Acoustic neuromas associated with sudden sensorineural hearing loss

Abstract

Background

Several studies have reported sudden sensorineural hearing loss (SHL) as an initial symptom of acoustic neuromas (ANs).

Aims/Objectives

We aimed to retrospectively evaluate the incidence of SHL and the associated rates of post-treatment improvement in patients diagnosed with ANs.

Materials and Methods

We analysed the clinical data of 686 patients with ANs and examined tumour size, correlation with hearing loss, number of episodes of SHL, extent of auditory recovery, and associated audiogram patterns.

Results

Among 686 enrolled patients with ANs, 232 (86 with intracanalicular and 146 with extrameatal tumours) experienced SHL at some point in their clinical history. The incidence of SHL was not significantly associated with tumour size. Of the 172 patients analysed, 119, 44, and 9 patients experienced one, two, or three or more episodes of SHL, respectively. Confirmed auditory recovery occurred in 61%, 45.3%, and 33.3% of the patients after the first, second, or third or later episodes of SHL, respectively. The audiograms of 78/172 (45.3%) patients showed a typical trough-shaped pattern of hearing loss at medium-level frequencies.

Conclusions and Significance

Among patients with ANs, 7.7% experienced two or more episodes of SHL. The recovery rate of SHL decreased with each successive occurrence.

Chinese Abstract

背景：若干项研究报告, 突发性感觉神经性听力损失 (SHL) 是听神经瘤 (ANs) 的初始症状。

目的：我们旨在回顾性评估被诊断为 ANs 的患者的SHL 的发生率和治疗后改善率。

材料与方法：我们分析了 686 例 ANs 患者的临床资料, 并检查了肿瘤大小、与听力损失的相关性、SHL 发作次数、听力恢复程度、和相关的听力图模式。

结果：在 686 名 ANs 患者中, 232 名（86 名内耳道肿瘤和 146 名外耳道肿瘤）在其临床病史的某个时间点经历过 SHL。 SHL发生率与肿瘤大小无显著相关性。在所分析的 172 名患者中, 119、44 和 9 名患者分别经历过一次、两次或三次或更多次 SHL 发作。确认听觉恢复在第一次、第二次或第三次或更晚发作后分别发生于 61%、45.3% 和 33.3% 的SHL患者。172名患者中有78名（45.3%）的听力图呈典型的槽形模式中等频率的听力损失。

结论和意义：在 ANs 患者中, 7.7% 的患者经历了两次或两次以上的SHL。 SHL的恢复率随着每次连续发生而降低。

Keywords:

• Acoustic neuroma
• sudden hearing loss
• audiogram pattern
• recovery

Introduction

Acoustic neuromas (ANs) are benign tumours arising from the vestibular nerve. Initial symptoms of ANs include progressive unilateral sensorineural hearing loss, tinnitus, and vertigo. While acoustic nerve tumours are generally associated with progressive sensorineural hearing loss, cases that present with sudden sensorineural hearing loss (SHL) have been reported as early as 1917 [1]. Similarly, atypical cases in which sudden onset of deafness in patients improves after treatment with steroids have been reported in Japan and elsewhere [2–4]. Treatment is discontinued once hearing is recovered; thus, ANs may not be identified as a cause of sudden onset deafness. Moreover, smaller ANs may not interfere with hearing [5,6]. Therefore, ANs may not be detectable early because of hearing recovery. Moreover, while recurrence of SHL has been reported [7], the extent of improvement with treatment and the recurrence rates are not well known. The relationship between ANs and SHL needs to be clarified to improve hearing preservation. In the present study, we aimed to retrospectively evaluate the incidence of SHL and associated rates of improvement following conservative management in patients diagnosed with ANs.

Materials and methods

The study protocol was approved by the Institutional Review Board of Nagoya City University, and the requirement for informed consent was waived (approval number: 60-20-0015). This study included patients who were diagnosed with ANs and underwent follow-up evaluations at Nagoya City University Hospital between June 2000 and March 2019.

