Introduction

Intratympanic injections for inner ear disease were first described by Trowbridge in 1944 [1] and a variety of treatment protocols have since been suggested. The rationale of locally delivered steroids is to allow the drug to reach tissues of interest with a high dosage, yet avoiding the adverse effects of systemic administration. The semi-permeable properties of the round window membrane allow intratympanic steroids to access the perilymph by pinocytosis and diffusion [2]. This bypassing of the hemato-cochlear barrier results in up to 1.270-fold higher steroids concentrations into the perilymph [3] when compared to systemic administration.

Corticosteroid receptors have been identified in both cochlear and vestibular tissues, suggesting that gene expression can be altered to produce anti-inflammatory effects [4]. Dexamethasone intratympanic injections have been shown to increase cochlear blood flow by 29 % [5] and increase the expression of aquaporin-1 [6], a key regulator in perilymphatic fluid homeostasis. Therefore, the inner ear’s physiology can be modulated by steroids and this explains why neurotologic disorders that have an inflammatory origin may respond to this drug.

This article reviews published evidence on the use of locally delivered steroids for treatment of inner ear pathologies suspected to have an inflammatory or autoimmune physiopathology. The objective is to identify inner ear diseases for which patients can benefit from ITSI.

Method

After approval from our institutional review board committee, PubMed and Medline Databases were searched for all human prospective randomized clinical trials with a treatment arm receiving ITSI. The following research terms were used: steroids, corticosteroids, prednisolone, methylprednisolone, dexamethasone glucocorticoids, intratympanic, transtympanic, inner ear disease, labyrinth diseases. Inner ear diseases of interest were identified through article titles and individually added to the search protocol. Authors reviewed all potential papers based on available abstracts and complete articles analysis. Articles’ references were also screened for potential additional studies. When multiple articles had been published by the same author for a growing series of patients, only the latest article, hence the one with the largest number of participants was included. Articles in languages other than English and French were also excluded. Database search included studies published between 1946 and December 2014.

Results

Databases search identified four inner ear diseases for which ITSI was used in a prospective randomized clinical trial: Ménière’s Disease, tinnitus, noise-induced hearing loss (NIHL) and idiopathic sudden sensorineural hearing loss (ISSNHL). After duplicates removal and screening, 29 articles remained for analysis.

For Ménière’s Disease, 6 articles were identified and listed in Table 1. When feasible, extraction of study data followed the recommendations of the Committee on Hearing and Equilibrium of the American Academy of Otolaryngology-Head and Neck Surgery [7]. Of the selected articles, three [8–10] found no benefit of ITSI over placebo. One of these studies [10] compared a dexamethasone poloxamer gel to a placebo and found no statistical significance (p = 0.086) in the reduction of number of vertigo-days per month (−0.201 vs −0.124) and no statistical significance (p = 0.055) in the reduction of the tinnitus handicap inventory score (−15.0 vs −4.0).

Table 1 Intratympanic steroid injections in Ménière’s disease
Full size table

One study [11] found that low-dose intratympanic gentamycin injections (ITGI) provided better vertigo control over ITSI and another found that vertigo was similarly controlled with ITGI, ITSI and endolymphatic sac decompression (ESD) [12].

In the remaining study, Garduno-Aaya et al. [2] found better vertigo control with ITSI over placebo (82 vs 57 %), a statistically significant improvement in subjective hearing (35 vs 10 %) and a statistically significant improvement in subjective tinnitus loudness and aural fullness (48 vs 20 %).

For tinnitus control, five prospective randomized trials were identified and listed in Table 2. Three studies [13–15] compared ITSI to placebo and found no benefit in tinnitus score improvement. One study [16] compared intratympanic dexamethasone, intratympanic prednisolone and oral carbamazepine and found no benefit in the ITSI groups. The study by Shim et al. [17] is the only study where tinnitus was improved in a statistically significant manner with ITSI (25.8 vs 9.8 % in control group).

