Update on Imaging of Hearing Loss

Introduction

In the imaging evaluation of hearing loss, the radiologic examination is complementary to the physical examination. The clinical evaluation includes not only an accurate history of the hearing loss itself (unilateral or bilateral, slow onset or fast onset, etc.) but also the patient’s medical history (possible trauma, ototoxic drugs, associated facial nerve palsies, etc.). Hearing loss can be described as sensorineural, conductive, or mixed, and this will guide the choice of imaging modality to correlate with the clinical history and symptoms. In this chapter, we will review the more common hearing loss entities, and divide these pathologies into those presenting with sensorineural hearing loss (SNHL), conductive hearing loss (CHL), and mixed hearing loss (MHL). The pathologies for SNHL and CHL will be further grouped and discussed as either congenital or acquired entities. In very simplistic terms, CHL is frequently initially evaluated with CT, whereas SNHL is most frequently initially evaluated with MRI. When reviewing the imaging studies, a systematic “outside-in” approach along the hearing pathway is helpful, from the external anatomic structures to the internal structures.

Imaging Modalities

Thin-section CT and MRI provide complementary information in evaluating skull base and temporal bone (TB) pathologies. On evaluation of a patient with hearing loss, CT of the TB is particularly helpful to evaluate the osseous structures (external auditory canal [EAC], middle ear, and otic capsule), whereas MRI is useful to evaluate the inner ear structures, internal auditory canal (IAC), cerebellopontine angle (CPA), and brainstem.

CT of the TBs can be performed with the patient lying supine, in the anatomic axial plane. Multidetector CT scanners acquire submillimeter slices with low radiation dose and reduced scan time. It is important for the imager to be wary of low-dose techniques, as these will limit the accurate evaluation of the normally dense TB osseous structures. Submillimeter multiplanar reconstructions can be performed with minimal overlap. Although parameters vary by institution, suggested TB CT parameters are described in Box 8.1 . This technique is capable of revealing a broad spectrum of EAC and middle ear lesions that may not be apparent on the basis of clinical findings alone. Radiation dose in TB CT imaging can be high because of the requirement of high spatial resolution. Leng et al. demonstrated an ultrahigh-resolution scan mode with an iterative reconstruction algorithm to improve the quality of the TB CT image with decreased radiation dose. Axial acquisitions with multiple plane reconstructions can effectively evaluate the osseous TB anatomy and pathology ( Fig. 8.1 ).

MRI is a fundamental method of evaluating the inner ear structures as well as the cerebrospinal fluid (CSF) spaces, IAC, CPA, and adjacent brain parenchyma. Dedicated head and temporomandibular joint coils are important for achieving an optimal signal to noise ratio. Screening MRI (noncontrasted thin-section CSF bright sequences) of the CPA/IAC can be performed in the initial evaluation of SNHL in the appropriately screened patient (i.e., older patient with progressive asymmetric hearing loss). When there is a suspected or known retrocochlear lesion, MRI of the CPA/IAC without and with contrast can be performed. In addition, in cases of complicated SNHL, or when there are other symptoms such as cranial neuropathy, long tract signs, and/or headache, imaging should include whole brain and posterior fossa sequences (axial T2 and/or fluid-attenuated inversion recovery [FLAIR]). Although parameters vary by institution, suggested screening MRI IAC parameters are described in Box 8.2 . Imaging is critical in the assessment of hearing loss, and when a pathology is identified, the most commonly encountered tumor is vestibular schwannoma (VS). One diagnostic review and meta-analysis concluded that nonimaging screening protocols, including pure-tone audiometry, auditory brainstem response, clinical symptoms, electronystagmography, caloric irrigation, and hyperventilation tests, were not accurate in detecting VSs.

MRI is widely used for the evaluation of asymmetric SNHL. There are, however, drawbacks of risks, cost, and time associated with MRI. Non-contrast-enhanced MRI studies have been shown to be adequate in the detection of CPA/IAC masses using thin CSF bright sequences. A prospective, blinded study using fast spin echo (FSE) T2-weighted (T2W) MRI demonstrated a threefold decreased cost while maintaining 98% sensitivity for schwannomas. A screening high-resolution FSE T2W MRI was also shown to be adequate to detect other pathologies, including other CPA/IAC lesions, inner ear lesions, and intraaxial lesions (such as infarctions, multiple sclerosis, mesial temporal sclerosis, and colloid cysts). Daniels et al. concluded that this high-resolution FSE screening technique, used in conjunction with appropriate clinical prescreening and referral, can provide an equally sensitive method of evaluating unilateral SNHL compared with contrast-enhanced T1-weighted (T1W) MRI while reducing costs and providing distinct advantages in evaluating non-VS causes of SNHL. They also showed that false positives on contrast-enhanced T1W MRI (true negatives on FSE T2W MRI) can lead to unnecessary diagnostic tests, intervention, and increased costs of care. Similarly, in a study comparing T2W MRI with contrast-enhanced MRI in 146 patients with asymmetric SHNL, Verret et al. demonstrated that only T2W MRI instead of contrast-enhanced MRI would have decreased costs over $100,000. With the addition of multiple planes (i.e., axial and coronal) and advances in MRI techniques (e.g., constructive interference in steady state [CISS], fast imaging employing steady-state acquisition [FIESTA], and sampling perfection with application optimized contrast using different flip angle evolutions [SPACE]), 100% sensitivity and excellent interrater agreement can be achieved with high-resolution MRI protocols.

Contrast-enhanced MRI has been shown by other studies to provide useful information for nonneoplastic lesions. Annesley-Williams et al. found that 2- to 5-mm lesions may be missed on two-dimensional (2D) FSE T2W MRI and that contrast-enhanced MRI may be needed to further investigate findings on three-dimensional (3D) and 2D MRI (9% and 15%, respectively). However, in considering patient care related to nonneoplastic CPA/IAC pathologies, such as inflammatory pathologies that are typically acute in onset and of viral etiology, these cases are treated regardless of the MRI findings.

Imaging pitfalls should also be considered when evaluating MRI. For example, the petrous apex marrow adjacent to the IAC can enhance on postcontrast T1W MRI (especially bright on cases without fat saturation) and be erroneously interpreted as a CPA/IAC tumor. Similarly, asymmetric aeration of the petrous apices with unilateral benign trapped fluid can also be mistaken for pathology. The enhancement of a prominent anterior inferior cerebellar artery (AICA) loop may be mistaken for a small VS. Other pseudomasses include the choroid plexus protruding through the lateral recesses of the fourth ventricle and the cerebellar flocculus normally projecting into the posterolateral aspect of the CPA cistern. The vestibular nerve ganglion (Scarpa ganglion) in the IAC should also not be mistaken as a small VS. If there is a punctate (<2 mm) enhancing lesion in the CPA/IAC, it can safely be followed over time. Pathologies arising from the jugular foramen may also extend superiorly into the CPA/IAC region.

Sensorineural Hearing Loss

SNHL indicates dysfunction of the cochlea, cochlear nerve, and/or brain and can be measured with a tuning fork touched to the top of the head, by conduction through bone, so that the sound will localize to the normal side in these patients. The structures of note include the cochlea and the cochlear nerve from the origin nucleus through the CPA/IAC cistern into the modiolus. In addition to CT of the TBs, thin-section MRI with T1W pre- and postcontrast with fat-saturation sequences, T2W sequences, and 3D CISS/3D T2 FSE/3D fourier transformation CISS/SPACE sequences are essential for the thorough evaluation of SNHL.