We retrospectively collected data and investigated the following clinical features of patients with ANs: tumour size and correlation with hearing loss, number of episodes of SHL, extent of auditory recovery, and changes in associated audiogram patterns. The number of episodes of SHL was categorized into 1, 2, and ≥3, and 2 and ≥3 episodes were defined as multiple episodes. Patients were divided into the following 4 groups based on the size and localized extent of the AN: i) intracanalicular tumours, ii) small tumours with ≤10 mm extracanalicular extension, iii) medium-sized tumours with 11–29 mm of extracanalicular extension, and iv) large tumours (≥30 mm). The majority of patients were treated with systemic steroids for SHL. Therefore, SHL recovery was defined as both partial and complete recovery. Hearing recovery was evaluated using the criteria of the Japan Ministry of Health, Labor and Welfare research group [8]. Complete recovery was defined as average hearing (250, 500, 1000, 2000, and 4000 Hz) up to 20 dB or improvement to the same degree of hearing in the unaffected ear. Partial recovery was defined as ≥10 dB improvement or improvement of hearing symptoms. Patients with neurofibromatosis type 2 and those with insufficient SHL data were excluded from the study.

Statistical analysis

Data are expressed as mean ± standard deviation. We analysed the relationship between tumour size and SHL using the chi-square test. Statistical analysis was performed using the JMP software (version 11.0.0, SAS Institute Inc., Cary, NC, USA). p < .05 was set as statistical significance.

Results

Of the 686 study participants, 232 (33.8%) experienced SHL. Among the remaining patients, while 405 (59.0%) did not experience SHL, the status of 49 (7.1%) patients was unknown. A total of 86 (37.1%) and 146 (62.9%) patients with SHL had intracanalicular and extrameatal tumours, respectively. The extrameatal tumours varied in size, with 100, 40, and 6 small (≤10 mm), medium (11–29 mm), and large (≥30 mm) tumours, respectively. The non-SHL group showed similar distribution; 147 (36.3%) patients had intracanalicular tumours. The 258 (63.7%) extrameatal tumours were 166, 77, and 15 small (≤10 mm), medium (11–29 mm), and large (≥30 mm) in size, respectively. The incidence of SHL was not significantly associated with tumour size (p = .79) (Table 1). Among the 232 patients who presented with SHL, the examination findings for 172 patients (93 women, 79 men) were available. The remaining 60 patients were unknown and therefore, excluded from the analysis. The mean age of the study patients was 48.9 ± 12.1 years (range, 25–76 years). A total of 90 right and 82 left ears were examined. The average hearing level at the initial consultation (tested at 5 different frequencies: 250, 500, 1000, 2000, and 4000 Hz) was 39.4 ± 24.6 dB on the affected side and 13.9 ± 10.2 dB for the contralateral unaffected ear. There were 63 cases of intracanalicular tumours. The extrameatal tumours consisted of 77, 29, and 3 small (≤10 mm), medium (11–29 mm), and large (≥30 mm) tumours, respectively.

Table 1. Tumour size and sudden hearing loss.

Tumour size	SHL+ (n = 232)	SHL- (n = 405)
Intracanalicular	86 (37.1%)	147 (36.3%)
Small	100 (43.1%)	166 (41.0%)
Medium	40 (17.2%)	77 (19.0%)
Large	6 (2.6%)	15 (3.7%)

The incidence of SHL was not significantly associated with tumour size (p = .79).

SHL: sudden sensorineural hearing loss.

The audiograms of 78/172 (45.3%) patients showed a typical trough-shaped pattern of hearing loss at medium-level frequencies. However, 22 (12.8%), 20 (11.6%), 16 (9.3%), 14 (8.1%), 10 (5.8%), and 12 (7.0%) patients showed a sloping-type hearing loss pattern (at higher frequencies), high-frequency steep curves, low-frequency hearing loss, horizontal patterns, patterns suggestive of deafness, and other audiometry patterns, respectively (Table 2).

Table 2. Audiogram patterns of sudden sensorineural hearing loss at the initial visit.

Hearing type	SHL+ (%)
Trough-shaped pattern of hearing loss at medium-level frequencies	78 (45.3%)
Sloping type of hearing loss patterns (at higher frequencies)	22 (12.8%)
High-frequency steep curves	20 (11.6%)
Low-frequency hearing loss	16 (9.3%)
Horizontal patterns	14 (8.1%)
Other types of audiometry patterns	12 (7.0%)
Patterns suggestive of deafness	10 (5.8%)

The audiograms of 78/172 (45.3%) patients showed a typical trough-shaped pattern of hearing loss at medium-level frequencies.

SHL: sudden sensorineural hearing loss.

