A Prospective Study of Audiological Manifestations in Patients of Allergic Rhinitis

Abstract

Allergic rhinitis is a type-I hypersensitivity reaction of the nasal mucosa, primarily mediated by immunoglobulin E (IgE) with complex etiological factors.Allergic rhinitis may involve the inner ear. The scientific basis for this is poorly understood. However, the inner ear has been found to demonstrate both cellular and humoral immunity, and the seat of immuno-activity appears to reside in the endolymphatic sac and duct. To assess the audiological profile of patients with allergic rhinitis. 100 Study group patients and 50 control group subjects underwent detailed audiological assessment. Present study revealed high frequency sensorineural hearing loss with prolongation of Wave I and shortened wave I–III and Wave I–V interpeak latencies on ABR and abnormal DPOAE findings, compared with controls which indicate inner ear involvement (cochlear pathology). Individuals with allergic rhinitis are more prone to hearing abnormalities which can be detected even before any symptoms of hearing impairment are present. However, the exact pathophysiology of inner ear damage in patients of airway allergy is poorly understood and therefore, additional studies in this area are required with a larger sample population to assess the benefits of hearing assessment in patients of allergic rhinitis for early detection of hearing loss.

Introduction

Allergic rhinitis is a global health problem. Worldwide prevalence of allergic rhinitis ranges from 0.8 to 39.7% with UK having a high prevalence (> 20%) [1]. The Allergic Rhinitis and its Impact on Asthma (ARIA) 2008 updated document estimates that there are 500 million subjects in this world who suffer with allergic rhinitis (AR) [2].

Allergic rhinitis (AR) is a type-I hypersensitivity reaction of the nasal mucosa, primarily mediated by immunoglobulin E (IgE) with complex etiological factors. The clinical symptoms of allergic rhinitis are nasal obstruction, watery rhinorrhea, sneezing and itching at the nose, palate and nasopharyngeal region [3]. The strong genetic component to the allergic response, is driven through mucosal infiltration and action on plasma cells, mast cells, and eosinophils. The head and neck are the most commonly affected target organs of the allergic reaction [4].

It is believed that allergy may affect the outer, the middle or the inner ear [5]. Various authors have studied the relationship between otitis media and allergic rhinitis. Most of them agree that allergies probably contribute to the development of otitis media with effusion, with other major risk factors being bacterial infection and Eustachian tube obstruction [6]. Epidemiologic studies consistently identify allergy as a risk factor for otitis media. It is also suggested that the observed relationship between allergy and otitis media with effusion is caused by mediators of inflammation and cytokines and colony-stimulating factors released by mucosal mast cells and other inflammatory and epithelial cells in the nose and nasopharynx. These mediators produce blockage of the Eustachian tube through a number of mechanisms. Both viral upper respiratory infection and nasal allergic reaction provoke nasal inflammation, which in turn causes Eustachian tube Dysfunction leading to decreased ventilation. This causes retraction of tympanic membrane and conductive hearing loss [7].

Allergic rhinitis may involve the inner ear but the scientific basis for this is poorly understood. However, the inner ear has been found to demonstrate both cellular and humoral immunity, and the seat of immuno-activity appears to reside in the endolymphatic sac and duct [8].

In view of scanty literature on inner ear manifestations of allergic rhinitis, the present study was conducted with the aim to study the correlation of allergic rhinitis with inner ear functions.

Materials and Methods

Methodology

The study was conducted in the Department of Otorhinolaryngology and Head and Neck Surgery, SMGS Hospital, Government Medical College, Jammu w.e.f November 2018–October 2019. The study was conducted after taking approval of institutional Ethical committee. The patients who attended ENT outpatient department with symptoms of allergic rhinitis were enrolled in the study.

The study group consisted of 100 patientswith clinical diagnosis of allergic rhinitis on the basis of signs and symptoms with the age group of 10–55 years. The control group consisted of 50 subjects of similar age group as the study group who were relatives and friends accompanying the study group patients; these control subjects had thus been exposed to a similar environment but were not suffering from allergic rhinitis or any systemic disease.

Exclusion criteria included patients of age less than 10 years or more than 55 years, history ofacoustic trauma, or any intake of ototoxic medications, any systemic and metabolic disorders causing hearing loss, any perforation in tympanic membrane or other middle ear pathology or any history of previous ear surgery.

All study and control group subjects underwent a detailed history and local and systemic examination was done. Audiological assessment was done in the speech and hearing unit attached to the Department. Audiologicaltesting, included pure tone audiometry with extended high frequencies (0.250–12 kHz), tympanometry, otoacoustic emission (OAE) and auditory brainstem response (ABR) testing.

