Auditory and Vestibular Recovery in Otosclerosis Surgery: Clinical Efficacy and Predictive Modeling of Intraoperative Corticosteroid Use.

Abstract

To evaluate the effect of intraoperative local corticosteroid application on hearing, tinnitus, and dizziness outcomes after stapes surgery for otosclerosis, and to assess predictive modeling for treatment response. Retrospective multicenter cohort study. Three tertiary referral hospitals. Two hundred primary stapedotomy patients were analyzed: Group 1 (n = 100) received topical triamcinolone acetonide (40 mg/mL) applied to the oval window/footplate; Group 2 (n = 100) underwent standard surgery without steroids. Preoperative and 3-month postoperative evaluations included pure tone thresholds, air-bone gap (ABG), speech discrimination score (SDS), Tinnitus Handicap Inventory (THI), and Dizziness Handicap Inventory (DHI). Machine learning algorithms identified predictors of a favorable response. The steroid group achieved significantly greater hearing gains at all frequencies, most pronounced at 8000 Hz (28.13 ± 6.7 vs 19.09 ± 5.4 dB; P < .001), higher SDS (+6.37% vs -1.52%; P = .006), and superior ABG closure (P = .004). THI and DHI scores improved significantly (P < .001). Machine learning models achieved an AUC of 0.92 for outcome prediction. Intraoperative corticosteroids significantly enhance auditory and vestibular recovery in stapes surgery and, combined with predictive modeling, may support personalized patient care.