CT Scans Provide Insight Into Possible Cause of Pediatric Sensorineural Hearing Loss

Newswise — The recent development of imaging techniques provide more and more detailed acquisitions, especially in small-sized structures such as the inner ear. Magnetic resonance imaging (MRI) can isolate the vestibular and the tympanic ramps of the cochlea as well as the vestibulo-cochlear bundle and its branches in the internal auditory canal. This high resolution can be very useful in order to assess sensorineural hearing losses (SNHL).

High-resolution computed tomography (CT) and MRI are frequently done in order to investigate the cause of a SNHL and before cochlear implantation. These techniques allow a precise identification of congenital malformations, temporal bone fractures, ossified cochleae, and anomalies of the cochlear nerve, but radiologic interpretation is still dependent on visual and qualitative inspection. Quantitative standardized measurements of inner ear structures would greatly improve the diagnostic accuracy of radiologic studies; although many studies have addressed this issue, there is still a need to have these measurements defined and validated.

A new study set out to establish the normal measurements in children without SNHL, to compare the values with those obtained in a large group of children presenting with a SNHL, and then to identify radiological predictive factors of SNHL. The authors of Computed Tomography Measurements of the Normal And Pathologic Cochlea in Children are Natacha Crozat-Teissier, M. Elmaleh, and T. Van Den Abbeele, all from the Service ORL Hopital Robert Debre 48 Boulevard Serurier, Paris, France. Their findings will be presented at The Twentieth Annual Meeting of the American Society of Pediatric Otolaryngology (ASPO) http://www.aspo.us/ being held May 27-30, 2005, at the J.W. Marriott Las Vegas Resort in Las Vegas, NV.

Methodology: A retrospective study of CT scans was conducted between February 1999 and July 2004, in Robert Debré Hospital, a tertiary care university pediatric center in Paris. The indications were either an inflammatory pathology of the middle ear (cholesteatoma, infection), a head trauma with suspected temporal bone fracture, or the assessment of a SNHL (causal investigation or before cochlear implantation).

Cochleas from 162 different children (282 ears, 145 right ears and 137 left ears) were included in the study. The difference between the number of right and left ears was due to the fact that, in some instances, only one of the two ears had been saved on optical disks. Children aged from a few days to 16 years-10 months (average 81 ± 49 months). The first group of children included 71 children with unilateral or bilateral SNHL. The 91 remaining children (157 ears), did not have any SNHL, and formed the control group named conductive hearing loss (CHL), since the children had for the most acquired middle ear pathologies.

The researchers designed a strict measuring protocol in order to limit the inter and intra-observer variability of the measurements. Three cochlear dimensions were measured: the width of the canal for the cochlear nerve (WCN) at the entry of the cochlea, the height of the cochlea (HC) and the width of the cochlea (WTC). The measurements of the canal were always performed tangentially to the two inferior extremities of the X-shaped modiolus. The height was defined by the length of the orthogonal projection passing through the middle of the canal measure. The measurement of the first turn of the cochlea was parallel to the latter. Inverted contrast was used to assess more precisely the limits.

Results: In this study, the researchers compared several cochlear measurements of normal and pathologic cochlea. This allowed the establishment of normative measurements. Additionally, the findings display the absence of variability with age, considering the precision limits of the CT scan technique. No intra-subject or inter-subject difference between right and left ears was significant, except for WCN. The criterions for selection of the reference plane and the landmarks were highly reliable for the measurement of the height of the cochlea (HC) and the width of the first cochlear turn (WTC).

The study results revealed significant differences of cochlear measurements between children with SNHL and normal children concerning width of the cochlear nerve and width of the first cochlear turn. This congenital origin of a SNHL can be suggested when the measure of the WCN blow 1.8 mm or above 2.5 mm and/or the WTC is found to be below 5.4 mm. A WCN blow 1.8 mm alerts to a possible absence of the cochlear nerve.

For the group of children with intermediate measurements, the researchers could not determine an associated SNHL. In fact, in this group, there are children with a congenital SNHL, children who became deaf after meningitis or a temporal bone fracture. In case of a possible involution of the nerve after these events, there is no change in the diameter of the bony canal.

In this study, the presence of vestibular or cochlear malformations is a predictive criterion for the presence of a smaller 1rst turn or canal. In children with a profound SNHL, there is frequently a malformation of the membranous labyrinth associated, suggesting an abnormal development of the otic vesicle may inhibit the normal production of nerve growth factor. This leads to an insufficient neural growth and therefore a smaller cochlear nerve. The diameter of the canal depends on the normal development of the nerve, and therefore, is smaller due to the lack of stimulus.

Conclusion: The researchers in this study have measured the cochlea in normal hearing and hearing impaired children and established morpho metrical criterions evoking a SNHL. They have confirmed that a measure of the cochlear nerve canal greater than 2.5 mm or less than 1.8 mm is frequently associated with a SNHL. This measure should lead to doing a MRI to confirm the presence of the cochlear nerve, especially in the case of a pre-cochlear implant assessment.

The findings also demonstrate a significant difference for the size of the first turn of the cochlea in case of congenital SNHL compared to the control group. This is true both when there is a malformation and when the cochlea seems radiologically normal. Therefore, the measure of the first turn can be an interesting element when trying to establish the etiology of a hearing loss.