Procedure

Audiological assessment was conducted in a sound-treated room which conformed to American National Standards Institute (ANSI) (1977) and International Organization for Standardization (ISO) standards for maximum permissible noise level [4].

Pure Tone Hearing thresholds were tested using Elkon 3N3 Multi-Diagnostic audiometer at speech frequencies between 0.25 and 12 kHz with TDH39P headphones.

An AT 235 By Interacousticstympanometer was used for tympanometry and acoustic reflex testing. A 226 Hz probe was used for tympanometry, with pressure varying from + 200 to − 300 daPa.

Otoacoustic emission and BERA testing were carried out using NeurosoftNeuro-audio device. Transient evoked OAE (TEOAE) testing was performed with a wide band click in continuous mode and with an intensity of 90 dB SPL. When measuring the Distortion product (DP) gram, the frequency separation of the primaries was f2/f1 = 1.22, with L1 and L2 set to 65 and 55 dB SPL, respectively. The parameter considered in TEOAE testing was a signal-to-noise ratio of more than 3 dB in at least three consecutive test frequencies (of 1, 1.5, 2, 3 and 4 kHz).

The parameters considered in distortion product OAE testing were:(1) a signal-to-noise ratio of more than 3 dB in three consecutive test frequencies, and (2) the amplitude of the signal in the 90th percentile of the normal distribution for the frequencies that were tested (i.e. 357, 499, 704, 1003, 1409, 2000, 2822, 3991and 5649 Hz).

Auditory brainstem responses were tested using the evoked potential system using NeurosoftNeuro-audio device. Silver-silver chloride button electrodes were used.

The parameters selected for recording included: (1) the filter bandwidth was adjusted to 100–3000 Hz; (2) the stimulus was clicks; (3) the stimulus rate was 19.3/second and its duration was 100 micro second/click; (4) a minimum of 2000 clicks was presented at each recording, (responses were repeated at each intensity level to ensure reproducibility); (5) wave forms were recorded at a sound intensity of 70–90 dBnHL, in both ears separately.

The site of electrode placement was cleaned thoroughly with a spirit swab to reduce the skin–electrode impedance to less than 5 kΩ. The non-inverting electrode was placed at the vertex, the inverting electrode on either mastoid, and the ground electrode on the forehead, using conduction gel. The surface impedance was adjusted below 5 kΩ to facilitate optimal recording.The following parameters were studied: the absolute latencies of waves I, III and V, and the Interpeak latencies of waves I–III, III–V and I–V.

Statistical Analysis

Data was entered into Microsoft Excel Sheet. Data was grouped in tables and were represented as means with standard deviation and percentages for qualitative and quantitative variables respectively. Statistical significance of the data was tested by IBM SPSS VERSION 21.

Results

A total of 100 patients of Allergic Rhinitis and 50 Control Subjects were included in the study. The mean age of the subjects in the study group was 32.91 ± 12.40 years and in the control group was 32.12 ± 13.12 years.In the study group, M:F ratio was 1:1.3 and in the control group, M:F ratio was 1.4:1.The presenting complaint of maximum number of cases was sneezing (37%) followed by watery rhinorrhea (17%) and nasal obstruction (16%)0.28% of the patients had associated complaints which included allergic conjunctivitis (6%), eczema (6%), asthma (5%) and ear itching (5%) andpruritis(4%)0.92% patients had raised IgE levels (> 150uIU/ml). On the contrary, 12% of controls had raised IgE levels. Computed Tomography findings includedbilateral inferior turbinate hypertrophy (17%), concha bullosa (15%),pansinusitis (15%), maxillary sinusitis with osteomeatal complex blockade (12%),sphenoiditis (12%) had deviated nasal septum (DNS) (12%).

PTA Findings

The mean value of hearing thresholds of various frequencies in study and control groups as depicted in the Table 1 showed high frequency sensorineural hearing loss in allergic rhinitis patients. The difference in the values of study and control groups was statistically nonsignificant. in the frequencies of 250, 500, 1 k, 2 k Hz whereas, the difference was statistically significant at higher frequencies of 4 k, 6 k and 8 k Hz. (p = 0.0001).

Table 1.

