Diagnostic value and prognostic significance of MRI findings in sudden sensorineural hearing loss
Jelena Todic,Nils Guinand,Vincent Lenoir,Pascal Senn,Minerva Becker
DOI: https://doi.org/10.1002/lio2.922
2022-09-15
Abstract:Purpose: We evaluated the clinical significance of magnetic resonance imaging (MRI) findings and their prognostic value for initial hearing loss and recovery in patients with sudden sensorineural hearing loss (SSNHL). Materials and methods: This retrospective study included consecutive adult patients with unilateral SSNHL, contrast-enhanced MRI and audiometric testing evaluated in our institution between 2005 and 2017. MRI reports, patient data, treatment, and audiometric tests were reviewed, with the relationship between MRI findings and hearing loss/recovery analyzed. Results: Overall, 266 patients were included. Additional symptoms comprised tinnitus (114/266; 43%), vertigo (45/266; 17%), ear pain (26/266; 10%), and ear pressure (6/266; 2%). At least one cardiovascular risk factor (hypertension, diabetes, hypercholesterolemia, cardiopathy, and active smoking) existed in 167/266 (63%) patients. Corticosteroid treatment was followed by 198/266 (74%) patients while contraindications/refusal/compliance precluded treatment in 68/266(26%). Complete, partial or slight hearing recovery occurred in 167/266 (63%) patients. Three MRI patient groups were identified: a group with normal MRI examinations or incidentalomas (128/266; 48%), a group with peripheral auditory system (PAS) lesions (95/266; 36%), and a group with central nervous system (CNS) lesions (43/266; 16%). PAS lesions included lesions from the cochlea to the brain stem (e.g., schwannoma, meningioma, labyrinthitis, intracochlear hemorrhage, vestibulocochlear neuritis), whereas CNS lesions corresponded in 42/43(98%) of cases to leukoaraiosis and other vascular lesions (e.g., stroke, hemorrhage, aneurysm, venous sinus thrombosis, and cavernoma). Belonging to one of the three MRI groups did not influence the degree of initial hearing loss, affected frequencies or treatment, p > .05. Gender and cardiovascular risk factors did neither affect initial hearing loss nor recovery. However, age > 70 years negatively affected initial hearing loss in all frequencies, as well as recovery in all frequencies except 1000 Hz. Also, poor recovery of initial high-frequency hearing loss (>1000 Hz) was significantly associated with CNS lesions. Conclusion: Age > 70 years and CNS lesions depicted by MRI independently predicted poor auditory recovery, albeit in different frequencies. Lay summary: In patients with sudden hearing loss, older age (above 70 years) predicts poorer hearing recovery than in younger patients in most hearing frequencies. In addition, abnormalities of brain tissue revealed by MRI predict poorer hearing recovery at high frequencies. Level of evidence: Level III.