Introduction
Idiopathic sudden sensorineural hearing loss (SNHL) is defined as SNHL of at least 30 dB in three consecutive frequencies occurring within 3 days or less. Idiopathic sudden SNHL presents mostly as an otological disease of unknown cause.Reference Tovi, Ovadia, Eliashar, de Jong and Gross1,Reference Chen, Zhang, Zhang, Wang, Hu and Wu2 Various causes such as viral, bacterial or protozoal infections, vascular events, metabolic conditions, and autoimmune mechanisms have been suggested.Reference Chen, Zhang, Zhang, Wang, Hu and Wu2
Both humoral and cellular immunity could cause immune response damage. Autoimmune damage can be classified as type I–IV allergic reactions. It has been shown that type II–IV allergic reactions are potential mechanisms that lead to inner-ear damage in SNHL.Reference Li, You, Ma, Li, Li and Sun3 A number of inner-ear antigens have been suggested as targets for autoantibodies: type II collagen, beta-actin, cochlin, beta-tectorin and prestin.Reference Tovi, Ovadia, Eliashar, de Jong and Gross1 Prestin, which is found in the lateral plasma membrane of outer hair cells, plays a central role in cochlear sensitivity and tuning.Reference Zheng, Shen, He, Long, Madison and Dallos4 Damage to outer hair cells is one of the earliest events that lead to hearing loss. Therefore, prestin could be an appropriate biomarker of inner-ear function and possibly hearing loss.Reference Parham5
Prestin has been reported to be a biochemical marker for early diagnosis of SNHL in an animal model of noise-induced hearing loss and cisplatin ototoxicity.Reference Naples, Cox, Bonaiuto and Parham6,Reference Parham and Dyhrfjeld-Johnsen7 So far, two human studiesReference Tovi, Ovadia, Eliashar, de Jong and Gross1,Reference Sun, Xuan, Zhou, Yuan and Xue8 have been conducted to confirm this theory. Prestin concentration in serum was found to be significantly higher in idiopathic sudden SNHL patients than in normal controls (p < 0.001). In 60 per cent of idiopathic sudden SNHL patients who responded to treatment, prestin concentrations were lower after than before treatment. However, the difference was not statistically significant.Reference Sun, Xuan, Zhou, Yuan and Xue8 Another study,Reference Tovi, Ovadia, Eliashar, de Jong and Gross1 in which anti-prestin autoantibodies were analysed by enzyme-linked immunosorbent assay in the plasma of idiopathic sudden SNHL patients, found no statistically significant differences between prestin autoantibodies and demographic or audiological parameters.
The current study aimed to compare the prestin levels in unilateral idiopathic sudden SNHL patients with controls, and to determine whether any relationships exist between prestin levels and certain clinical features.
Materials and methods
In this prospective study, patients diagnosed with sudden SNHL presenting to the otolaryngology clinic of a university hospital were enrolled between January 2020 and June 2020. The study inclusion criteria were adults (aged over 18 years) diagnosed with sudden SNHL within 7 days from the onset. The exclusion criteria included hearing loss with an identified cause such as Ménière's disease, retrocochlear diseases and autoimmune hearing loss. Controls were selected from healthy normal adults with no history of hearing loss. The control group was individually matched to cases according to age group and sex.
The therapeutic protocol consisted of 1 mg/kg oral prednisolone for one week, which was tapered by 10 mg every other day. If systemic steroid was contraindicated (e.g. because of uncontrolled diabetes mellitus), intratympanic dexamethasone injection (0.5 ml at a dose of 4 mg/ml every other day for 12 consecutive days) was administered.
All participants underwent audiometry, routine serological tests and blood tests for serum prestin level at the time of study entry (T0). In the case group, audiometry and prestin measurements were also performed at the end of treatment (day 14, T1) and day 30 (T2). Moreover, all patients underwent magnetic resonance imaging with gadolinium of the inner ear and cerebellopontine angle.
