Conductive hearing loss (CHL) occurs when sound transmission through the external or middle ear is impaired, resulting in an air–bone gap on pure-tone audiometry. Unlike sensorineural hearing loss, which reflects cochlear or neural pathology, CHL is typically mechanical in origin. The major aetiologies include otosclerosis (characterised by stapes fixation because of aberrant otic capsule remodelling), chronic otitis media with or without cholesteatoma, ossicular chain discontinuity or congenital malformations and middle ear trauma. Less commonly, neoplastic or vascular lesions, such as paragangliomas, may be implicated.¹ Additionally, inner-ear ‘third-window’ phenomena, particularly superior semicircular canal (SSC) dehiscence, can produce a conductive component by creating an abnormal pathway for sound energy dissipation.²

High-resolution CT (HRCT) and MRI are the cornerstone imaging tools for CHL. CT offers excellent depiction of the ossicular chain and bony labyrinth, whereas MRI provides complementary soft-tissue and vascular detail.³ CT angiography (CTA) has a more specialised role, primarily in delineating vascular architecture and aiding in the diagnosis of vascular middle ear lesions.⁴ This review synthesises current evidence on the diagnostic performance, technical protocols and clinical applications of CT, MRI and CTA in both adult and paediatric CHL. It also proposes tailored protocol recommendations and a comparative modality table to support a structured, evidence-based approach to evaluation and diagnosis.

High-resolution CT of the temporal bone is the first-line imaging modality for unexplained CHL when otoscopic and microscopic examinations are inconclusive.³ Using submillimetre isotropic voxels, HRCT provides micrometric delineation of the external auditory canal (EAC), tympanic cavity, ossicular chain, mastoid air cell system, facial nerve canal, jugular bulb and osseous labyrinth. This precision enables identification of fenestral otosclerosis, such as radiolucent foci at the fissula ante fenestram and round window niche obliteration.⁵ The technique also detects ossicular erosions, congenital fusions or dysplasias and occult traumatic injuries – including non-displaced fractures, incudostapedial joint dislocation or ligamentous avulsion. Tailored Pöschl and Stenvers reconstructions allow visualisation of SSC dehiscence (Figure 1) and other third-window lesions, while defects of the tegmen tympani or antrum, with associated meningocele or cerebrospinal fluid leak, are readily identified. In the study by Belden et al., SSC dehiscence was detected in all 36 patients with the clinical syndrome, including bilateral involvement in six patients; six additional patients without clinical signs appeared dehiscent on 1.0 mm CT, but intact bone was confirmed on 0.5 mm scans.⁶

High-resolution CT further characterises mastoid air cell pneumatisation patterns, which inform surgical planning and influence postoperative ventilation outcomes. Postoperatively, HRCT confirms stapes piston or ossicular prosthesis placement and assesses footplate fenestration following stapedotomy or ossiculoplasty.⁷

In well-aerated middle ear and mastoid spaces with an intact ossicular chain, HRCT demonstrates high negative predictive value for excluding occult cholesteatoma, granulation tissue or hypervascular tumours, thereby guiding clinical decision-making towards surveillance or surgical intervention only when indicated. In the study by Uz Zaman et al., HRCT exhibited a sensitivity of 100%, specificity of 88.1%, positive predictive value of 92.1% and negative predictive value of 100% for detecting cholesteatoma.⁸ Finally, advances in photon-counting and ultra-high-resolution detector technologies promise faster acquisition, reduced radiation dose and enhanced depiction of osseous detail, reinforcing HRCT as the benchmark modality for temporal bone imaging.⁹

MRI is not typically the first-line investigation for pure CHL, as its spatial resolution is insufficient to depict the minute bony landmarks of the ossicular chain or the oval- and round-window complex. It becomes indispensable, however, when a soft-tissue process is suspected, the full extent of disease must be mapped, or a mixed sensorineural component is present. MRI’s superior soft-tissue contrast allows direct visualisation of fluid, inflammation, fibrosis and enhancement patterns that CT cannot demonstrate.²

