Findings: The mean age of the individuals is 22.52 (SD: 12.37, Range: 5 - 55) months. CNC in male patients is 3.88 (SD: 0.10) mm and in female patients 3.71 (SD: 0.09) mm, and there was no difference. The mean CNC was 3.81 (SD: 0.41) mm. The mean CNSA we found was 1.08 (SD: 0.23) mm2. A moderate positive correlation was found between age and CNC (r=0.38, p=0.03), CNSA (r=0.39, p=0.03). Although not significant on the t-test (p=0.06), children under 24 months (1.02, SD: 0.15 mm2) had a smaller CNSA than those over 24 months (1.17, SD: 0.32 mm2).
Conclusion: This study provides weak evidence that myelination continues in its normal course in children with a congenital non-syndromic sensorineural profound hearing loss despite the absence of a hearing signal.
Limitations: Retrospective design.
Ethics committee approval: Exist (DN:2018/0177).
Funding for this study: Not applicable.