Pure tone averages (PTAs) from 0.25 kHz to 4 kHz were assessed. The PTAs were measured by the mean threshold value at 0.5, 1, 2 and 4 kHz. The types of initial audiogram were classified as: upward-sloping (hearing loss more severe in low frequencies), downward-sloping (hearing loss more severe in high frequencies), flat (no more than 10 dB deviation on PTA thresholds) or profound (PTA thresholds worse than 70 dB on all frequencies). Pre-treatment hearing level was categorised by a proposed grading system, modified from Siegel's criteria,Reference Cheng, Chu, Tu, Shiao, Wu and Liao9 according to average thresholds of: 25 dB or less (grade 1), 26–45 dB (grade 2), 46–75 dB (grade 3), 76–90 dB (grade 4), or greater than 90 dB (grade 5). We categorised the hearing outcome into three groups: significant recovery, slight recovery and unchanged. Significant recovery was defined as thresholds improved to the same level as the unaffected ear, or improved by 30 dB or more on average. Slight recovery was hearing improvement between 10 and 30 dB on average. Hearing recovery less than 10 dB was considered unchanged.
For prestin measurement, blood samples were collected using ethylenediamine tetra-acetic acid as an anticoagulant, and centrifuged for 15 minutes at 1000 revolutions per minute within 30 minutes of collection. The samples were stored in a −80°C refrigerator until the time of assay. Prestin concentration was measured in plasma using a human prestin (catalogue number: E4170Hu) enzyme-linked immunosorbent assay kit (Bioassay Technology Laboratory, Shanghai, China), as described in the manufacturer's instruction manual. Briefly, samples and standards are pipetted into the wells in which a specific antibody has been pre-coated onto a microplate. After incubation, streptavidin-conjugated horseradish peroxidase is added to bind to the biotinylated prestin antibody. Following washing, a substrate solution is added and colour develops. The intensity of the colour was measured at 450 nm (by a BioTek ELx808™ Spectrophotometer).
Quantitative variables were described in terms of means and standard deviations (SDs). Descriptive analyses for categorical variables included absolute frequencies and percentages. Hearing thresholds and prestin levels were log-transformed to normalise distributions. All the analyses reported in the manuscript are based on the log-transformed variables unless indicated otherwise. Back-transformation converted log-transformed means into geometric means on the original scale, which were easier to interpret. In addition, the differences between means on the log scale were back-transformed. The difference of means on the log scale conveniently became a ratio of the means on the original scale. Outcomes were back-transformed for presentation. A one-way analysis of variance was performed to assess the relation between serum prestin level and grade of hearing loss, as well as recovery outcome. All statistical tests were two-sided, and a p-value of less than 0.05 was set as the level of statistical significance. Statistical analyses were conducted using Stata software version 14.0 (StataCorp, College Station, Texas, USA).
Results
Between January 2020 and June 2020, 25 patients and 25 age- and sex-matched controls were recruited in this study. Twenty individuals were male and 30 were female, with a mean (± SD) age of 48.2 ± 14.6 years (range, 19–77 years). The sudden SNHL was right-sided in 17 patients (68.0 per cent) and left-sided in 7 patients (32.0 per cent). None of the individuals had a previous history of sudden SNHL. Seven patients (28.0 per cent) had diabetes, 11 (44.0 per cent) had hypertension and 5 (20 per cent) had dyslipidaemia. According to National Cholesterol Education Program Adult Treatment Panel III criteria,10 seven patients (28.0 per cent) and two controls (8.0 per cent) had metabolic syndrome. Table 1 shows the demographic characteristics of the two groups in detail.
Characteristics of study subjects
Data represent numbers (percentages) for categorical variables and means (standard deviations) for continuous variables. *n = 25; †n = 25
Regarding audiometry types, 7 (28.0 per cent) were upward-sloping, 10 (40.0 per cent) were downward-sloping and 8 (32.0 per cent) were flat. Five patients (20.0 per cent) had profound hearing loss. According to the modified Siegel's criteria, pre-treatment hearing level was determined to be grade 2 in 13 patients (52.0 per cent), grade 3 in 9 patients (36.0 per cent), and grades 4–5 in 3 patients (12.0 per cent).
Mean (± SD) time between onset and treatment initiation was 2.3 ± 2.2 days. Oral steroid therapy was administered to 8 subjects (32.0 per cent), as mentioned in our standard treatment protocol, and 17 patients (68.0 per cent) were treated with intratympanic corticosteroids. After 30 days, significant recovery occurred in 5 patients (20.0 per cent), slight recovery in 5 patients (20.0 per cent) and unchanged hearing in 15 patients (60.0 per cent).