Mean Hearing Threshold In Study (n=200 ears) and Control Group (n=200 ears)

Frequency	Study group (dB)	Control group (dB)	P-value
250 Hz	11.24 $\pm$ 3.45	10.86 $\pm$ 2.98	0.07 (NS)
500 Hz	13.29 $\pm$ 4.56	12.24 $\pm$ 4.12	0.12 (NS)
1 K Hz	14.76 $\pm$ 3.98	14.09 $\pm$ 3.75	0.45 (NS)
2 K Hz	15.86 $\pm$ 2.78	14.91 $\pm$ 3.20	0.017 (NS)
4 K Hz	26.77 $\pm$ 7.43	18.59 $\pm$ 4.52	0.0001 (S)
6 K Hz	30.97 $\pm$ 9.56	20.98 $\pm$ 7.54	0.0001 (S)
8 K Hz	37.64 $\pm$ 7.41	23.71 $\pm$ 8.41	0.0001 (S)

Tympanometry Findings

In the study group, 114 (72%) ears had Type A tympanogram and 46 (23%) ears had Type B tympanogram.In the control group, 84 (84%) ears had Type A tympanogram and 10 (10%) had Type B tympanogram.In the study group, 144 (27%) ears had absent reflexes and in the control group 18 (18%) ears had absent stapedial reflexes on tympanometric studies. The difference was statistically significant (p = 0.001).

DPOAE Findings

The difference in the mean values of signal-to noise on DPOAE was statistically significant between study and control group at 1481, 2222, 2963, 5714 and 800 Hz. However, the difference was not statistically significant at 988 Hz. These results suggest outer hair cell dysfunction in patients with allergic rhinitis as depicted in Table 2.

Table 2.

Mean Values For DPOAE of SIGNAL-TO-NOISE Ratio Across Various Frequencies

Frequencies	Study group	Control group	P-value
988 Hz	8.1 $\pm$ 4.2	9.4 $\pm$ 5.8	0.07 (N.S)
1481 Hz	14.1 $\pm$ 3.1	12.3 $\pm$ 6.2	0.01 (S)
2222 Hz	1.4 $\pm$ 0.2	6.6 $\pm$ 2.4	0.001(S)
2963 Hz	-0.08 $\pm$ 4.3	18.2 $\pm$ 8.3	0.0001(S)
5714 Hz	2.4 $\pm$ 4.2	17.8 $\pm$ 7.4	0.0001(S)
8000 Hz	3.8 $\pm$ 2.3	21.2 $\pm$ 8.2	0.0001(S)

The significance of the bold values is that the difference between the study group and control group value is statistically significant

ABR Findings

In the study group, 115 ears (57.5%) had prolonged Wave I latency on BERA as compared to controls in whom only 06 ears (6%) had prolonged Wave I latency. The difference was statistically significant (p = 0.0001). Wave III and Wave V were prolonged in 03 (1.5%) and 05 ears (2.5%), respectively in the study group and Wave III was prolonged only in 01 ear (1%) in control group and Wave V was prolonged in none of the control group population. The difference was statistically significant for Wave V prolongation and not for Wave III prolongation.

Shortening of Wave I-III interpeak latencies was seen in 106 (53%) ears in the study group and 06 (6%) ears in the control group. Wave I-V shortening was seen in 112 (56%) ears and 06 (6%) ears in controls. Shortening of Wave III-V was seen only 3 (1.5%) ears in study group and 01 (1%) ear in control group. The difference between interpeak latencies in both study and control groups was statistically significant (p = 0.0001) for shortening of Wave I-III and Wave I-V and not for shortening of Wave III-V.

The mean value of Absolute Wave I Latency in the study group was 2.43 ± 0.21 and in control group was 1.68 ± 0.13. The difference between the two was statistically significant (p = 0.003). The absolute Wave III and Wave V latencies mean value was 3.61 ± 0.16 msecand 5.69 ± 0.23 msecrespectively and that of control group was 3.67 ± 0.11 msecand 5.61 ± 0.16 msecand the difference was statistically non-significant. The mean value of IPL of Wave I-III in the study group was 1.18 ± 0.11msec and that of control group was 1.99 ± 0.14 ms. The difference was statistically significant (p = 0.004).

The mean value of IPL of Wave III-V in the study group was 2.08 ± 0.18 and that in the control group was 1.94 ± 0.16 ms. The difference was statistically non-significant.

The mean value of IPL of Wave I-V in the study group was 3.26 ± 0.13 msecand that in the control group was 3.93 ± 0.21 ms. The difference was statistically significant (p = 0.004).

These findings indicate the cochlear involvement in patients with allergic rhinitis. These findings are depicted in Table 3.

Table 3.

