Audiology in the sudden hearing loss clinical trial

Abstract

Objective: To report the pretreatment and posttreatment population characteristics and the overall stability of the audiologic outcomes found during the Sudden Hearing Loss Clinical Trial (ClinicalTrials.gov: Identifier NCT00097448).

Study design: Multicenter, prospective randomized noninferiority trial of oral versus intratympanic (IT) steroid treatment of sudden sensorineural hearing loss (SSNHL).

Setting: Fifteen academically based otology practices.

Patients: Two hundred fifty patients with unilateral SSNHL presenting within 14 days of onset with 50 dBHL or greater pure tone average hearing threshold in the affected ear.

Intervention: Either 60 mg/d oral prednisone for 14 days with a 5-day taper (121 patients) or 4 IT doses for 14 days of 40 mg/ml methylprednisolone (129 patients).

Main outcome measure: Primary end point was change in hearing [dB PTA] at 2 months after treatment. Noninferiority was defined as less than 10 dB difference in hearing outcome between treatments. In this article, pretreatment and posttreatment hearing findings will be reported in detail.

Results: A general (and stable) effect of treatment and a specific effect of greater improvement at low frequencies were found in both treatment groups.

Conclusion: Hearing improvements are stable, and a significantly greater improvement occurs with lower frequency after either oral or IT steroid treatment of SSNHL.

Figures

Figure 1

Box (quartile) and whisker (range) plots are shown describing the pure tone average (PTA4; .5,1.2.4 KHz) for both study arms and for all participants combined. The left-most open boxes show the unaffected ear values. The gray filled boxes show (left to right) the pre-treatment findings on study entry, the 2 month post-treatment findings and the 6-month findings. There was a significant (***; p

Figure 2

Similar box-and-whisker plots to those in figure 1 are shown describing the word recognition findings (percent correct; CID W-22 50-item, recorded). These are also arranged with the entry values for unaffected ears on the left and then the pre-, post and longterm scores. Here again, there was a significant (***; p

Figure 3

Pre- and 2 month post-treatment hearing status for participants in this clinical trial are displayed using the proposed AAO-HNS Hearing Loss Scale developed for this purpose. Decade bins are formed vertically for hearing thresholds (dBPTA4) and horizontally for word recognition scores (percent correct). In the graph on the left no values less than 50 dB PTA4 appear since this was the study inclusion criterion.

Figure 4

Each individual case can be seen, plotted horizontally by the severity of hearing loss on entry and vertically by the amount of improvement (dB) seen at the 2 month post treatment evaluation. Oral cases are shown using filled circles and intratympanic using open triangles. Losses of 50 dBPTA4 are unlikely to improve beyond 50 dB and this gives rise to the diagonal improvement limit. The bottom shaded area contains those cases where no improvement (>10 dB) was seen.

Figure 5

Pre-treatment audiogram slopes in the affected ears are shown by subtracting the threshold at 500 Hz from that at 4000 Hz. A high frequency slope will result in a positive value along the horizontal axis. Most cases had flat losses.

Figure 6

Box-and-whisker plots are shown describing the improvement at each of the PTA4 frequencies between pre-treatment and 2 months post treatment with both study arms combined. The central tic shows the median for each set of improvement values and shows a significant trend for better improvement at lower frequencies (SAS GLM, PROC MIXED; p