Introduction
Sudden sensorineural hearing loss is defined as hearing loss of 30 dB or greater across three contiguous audiometric frequencies within 72 hours.Reference Loughran1 It is an otological emergency with the potential to become permanent.Reference Jung and Hoegerl2 The annual incidence is 5 to 20 per 100 000 individuals in Western countries, with a higher prevalence among individuals aged 50–60 years.Reference Byl3, Reference Kuhn, Heman-Ackah, Shaikh and Roehm4 Patients usually complain of sudden hearing loss, aural fullness, tinnitus and vertigo, typically right after awakening.Reference Tripathi and Deshmukh5
The presence of vestibular symptoms is a clue to the extension of the pathology to the vestibular organs, possibly due to its anatomical proximity and the same embryonic origin as the cochlea.Reference Lim, Jeong, Han, Rah, Cha and Choi6 The prevalence of vestibular symptoms has been estimated to be between 30 and 60 per cent of cases.Reference Fetterman, Saunders and Luxford7–Reference Chiossoine-Kerdel, Baguley, Stoddart and Moffat10 The probable aetiologies encompass neurological disorders, traumas and neoplasms, while the exact pathophysiology remains unclear in most cases.Reference Lazarini and Camargo11 Some explanations include infections, labyrinthine membrane rupture, impaired immune-mediated processes and vascular disorders.Reference Kuhn, Heman-Ackah, Shaikh and Roehm4, Reference Lazarini and Camargo11 Vasospasm, haemorrhage or thrombosis can damage the inner-ear structures responsible for hearing and balance through reduced blood flow, hypoxia and decreased metabolic activity.Reference Sone, Kobayashi, Yoshida and Naganawa12, Reference Tsuzuki and Wasano13
The diagnosis is based on history and clinical evaluations, and confirmed by pure tone audiometry (PTA).Reference Jung and Hoegerl2 After the clinical diagnosis, magnetic resonance imaging (MRI) is suggested to rule out retro-cochlear pathologies.Reference Chandrasekhar, Tsai, Schwartz, Bontempo, Faucett and Finestone14–Reference Todic, Guinand, Lenoir, Senn and Becker16 Although treatment is very controversial, it is recommended that all patients receive corticosteroids either orally or locally via intratympanic injection.Reference Chandrasekhar, Tsai, Schwartz, Bontempo, Faucett and Finestone14, Reference Crane, Camilon, Nguyen and Meyer17–Reference Shemirani, Schmidt and Friedland19
Spontaneous recovery has been estimated to occur in 32–65 per cent of cases.Reference Chaushu, Ungar, Abu Eta, Handzel, Muhanna and Oron20–Reference Mattox and Simmons22 Various factors affect the prognosis, such as age, hearing loss severity, audiogram pattern, the period between disease onset and treatment, and some systemic disorders and conditions.Reference Tripathi and Deshmukh5, Reference Lee and Chung23 The correlation between vestibular dysfunction and hearing loss prognosis, as well as the patterns of vestibulocochlear lesions, continues to be the subject of recent studies, which utilise caloric tests and cervical vestibular-evoked myogenic potential tests.Reference Lee and Chung23, Reference Yu and Li24 The caloric test investigates the function of the lateral semicircular canal and superior vestibular pathway pathologies,Reference Shepard and Jacobson25 while the cervical vestibular-evoked myogenic potential test assesses saccular function and the inferior vestibular pathway.Reference Colebatch, Rosengren and Welgampola26
A recent meta-analysis showed that vestibular function is impaired in sudden sensorineural hearing loss regardless of the presence of dizziness.Reference Yu and Li24 Moreover, while some studies have indicated that vertigo is associated with more severe hearing loss and a worse prognosis,Reference Suzuki, Mori, Hashida, Shibata, Nguyen and Wakasugi27–Reference Cho and Cho29 others have found no significant difference.Reference Cho and Cho29, Reference Wen, Chen and Wu30
Accordingly, this study aims to explore the correlation between hearing loss and vestibular dysfunction in sudden sensorineural hearing loss through clinical diagnostic tests before and after standard treatment. In addition, it seeks to consider the prognostic role of vestibular dysfunction.