The prestin levels in patients had a broad range (93.4–509.4 pg/ml). Baseline prestin levels are shown in Figure 1. The control subjects exhibited normal ranges, from 98.7 to 255.9 pg/ml. The geometric mean of prestin level (95 per cent confidence interval (CI)) in the case and control groups at T0 (time of study entry) was 227.7 (178.5–290.4) pg/ml and 130.5 (118.0–144.5) pg/ml, respectively. There was a statistically significant difference between the two study groups (p < 0.001), giving a ratio of the means of 1.74 (with a 95 per cent CI of 1.38–2.20).
Baseline prestin concentrations in patients with idiopathic sudden sensorineural hearing loss, according to hearing level grades (a) and audiometric types (b). Thirteen patients had hearing level grades of 1–2 and 12 patients had grades of 3 or higher.
In order to gain reliable statistical results, hearing level classified according to two subgroups: lower than grade 3 and grade 3 or higher. There was no significant difference between the geometric means of prestin level in the two subgroups (205.7 pg/ml vs 249.4 pg/ml, respectively; p = 0.42).
The blood examination performed at day 14 (end of treatment, T1) and at day 30 (T2) demonstrated a downward trend in the geometric means of prestin level in the case group (214.0 pg/ml and 180.1 pg/ml, respectively; p < 0.001). However, the last recorded prestin level (T2) in patients was significantly higher than the baseline level in the control group (180.1 pg/ml vs 130.5 pg/ml, ratio of the means 1.38; p < 0.01).
We combined significant and slight recovery to gain reliable statistical results. In addition, hearing level was again categorised as lower than grade 3 or grade 3 or higher. Figure 2 displays the geometric means of the prestin concentration at different times according to audiogram type and recovery outcome. The ratio of the means for prestin level at T0 and T2 did not show any association with audiogram type or recovery outcome. Regardless of the recovery outcome or audiometric type of hearing loss, prestin concentration tended to decrease at T2 in all patients.
Geometric means of prestin concentrations at different times in patients with or without response to treatment (a), and with different audiometry types (b).
Of the 25 patients, 17 (68 per cent) had prestin levels greater than 150 pg/ml, which were above the upper limit of 95 per cent CI for prestin levels in the controls (Figure 3). According to the trend in prestin levels between T0 and T2, patients were classified into three groups. In the first group, the prestin level was less than or equal to 150 pg/ml at both T0 and T2 (low-low, n = 8). The rate of good recovery in this group was 50 per cent. In the second group, the prestin level was above 150 pg/ml at T0 and decreased to less than 150 pg/ml at T2 (high-low, n = 11). Of these, 45.5 per cent had a good recovery. Finally, the prestin level was above 150 pg/ml at both T0 and T2 in the third group (high-high, n = 6), and none of them showed good recovery (Figure 4).
Box plot showing baseline prestin levels in idiopathic sudden sensorineural hearing loss (SNHL) patients and controls.
Audiometric recovery after treatment in patients with different prestin trends between time of study entry (T0) and day 30 (T2).
Discussion
Sudden SNHL has many possible causes. Histopathological studies of human temporal bone have suggested viral, vascular and immunological causes, and different pathologies, such as tectorial membrane, spiral ganglion cells and stria vascularis. However, damage to the outer hair cells is believed to be one of the important events leading to idiopathic sudden SNHL, especially in patients with high-frequency hearing loss.Reference Sun, Xuan, Zhou, Yuan and Xue8,Reference Ishai, Kamakura and Nadol11,Reference Ungar, Handzel and Santos12 This study revealed that only 68 per cent of patients had a higher prestin level than controls. This is similar to the finding by Sun et al.,Reference Sun, Xuan, Zhou, Yuan and Xue8 who reported higher prestin concentration than the controls in only half of the patients. Therefore, outer hair cells appear to play a central role in approximately two-thirds of idiopathic sudden SNHL cases. In these patients, idiopathic sudden SNHL is associated with the disruption of outer hair cells, followed by the release of prestin into circulation. Because of different degrees of severity of outer hair cell damage in idiopathic sudden SNHL patients, the prestin level is increased to various degrees compared with controls. The current study demonstrated a wide range of prestin levels (93–509 pg/ml) in patients.