Mean Values of BERA Parameters in Study (n=200 ears) and Control Group (n=200 ears)

Parameters	Study group (msec)	Control group (msec)	P-value
Wave I latency	2.43 $\pm$ 0.21	1.68 $\pm$ 0.13	0.003 (s)
Wave III latency	3.61 $\pm$ 0.16	3.67 $\pm$ 0.11	0.72 (NS)
Wave V latency	5.69 $\pm$ 0.23	5.61 $\pm$ 0.16	0.91 (NS)
Wave I–III IPL	1.18 $\pm$ 0.11	1.99 $\pm$ 0.14	0.004 (S)
Wave III–V IPL	2.08 $\pm$ 0.18	1.94 $\pm$ 0.16	0.67 (NS)
Wave I–V IPL	3.26 $\pm$ 0.13	3.93 $\pm$ 0.21	0.004 (S)

The significance of the bold values is that the difference between the study group and control group value is statistically significant

Discussion

In the present study, 100 patients of allergic rhinitis and 50 controls attending outpatient department of Otorhinolaryngology Jammu were evaluated for their inner ear functions.

On Pure tone audiometry, the mean pure tone thresholds were poorer at frequencies of 4 k, 6 k and 8 k Hz in the study group as compared to the control group. The mean value ± S.D of hearing thresholds of study group patients were 26.77 ± 7.43 dB at 4 k Hz, 30.97 ± 9.56 dB at 6 k Hz and 37.64 ± 7.41 dB at 8KHz frequency respectively. The difference between study and control group patients was statistically significant at these frequencies. So, it can be interpreted that patients with allergic rhinitis had sensorineural hearing loss at these frequencies as compared to controls.

Similar observations were made by Singh S et al. [4] who reported that allergic rhinitis patients had sensorineural hearing loss that was worse in the high frequency region. Similarly, ina study by Sekhon GS et al. [9] it wasconcluded that in the pure tone audiometry, there was statistically significant hearing loss in mean thresholds between study and control groups, for both ears, at 4000 Hz and 8000 Hz. On the contrary, Nursoy MA et al. [10] showed no statistically significant difference in the hearing thresholds of study and control groups between frequencies ranging from 250 to 16,000 Hz. Kumar S et al. [3] concluded that patients of perennial allergic rhinitis were having more hearing impairment (28.41%) than seasonal allergic rhinitis (15.4%), which was mostly conductive type. These findings were also contradictory to the present study.

On tympanometry in the present study, majority of cases i.e., 114 (72%) ears had Type ‘A’ Tympanogram and majority of control groups i.e., 84 (84%) ears also had Type ‘A’ Tympanogram. Type ‘B’ Tympanogram was seen in 46 (23%) ears in the study group and 10 (10%) ears in the control group population. Acoustic reflex was absent in 144 (72%) ears in study group and 18 (18%) ears in control group.Dwarakanath VM et al. [11] in their study concluded that Tympanometry showed “A” type with compliance ranging between 0.3 cc and 1.6 cc, peak pressure between − 100 daPa and + 60 daPa and ear canal volume between 0.9 cc and 2.0 cc indicating normal middle ear functions bilaterally among all the participants which was similar to the present study.However, other studies of Pelikan Z et al. [12] and Anazy FHA et al. [13] showed significant changes in the middle ear.

The difference of Mean signal-to-noise ratio on DPOAE was statistically significant between study and control group at 1481, 2222, 2963, 5714 and 8000 Hz. However, the difference was not statistically significant at 988 Hz. These results indicate reduced outer hair cell function which can be attributed to changes in outer hair cell function due to allergens. The presence of an allergen will trigger the endolymphatic sac in the inner ear that can process antigens and produce its own local antibody response, the resulting inflammatory mediators and toxic products may interfere with hair cell function and thus affect hearing.

A study by Sekhon GS et al. [9] demonstrated abnormal DPOAEs in the study group patients with statistically significant difference in the signal to noise ratios in study and control groups at most of the frequencies. Singh S et al. [4] found abnormal transient evoked otoacoustic emissions in all and abnormal DPOAE in 27 out of 30 (90%) patients in their study, thus indicating involvement of outer hair cell dysfunction of cochlea similar to present study. Dwarakanath VM et al. [11] also showed similar results. However, Nursoy MA [10] found no statistically significant difference between the study group and the control group in terms of their signal noise ratios in all frequencies of DPOAE which is contrary to the present study.