In total, 119 (69.2%), 44 (25.6%), and 9 (5.2%) patients experienced one, two, and three or more episodes of SHL, respectively. Approximately 30% of the patients had multiple episodes (≥2). Confirmed auditory recovery occurred in 105 patients (61.0%) after the first SHL episode. However, those who experienced multiple episodes of SHL showed a significantly lower recovery rate (27 [43.5%], p = .017, chi-square test). Furthermore, only 24 (45.3%) and 3 (33.3%) patients experienced hearing recovery after the second and third episodes, respectively, which is significantly lower in the former group compared to the first episode (p = .043 and p = .098, respectively, chi-square test (Table 3). Thus, the recovery rate decreased in cases with multiple episodes, presumably, as the frequency of occurrence of SHL increased.

Table 3. Recovery rate in each episode.

Episode	SHL recovery (%)	SHL no recovery (%)	p Value
First	105 (61.0%)	67 (39.0%)
Second	24 (45.3%)	29 (54.7%)	.043*	.017*
Third	3 (33.3%)	6 (66.7%)	.098

The hearing of 105 patients (61.0%) was recovered after the first SHL episode. Only 24 (45.3%) and 3 (33.3%) patients experienced hearing recovery after the second and third episodes, respectively. The recovery rate decreased as the frequency of occurrence of SHL increased. The differences in recovery rate between the first episode and second, third, and multiple (second and third) episodes were assessed using the chi-square test. *p ≤ .05.

SHL: sudden sensorineural hearing loss.

Discussion

We found that 33.8% of patients experienced SHL, and 7.7% experienced multiple episodes of SHL. Even more surprisingly, the recovery rate of SHL decreased with each successive occurrence.

Friedman et al. reported that 0.8–47.5% of SHL cases were associated with a diagnosis of AN, while 3–26% of patients diagnosed with ANs reported a history of SHL [9]. In our study, 232 (33.8%) patients with ANs experienced SHL, which is higher than the previously reported rate. This difference may be due to a historical difference in available technology. Earlier reports were compiled by Friedman prior to 2000, and the use of magnetic resonance imaging (MRI) was not widespread at the time. In the present study, MRI allowed us to collate newer data, including patients who were diagnosed with ANs. This resulted in higher enrolment of affected patients.

The relationship between SHL and tumour size remains controversial. Some reports have suggested that SHL is observed more frequently in patients with smaller-sized tumours [4,10]. However, our results showed no significant association between SHL and tumour size, which corroborates the findings of previous studies [11,12].

In their report, Gimsing [13] identified the following types of hearing loss audiogram patterns occurring due to an acoustic nerve tumour: high-frequency sloping, horizontal, or trough-shaped. Suzuki et al. [14] reported basin-shaped (same mean as trough-shaped) curves, horizontal-type patterns, those associated with total deafness, high-frequency steep curves, and high-frequency sloping curves during audiometry examinations of affected patients with asymmetric sensorineural hearing loss. Kanzaki et al. [15] reported that the audiogram pattern of AN showed flat type, high-frequency sudden loss, high-frequency gradual-loss, and U-shaped (trough-shaped) audiometric configurations frequently in patients with small tumours. Many studies have reported the occurrence of trough-shaped audiograms in patients with auditory nerve tumours. Similarly, several studies reported that the audiograms of patients showing SHL with AN were trough-shaped [2,4,16].

In our study, a trough-shaped audiogram was the most common pattern (n = 78, 45.3%), followed by the sloping-type hearing loss pattern (n = 22, 12.8%) and high-frequency steep curves (n = 20, 11.6%). Sando [17] studied the internal auditory canal in which the apical nerve fibres were located in the centre, and the basal nerve fibres were located on the inferior outer surface. These findings indicated nerve compression or conduction block of the cochlear nerve and a trough-shaped hearing pattern on audiometry. Therefore, patients presenting with acute sensorineural hearing loss and a trough-shaped audiogram should undergo MRI to rule out AN from differential diagnoses.

