Prognostic analysis and nomogram prediction model construction of severe full-frequency sudden sensorineural hearing loss

AbstractObjectivesThis stidy aimed to identify factors influencing prognosis in severe full-frequency sudden sensorineural hearing loss and develop an early prognostic tool.MethodsA total of 194 patients with severe full-frequency sudden sensorineural hearing loss were analyzed. Univariate analysis screened variables, followed by multivariate logistic regression to construct a nomogram prediction model. Model performance was evaluated using a receiver operating characteristics curve.ResultsSignificant differences (p < 0.05) were found between ineffective and effective treatment groups in gender, age, contralateral ear hearing, duration of hearing loss, dizziness/vertigo, hypertension, platelet count and fibrinogen levels (Fib1, Fib2, Fib3). The nomogram model, incorporating these factors, showed good calibration. The receiver operating characteristics curve analysis revealed an area under the curve of 0.880 (95 per cent confidence interval: 0.829–0.931), with sensitivity of 80.7 per cent and specificity of 87.5 per cent.ConclusionThe nomogram model, integrating 11 factors, effectively predicts prognosis in severe full-frequency sudden sensorineural hearing loss.