Introduction
Sudden sensorineural hearing loss (SSNHL) is defined as a hearing reduction of 30 dB or more, in at least three consecutive frequencies, occurring within 3 days of symptom onset. 1 The cause is evident in about 10–15 per cent of cases, with 1 per cent having retrocochlear causes such as acoustic neuroma, demyelinating disease and stroke.Reference Mattox and Simmons 2 , Reference Hughes, Freedman, Haberkamp and Guay 3
Sudden sensorineural hearing loss affects 5–20 per 100 000 people per year.Reference Byl 4 The spontaneous remission rate has been reported as 45–65 per cent.Reference Mattox and Simmons 2 , Reference Byl 4 Vascular occlusion, viral infections, labyrinthine membrane breaks, immune-associated mechanisms and abnormal stress response in the cochlea have been proposed as mechanisms for idiopathic SSNHL.Reference Alimoglu, Inci, Edizer, Ozdilek and Aslan 5
Many therapeutic modalities have been employed to treat the condition. Oral corticosteroids and intratympanic steroids are the most commonly preferred options. Hyperbaric oxygen is utilised as an adjunct first-line treatment in Europe. This study aimed to examine the usefulness of hyperbaric oxygen treatment for SSNHL after oral corticosteroid therapy failure in a retrospective case note study.
Materials and methods
Case notes of patients who were followed up because of idiopathic SSNHL between 2005 and 2011 in a tertiary care centre were examined retrospectively.
Patients
The criteria for SSNHL diagnosis were a hearing reduction of 30 dB or more, in three consecutive frequencies, within 3 days of symptom onset. 1
Patients were included in this study if their mean pure tone averages (PTAs) improved by less than 15 dB HL or deteriorated after a 14-day course of oral corticosteroid therapy, and if they had received hyperbaric oxygen therapy as a salvage treatment after oral corticosteroid therapy.
Patients were excluded if their oral corticosteroid treatment was started more than 10 days after the onset of symptoms, or if hyperbaric oxygen treatment was started more than 35 days after the onset of symptoms. Patients were also excluded if: a cause for the hearing loss was evident, their hearing loss was bilateral, they were aged less than 12 years, or pre-treatment and post-treatment audiograms were not available.
The oral corticosteroid treatment protocol for SSNHL in our clinic was 1 mg/kg body weight of oral prednisolone or equivalent for 7 days, with a 10 mg taper every 2 days. Oral corticosteroid therapy lasted about 14 days. Hyperbaric oxygen treatment consisted of 20–30 sessions of 120 minutes at 2.5 atmospheres absolute pressure, over a treatment period of 14–21 days.
Evaluation criteria
Pure tone average was calculated as the arithmetic mean of thresholds at 0.5, 1, 2 and 4 kHz. The PTAs before oral corticosteroid therapy, and before and after hyperbaric oxygen treatment were recorded. Treatment outcome was evaluated using: Siegel's criteriaReference Siegel 6 before and after hyperbaric oxygen therapy, and mean PTA gain and gains at 0.5, 1, 2, 4 and 8 kHz (by subtracting post-hyperbaric oxygen therapy threshold values from pre-therapy threshold values).
Treatment outcomes according to Siegel's criteria can be categorised as follows: (1) complete recovery – final threshold of more than 25 dB; (2) partial recovery – gain of more than 15 dB, with a final hearing threshold of 25–45 dB; (3) slight recovery – gain of more than 15 dB, with a final threshold of more than 45 dB; and (4) no improvement – gain of less than 15 dB, with a final hearing threshold of more than 75 dB.Reference Siegel 6
A Wilcoxon signed rank test was used to compare mean values before and after salvage hyperbaric oxygen therapy. Statistical analysis was performed with the use of SPSS software (SPSS, Chicago, Illinois, USA). A p value of less than 0.05 was considered significant.
The protocol of this study was approved by the institutional review board.
Results
Thirty-six patients, all unilateral cases, were found to meet the criteria. Excluded patients and reasons for exclusion are shown in Table I. Of these 36 patients who met the criteria, 12 (33.3 per cent) were male and 24 (66.3 per cent) were female. Mean age was 41.42 ± 15.90 years (range, 12–75 years). The time to onset of oral corticosteroid therapy was 2.67 ± 2.91 days (range, 0–10 days). The time to onset of hyperbaric oxygen therapy was 19.19 ± 6.86 days (range, 11–35 days).
Reasons for exclusion
SSNHL = sudden sensorineural hearing loss
Audiogram characteristics before and after hyperbaric oxygen treatment are shown in Tables II and III. Mean gains after hyperbaric oxygen treatment, at each frequency, and PTA gain, are shown in Table IV and in Figure 1.
Gain at each frequency and pure tone average (PTA) gain. The PTA represents the mean of thresholds at 0.5, 1, 2 and 4 kHz.