Table 2 Intratympanic steroid injections in tinnitus treatment
Full size table

In the treatment of NIHL, only one randomized clinical trial on humans was available [18] (Table 3). Of the 27 patients receiving combination therapy (systemic and ITSI), 51.9 % improved their pure tone average (PTA) by more than 15 dB and 66.7 % improved their speech discrimination score (SDS) by 15 % or more. Recovery rates for PTA and SDS were significantly better in the combination therapy group when compared to systemic therapy alone.

Table 3 Intratympanic steroid injections in noise-induced hearing loss
Full size table

Ten studies investigated ITSI as first-line therapy for ISSNHL recovery (Table 4). Two studies [19, 20] found that ITSI therapy was equivalent to systemic therapy in hearing level (HL) improvement, and one [21] found that combination therapy was similar to systemic steroids. Of the two studies [22, 23] comparing systemic, ITSI and combination therapy, one found that combination therapy was superior to systemic treatment alone (87.5 vs 44.4 %). As for the four remaining studies, three [24–26] found that combination therapy was superior to systemic therapy and one [27] favored ITSI over placebo.

Table 4 Intratympanic steroid injections as first-line therapy of idiopathic sudden sensorineural hearing loss
Full size table

For salvage therapy in ISSNHL, 8 studies were identified (Table 5). All but one found that ITSI therapy was superior to control, placebo or systemic therapy.

Table 5 Intratympanic steroid injections as salvage therapy of idiopathic sudden sensorineural hearing loss
Full size table

Discussion

A meta-analysis of the identified literature was not performed due to the heterogeneous nature of the data. Studies used different treatment protocols, definitions of SSNHL and Ménière’s disease, definitions of outcome criteria, and timeframes for follow-up. Reported studies described the use of different steroids: dexamethasone or methylprednisolone. A single study [16] in our review compared methylprednisolone to dexamethasone injections and found no difference in tinnitus control between groups. In their pharmacokinetics studies in 1999, Parnes et al. [28] compared these two molecules. They found that methylprednisolone was superior to dexamethasone as peak concentrations were higher and remained higher for longer duration. Other authors have argued that higher concentrations of methylprednisolone were sampled in the endolymph due to decreased absorption by cochlear and vestibular tissues. To this day, no clinical data favors one over the other.

Given the natural evolution of Ménière’s disease, it has been suggested that study protocols always include a placebo group. Of the listed studies in Table 1, only 3 included a placebo group. The studies by Lambert et al. [10] and Garduno-Aaya et al. [2] were both of Level 1 evidence [29], but only the latter suggested benefits in vertigo and tinnitus control over placebo. The study by Silverstein et al. [8] found no difference between ITSI and placebo, but was criticized for being a crossover study. No benefits were found at 3 weeks, before the crossover, the only time point unbiased by the potential carry-over effect.

Paragache et al. [9] compared ITSI to medical therapy, comprising salt, caffeine, nicotine and alcohol restriction with cinnarizine and betahistidine hydrochloride. No difference was measured between the two groups in vertigo, tinnitus or hearing loss recovery (Table 1). However, the patients in the ITSI group were instilled the lowest dexamethasone concentration of all reported studies in this review: 20 times less than the usually used concentration of 4 mg/mL. With the expected dose–response relationship of inflammatory and autoimmune diseases to steroids, one can expect that the used steroid concentration was too low to produce any therapeutic effect.

The two remaining randomized studies interested in Ménière’s disease compared ITSI to ITGI. Together with our previously published local experience [30], ITGI seems to offers better vertigo control over ITSI. However, given the potential cochlear toxicity of gentamycin, ITGI should only be considered for patients with non-serviceable hearing.

In tinnitus therapy, one study [17] found benefits of ITSI over a control group. Patients were selected for having unilateral idiopathic tinnitus for less than 3 months. The authors hypothesized that in the early stage of disease; the cochlear lesion causing tinnitus can be reversed. Because minimal plastic change has occurred in the central auditory pathway, early administration of ITSI may enable cochlear lesion recovery and restore neural hyperactivities of the central auditory pathway. Unfortunately, there was no placebo group in this study so results were not compared to the natural evolution of the disease. Also, patients’ last follow-up was at 3 months, so long-term benefits were not assessed. Yet, this study suggests that patients might benefit from a reduced time between tinnitus onset and therapy.

