2.1

Patients

In this study, we prospectively enrolled 132 patients who complained of an acute onset of unilateral idiopathic sudden sensorineural hearing loss (ISSNHL) within 7 days, and showed a mean hearing level greater than 30 dB between January 2013 and March 2014. All patients were treated with steroids alone. Age, gender, presence of diabetes or hypertension, presence of tinnitus and/or dizziness, interval between onset and treatment, initial hearing level, and final hearing level at three months from the initial visit were documented. Following a medical chart review, we gathered data from 421 patients who were diagnosed with unilateral ISSNHL and were treated with a combination of steroids and alprostadil injection between January 2005 and December 2012.

The following patients were excluded in order to exclude the possibility of Meniere’s disease or another autoimmune disease: patients with an age of less than 15 years; patients with concomitant vestibular schwannoma, meningitis, myelitis, vasculopathy, or neuropsychiatric disease; patients who could not be observed for at least 3 months; and patients with a history or recurrence of sudden hearing loss, or fluctuating symptoms including tinnitus, ear fullness, and dizziness during the study period. The Institutional Review Board of University Hospital approved this study.

2.2

Treatment

All patients were hospitalized for one week, and an initial dose of oral methylprednisolone (48 mg, MPD) was administered for 4 days. The dose was tapered 8 mg every 2 days . After interviewing patients, additional intratympanic dexamethasone (IT-DEX) injections were administered as a salvage treatment or concurrently with oral MPD treatment in case of insufficient recovery. IT-DEX was administered three times over 2 weeks by tilting the patients’ heads to the contra-lateral side at approximately 45°. Before the injection, all patients were instructed to maintain this posture for 30 min and not to swallow at all during this time . Each patient received 0.5 mL of dexamethasone (5 mg/mL, Yuhan Corp., Korea) through the antero-superior quadrant of the tympanic membrane by an otologist using a 26-gauge spinal needle and a 1-mL syringe.

A retrospective analysis of alprostadil treatment data collected between 2005 and 2012 was performed. A combined continuous infusion of 10 μg/d alprostadil (5 μg/mL, Tanabe Mitsubishi Pharma, Japan) was applied for 7 days with oral MPD treatment.

2.3

Classification

We classified the patients into two groups. Group I patients corresponded to the historical cohort that visited the university hospital between 2005 and 2012 and that received combination therapy. Group II patients were enrolled prospectively between January 2013 and March 2014, and demonstrated more than 30 dB of hearing loss in three sequential frequencies by initial pure tone audiometry. Group II patients were treated with steroid treatment alone, including IT-DEX.

2.4

Calculations of hearing and assessment of recovery

Pure tone averages (PTAs) were calculated using the arithmetic mean of the hearing levels at whole frequencies (250 Hz, 500 Hz, 1 kHz, 2 kHz, 4 kHz, and 8 KHz). For assessment of improvement, initial and final hearing levels were classified into six groups: normal (PTA ≤ 25 dB), mild (25 < PTA ≤ 40 dB), moderate (40 < PTA ≤ 70 dB), severe (70 < PTA ≤ 90 dB), and profound (90 < PTA) . Improvement was defined as positive changes in hearing after treatment, whereas no change or worsening of hearing was defined as no improvement . For measurement of recovery, calculations were performed using four types hearing recovery formula and compared using a receiver operating characteristics (ROC) curve based on the predictability of improvement. The four measurements used were hearing improvement rates (HIRs) , PTA recovery percentages , relative gain (RG) , and Siegel’s criteria . HIRs were calculated as the hearing gain divided by the initial hearing difference between the affected side and the healthy side, multiplied by 100 . The PTA recovery percentage was calculated as difference between initial inter-aural PTA difference and final inter-aural PTA difference, divided by initial inter-aural PTA difference, and then multiplied by 100 . The RG was calculated as the hearing gain divided by the initial PTA of the affected side, multiplied by 100 . By analyzing the ROC curve, the best hearing formula for predicting recovery was chosen among these four methods and the cut-off value was calculated.

2.5

Statistical analyses

Independent t -tests were used to assess the differences between groups. Chi-squared analysis was conducted to compare the differences between nominal variables. Estimations used in the propensity score included: (1) the age, sex, accompanying hypertension and diabetes, presence of tinnitus or dizziness, initial hearing levels of lesions and healthy sides, IT-DEX treatment as covariates; (2) treatment assignment (group I or II) as the outcome variable; and (3) logistic regression was performed, and propensity scores were calculated. The 3:1 nearest-neighbor method was utilized for matching. Next, a test of the balance of the covariates was performed, and the treatment effects were finally compared using paired t -tests. All statistical analyses were performed using SPSS software (ver. 18.0, SPSS Inc., Chicago, IL, USA) and R Commander Plug-in for the EZR Package (RcmdrPlugin.EZR), and a p -value of < 0.05 was considered significant .