In CHL, MRI is generally reserved for suspected cholesteatoma (Figure 6 and Figure 7) or middle ear tumours. For cholesteatoma – especially in complex or postoperative cases – dedicated sequences confirm residual disease and differentiate it from fibrosis. Non-echo-planar DWI has revolutionised evaluation by detecting restricted diffusion of keratin debris; cholesteatomas appear hyperintense with low apparent diffusion coefficient (ADC) values, enabling detection of foci as small as 2 mm – 3 mm.²⁴ This non-invasive method often obviates the need for second-look surgery, as negative non-echo-planar DWI findings can reliably exclude residual or recurrent disease and allow patients to be followed radiologically rather than undergoing re-exploration.^25,26

Combined with CT, MRI also reveals inner ear or intracranial complications – such as labyrinthitis, abscess or meningeal enhancement – that are crucial for surgical planning. Contrast-enhanced MRI is preferred for glomus tympanicum and jugulare tumours (Figure 8). These lesions typically appear as lobulated, avidly enhancing masses engulfing ossicles and vessels, exhibiting the characteristic ‘salt-and-pepper’ T1 pattern.¹⁴ Dynamic post-gadolinium imaging shows rapid arterial filling with early washout, differentiating paraganglioma from schwannoma or meningioma. Fat-suppressed post-contrast sequences delineate tumour extension from the hypotympanum into the jugular bulb, superiorly along the sigmoid sinus, medially into the petrous carotid canal and inferiorly into the carotid space. MRI further depicts marrow invasion, venous sinus thrombosis, extradural intracranial extension and perineural spread – findings essential for staging, selecting the surgical corridor and planning preoperative embolisation. Although HRCT remains unrivalled for cortical bone evaluation, MRI more accurately defines tumour margins, vessel encasement and soft-tissue infiltration, thereby guiding multidisciplinary management and reducing operative morbidity.²⁷

MRI is also indicated in mixed hearing loss (conductive and sensorineural components) to evaluate the cochlea, vestibulocochlear nerve and brain for underlying pathology, such as cochlear otosclerosis, cochlear nerve deficiency or retro-cochlear lesions. Active otosclerotic disease extending to the cochlea may sometimes be appreciated on MRI, demonstrating enhancement of the otic capsule or cochlear inflammation, although CT remains the first-line modality for otosclerosis.²⁸

In paediatric CHL, MRI complements CT by revealing inner ear malformations – such as cochlear aplasia or an enlarged vestibular aqueduct – and by confirming cochlear nerve presence and calibre, a critical consideration for cochlear implantation. While most of these malformations primarily cause sensorineural hearing loss, mixed patterns may occur, making MRI an important adjunct. For purely conductive congenital anomalies, MRI mainly serves to rule out syndromic inner ear abnormalities, whereas CT remains superior for evaluating ossicles and middle ear anatomy.²⁹

CT angiography is a specialised form of CT that visualises blood vessels following intravenous contrast administration. In the evaluation of CHL, CTA is not routinely indicated and is generally not part of first-line imaging unless specific findings suggest its necessity. The American College of Radiology guidelines note that there is no evidence supporting routine CTA for uncomplicated CHL. However, CTA becomes highly useful when a vascular anomaly or tumour is suspected as the underlying cause of hearing loss.

A classic example is a middle ear paraganglioma (glomus tympanicum or glomus jugulare). While MRI can demonstrate the presence of the tumour, CTA confirms its hypervascular nature, delineates its arterial supply (for example, feeders from the ascending pharyngeal or occipital arteries) and assesses involvement of adjacent structures such as the jugular bulb or carotid artery. Moreover, in a single acquisition, CTA can demonstrate the tumour blush and map arterial feeders – information indispensable when planning endovascular embolisation or surgical resection.¹⁴

CT angiography is also valuable for identifying rare vascular malformations causing CHL. Congenital anomalies such as an aberrant internal carotid artery or a persistent stapedial artery in the middle ear can present with pulsatile tinnitus and CHL. High-resolution CT may show an unusual tubular structure or dehiscent bony canal, but CTA definitively identifies the vessel and outlines its course. Likewise, a high-riding or dehiscent jugular bulb can encroach upon the middle ear space; CTA reveals venous anatomy and any relevant variants. In many cases, these diagnoses can be sufficiently established on non-contrast CT, which shows absent bone or abnormal canals, and contrast-enhanced imaging is reserved for surgical planning– for example, to distinguish a dehiscent jugular bulb (venous) from a glomus tumour (arterial).³⁷

