Comparison of two different steroid treatments with hyperbaric oxygen for idiopathic sudden sensorineural hearing loss

The purpose of the study was to assess the efficacy of the association of intratympanic (IT) steroid and hyperbaric oxygen (HBO) therapy in patients presenting with idiopathic sudden sensorineural hearing loss (ISSNHL), and to compare this protocol with another consisting of intravenous (IV) steroid administration and HBO therapy. A total of 80 patients diagnosed with ISSNHL were included in this prospective trial. Patients were divided into three categories: a mild-to-moderate ISSNHL group with a pure-tone average (PTA) ≤60 decibels (dB), a severe ISSNHL group with a PTA of 60–80 dB, and a profound ISSNHL group with a PTA ≥81 dB. The first protocol consisted of 20 sessions of HBO therapy together with IV methylprednisolone 1 mg/kg body weight and a 10 mg taper every 3 days for 10 days. The second protocol consisted of HBO therapy for 20 sessions, together with an IT injection of dexamethasone at a dose of 4 mg/mL, 0.5–0.7 mL once a day for 7 consecutive days, performed 3 h before the HBO therapy. In the mild-to-moderate ISSNHL patients, the mean hearing gain and successful treatment rate was 19 (0–27) dB and 78.9 %, respectively in the IT + HBO treatment group, and 18 (3–44) dB and 70.5 % in the IV + HBO therapy group. In the severe ISSNHL patients, the mean hearing gain and successful treatment rate was 33 (1–54) dB and 81.8 %, respectively in the IT + HBO treatment group and 33.5 (7–57) dB and 58.2 % in the IV + HBO group. In the profound ISSNHL patients, the mean hearing gain and successful treatment rate was 36 (4–69) dB and 40 %, respectively in the IT + HBO therapy group, and 39.5 (0–92) dB and 72.7 % in the IV + HBO treatment group. The results demonstrated that patients with severe hearing loss success rate was superior in the group submitted to IT + HBO treatment, conversely IV + HBO therapy may be benefit for patients with profound hearing loss. Nevertheless, these clinical results were not statistically significant.