Materials and methods
This retrospective cohort study was conducted between 2021 and 2023 at a national referral tertiary hospital’s ENT clinic. Patients with sudden hearing loss (within less than 72 hours) of more than 30 dB in three consecutive audiometric frequencies with a duration of less than 15 days from symptom onset were included if they had no history of head trauma, meningitis, Ménière’s disease, noise-induced hearing loss, otological surgery, chronic otitis media or autoimmune disorders.Reference Loughran1 Those with any pathological finding on their MRI (Philips Ingenia 1.5T MRI system, Netherlands, cerebellopontine angle with contrast) were excluded.
One of the following treatments was prescribed: daily oral (PO) prednisolone (1 mg per kilogram, maximum dose 80 mg) for 2 weeks, which thereafter gradually tapered. For patients with any accompanying co-morbidity that could be considered relative contraindications for systemic corticosteroids, such as diabetes, glaucoma, active peptic ulcer and cataract, intratympanic injection of dexamethasone or methylprednisolone was indicated (IT).Reference Crane, Camilon, Nguyen and Meyer17 A second audiometry was performed for patients receiving PO treatment after 10 days, and their regimen was changed to IT if no progress was observed.
Patients’ demographic data, including age, gender and co-morbidities, were recorded. Patients were examined regarding nystagmus and were subjectively asked about vertigo. The caloric test, the cervical vestibular-evoked myogenic potential test and PTA were performed at the first visit. Patients were followed by PTA three months after the onset of their treatment. The hearing threshold was calculated at frequencies between 250 and 8000 Hz by PTA, and the average was assessed at 500, 1000, 2000 and 4000 Hz. The severity of hearing loss was determined based on American Speech and Hearing Association criteria.Reference Loughran1
Pure tone audiometry improvement based on Siegel criteria was classified as follows: complete recovery (patients whose final hearing level was better than 25 dB), partial recovery (patients who showed more than 15 dB of gain with a final hearing level between 25 and 45 dB), slight recovery (patients who showed more than 15 dB of gain with final hearing level poorer than 45 dB) and no improvement (patients who showed less than 15 dB of gain).Reference Siegel31
The study protocol was explained to the patients and informed consent was obtained from them or their legal guardians. All procedures in this study followed the 1964 Helsinki Declaration and its later amendments, and were approved by the Institutional Review Board and Ethics Committee of Iran University of Medical Sciences (IR.IUMS.FMD.REC.1401.081). Data are accessible on request from the first author.
Data were presented as mean and standard deviation for quantitative variables and as number and percentage for qualitative variables. An independent t-test was used to compare the means between two groups of patients. We performed the paired t-test to compare hearing improvement before and after treatment. To compare hearing between more than two categorisations, an analysis of variance test was used. We used the chi-square test to compare the expected and observed frequencies of hearing improvement based on Siegel criteria and other categorical variables. When the Kolmogorov–Smirnov test indicated that the distribution was not normal, non-parametric equivalent tests were used. The statistical variables were analysed using SPSS (version 22.0, IBM Corporation, Armonk, NY, USA) and p values less than 0.05 were considered statistically significant.
Results
We enrolled 44 eligible patients with an average age of 52.1 ± 6.2 years (range, 13–79 years) in the final analysis, of which 23 (52.2 per cent) were female. Seventeen cases (38.6 per cent) reported at least one of the following co-morbidities that affected their treatment regimen: diabetes mellitus (n = 11, 25 per cent), hypertension (n = 15, 34 per cent), peptic ulcer (n = 2, 4.5 per cent) and ischaemic cardiac disease (n = 5, 11.4 per cent). Hearing loss was detected in the right ear in 24 patients (54.5 per cent) and in the left ear in 20 patients (45.5 per cent).