Mean thresholds before hyperbaric oxygen treatment
Pure tone average represents the mean of thresholds at 0.5, 1, 2 and 4 kHz. PTA = pure tone average; SD = standard deviation
Mean thresholds after hyperbaric oxygen treatment
Pure tone average represents the mean of thresholds at 0.5, 1, 2 and 4 kHz. *p < 0.05 (compared with data collected before hyperbaric oxygen salvage therapy). PTA = pure tone average; SD = standard deviation
Mean gains after hyperbaric oxygen treatment
Pure tone average represents the mean of thresholds at 0.5, 1, 2 and 4 kHz. PTA = pure tone average; SD = standard deviation
Of 3 patients who recovered completely, 1 did so after 30 sessions of hyperbaric oxygen treatment (mean gain of 51.25 dB). Another patient achieved complete recovery after 20 sessions (mean gain of 27.5 dB). A further patient achieved partial recovery after 10 sessions (mean gain of 36 dB), followed by complete recovery after 20 sessions (mean gain of 45 dB).
Analysis
The thresholds at frequencies 0.25, 0.5, 1, 2, 4 and 8 kHz and the PTA values before and after hyperbaric oxygen salvage therapy were compared. The thresholds at 0.25 kHz (Z = −4.125, p = 0.0), 0.5 kHz (Z = −4.166, p = 0.0), 1 kHz (Z = −4.090, p = 0.0), 2 kHz (Z = −2.786, p = 0.005) and 4 kHz (Z = −2.530, p = 0.011), and PTAs (Z = −3.957, p = 0.0), were statistically significantly different.
Discussion
The cause of SSNHL is unknown in 85–90 per cent of cases.Reference Mattox and Simmons 2 Therefore, the treatment is empirical. The most commonly used treatments are oral corticosteroids, intratympanic corticosteroids and hyperbaric oxygen therapy. When primary treatment fails, intratympanic steroids or hyperbaric oxygen are generally offered if they were not administered as primary treatment.
Given the possibility that SSNHL is of ischaemic origin, oxygen may be a beneficial therapeutic option. Oxygen diffusion across the blood brain barrier reaching target tissues is the rationale behind hyperbaric oxygen treatment; furthermore, this treatment is well tolerated.Reference Mazurek, Haupt, Georgiewa, Klapp and Reisshauer 7 , Reference Bennett, Kertesz and Yeung 8 The purpose of hyperbaric oxygen therapy in SSNHL treatment is to increase partial oxygen pressure in blood and inner-ear fluids. Nagahara et al., in 1983, measured perilymphatic oxygen tension through an electrode inserted in the perilymphatic space and found it to be decreased in some patients with SSNHL.Reference Nagahara, Fisch and Yagi 9 Kubo et al. suggested that decreased cochlear blood flow and subsequent capillary oedema may be a possible mechanism of cochlear hypoxia.Reference Kubo, Matsunaga, Asai, Kawamoto, Kusakari and Nomura 10 If 100 per cent oxygen is inhaled at 2 atmospheres absolute pressure, the amount of dissolved oxygen increases from 0.32 per cent at normal atmospheric pressure to 4.44 per cent. Hyperbaric oxygen treatment also decreases cyclo-oxygenase 2 and prostaglandin E2 production, and serves as an anti-inflammatory agent equivalent to 20 mg/kg diclofenac.Reference Muzzi, Zennaro, Visentin, Soldano and Sacilotto 11
Some previous studies have investigated the role of hyperbaric oxygen utilised as a salvage treatment after systemic corticosteroid therapy failure. In 2005, Horn et al. reported on nine patients who underwent hyperbaric oxygen salvage treatment after two weeks of corticosteroid and antiviral therapy.Reference Horn, Himel and Selesnick 12 Two patients improved dramatically and one patient improved in word recognition score only.
In 2010, Muzzi et al. reported on 19 patients who received 30 sessions of hyperbaric oxygen therapy after unsuccessful medical treatment.Reference Muzzi, Zennaro, Visentin, Soldano and Sacilotto 11 Details of the medical treatment were not provided. Higher gains were observed in lower frequencies (mean gains of 13.42 dB, 11.84 dB, 8.95 dB, 6.05 dB, 8.42 dB and 3.16 dB at 0.25 kHz, 0.5 kHz, 1 kHz, 2 kHz, 4 kHz and 8 kHz, respectively). Recovery data were not presented categorically but as relative decreases in thresholds. Absolute mean gains were similar to our gain results.
In a prospective randomised study, Cvorovic et al. examined the effect of intratympanic steroids and hyperbaric oxygen therapy after unsuccessful medical treatment (6 days of systemic corticosteroid therapy).Reference Cvorovic, Jovanovic, Milutinovic, Arsovic and Djeric 13 Fifty patients received either 20 sessions of hyperbaric oxygen therapy or intratympanic dexamethasone. Both treatments were found to be effective in decreasing threshold values.
