A New Methodology for Evaluation of Large Vestibular Aqueduct Syndrome in CT and MRI Images: Defining the Syndrome and Finding Clinical Correlation.

HYPOTHESIS: The goal of the research is to propose criteria to define large vestibular aqueduct (LVA)/large endolymphatic sac anomaly (LESA) using our previously developed method based on analysis of CT/CBCT and MRI, and to propose a new classification of the clinical course of hearing loss. BACKGROUND: Previous studies did not provide results that would conclusively relate audiological and radiologic data. METHODS: A retrospective analysis of radiologic images and clinical data was performed at a tertiary center. The dataset involved 173 patients (315 ears), who were diagnosed with LVA/LESA based on CT/CBCT and/or MRI of the temporal bone. The retrospective radiologic analysis was performed in the control groups on CT/CBCT images: 93 (100 ears) patients with normal hearing and 50 (100 ears) patients with various hearing loss. Measurements in the control and LVA groups were compared. RESULTS: Logistic regression model suggests that variables that correlate significantly with hearing loss are length of vestibular aqueduct (VA) (P<0.05) and operculum diameter (P<0.05) obtained from CT/CBCT images, and "sac_volume_to_dura_contact ratio" (P<0.05) measured in MRI. The decision tree model showed that CT opening [maximum perpendicular width of the opening of the VA (closest to the vestibule)] (<0.075 mm) is a significant factor differentiating the control and LVA groups. CONCLUSION: The presence of the CT opening can be used for screening and diagnosis of the LVA in bony images. The presence of contact between the sac and dura on MRI can confirm the diagnosis of LESA. MRI is superior for LVA/LESA diagnosis. The shorter and brighter the vestibular aqueduct, the smaller the "sac_volume_to_dura_contact ratio," which indicates a worse prognosis of hearing loss.