The median time between the symptom onset and the start of treatment was 7 days (interquartile range = 5 days). It was less than 7 days in 24 cases (54.5 per cent). No statistically significant difference was detected between those with immediate treatment (less than 7 days) and those receiving treatment after 7 days in their PTA results at the beginning of treatment (55.10 ± 20.18 vs 55.03 ± 20.06, p = 0.74) and after treatment (44.37 ± 25.42 vs 46.37 ± 24.69, p = 0.79). Accordingly, improvement in the PTA test was not significantly different between the two groups. (10.72 ± 16.26 vs 6.68 ± 13.45, p = 0.38). None of the co-morbidities significantly impacted the primary and secondary PTA tests.
Twenty patients (45.4 per cent) received a combination of PO and IT, while 16 patients (36.36 per cent) received IT and 8 (18.18 per cent) received PO. No significant difference in PTA improvement was found between patients receiving different regimens (PO and IT, 8.90 ± 16.67; IT, 8.57 ± 15.69; PO, 9.21 ± 1.03; p = 0.99). The chi-square test showed that patients receiving combination therapy were significantly more frequent than expected in the category of complete recovery based on Siegel criteria (p = 0.001).
The caloric and cervical vestibular-evoked myogenic potential tests were abnormal in 16 cases (36.4 per cent) and 14 cases (31.8 per cent) at the first visit, respectively. It should be noted that 13 patients (29.5 per cent) had impairments in both tests. Sixteen patients (36.4 per cent) complained of vertigo at the first visit, which reduced to 8 patients (18.2 per cent) post treatment. Four patients with vertigo had normal caloric and cervical vestibular-evoked myogenic potential tests. The two cases with spontaneous nystagmus at the first visit had no nystagmus at the follow up.
Table 1 shows the results of PTA for each evaluated frequency at the first visit and post treatment. There was significant improvement at all frequencies. Speech discrimination scores were reported between poor and 100 at the first visit and post treatment (median, 88 per cent for both time points). The mean speech discrimination score improvement was 3.63 ± 12.58 (range, −12 to 64).
Pure tone audiometry results for different evaluated frequencies at the first visit and post treatment
SD = standard deviation; CI = confidence interval.
Table 2 shows the severity of hearing loss based on American Speech and Hearing Association criteria at the first visit and the follow up. According to these, at follow up, 16 patients (36.4 per cent) experienced at least one grade improvement to a less severe hearing loss status, 24 cases (54.5 per cent) remained at the same level of hearing loss and 4 cases (9.1 per cent) progressed to a more severe category of hearing loss. Overall, 21 cases (47 per cent) had normal hearing or very mild hearing loss.
The severity of hearing loss based on American Speech and Hearing Association criteria at the first visit and post treatment
Based on the Siegel criteria, 10 patients (22.7 per cent) showed complete recovery, 4 patients (9 per cent) had partial recovery and 2 patients (4.5 per cent) had slight recovery. Twenty-eight cases (63.6 per cent) had a hearing improvement of less than 15 dB. Patients with nystagmus, vertigo, abnormal caloric or abnormal cervical vestibular-evoked myogenic potential tests at the first visit had significantly higher PTA at the follow up (Table 3). The chi-square test showed hearing improvement based on Siegel criteria was significantly different between patients with and without vertigo and nystagmus, and between patients with normal and abnormal caloric and cervical vestibular-evoked myogenic potential tests at the first visit and the follow up (Table 4).
The mean of primary and secondary pure tone audiometry between patients with and without nystagmus and vertigo, and patients with normal and abnormal caloric and cervical vestibular-evoked myogenic potential tests
a PTA = pure tone audiometry; T0b = at the first visit; T1c = post-treatment; Abnld = abnormal; Nle = normal.
Vestibular test results and their relation to hearing improvement based on Siegel criteria
The chi-square test showed patients with abnormal cervical vestibular-evoked myogenic potential tests, caloric tests and vertigo were more frequently seen in the group with no PTA improvement based on American Speech and Hearing Association criteria than in those with at least 1 grade improvement (n = 14 out of 28 (50 per cent) vs 0 out of 16 (0 per cent), p = 0.001 for abnormal cervical vestibular-evoked myogenic potential; n = 16 out of 28 (57.14 per cent) vs 0 out of 16 (0 per cent), p < 0.001 for the caloric test and vertigo.
