MRI of the inner ear: comparison of axial T2-weighted, three-dimensional turbo spin-echo images, maximum-intensity projections, and volume rendering

The superiority of 3D-CISS sequence in displaying the cisternal segment of facial, vestibulocochlear nerves and their abnormal changes

OBJECTIVE

To select the best imaging method for clinical otologic patients through evaluating 3D constructive interference of steady state (CISS) image quality in visualizing the facial, vestibulocochlear nerves (CN:VII-VIII) and their abnormal changes.

METHODS

The CN:VII-VIII as well as inner ear structures in 48 volunteers were examined using 3D-CISS and 3D turbo spin echo (TSE) sequences respectively, and displayed to the full at the reformatted and maximum intensity projection (MIP) images. The nerve identification and image quality were graded for the CN:VII-VIII as well as inner ear structures. Statistical analysis was performed using the Wilcoxin test, p<0.05 was considered significant. In addition, 8 patients with abnormality in facial or vestibulocochlear nerves were also examined using 3D-CISS sequence.

RESULTS

The identification rates for the cisternal segment of facial, vestibulocochlear nerves and corresponding membranous labyrinth were 100%. Abnormal changes of the facial or vestibulocochlear nerves were clearly shown in 8 patients, among them 1 was caused by bilateral acoustic neurinoma, 1 by cholesteatoma at cerebellopontine angle, 1 by arachnoid cyst, 1 by neurovascular adhesion, 4 by neurovascular compression.

CONCLUSION

With 3D-CISS sequence the fine structure of the CN:VII-VIII and corresponding membranous labyrinth can be clearly demonstrated; lesions at the site of cerebellopontine angle can also be found easily.

Anatomic study of maximum intensity projection of the membranous labyrinth and the internal auditory meatus - MRI scan in 16 Chinese adults

CONCLUSION

Three-dimensional reconstruction of maximum intensity projection (MIP) might document objectively, stereoscopically and directly the minute structures of the membranous labyrinth and internal auditory meatus. In this study, we establish magnetic resonance imaging (MRI) measurement criteria of the inner ear in Chinese adults.

OBJECTIVE

The goal of this study was to provide an anatomic basis for otolosurgery and neurosurgery in Chinese adults.

MATERIALS AND METHODS

Sixteen healthy volunteer subjects were scanned by a GE-signa 1.5T MRI scanner. All original images were transferred to an MRI workstation and all the structures of the inner ear were reconstructed, rotated at various angles and measured with an MIP program.

RESULTS

Anatomic structures of the membranous labyrinth and internal auditory meatus were well demonstrated in MIP images in all volunteers. All inner ear structures including utricle, saccule, cochlear duct, internal auditory meatus and three semicircular ducts produced high intensity signals.

T2-weighted turbo spin-echo images, maximum-intensity projections, and three-dimensional volume-rendering for delineation of pathologies and anatomic details of the inner ear

PURPOSE

The aim of the current study was to compare 3-dimensional T2-weighted turbo spin-echo (TSE) axial slices, maximum-intensity projections (MIPs), and volume-rendered images with respect to the delineation of pathologic findings and anatomic structures of the inner ear. Each of the compared image types was available at 2 different resolutions.

MATERIALS AND METHODS

In 64 patients, 3-dimensional T2-TSE images were acquired with a matrix of 128 x 100 and 230 x 230 (field-of-view 90 x 90 mm). Direct axial images, MIPs, and volume-rendered images from the 2 datasets were evaluated in terms of the delineation of pathologic findings and anatomic structures. Analysis was performed by 2 radiologists in consensus.

RESULTS

Though axial images acquired with a matrix of 230 x 230 showed more anatomic details than images acquired with a matrix of 128 x 100, pathologic findings (33 in total) were assessable with the same confidence on images of either resolution. Pathologies completely surrounded by fluid were best assessed on axial slices. The delineation of pathologies not embedded in a fluid was almost equally good in axial slices, MIPs, or 3-dimensional volume-rendered images.

CONCLUSION

Reading the axial images cannot be replaced by viewing postprocessed reconstructions for initial diagnosis, because only some of the pathologies can be delineated on MIPs and volume reconstructions. However, reconstructions can be valuable for assessing the extent of pathologies and their spatial relation to the anatomic structures.

Clinical usefulness of 3-dimensional computed tomography laryngography in laryngeal and hypopharyngeal cancer

This study was designed to evaluate the clinical usefulness of 3-dimensional computed tomography (CT) laryngography for the determination of subglottic extension in laryngeal cancer and for the involvement of the pyriform sinus apex in hypopharyngeal cancer. Twenty-one patients with laryngeal cancer and 19 patients with hypopharyngeal cancer underwent spiral CT during quiet breathing, /E/ phonation, Valsalva's maneuver, and modified Valsalva's maneuver. Three-dimensional volume rendering with CT data was performed on a workstation. Three-dimensional CT laryngography was compared to laryngoscopy, axial CT, endoscopy, and/or histopathologic findings. Three-dimensional CT laryngography was useful for showing subglottic extension in laryngeal cancer and involvement of pyriform sinus apex in hypopharyngeal cancer. Three-dimensional CT laryngography with quiet breathing was most informative for the determination of subglottic extension, whereas 3-dimensional CT laryngography with modified Valsalva's maneuver was most informative for the diagnosis of pyriform sinus apex involvement. In patients with vocal cord palsy, the contour of the vocal cord was the same during all 4 maneuvers. Three-dimensional CT laryngography during quiet breathing, /E/ phonation, Valsalva's maneuver, and modified Valsalva's maneuver may be useful in assessing laryngeal and hypopharyngeal cancers. This technique may provide important information not provided by conventional CT images.

