Otitis media is defined as the presence of an inflammatory process in middle ear topography and comprises a spectrum of related diseases covering lesions in several phases: acute, subacute, and chronic.1 According to studies developed by Paparella et al.,1–3 otitis media may be part of a continuum, in which the maintenance of the inflammatory process can lead to tissue changes and structural lesions progressively.

It is estimated that its incidence is 10.85% or 709 million cases per year worldwide for Acute Otitis Media (OMA) and 31 million new cases per year for Chronic Otitis Media (COM), and 22.6% of these cases occur in children under 5 years of age.4,5 The overall incidence of suppurative COM is estimated at 4.8 new cases per 1000 people/year, with a prevalence of around 65–300 million individuals.4,6

Due to the high incidence and prevalence rates, the non-lethal complications of otitis media (hearing loss, tinnitus, vestibular symptoms) should receive special attention, as they cause great morbidity to patients. About 60% of people with supple COM attend hearing loss worldwide, which corresponds to 39–200 million individuals.5

In the various presentations of otitis media, the most common type of hearing loss is conductive, either due to the presence of effusion, tympanic membrane perforation or ossicular changes. However, in recent decades, sensorineural hearing loss secondary to the inflammatory processes of the middle ear has been demonstrated.7–9 Inflammatory products and toxins from the middle ear cross the membrane of the round window, mainly affecting the basal turn of the cochlea,10 which can lead to inflammatory injury of its sensory and neural elements. The most severe location of these cochlear changes (basal turn) corresponds tonotopically to the most affected frequencies (acute frequencies).7,10,11

In general, all otitis media present with some degree of hearing loss, temporary or permanent. Despite such a common complaint, there is no study comparing the auditory thresholds in the various clinical presentations of otitis media, which is the main objective of our research.

The study was of the individual cross sectioned study (level 2). We selected patients from the otology outpatient clinic and the otorhinolaryngology emergency department of our institution. We divided the patients into non-suppurative CMO, suppurative CMO, cholesteatomatous CMO (suppurative and non-suppurative forms), Otitis Media with Effusion (OME) and Acute Otitis Media (AOM). Ears with otitis media were compared with ears of healthy individuals from the otological point of view. All patients and individuals in the control group were selected after applying the inclusion and exclusion criteria.

We included patients with 1) Diagnosis of otitis media, made on an outpatient basis or in the emergency room by an ENT physician; 2) Patients over 10 years of age; 3) Patients with AOM of up to 5 days of evolution.

We excluded patients who had: 1) History of otologic surgery or placement of ventilation tube; 2) Cognitive alterations; 3) Syndromic diseases or with external, middle or internal ear malformations; 4) Chronic exposure to noise; 5) History of primary cancers or metastases to the temporal bone; 6) History of systemic chemotherapy or radiotherapy in the head and neck region; 7) Autoimmune diseases; 8) History of exposure to aminoglycosides or other potentially ototoxic medications; 9) Cases of diabetes mellitus or other systemic diseases that may compromise the auditory thresholds; 10) History of head trauma or traumatic perforations of the Tympanic Membrane (TM).

We selected the individuals from the control group in a non-probabilistic manner, being a convenience sample. We applied the same exclusion criteria for this group and included only individuals with air and bone <25 dBNA and absence of bone air GAP. Only individuals and patients who signed and agreed with the informed consent form participated in the study.

Computed tomography of temporal bones was performed in patients with chronic otitis media to classify the type of clinical presentation. Some cases of OME and non-suppurative CMO were submitted to computed tomography for better diagnostic elucidation.

We performed conventional tonal audiometry (500–8000 Hz), vocal audiometry with IPRF and SRT evaluation and imitanciometry (when the tympanic membrane was intact) in patients with otitis media and in the control group. We considered hearing loss when the auditory thresholds were > 25 dBNA. Audiometric examinations were performed by an experienced audiologist.

Statistical analysis was performed with the statistical program R version 3.5.1. for Windows. To compare the groups of otitis media with each other, as well as to compare the various groups of otitis media with the control group, the Kruskal-Wallis test and the Dunn-Bonferroni post hoc test were used. We used the same statistical tests to analyze the statistical significance of the presence of tinnitus between the groups. Spearman's correlation was used to correlate numerical variables. Values of p < 0.05 determined the statistical significance.