The causes of CHL differ between paediatric and adult populations, influencing imaging strategies. In children, CHL often results from congenital abnormalities or sequelae of chronic otitis media. Common findings include ossicular malformations (e.g. stapes fixation, absent incus), EAC atresia or syndromic deformities such as Treacher Collins or CHARGE syndrome. These conditions rely primarily on CT for surgical planning, as most anomalies involve bony structures that are well visualised on imaging. Studies show that CT frequently identifies aetiologies such as ossicular dysplasia, fixation or malformation. Congenital cholesteatoma may also occur behind an intact tympanic membrane, typically presenting as a pearly white anterosuperior mass causing unilateral CHL with a normal drum.³⁸ High-resolution CT (≤ 0.5 mm) or cone-beam CT is used to map the precise quadrant of involvement, detect ossicular erosion and assess mastoid extension. Non-echo-planar diffusion-weighted MRI provides essential soft tissue contrast, reliably distinguishing cholesteatoma from effusion or granulation tissue. It is now the modality of choice for both preoperative delineation and detection of residual disease as small as 3 mm – 5 mm. Congenital cholesteatoma can also be staged with the four-tier Potsic classification, which correlates closely with residual disease risk, guiding operative planning and postoperative surveillance.³⁹

In adults, CHL is most often acquired. Otosclerosis typically arises in young to middle-aged adults and is best detected with submillimetre HRCT, which demonstrates high diagnostic accuracy for fenestral or cochlear foci of demineralisation.⁴⁰ Chronic otitis media and cholesteatoma often begin in adolescence and persist into adulthood. Ossicular disruption following trauma peaks in young adults. Highly vascular glomus tympanicum or jugulare paragangliomas usually present in the fifth to sixth decades with pulsatile tinnitus and a retrotympanic mass.⁴¹ ‘Third-window’ defects, such as SSC dehiscence, although congenital, typically become symptomatic later in adulthood.²

Because younger patients carry a higher lifetime risk of radiation–induced malignancy, any unexplained – or especially unilateral – adult conductive deficit not clearly attributable to otitis is routinely investigated first with HRCT to screen for otosclerosis, ossicular discontinuity, third-window lesions or other structural causes. MRI is then added selectively to characterise soft tissue masses, map cholesteatoma extent or delineate vascular tumours, complementing rather than replacing CT in the diagnostic work-up.

Table 1 summarises the roles of CT, MRI and CTA in diagnosing common causes of CHL, highlighting the preferred imaging modality for each condition and outlining the unique contributions and advantages of each technique.

As shown, CT and MRI are complementary. CT excels at depicting bony anatomy and serves as the primary tool for the initial diagnosis of CHL, while MRI is superior for characterising soft tissue lesions and detecting complications. CT angiography is used selectively, mainly to illuminate vascular lesions. Understanding these distinctions enables clinicians to select the most appropriate modality at each stage of the CHL work-up.

High-resolution CT of the temporal bone remains the primary imaging modality for evaluating CHL because of its submillimetre resolution, which allows detailed visualisation of bony structures – from stapes fixation in otosclerosis to ossicular discontinuities and complex congenital malformations. MRI serves as an essential complementary tool for soft-tissue characterisation: non-echo-planar diffusion-weighted and contrast-enhanced sequences facilitate detection or exclusion of cholesteatoma, delineation of tumours and inflammatory disease and evaluation of sensorineural components in mixed hearing loss cases. Although not routinely performed, CTA plays a valuable role in assessing suspected vascular middle ear lesions by revealing feeding vessels and their anatomic relationships.

In paediatric patients, imaging protocols emphasise radiation dose reduction, reserving CT for cases where findings will impact clinical management. In adults, a broader differential diagnosis and lower lifetime radiation risk support more frequent CT use, with MRI or CTA added selectively for soft-tissue or vascular clarification. Current appropriateness guidelines recommend temporal bone CT as the first-line modality, with MRI or CTA reserved for further evaluation as indicated.

Looking forward, emerging technologies such as ultra-HRCT and advanced non-echo-planar DWI (e.g. PROPELLER) promise improved detection of millimetre-scale lesions. Experimental CT–MRI fusion techniques may further enhance diagnostic specificity and surgical planning. With continued advancements and optimised multimodal imaging protocols, clinicians can anticipate even greater accuracy and confidence in diagnosing the full spectrum of CHL aetiologies in daily practice.