Precision and prediction matter: investigating hearing recovery measurements and prognosis in sudden sensorineural...

Sudden sensorineural hearing loss (SSNHL) is defined as 30 decibels (dB) hearing loss in 3 consecutive frequencies occurring within 72 h. Pure-tone average of four frequencies (PTA4) is commonly used to evaluate hearing levels but may not accurately reflect the recovery. We aimed to identify prognostic factors for recovery and to evaluate how recovery should be assessed, by comparing PTA4 with an individual pure-tone average (iPTA), including solely the hearing thresholds for the affected frequencies.

Demographic, clinical, and audiologic factors were analyzed using multivariable linear and logistic regression models. A Bland-Altman plot was used to compare recovery measurements based on iPTA and PTA4.

In this cohort, the mean age was 57 years (range 19-91 years). Dizziness was a prominent negative predictive factor (logistic regression: iPTA OR 0.09 95% CI 0.02-0.38, for full recovery; linear regression: iPTA 14.4 dB poorer recovery). Tinnitus correlated with, on average, 4.9 dB poorer recovery (P=0.043). Each day of delayed assessment was linked to a 0.84 dB reduction in recovery (P <0.001; OR 0.92, 95% CI 0.87-0.98). Comparing PTA4 with iPTA, the Bland-Altman plot showed -2.4 dB mean difference with wide limits of agreement, ranging from approximately -17 to 13 dB. In cases of frequency range-specific hearing loss, hearing recovered 8.7 dB more by using iPTA than PTA4 (P=0.003).

Dizziness, tinnitus, and increasing disease duration until assessment are negative prognostic factors. Compared to PTA4, iPTA better reflects actual hearing recovery, particularly in frequency range-specific hearing loss.