The study was approved by the Research Ethics Committee (number 0364/2017) and consent was obtained from all the participants that agreed to join in the research.

Hearing loss secondary to inflammatory processes of the middle ear is a topic widely addressed in the literature, with descriptions of sensorineural hearing loss dating from 1941.12 However, it was only from the 1970s that it began to prove that sensorioneural loss after otitis media may be the result of a cochlear insult.13–16 Although there are several articles in the literature addressing sensorineural losses in otitis media, there is still no study comparing this type of hearing loss in its various clinical presentations.

In our sample of 151 ears, we observed that conductive hearing loss corresponded to 58.28% of our sample and was more frequent in the non-suppurative And OME groups. As they are ears with lower inflammatory activity, there is a lower chance of intracochlear injury compared to the ears of the other groups. Mixed hearing loss occurred in 40.4% of all ears, and were more frequent in the groups Suppurative CMO, cholesteatomatous CMO (suppurative and non-suppurative) and AMO — more active groups from the inflammatory and infectious point of view. These findings are parallel in the literature: several studies have demonstrated, in temporal bones, that the degree of injury in the sensory structures of the cochlea has an intimate relationship to the degree and severity of inflammatory pathological changes in the middle ear.10,13,14,17

In addition, two ears with AMO had pure sensorineural hearing loss at higher frequencies. This result goes according to Song et al.18 and Margolis and Nelson,19 who describe cases of pure sensorineural hearing loss after AMO. It is possible that the pathophysiological mechanism for these cases is the passage of harmful agents from the middle ear to the basal gyration of the cochlea through the membrane of the round window, considering the tonotopic relationship of this region with the frequencies reached. As reported in the report by Margolis and Nelson,19 at the beginning of the picture there was no effusion and, therefore, there was no presence of air-bone GAP.

Regarding symptomatology, tinnitus was the second largest complaint reported by patients. Tinnitus is more prevalent in the ears with a higher degree of infectious/inflammatory activity. This indicates a probable lesion to the cochlear neuroepithelium, as discussed by Tailor et al.,20 since tinnitus is more intense in groups with more severe inflammatory conditions. In addition, we found worse bone thresholds at 500–4000 Hz in patients with tinnitus when compared to patients without tinnitus. We also observed significantly worse air thresholds at frequencies of 4000 Hz, 6000 Hz and 8000 Hz in tinnitus patients. This leads us to believe that the worst air thresholds in acute frequencies in patients with tinnitus may reflect more intense neural injury in the basal spin of the cochlea.21

In relation to vocal audiometry, we found worse RTS values in the groups with higher inflammatory/infectious activity. Only 10 ears of 151 (6.62%) had IPRF < 88%, and we could observe worse values of IPRF in the suppurative and cholesteatomatous CMO (non-suppurative form). The fact that RTs has greatly altered values between groups of otitis media compared to the low frequency of changes in the IPRF suggests that the cochlear lesion secondary to otitis media occurs primarily in sensory epithelium (hair cells) and, only in more severe and late cases, affect neural structures.22 Speech recognition changes are secondary to the degeneration of internal hair cells and neural cells, which occurs later.23 This situation is explained by the degeneration of external hair cells and vascular stria in comparison to the lesion in internal hair cells in the various forms of otitis media.17 The authors also show that the degree of injury to internal hair cells was closely correlated with the severity and chronicity of the inflammatory process in the middle ear, similarly to our clinical findings.

Moreover, it was possible to observe that the presence of tinnitus is related to significantly higher LRF values compared to patients without tinnitus, as shown in Fig. 5. This finding may be the result of a cochlear lesion which, in turn, may be secondary to a deficit of the cochlear ciliary epithelium.24 Another possible mechanism for the onset of tinnitus would be that proposed by Liberman et al.,25 who proved that otitis media and chronic hearing deprivation may occur with cochlear sinaptopathy and impairment of aferens and aferental innervation of hair cells.