• Prestin has been used as a biochemical marker for early sensorineural hearing loss (SNHL) in animal models of noise-induced hearing loss and cisplatin ototoxicity
• So far, two studies have attempted to confirm this link in humans, but the results were contradictory
• This study found no correlations between idiopathic sudden SNHL prestin levels and clinical findings or treatment outcomes
• A decrease in post-treatment prestin levels in those with elevated baseline prestin (over 150 pg/ml) was associated with a 50 per cent chance of good recovery
As seen in Table 2, this study showed a lower probability of recovery with a higher hearing grade at baseline. However, prestin level showed no association with these variables at the 95 per cent confidence level. Previous studies have documented an association between the severity of initial hearing loss and poorer prognosis in idiopathic sudden SNHL.Reference Kang, Yang, Shim, Song, Kim and Lim13,Reference Conlin and Parnes14 It is generally accepted that almost all patients with hearing loss greater than 90 dB do not recover, regardless of the therapy.Reference Conlin and Parnes15
Changes over time in audiometric findings and serum prestin levels for sudden SNHL patients
SNHL = sensorineural hearing loss; CI = confidence interval
Although the sample size was too small to be conclusive, only four of eight subjects with baseline prestin levels of 150 pg/ml or lower showed good recovery. Further analysis based on a large sample size is needed to confirm this finding. Of the 17 patients with elevated baseline prestin levels (greater than 150 pg/ml), 30 days after idiopathic sudden SNHL, 11 patients (64.7 per cent) had prestin levels which reached that of controls. About half of this subgroup (45.5 per cent) showed good recovery (in terms of PTA) with corticosteroid use, while hearing threshold improvements were not recorded in the subjects with elevated prestin levels at day 30 (T2).
The present study has several limitations. First, the sample size was small. This study aimed to explore the possibility of using prestin as a biomarker in idiopathic sudden SNHL. Further studies are needed to compare prestin levels at varying times. The findings suggest it is worthwhile pursuing this aim and estimating the required sample size for these enlarged studies. Second, we only measured hearing thresholds using PTA. The use of objective tools such as auditory brainstem response testing may have allowed us to demonstrate the location of injury in idiopathic sudden SNHL cases, and accurately determine changes in prestin level and hearing thresholds. Thus, there is room for additional work to confirm our results. Third, the present study revealed time-dependent changes in prestin following idiopathic sudden SNHL. Previous animal studiesReference Naples, Cox, Bonaiuto and Parham6,Reference Parham, Sohal, Petremann, Romanet, Broussy and Tran Van Ba16 showed that changes can be slow, as in noise-induced hearing loss, or rapid, as in cisplatin ototoxicity. Furthermore, with noise, prestin concentrations tend to decline, whereas with cisplatin, there is temporary rise in prestin concentrations. We know little about prestin changes after idiopathic sudden SNHL. Thus, lack of uniformity in terms of time of study entry, after the onset of idiopathic sudden SNHL, could confound the interpretation of the results. Finally, the treatment protocol at our clinic involves systemic or intratympanic corticosteroids, which excludes other commonly used methods.
Conclusion
This study showed that damage to outer hair cells occurred only in some patients with idiopathic sudden SNHL. Although hearing loss severity had a negative association with recovery, there was no correlation between the prestin concentrations in the idiopathic sudden SNHL group and clinical findings or treatment outcomes. Interestingly, we found that the change in prestin level after treatment in the subgroup with high baseline prestin levels was indicative of response to treatment. About 46 per cent of these patients showed good recovery, while in patients without decreasing prestin levels, good recovery was not observed. These findings support previous studies which suggest prestin may serve as an inner-ear biomarker that could be a potential indicator of the prognosis of idiopathic sudden SNHL.
Acknowledgements
The authors wish to thank all subjects for participating in the study. This study was supported by a grant (number: 99022901) of the Guilan University of Medical Sciences (‘GUMS’), Iran.
Competing interests
None declared
