Influencing factors of the long-term prognosis for residual tinnitus after SSNHL
Based on the results of two follow-ups, we divided the final tinnitus prognoses of all the subjects into cured (including grades 0 and 1) and uncured. First, the univariate Cox regression analysis was used to screen the possible influencing factors (Table 3). The analysis showed that the laterality of SSNHL and the initial tinnitus degree may be related to the long-term prognosis of tinnitus (P < 0.05). Then, the duration before treatment (0.05 < P < 0.10), the laterality of SSNHL, and initial level of tinnitus, as well as hearing recovery at discharge, which was not statistically remarkable but was clinically remarkable, were all included in the multivariate Cox regression analysis. The results showed that the initial level of tinnitus was an independent factor for the long-term prognosis of residual tinnitus after SSNHL.
Finally, the healing rates of patients with different initial tinnitus levels were calculated using the Kaplan–Meier method (Figure 2). Moreover, the median healing time of patients with different initial tinnitus levels can be used to predict the healing time of tinnitus for patients with different initial levels (Table 4). The analysis showed that tinnitus at all levels tended to heal with time. A higher initial degree of tinnitus corresponded to a prolonged healing time. The median times for tinnitus healing for patients with initial tinnitus levels of 1 to 5 were as follows: 1.00 ± 0.00 months, 21.00 ± 4.50 months, 33.00 ± 3.76 months, and 37.00 ± 11.50 months, respectively. The median recovery time for all cases was 23.00 ± 3.80 months (Table 4).