Regarding tonal audiometry, the groups of otitis media presented worse air and bone thresholds in all frequencies, when compared with the ears of the control group (Fig. 7). In addition, it is possible to observe a descending characteristic of the thresholds, a fact also observed by other authors.18,26–28 As the acute frequencies are the most affected, we can infer that the basal turn of the cochlea is the most affected by the infectious-inflammatory processes that occur in the middle ear.10,14

We observed worse bone thresholds in the most active ears from the infectious and inflammatory point of view, such as ears with suppurative CMO, cholesteatomatous CMO and OMA. This finding corroborates that of English et al.,9 who found worse bone thresholds in patients with CMO with discontinuity of the ossicular chain and with mastoid extension (more active CMOs), when compared with the non-suppurative form of the CMO. We did not observe a significant difference in air and bone thresholds in patients with suppurative CMO and cholesteatomatous CMO. That is, in our study cholesteatoma was not related to worse auditory results, as reported by other authors.26,27,29 In addition, there was no difference between the types of cholesteatoma (suppurative and non-suppurative) with regard to auditory thresholds. These findings suggest that auditory sequelae seem to be related to the local inflammatory process itself and bone destruction.

A curious observation of our study is that the ears with AMO had worse bone thresholds at 3000 and 4000 Hz. This is explained by the inflammatory lesion of the basal turn of the cochlea secondary to acute middle ear infection. It is proven that, in the acute situations, the membrane of the round window presents an increase in its permeability, with a subsequent decrease of its semipermeable nature in the chronicity of the picture.7,30,31 In addition, previous studies have shown that in the initial phase of AMO temporary functional changes may occur in the cochlea, called “Temporary Threshold Shifts” (TTS), which may recover after resolution of the acute infectious process.32,33 According to our methodology, we included patients with AMO of up to 5 days of evolution, which may have led to these observed results.

In addition, it was possible to observe worse bone thresholds with advancing age in chronic cases of otitis media, in agreement with several authors.29,34 Probably chronic hearing deprivation,25 acute episodes of AMO and exacerbations of CMO over the years lead to cochlear insults, which can add to age-related auditory degeneration. The latter factor influences bone audiometric thresholds, because we observed that older individuals with normal ears have higher bone-based auditory thresholds. However, it is noted that bone auditory thresholds are within normal limits, as they do not exceed 25 dBNA.

We did not observe a correlation between disease duration and worse bone thresholds in the ears with CMO and OME (p = 0.1362), in agreement with some authors.11,27 However, the relationship of duration of the disease and increase of bone thresholds has been previously reported.28,34 Perhaps our results did not point to such correlation due to a small sample of patients in each group. Another hypothesis for the lack of correlation would be the probable thickening of the round window membrane, acting as a protective factor for the cochlea against inflammatory aggressions of the middle ear.7

Our study has some limitations. Although we have a group of 116 patients, when we divide in the groups of the different clinical presentations of otitis media, this number becomes reduced. Moreover, because it is a cross-sectional study, we could not follow the auditory evolution with the treatment (clinical or surgical), as well as we could not define whether the hearing loss is temporary or permanent.

Despite the limitations, this is the first study to compare the auditory thresholds in the different clinical presentations of otitis media. We adopted many exclusion criteria, being very careful in the selection of our patients. We chose to include children over 10 years of age, who are more collaborative, contributing to more reliable responses during the audiometry exam. Moreover, at this age, the auditory system already has greater development.36 We excluded patients who underwent previous otological surgery, due to the increased risk of sensorineural hearing loss after such surgeries.36,37 As our exclusion criteria were very strict, there was a reduction in patients who could participate in the study. In addition, all our exams were performed by an experienced audiologist, ensuring a reliable result.

Our results confirm the auditory impairment (airway and bone) in all types of otitis when comparing them with the control ears. Bone hearing loss is more pronounced in patients with a higher degree of suppuration and in acute conditions, especially at higher frequencies.

We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed.

Ana Luiza Pappi Kasemodel de Araújo, Francisco Polanski Cordeiro and Rafael da Costa Monsanto received funding for the Improvement of Higher Education Personnel (CAPES). Norma de Oliveira Penido received funding from the São Paulo reserarch Foundation – Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), also from National Council for the Scientific and Technological Development (CNPq).

35.