Driven equilibrium (drive) MR imaging of the cranial nerves V–VIII: comparison with the T2-weighted 3D TSE sequence

PURPOSE

The aim of this study is to evaluate the efficacy of the driven equilibrium radio frequency reset pulse (DRIVE) on image quality and nerve detection when used in adjunction with T2-weighted 3D turbo spin-echo (TSE) sequence.

MATERIALS AND METHODS

Forty-five patients with cranial nerve symptoms referable to the cerebellopontine angle (CPA) were examined using a T2-weighted 3D TSE pulse sequence with and without DRIVE. MR imaging was performed on a 1.5-T MRI scanner. In addition to the axial resource images, reformatted oblique sagittal, oblique coronal and maximum intensity projection (MIP) images of the inner ear were evaluated. The nerve identification and image quality were graded for the cranial nerves V-VIII as well as inner ear structures. These structures were chosen because fluid-solid interfaces existed due to the CSF around (the cranial nerves V-VIII) or the endolymph within (the inner ear structures). Statistical analysis was performed using the Wilcoxon test. P < 0.05 was considered significant.

RESULTS

The addition of the DRIVE pulse shortens the scan time by 25%. T2-weighted 3D TSE sequence with DRIVE performed slightly better than the T2-weighted 3D TSE sequence without DRIVE in identifying the individual nerves. The image quality was also slightly better with DRIVE.

CONCLUSION

The addition of the DRIVE pulse to the T2-weighted 3D TSE sequence is preferable when imaging the cranial nerves surrounded by the CSF, or fluid-filled structures because of shorter scan time and better image quality due to reduced flow artifacts.

Common crus aplasia: diagnosis by 3D volume rendering imaging using 3DFT-CISS sequence

AIM

The purpose of this study was to evaluate the findings of three-dimensional (3D) volume rendering (VR) imaging in common crus aplasia (CCA) of the inner ear.

MATERIALS AND METHODS

Using 3D VR imaging of temporal bone constructive interference in steady state (CISS) magnetic resonance (MR) images, we retrospectively reviewed seven inner ears of six children who were candidates for cochlear implants and who had been diagnosed with CCA. As controls, we used the same method to examine 402 inner ears of 201 patients who had no clinical symptoms or signs of sensorineural hearing loss. Temporal bone MR imaging (MRI) was performed with a 1.5 T MR machine using a CISS sequence, and VR of the inner ear was performed on a work station. Morphological image analysis was performed on rotation views of 3D VR images.

RESULTS

In all seven cases, CCA was diagnosed by the absence of the common crus. The remaining superior semicircular canal (SCC) was normal in five and hypoplastic in two inner ears, while the posterior SCC was normal in all seven. One patient showed bilateral symmetrical CCA. Complicated combined anomalies were seen in the cochlea, vestibule and lateral SCC.

CONCLUSION

3D VR imaging findings with MR CISS sequence can directly diagnose CCA. This technique may be useful in delineating detailed anomalies of SCCs.

Segmental anatomy of the temporal bone

More of us suffer from hearing loss than from visual impairment, coronary artery disease, and cancer combined. Auditory dysfunction is detected in less than one third of newborns afflicted and is underdiagnosed in the elderly. Despite biomedical and technological advances, our understanding of vestibular dysfunction has grown little. Forming part of the lateral and inferior surfaces of the skull, the temporal bone contains the organs of hearing and balance. Diagnosis of diseases in these important regions requires adequate clinical information and a thorough knowledge of normal temporal bone anatomy to recognize abnormal findings when present. This article dissects the temporal bone into segments, describes the relevant radiologic anatomy of each, and discusses the imaging appropriate to specific concerns.

A post-processing protocol for three-dimensional visualization of the inner ear using the volume-rendering technique based on a standard magnetic resonance imaging protocol

A superior diagnostic quality compared to other post-processing (PP) techniques for three-dimensional (3D) inner ear imaging has been attributed to volume rendering (VR). We defined and assessed a VR protocol for 3D visualization of the inner ear in a routine imaging setting. Following definition of a VR protocol by using lower threshold values, surface shading, perspective views and related parameters, standardized 3D views of the inner ear were generated and evaluated in 32 patients suffering from sensorineural or combined hearing loss. Comprehensive inner ear visualization was achieved in 28 patients by means of two 3D shaded-surface views. Incomplete data acquisition (1/32), motion and pulsation artefacts (2/32) and interposed fluid-retaining mastoid cells (1/32) were responsible for non-diagnostic image quality in 4/32 patients. In 5/32 patients modifications of the PP protocol involving the threshold value and depth-cueing parameters helped to establish diagnostic image quality. Mean post-processing time amounted to 5.8 min per site. 3D imaging with the VR technique is suitable for routine inner ear assessment if direct VR, predefined PP protocols and standardized labyrinthine views are used.