2.1

Patients

In this study, we prospectively enrolled 132 patients who complained of an acute onset of unilateral idiopathic sudden sensorineural hearing loss (ISSNHL) within 7 days, and showed a mean hearing level greater than 30 dB between January 2013 and March 2014. All patients were treated with steroids alone. Age, gender, presence of diabetes or hypertension, presence of tinnitus and/or dizziness, interval between onset and treatment, initial hearing level, and final hearing level at three months from the initial visit were documented. Following a medical chart review, we gathered data from 421 patients who were diagnosed with unilateral ISSNHL and were treated with a combination of steroids and alprostadil injection between January 2005 and December 2012.

The following patients were excluded in order to exclude the possibility of Meniere’s disease or another autoimmune disease: patients with an age of less than 15 years; patients with concomitant vestibular schwannoma, meningitis, myelitis, vasculopathy, or neuropsychiatric disease; patients who could not be observed for at least 3 months; and patients with a history or recurrence of sudden hearing loss, or fluctuating symptoms including tinnitus, ear fullness, and dizziness during the study period. The Institutional Review Board of University Hospital approved this study.

2.2

Treatment

All patients were hospitalized for one week, and an initial dose of oral methylprednisolone (48 mg, MPD) was administered for 4 days. The dose was tapered 8 mg every 2 days . After interviewing patients, additional intratympanic dexamethasone (IT-DEX) injections were administered as a salvage treatment or concurrently with oral MPD treatment in case of insufficient recovery. IT-DEX was administered three times over 2 weeks by tilting the patients’ heads to the contra-lateral side at approximately 45°. Before the injection, all patients were instructed to maintain this posture for 30 min and not to swallow at all during this time . Each patient received 0.5 mL of dexamethasone (5 mg/mL, Yuhan Corp., Korea) through the antero-superior quadrant of the tympanic membrane by an otologist using a 26-gauge spinal needle and a 1-mL syringe.

A retrospective analysis of alprostadil treatment data collected between 2005 and 2012 was performed. A combined continuous infusion of 10 μg/d alprostadil (5 μg/mL, Tanabe Mitsubishi Pharma, Japan) was applied for 7 days with oral MPD treatment.

2.3

Classification

We classified the patients into two groups. Group I patients corresponded to the historical cohort that visited the university hospital between 2005 and 2012 and that received combination therapy. Group II patients were enrolled prospectively between January 2013 and March 2014, and demonstrated more than 30 dB of hearing loss in three sequential frequencies by initial pure tone audiometry. Group II patients were treated with steroid treatment alone, including IT-DEX.

2.4

Calculations of hearing and assessment of recovery

Pure tone averages (PTAs) were calculated using the arithmetic mean of the hearing levels at whole frequencies (250 Hz, 500 Hz, 1 kHz, 2 kHz, 4 kHz, and 8 KHz). For assessment of improvement, initial and final hearing levels were classified into six groups: normal (PTA ≤ 25 dB), mild (25 < PTA ≤ 40 dB), moderate (40 < PTA ≤ 70 dB), severe (70 < PTA ≤ 90 dB), and profound (90 < PTA) . Improvement was defined as positive changes in hearing after treatment, whereas no change or worsening of hearing was defined as no improvement . For measurement of recovery, calculations were performed using four types hearing recovery formula and compared using a receiver operating characteristics (ROC) curve based on the predictability of improvement. The four measurements used were hearing improvement rates (HIRs) , PTA recovery percentages , relative gain (RG) , and Siegel’s criteria . HIRs were calculated as the hearing gain divided by the initial hearing difference between the affected side and the healthy side, multiplied by 100 . The PTA recovery percentage was calculated as difference between initial inter-aural PTA difference and final inter-aural PTA difference, divided by initial inter-aural PTA difference, and then multiplied by 100 . The RG was calculated as the hearing gain divided by the initial PTA of the affected side, multiplied by 100 . By analyzing the ROC curve, the best hearing formula for predicting recovery was chosen among these four methods and the cut-off value was calculated.

2.5

Statistical analyses

Independent t -tests were used to assess the differences between groups. Chi-squared analysis was conducted to compare the differences between nominal variables. Estimations used in the propensity score included: (1) the age, sex, accompanying hypertension and diabetes, presence of tinnitus or dizziness, initial hearing levels of lesions and healthy sides, IT-DEX treatment as covariates; (2) treatment assignment (group I or II) as the outcome variable; and (3) logistic regression was performed, and propensity scores were calculated. The 3:1 nearest-neighbor method was utilized for matching. Next, a test of the balance of the covariates was performed, and the treatment effects were finally compared using paired t -tests. All statistical analyses were performed using SPSS software (ver. 18.0, SPSS Inc., Chicago, IL, USA) and R Commander Plug-in for the EZR Package (RcmdrPlugin.EZR), and a p -value of < 0.05 was considered significant .