Discussion
In this study, we assessed 44 sudden sensorineural hearing loss patients with vestibular dysfunction regarding hearing improvement three months after steroid therapy, considering their demographic data, background and the presence or absence of vertigo and nystagmus.
The most frequent co-morbid disorder among our study population was hypertension (34 per cent), followed by diabetes mellitus (25 per cent). This result aligns with other studies reporting these as the two most common major co-morbid disorders.Reference Lin, Lee, Yu and Lin32, Reference Xie, Dai, Liu, Liu, Hellström and Duan33 Although this was not an epidemiological study, we noted that the prevalence of hypertension in our patients was higher than the average prevalence of hypertension in Iranian adults (26.26 per cent)Reference Mohammadi, Hassanipour, Delam, Nikbakht, Far and Firoozi34 and still higher when considering their age group (60.9 vs 50 per cent for those aged 55 years and over).Reference Mirzaei, Moayedallaie, Jabbari and Mohammadi35 Similarly, the prevalence of diabetes mellitus was higher in our study population (25 vs 10 per cent in Iranian adults and 42.3 per cent in our patients over 45 years vs 34.28 per cent for the same age group of Iranians).Reference Moradpour, Rezaei, Piroozi, Moradi, Moradi and Piri36
In a study of 739 patients with sudden sensorineural hearing loss, Xie et al.Reference Xie, Dai, Liu, Liu, Hellström and Duan33 concluded that cardiovascular and metabolic disorders are closely correlated with the occurrence and severity of sudden sensorineural hearing loss. Theoretically, these co-morbidities can induce microvascular damage, disrupting the inner-ear supply and leading to sudden sensorineural hearing loss. In a study by Lin et al.Reference Lin, Lee, Yu and Lin32 on the impact of different co-morbid disorders on the prognosis of sudden sensorineural hearing loss, only diabetes mellitus and hypercholesterolaemia were associated with a lower probability of hearing improvement. The present study found no correlation between co-morbid disorders and hearing or vestibular improvement due to the limited sample size and confounding factors.
In the present study, the quantitative amount of PTA improvement was higher in patients receiving immediate therapy (within 7 days) than in those who initiated treatment after 7 days. However, the difference was not statistically significant. This could be attributed to the fact that all patients started therapy within the first two weeks. In a recent study of 666 patients with sudden sensorineural hearing loss, whose treatment regimen was similar to our study (either oral treatment (57.5 per cent) or PO + IT (42.5 per cent) if no improvement was detected after 7 days of oral corticosteroid), the exact time frame of therapy after which hearing gain significantly decreased was found to be 2 weeks.Reference Chen, Eligal, Menahem, Salem, Sichel and Perez37
The recovery rate was 31.7 per cent, considering the sum of complete and partial recovery based on Siegel criteria. At the end of the study, 47 per cent of patients had normal hearing or very mild hearing loss based on American Speech and Hearing Association categorisation. In a study by Lee et al. on 453 SSNHL cases the sum of complete and partial recovery was 44.8% after a one-month follow up.Reference Lee, Jeon, Baek, Kim, Rah and Choi38 In another study, Edizer et al. examined 453 sudden sensorineural hearing loss patients with a mean age of 50.3 years. The sum of complete and partial recovery based on Siegel criteria was 44.8 per cent after a 1-month follow up.Reference Edizer, Çelebi, Hamit, Baki and Yiğit39 In 205 sudden sensorineural hearing loss patients with an average age of 45.9 years, the sum of complete and partial recovery based on Siegel criteria was 39 per cent 6 months after the end of treatment.
In a study by Kang et al.Reference Kang, Yang, Shim, Song, Kim and Lim40 on 494 sudden sensorineural hearing loss patients with an average age of 50.9 years, the sum of complete and partial recovery based on Siegel criteria was 46.5 per cent 2 months after the initiation of treatment. These differences could be due to the variations in study populations, including age, ethnicity, severity of the primary hearing loss, co-morbid disorders, follow-up durations and, most importantly, different treatment regimens. Because our patients were not randomly categorised for the type of treatment they received, it is impossible to interpret the comparison of hearing improvement between them. Based on this study and the literature, we can conclude that more than a third of patients experience recovery.
