Introduction

Most of the acute sensorineural hearing loss is universally diagnosed as sudden hearing loss (SHL). However, the pathogenesis of SHL remains unknown and proposed theories include vascular occlusion, membrane brakes, and viral cochleitis [1]. Treatment of SHL is widely variable but no valid randomized controlled trial exists to determine effective treatment of SHL [2–4].

Labyrinthine window rupture (LWR) is one cause of acute sensorineural hearing loss and mimics SHL [5–12]. Exploratory tympanotomy for SHL with inspection and sealing of the labyrinthine windows is therapeutic option if conservative treatment for SHL fails to improve the hearing level [6–12]. However, early surgical intervention appears to be rewarded with good hearing recovery [8, 9, 11]. One reason for preferring conservative treatment at the early stage is absence of diagnostic tests for LWR. Therefore, new noninvasive diagnostic tools must be developed to improve the management of LWR [13].

LWR is a condition that allows the egress of perilymph from the inner ear to the middle ear. Temporal bone CT can detect even a small amount of fluid in middle ear cavity. Fourteen out of 16 cases were true LWR when CT showed more than one-third of their round window recess was filled up with fluid [14]. Air fluid level in the middle ear documented by CT is a suggestive image of LWR [15]. Therefore, we have utilized CT to make diagnose of LWR since 2006. Exclusion of LWR from SHL and early surgical intervention in LWR have improved hearing prognosis of both LWR and SHL.

Materials and methods

A retrospective review was made of 48 patients, seen between May 2006 and May 2010, who attended the Miyakonojo Hospital with symptoms of a rapid loss of hearing of 60 dB or more. Twenty-four patients had been treated with corticosteroid before attending the Miyakonojo Hospital. The severity of hearing loss was evaluated with the average hearing level at five frequencies (250 Hz, 500 Hz, 1 kHz, 2 kHz, 4 kHz) on the pure tone audiogram. The severity was categorized into three grades: severe deafness (60–89 dB), profound deafness (90–110 dB) and total deafness (111 dB) that was no response on audiometry at all frequencies.

All patients were examined with dual slice CT scanner of a G.E. Hi speed NX/is (General Electronic, Milwaukee, WI, USA). All but one patient underwent CT examination on the same day of their hospital attendance. Standard temporal bone CT was performed in the axial projection with a bone algorithm. Slices were 1 mm thick and contiguous. The images were carefully surveyed whether there was any space-occupying lesion in the tympanic sinus (Figs. 1, 2, 3) or in the round window recess (Figs. 4, 5, 6).

Fig. 1
figure 1

Axial CT shows air fluid level in the tympanic sinus (arrow)

Full size image
Fig. 2
figure 2

Axial CT shows hemispherical lesion on the sidewall of the tympanic sinus (arrow)

Full size image
Fig. 3
figure 3

Axial CT shows space-occupying lesion at the orifice of the tympanic sinus (arrow)

Full size image
Fig. 4
figure 4

Axial CT shows focal bulging of the round window membrane (arrow). Double arrow shows air fluid level over the orifice of the left tympanic sinus

Full size image
Fig. 5
figure 5

Axial CT shows space-occupying lesion in the round window recess (arrow)

Full size image
Fig. 6
figure 6

Axial CT shows hemispherical lesion in the round window recess (arrow)

Full size image

Our indications for exploratory tympanotomy were same to literatures [9–12]. They were progressive hearing loss, acute hearing loss with vertigo and acute hearing loss with the presence of positional nystagmus in a spine position. Space-occupying lesions around the labyrinthine windows were a pre-condition for exploratory tympanotomy.

Thirty-five ears of 34 patients who made the decision to have exploratory tympanotomy were operated for LWR. There were 19 males and 15 females in the patient group. Subjects ranged in age from 9 to 80 years. The median age was 47.4 years.

Sixteen ears of LWR were severe deafness, 15 ears were profound deafness and four ears were total deafness. Nine patients had vertigo and nine patients had balance disturbance. There was a positive history of antecedent head trauma in nine patients 3 months to 38 years before. Seven patients had a recent history of heavy lifting or straining. Nose blowing was in four patients and exposure to loud sound was in two patients. Past history of SHL was in one patient.