Yang et al. compared intratympanic steroid injection, hyperbaric oxygen treatment and both of these treatments combined in a retrospective controlled study comprising 103 patients.Reference Yang, Wu and Hwang 14 Their pre-salvage treatment protocol consisted of systemic corticosteroid therapy for 10 days, plasma expanders and radiopaque contrast material application. Final PTA was measured two months after diagnosis. Patients received 10 sessions of hyperbaric oxygen treatment over 2 weeks. The authors reported that both individual salvage treatments and the combination therapy resulted in higher gains in PTA and increased word recognition scores than the control group. They defined gains in PTA that were higher than 15 dB as good recovery, and found that only 22.2 per cent of cases achieved good recovery in the control group compared with 45.5–68.4 per cent in the salvage treatment groups. The mean hearing gain of the hyperbaric oxygen salvage group was 17.4 dB while that for the control group was 7.4 dB.
Psillas et al. administered 15 sessions of hyperbaric oxygen therapy as a salvage treatment in 15 patients, and compared their outcomes to those of a control group of 30 patients.Reference Psillas, Ouzounidou, Stefanidou, Kotsiou, Giaglis and Vital 15 The authors reported a greater ratio of improvements (46 per cent vs 13.3 per cent) and better PTA gains (12.1 dB vs 2.7 dB) in the hyperbaric oxygen therapy group compared with the control group.
Pezzoli et al. managed 23 patients with hyperbaric oxygen treatment and 21 patients with observation.Reference Pezzoli, Magnano, Maffi, Pezzoli, Marcato and Orione 16 The treated patients had a better hearing gain (15.6 ± 15.3 dB vs 5.0 ± 11.4 dB). The authors also reported a better recovery rate with 15 sessions of hyperbaric oxygen treatment compared with observation only (1 of the treated patients completely healed, 5 had good recovery, 10 had fair recovery and 7 were unchanged, whereas 3 of the observed patients had good recovery, 1 had fair recovery and 17 were unchanged). Good recovery was defined as recovery in PTA within 15 dB of the unaffected ear. Fair recovery was defined as more than 10 dB recovery but out of the 15 dB range. Poor recovery was defined as a less than 10 dB increase in PTA.
Our patients had statistically significant decreases in thresholds at 0.25–4 kHz, and in PTAs, after hyperbaric oxygen treatment.
Mean gains were more prominent in lower frequencies (0.25 kHz and 0.5 kHz), approaching 17 dB. The higher gains obtained in the lower frequencies could mean that these frequencies are more responsive to hyperbaric oxygen salvage therapy. The basal turn of the cochlea has been proposed to be more vulnerable to damage and may therefore be less prone to recovery.
The gains at 1 kHz and PTA were just exceeding 10 dB (Table IV), which is the lower limit that it is suggested should be reported, as stated in the American Academy of Otolaryngology – Head and Neck Surgery clinical practice guideline for sudden hearing loss. However, using a 10 dB HL change in PTA has been considered worrisome because this magnitude of change is within test–retest reliability of measuring pure tone thresholds,Reference Stachler, Chandrasekhar, Archer, Rosenfeld, Schwartz and Barrs 17 making the amount of change very close to 10 dB clinically unreliable.
Changes in Siegel recovery classification are another means of determining salvage treatment benefit. Three patients (8.33 per cent) had complete recovery, while one patient (2.7 per cent) had a more than 45 dB PTA gain and five patients (13.88 per cent) had a 15–45 dB PTA gain. Twenty-five patients (69.44 per cent) showed no improvement. None of the five total hearing loss patients in our group attained serviceable hearing that enabled them to use hearing aids.
Studies on salvage hyperbaric oxygen treatment administered after oral corticosteroid therapy failure report mostly mean gains, but categorical recovery is mentioned in a minority of studies.Reference Horn, Himel and Selesnick 12 , Reference Yang, Wu and Hwang 14 , Reference Pezzoli, Magnano, Maffi, Pezzoli, Marcato and Orione 16 We think that the latter is a more appropriate way of presenting findings.
-
• Hyperbaric oxygen therapy can be used as salvage treatment for sudden sensorineural hearing loss
-
• The mean gain seems to be marginal, but some patients may have better recovery
A study by Yeo et al. investigated 156 patients who received oral corticosteroids for 10 days (with a 5-day period for tapering), carbogen treatment and vasodilators.Reference Yeo, Lee, Jun, Park and Park 18 The authors reported recovery of hearing (at least a 15 dB PTA gain) to their final hearing level after: 10 days in 66 cases (42.3 per cent), 10–38 days in 43 cases (27.56 per cent) and 38–120 days in 12 cases (7.69 per cent). The high rate of continuing recovery after treatment at this magnitude may have contributed to the recovery rates after salvage therapy reported in other studies. Therefore, we think that larger studies should be conducted that investigate the natural course of the SSNHL, with or without salvage treatment.
Even though the benefit reported in other studies is higher, we were able to find some significant recovery gain in the lower frequencies and some recovery in outcome according to Siegel's criteria. Thus, some patients may have benefited from hyperbaric oxygen treatment.
Conclusion
Hyperbaric oxygen salvage therapy utilised after corticosteroid therapy failure may be beneficial in some patients. Studies on the natural course of the disease after systemic corticosteroid therapy are needed, and further studies should comprise a greater number of patients.