The rates of caloric test and cervical vestibular-evoked myogenic potential test abnormalities were 36.4 and 31.8 per cent at the first visit, respectively. Additionally, 36.4 per cent of patients reported vertigo as an accompanying symptom with hearing loss. The presence of these test abnormalities and vertigo was associated with a poorer prognosis. The mean PTA was significantly higher in patients with these vestibular abnormalities. However, because 13 out of 14 cases with an abnormal caloric test also had an accompanying abnormal cervical vestibular-evoked myogenic potential test, we could not determine which test abnormality could precisely affect the prognosis of sudden sensorineural hearing loss. This indicates profound damage to the semicircular canal and superior vestibular nerve, as well as to the saccule and inferior vestibular nerve, in 29.5 per cent of our patients simultaneously.
A systematic review and meta-analysis by Yu and LiReference Yu and Li24 assessed vestibular dysfunction in sudden sensorineural hearing loss. Including a total of 18 studies, they found the prevalence of lateral semicircular canal and superior vestibular nerve pathologies, detected by the caloric test, to be up to 44 per cent, and the prevalence of saccule and inferior vestibular nerve pathology, detected by the cervical vestibular-evoked myogenic potential test, to be up to 41 per cent. An abnormal caloric test was revealed as a negative prognostic factor. It is supposed that the superior vestibular nerve is more vulnerable to ischaemic labyrinthine damage compared with the inferior vestibular nerve because of the longer and narrower anatomical feature of the bony canal.Reference Gianoli, Goebel, Mowry and Poomipannit41 The presence of vertigo has been shown to increase the risk for vestibular organ damage in sudden sensorineural hearing loss.Reference Yu and Li24, Reference Hepkarsi, Kaya and Kirazli42
In a study by Lim et al. involving 263 sudden sensorineural hearing loss patients, those with abnormal caloric tests (45.2 per cent) showed less improvement in PTA.Reference Lim, Jeong, Han, Rah, Cha and Choi6 The frequency of vertigo was 36.5 per cent, very close to the rate of vertigo in our patients, and cases with vertigo had worse initial PTA and less PTA improvement after steroid therapy.
In a study by Wang et al.Reference Wang, Wang, Jing, Yu and Ye43 on 156 sudden sensorineural hearing loss patients, an abnormal cervical vestibular-evoked myogenic potential test was associated with a lower probability of recovery. The authors hypothesised that because saccules are supported by collateral blood supply, their damage can indicate more severe cochlear damage and less recovery is expected. We had four cases of vertigo in the absence of abnormal vestibular assessments. This finding was also previously reported by Yu and LiReference Yu and Li24 This observation suggests that vertigo may not independently determine sudden sensorineural hearing loss, and such symptoms can arise from the genetic and anatomical closeness of the cochlear and vestibular structures.Reference Korres, Stamatiou, Gkoritsa, Riga and Xenelis44, Reference Ryu, Yoon, Ahn, Kang, Choi and Lee45
Regarding the study limitations, sample size was limited, which did not allow us to control for confounding factors when comparing different groups pre and post treatment. The ocular vestibular-evoked myogenic potential test, which examines utricle lesions, was not performed. Regarding these limitations, we suggest that longitudinal multicentre studies with large enough sample sizes be conducted to better understand the effect of vestibular dysfunction on the prognosis of Idiopathic Sudden Sensorineural Hearing loss patients.
• Idiopathic sudden sensorineural hearing loss is a common disease worldwide
• There is no consensus on its treatment, risk factors or prognostic factors
• Vestibular system involvement has been considered a poor prognostic factor for years, but recent studies have expressed doubts about this and the exact effect of vestibular involvement is unknown
• The prognostic effect of vestibular involvement on hearing improvement during treatment was found and the most important vestibular tests were specified for this purpose
Conclusion
The presence of vestibular disorders simultaneously with sudden sensorineural hearing loss is associated with a lower rate of hearing improvement. Vestibular function assessment by caloric and cervical vestibular-evoked myogenic potential tests, as well as patient self-reporting of vertigo, may find prognostic value with further investigations.
Competing interests
None declared