The mean value of absolute Wave I latency in the study group was 2.43 ± 0.21 ms and in the control group was 1.68 ± 0.13 ms. The difference was statistically significant depicting the prolongation of Wave I latency. The mean value of interpeak latency of Wave I-III in the study group was 1.18 ± 0.11 ms and that of control group was 1.99 ± 0.14 ms. The difference was statistically significant showing shortening of Wave I-III IPL. The mean value of IPL of Wave I-V in the study group was 3.26 ± 0.13 ms and that in control group was 3.93 ± 0.21 ms. The difference was statistically significant showing shortening of Wave I-V IPL. Singh S et al. [4] found that there was statistically significant prolongation of Wave I latency in the study group, compared to the control group (p < 0.05). Also, there was statistically shortening of the Wave I-III absolute IPL (P < 0.05) and shortening of the wave I-V absolute IPL (P < 0.05) in the study group, compared with the control group. Findings were similar to our study. In contrast to our study, Sekhon GS et al. [9] found that in the study group, the mean inter peak latency I-III at 70–90 dB nHL varied from 2.18 ± 0.54 ms to 2.37 ± 0.60 ms. In control group, the mean inter peak latency at 70–90 dB nHL varied from 2.04 ± 0.54 ms to 2.31 ± 0.33 ms. The difference between two groups was not statistically significant. Also, in the study group, the mean inter peak I-V latency at 70–90 dB nHL varied from 1.33 ± 1.33 ms to 12.06 ± 55.417 ms. Comparatively in the control group the mean inter peak latency (wave I–V) varied from 1.33 ± 1.178 ms to 25.35 ± 109.4 ms. The difference between the study and control group was not statistically significant. The absolute latencies of wave I, III and V also showed no statistically significant difference between study and control groups in this study.

Present study revealed high frequency sensorineural hearing loss with prolongation of Wave I and shortened wave I–III and Wave I–V on ABR and abnormal DPOAE findings which indicate inner ear involvement (cochlear pathology).In literature, various mechanisms of inner ear damage due to allergic rhinitis have been postulated: The endolymphatic sac acts as a target organ during allergic reactions, resulting in inner ear changes seen in allergic rhinitis [5]. The endolymphatic sac has been shown to be capable of both processing antigen and producing its own local antibody response. It has a highly vascular subepithelial space containing numerous fenestrated blood vessels. Most immunologically competent cell types are found in the interosseous portion of the endolymphatic sac, because of its unique blood supply [14]. The sac’s peripheral and fenestrated blood vessels may allow entry of antigens, which could then stimulate mast cell degranulation in the perisaccular connective tissue [15]. The resulting inflammatory mediators and accumulation of toxic metabolic products may interfere with hair cell function. Production of circulating immune complexes (e.g. involving food antigens) which are deposited in the vascular basement membrane produce inflammation. Antigen–antibody complexes localized in and around blood vessel walls induce an inflammatory reaction mediated by complement activation resulting in the release of chemotactic factors that promote the migration of polymorphs and macrophages into the region. An increased serum concentration of circulating immune complexes has been described in both Ménière’s disease and allergic rhinitis [14]. The interaction between viral antigens and allergy mechanisms, and the deposition of circulating immune complexes in the stria, may both cause leakage of the blood-labyrinth barrier as a result of increased vascular permeability and disruption of ionic and fluid balance in the extracapillary spaces. This could facilitate the entry of autoantibodies into the inner ear [4].

In the present study, in addition, it has been postulated that middle ear mucosa (respiratory epithelium) reacts to the allergic response and histamine, vasoactive mediators, immune complexes may transfer from round window/oval window to inner ear and induce inner ear damage which may probably explain high frequency sensorineural in allergic rhinitis. This theory may be similar to the hypothesis which explains acute otitis media induced high frequency SNHL caused due to increased round window permeability because of inflammation, allowing passage of endotoxins and ionic disequilibrium within the cochlea which are detected by hair cells in the cochlear basal turn adjacent to the round window membrane [16].

Conclusion

It is concluded from the present study that poor pure-tone thresholds more at higher frequencies, abnormal DPOAE, abnormalities on ABR and the absence of middle ear pathology (“A” typetympanogram) clearly indicate the presence of cochlear involvement which can be attributed to the allergens which might have spread to the cochlea. These abnormalities can be detected even before any symptoms of hearing impairment are present. However, the exact pathophysiology of inner ear damage in patients with airway allergy is poorly understood and therefore, additional studies in this area are required with a larger sample population to assess the benefits of hearing evaluation in allergic rhinitis patients for early detection of hearing loss.