# The prediction of round window visibility at posterior-tympanotomy for cochlear implantation with “black bone” magnetic resonance imaging: Author lists

## TL;DR

This study shows that a new MRI technique can predict how visible a part of the ear is during surgery for cochlear implants, with better results in adults than children.

## Contribution

The study introduces the use of 'Black Bone' MRI for predicting round window visibility during cochlear implant surgery.

## Key findings

- The EAC angle predicted RW visibility in children with 73.4% sensitivity and 69.2% specificity.
- The modified RW niche angle was the best predictor in adults with 82.2% sensitivity and 71.4% specificity.
- Adults showed better overall prediction performance compared to children using BB MRI.

## Abstract

Although computed tomography (CT) can predict round window (RW) visibility at posterior tympanotomy pre-operatively, there is a trend towards the application of using magnetic resonance imaging (MRI) alone for cochlear implant (CI) planning. This study assessed the potential of a novel “Black Bone” (BB) MRI sequence to determine RW visibility during posterior tympanotomy.

Patients underwent BB MRI as part of pre-operative CI planning. Two independent radiologists performed 5 landmark-based MRI measurements. RW visibility was recorded during posterior tympanotomy as > 50% or < 50% visibility. Patients with undefinable landmarks or absent surgical grading were excluded. Mann-Whitney U or t-tests compared MRI measurements to RW visibility whilst step wise logistic regression determined significant predictors.

86 patients (40 male; median 10.5 years) and 129 implanted ears (43 unilateral, 43 bilateral; 52 adult, 77 paediatric) were evaluated. There were 109/129 ears with > 50% and 20/129 ears with < 50% RW visibility. The external auditory canal (EAC) angle was increased in paediatric patients with > 50% RW visibility (P = 0.033; AUROC 0.689; 73.4% sensitive, 69.2% specific with threshold 6.2o) with ICC = 0.598. The facial recess distance, facial nerve location, and modified RW niche angle were increased in adult patients with > 50% RW visibility but only modified RW niche angle was significant on stepwise regression (P = 0.016; AUROC 0.805; 82.2% sensitive, 71.4% specific with threshold 17.95o) with ICC = 0.578.

The visibility of RW at posterior tympanotomy may be determined with pre-operative BB MRI. The optimal predictive measurements differ between adult and paediatric patients and there is superior performance in adults.