Length of time from CT examination to exploratory tympanotomy was 1–8 days (Table 1). The median time interval was 2.1 days. Length of time from initiation of hearing loss to exploratory tympanotomy was 1–26 days. The median time interval was 8.5 days.

Table 1 Number of ears related to time intervals from CT to surgery in 35 ears of LWR
Full size table

Exploratory tympanotomy was performed by post auricular incision under general anesthesia. Annular bone removal was done with a currete. The oval and round windows recesses were then examined for perilymphatic leaks. An oval window leak was seen in nine ears, a round window leak in three ears and leakage from both oval and round windows in 23 ears. A large quantity of leak was seen in 57% of the ears and a small quantity of leak was seen in 43% of the ears. Displacement of the stapes footplate was not encountered. Both labyrinthine window recesses were packed with the temporalis fascia supported laterally by Gelfoam. Neither corticosteroid nor dextrane was used for post-operative treatment.

Fourteen patients were diagnosed as having SHL and received conservative treatments. However, there was a positive history of antecedent head trauma in four patients 10 days to 20 years before the onset. One patient had an event of heavy lifting 3 days before the onset. Two patients of total deafness, who had positive findings in temporal bone CT, did not agree to have exploratory tympanotomy. There were ten males and four females in patient group. Subjects ranged in age from 21 to 84 years. The median age was 61.6 years.

Five ears were severe deafness and 3 ears were profound deafness. They had neither vertigo nor balance disturbance. Two patients respectively received total 110 mg or 40 mg of oral prednisone. Four patients received oral adenosine triphosphoric disodium hydrate.

Six ears were total deafness. One patient had vertigo and five patients had balance disturbance. Two patients were treated with 125 mg of methylprednisolone sodium succinate intravenously on 3 days followed by a course of 25-mg prednisolone tapered over 11 days. One patient received total 135 mg of oral prednisone. Three patients received oral adenosine triphosphoric disodium hydrate.

Improvement in hearing was categorized into five grades. Grade “cured” is all five frequencies within normal level or recovery to the same level of the other ear. Grade “markedly recovered” is the improvement of 30 dB or more. Grade “slightly recovered” is the improvement of 10 dB or more but <30 dB. Grade “unchanged” is the improvement of <10 dB. Grade “worsened” is decline of 10 dB or more.

Results

The hearing prognosis of LWR was analyzed according to the severity of hearing loss (Table 2). Fifty percent of severe deafness and 73% of profound deafness were cured or markedly recovered. Of four ears of total deafness, one ear was markedly recovered.

Table 2 Recovery related to severity of hearing loss in 35 ears of LWR
Full size table

Lower frequencies tended to recover more favorably than higher frequencies. Ten of the 15 markedly recovered ears and 3 of the 10 slightly recovered ears showed an abruptly falling audiometric configuration [7]. Hearing loss did not return to normal where high and middle frequencies in the audiogram obtained just before the surgery were scale out.

The hearing prognosis of LWR was affected by the time of surgery (Table 3). Of the 15 markedly recovered ears, 14 ears were operated within 2 weeks from the onset. Of the five cured ears, four ears were operated within a week from the onset.

Table 3 Recovery related to time intervals from onset of symptoms to surgery in 35 ears of LWR
Full size table

Six ears were operated within 2 days from the onset (Table 3). Two of them were cured. Three of them with no audiometric response in middle and high frequencies were markedly recovered. The remaining one ear that was total deafness with balance disturbance was unchanged.

Hearing prognosis was analyzed according to both a quantity of leak and time of surgery (Table 4). Thirteen of 19 ears that received the tympanotomy within a week had a large quantity of leak, whereas 7 of 16 ears that underwent the tympanotomy on and after the 8th day had a large quantity of leak. Hearing recovery was observed in greater percentage of patients who underwent the tympanotomy within a week and had a large quantity of leak than patients who had a small quantity of leak. On the contrary, hearing recovery was observed in smaller percentage of patients who underwent the tympanotomy on and after the 8th day and had a large quantity of leak than patients who had a small quantity of leak.