The pathophysiological mechanisms underlying the development of SHL in patients with AN are unknown. Berenholz et al. [18] suggested the following possible mechanisms for SHL: accelerated tumour growth, haemorrhage into the acoustic tumour, a transient increase in tumour volume due to interstitial fluid pressure, biochemical changes within the inner ear, inflammatory immune reaction to the neoplasm, vascular compromise of labyrinthine vessels, and diminished fibre mass, and stated that one or more of these mechanisms will induce interruption or conduction block of the cochlear nerve. However, Friedman et al. [9] supported the nerve conduction block theory. They surmised that arterial compression should cause both auditory and vestibular symptoms, and that the latter are not observed in affected patients. They further theorized that the commonly observed trough-type audiogram patterns showed relative conservation of both high- and low-frequency hearing in patients with SHL, supporting the nerve conduction block hypothesis. The cochlear apical turn is the distal watershed region of cochlear blood flow and has a diluted blood supply. The lack of auditory recovery in patients with SHL indicates a more permanent mechanism of vascular occlusion, such as infarction, as opposed to reversible ischaemia or hypoxia. The cochlear nerve has tonotopic organization. High- and mid-frequency fibres are located on the outer surface of the nerve, especially mid-frequency fibres are located on the outermost part of the dorsal side. On the other hand, low-frequency fibres are located in the centre of the cochlear nerve. Therefore, mid-frequency fibres are easily affected by compression or ischaemia due to tumour compression compared to low-frequency fibres. Ogawa et al. [4] reported that the clinical presentation of AN patients with SHL is associated with a high incidence of trough-type audiogram configuration and normal caloric response, suggesting an underlying conduction blockade of the cochlear nerve. Higgs [19] reported that compression of the labyrinthine artery by the neuronal tumour mass or occurrence of haemorrhage into an intra-canalicular tumour could be a likely etiological factor for the development of SHL in patients with AN. However, Chay et al. [20] discussed the possibility that sudden deafness is secondary to vertebral basilar artery insufficiency and may occur due to reversible cochlear artery ischaemia. The occurrence of repeated episodes of sudden deafness supports the theory of nerve compression as a possible mechanism; however, further studies are required to gather confirmatory evidence.

Aslan et al. [11] found that 57% of patients presenting with SHL and diagnosed with ANs reported partial or complete recovery of hearing, while 36% experienced recurrent bouts. Berg et al. [3] reported that 23% of patients with SHL and vestibular symptoms recovered their hearing. However, previous studies have not investigated the number of relapses or the level of improvement following the recurrence of SHL in patients with ANs. Our study showed that SHL can occur repeatedly, with 119 (69.2%), 44 (25.6%), and 9 (5.2%) patients experiencing 1, 2, and ≥3 episodes, respectively, and 30% experiencing multiple episodes. Most of the confirmed cases were treated with corticosteroids. Berenholz et al. [18] described a patient with AN who experienced 5 episodes of SHL and recovered after receiving corticosteroid therapy. Other reports have also described improvements in SHL with steroid treatment. However, in our study, steroid treatment led to hearing recovery in 105 (61.0%), 24 (45.3%), and 3 (33.3%) patients who experienced 1, 2, and ≥3 episodes of SHL, respectively; thus, the rate of auditory recovery was inversely correlated with the number of SHL episodes. Therefore, although patients may initially respond to steroid treatment, the hearing loss progresses with each successive bout of SHL, hampering the overall extent of auditory recovery. This also indicates that symptomatic auditory improvement or recovery in patients experiencing SHL does not exclude a diagnosis of AN. Therefore, even in patients diagnosed with hearing loss, those who recover with steroid administration, ANs should not be excluded from the differential diagnoses. Identification of a trough-shaped curve on audiometry or clinical presentation with repeated bouts of SHL should raise the index of suspicion for auditory nerve tumours. Since the extent of improvement decreases with successive bouts of SHL, it is necessary to make an early diagnosis and consider aggressive treatment to preserve residual auditory function. We have known the importance of MRI examination for unilateral sensorineural hearing loss and trough-shaped audiogram; therefore, we strongly recommend MRI examination in patient with SHL and trough-shaped audiogram.

In the present study, hearing recovery was assessed based on information from a previous physician and interviews with patients. This is because this study was retrospective in nature and could not collect all the past audiograms. This may be a limitation of the present study.

In conclusion, SHL is more frequently associated with ANs than previously thought, and affected patients may experience recurrent episodes of SHL. Hearing loss can worsen with successive episodes. Predicting the occurrence of SHL is difficult because the natural history of acoustic nerve tumours has not yet been established. Despite the reported auditory improvement in affected patients, intermittent monitoring using regular audiometric examinations is recommended. In the future, a complete analysis of factors causing SHL recurrence is required to better understand progressive hearing loss in patients with ANs.

Acknowledgments

The authors thank Editage (www.editage.com) for English language editing.

Disclosure statement

The authors have no conflicts of interest to disclose.