Zhou et al. [18] found that combination therapy was superior to systemic steroids alone for patients exposed to noise trauma who had shown no spontaneous recovery within the first 72 h. Eighty-one percent of the recruited patients had been exposed to fireworks or military training noise. The therapeutic effects of steroids are thought to arise from their protective effects on injured cells [31, 32] by stabilization of cellular membranes, scavenging of oxygen free radicals and by inhibition of phospholipase A2. This first human study on ITSI therapy for NIHL shows promising results. However, sample size was small (53 patients) and patient selection was heterogeneous. Further studies are needed to confirm these benefits.

The use of ITSI for treatment of ISSNHL has been studied in two settings: as first-line and as salvage therapy. Given the unethical considerations of offering placebo as first-line therapy to patients suffering from ISSNHL, most studies compared combination to systemic therapy.

Four of these studies found that combination therapy was superior to systemic therapy alone and three found that combination therapy was equivalent to systemic treatment. Hence as first-line therapy of ISSNHL, adding ITSI to systemic therapy significantly improved patients’ outcome in more than half of the available studies. The used steroid concentrations in the positive studies ranged from methylprednisolone 12 mg/mL to dexamethasone 12 mg/mL (Table 4). Furthermore, Rauch et al. [19] designed a non-inferiority trial involving 205 patients and found that ITSI alone was not inferior to oral therapy. Hence, with recovery rates ranging from 55 % [22] to 96 % [27] the added benefits of ITSI need to be considered. The addition of ITSI to ISSNHL therapy may allow the use of lower systemic doses, thereby minimizing their adverse effects.

When considering salvage therapy in ISSNHL, ITSI groups had PTA improvement of 15 dB or more that ranged from 37.5 % [33] to 54.5 % [34]. The benefits on hearing levels were also significantly higher with ITSI in all but one study. Hence the available evidence supports the use of ITSI in salvage therapy.

In the six positive studies reported in Table 5, salvage therapy was administered, if no response was noted, 10–13 days after first-line systemic therapy. The used steroid concentration ranged from methylprednisolone 40 mg/mL to dexamethasone 5 mg/mL. Therefore, physicians should consider offering ITSI to patients suffering from ISSNHL after as little as 10 days into an unsuccessful first-line systemic therapy regiment.

Adverse events of ITSI therapy include ear pain at time of injection, caloric vertigo, dizziness, infection and persistent tympanic perforation. All of these side effects are either transient or easily curable. Pain and caloric vertigo can be, respectively, minimized with the use of fine needles and adequate steroid temperature at time of injection. Hence, when compared to systemic administration of steroids, intratympanic delivery is safe and can easily be managed by otolaryngologists.

Conclusion

Due to heterogeneity in treatment protocols and follow-up, a meta-analysis was not performed. Our review found only one article over six where ITSI therapy offers potential benefits to patients with Ménière’s disease in the control of tinnitus and vertigo. Patients affected with ISSNHL seem to benefit from ITSI in both first-line and salvage therapy. Only one human study was found on NIHL and its results showed a statistically significant improvement on hearing thresholds. Furthermore, our review showed that ITSI does not seem to be effective in the treatment of tinnitus.

Given the low adverse effects rates of ITSI therapy and good patient tolerability, local delivery should be considered as an interesting adjunct to the therapy of the idiopathic sudden sensorineural hearing loss and noise induced hearing loss. However, despite the number of published studies on this delivery modality, it is yet difficult to recommend a specific treatment protocols for these inner ear conditions. A tailored approach based on patient’s tolerance and response seems most appropriate. The inner ear diseases presented in this study are all thought arise from an inflammatory or an autoimmune process. Therefore, expect a dose–response relationship with ITSI therapy and future local delivery devices offering increased and prolonged release might improve recovery rates.