Table 4 Hearing recovery related to both a quantity of leak and time intervals from onset of symptoms to surgery
Full size table

Vertigo and balance disturbance had no influence on their hearing prognosis. Hearing recovery was not affected by positive histories of antecedent head trauma, heavy lifting and straining et al.

The hearing prognosis of SHL was analyzed according to the severity of hearing loss (Table 5). Of five ears of severe deafness, four ears were cured. Of three ears of profound deafness, two ears were cured and one ear was markedly recovered. Six ears of total deafness with vertigo or balance disturbance were neither cured nor markedly recovered.

Table 5 Recovery related to severity of hearing loss in 14 ears of SHL
Full size table

Discussion

There are three criteria to confirm diagnosis of LWR at surgery. They are actual observation of fluid leak [6, 7], direct inspection of the round window membrane rupture [8–11] and no simultaneous pressure transmission from the oval to the round window [11, 12]. The actual fluid leak in the labyrinthine window recesses was found in 71% [6] and 40% [7]. Our result was that a large quantity of leak was seen in 57% and a small quantity of leak was seen in 43%.

A large quantity of leak was more frequent in ears operated within a week than in ears operated on and after the 8th day. Ears with a large quantity of leak had a tendency toward poor hearing recovery when exploratory tympanotomy was delayed. Temporal bone CT could not predict a quantity of perilymphatic leak. Therefore, ears with space-occupying lesions around the labyrinthine windows should be operated early when the patients have suggestive symptoms of LWR.

Arndt [8] carried out 45 exploratory tympanotomy among 63 patients of SHL. Mertens and Rudert [9] performed 39 exploratory tympanotomy among 247 patients of SHL. We operated 35 ears for LWR while we conservatively treated 14 ears of SHL. Incidence of perilymphatic fistula surgery for acute sensorineural hearing loss is very limited in the United States [13]. However, we think LWR might not be a rare cause of acute sensorineural hearing loss.

Indication of exploratory tympanotomy for SHL is progressive hearing loss or no response to conservative treatment [8–12]. A term of conservative treatment is 1 day to 2 weeks [6–12]. Early exploratory tympanotomy is not recommended because it might decrease the chance for inner ear healing and spontaneous remission might occur in the first days [5]. To avoid unnecessary exploratory tympanotomy, we utilize temporal bone CT to detect perilymph in the middle ear cavity. We choose conservative treatment when CT shows no space-occupying lesion around the labyrinthine windows.

Most patients of LWR in literatures are operated within 2 weeks from the onset [9–12]. Recovery rate is better in patients operated within the first week than those within the second week [9, 11]. Exploration of the round window should be done within the first few days [8]. Our hearing prognosis related to time of surgery is same to the previous reports.

According to the literatures, complete recovery rate of hearing loss by sealing of the labyrinthine windows is 7% [8] and 10% [12]. Our result is 14%. Needed conditions of the complete recovery are both early surgical intervention and detectable hearing loss in middle and high frequencies in the audiogram obtained just before the surgery.

Treatments of LWR are strict bed rest or exploratory tympanotomy with sealing of both round and oval windows [5–13]. Fifty-seven percent of LWR over 60 dB was either cured or markedly recovered after surgical closure of both labyrinthine windows. If they had received conservative treatments, their hearing prognosis would have been <50% [16–18].

Our hearing prognosis of SHL is better than the literatures [16–18]. Eighty-eight percent of severe and profound deafness were cured or markedly recovered. Hearing prognosis of total deafness with vertigo or balance disturbance was not good, which was equal to literatures [16–18].

In conclusion, presence of space-occupying lesion in the tympanic sinus and/or the round window recess strongly supports the diagnosis of true LWR with symptoms suggesting LWR. Since delayed surgery may permit cochlear deterioration, early surgical intervention is needed for excellent postoperative hearing of LWR. Surgical intervention in LWR, that resultantly excludes LWR from SHL, will make hearing prognosis of SHL better than general treatments [2–4].