Idiopathic sudden sensorineural hearing loss (ISSNHL) is a sensorineural hearing dysfunction of unknown etiology characterized by a sudden onset and mostly unilateral hearing loss. A recent epidemiological survey in Japan showed that the incidence of ISSNHL occurs in 60.9 per 100,000 of the population (Nakashima et al., 2014). The pathogenesis of ISSNHL is still unknown; however, administration of corticosteroids, hyperbaric oxygen therapy, vasodilators, and intratympanic steroid injections are often used to treat ISSNHL (Kitoh et al., 2017; Chandrasekhar et al., 2019; Herrera et al., 2019). Although these treatments have been proposed to constitute the possible intracochlear mechanisms of the disease, the prognosis of ISSNHL remains insufficient (Kitoh et al., 2017).

Molecular hydrogen (H₂) is an antioxidative agent that can prevent oxidative stress-induced diseases, such as brain infarction (Ono et al., 2017), myocardial infarction (Katsumata et al., 2017), post-cardiac arrest syndrome (Tamura et al., 2020), respiratory function in post-COVID-19 patients (Botek et al., 2022), and rheumatoid arthritis (Ishibashi et al., 2014), in human clinical trials. In addition, H₂ has been reported to be effective in reducing hearing loss due to transient cochlear ischemia (Ogawa et al., 2018), loud sound (Lin et al., 2011; Zhou et al., 2012; Chen et al., 2014; Kurioka et al., 2014; Fransson et al., 2021), and ototoxic agents (Qu et al., 2012; Kikkawa et al., 2014; Fransson et al., 2017). Ogawa et al. (2018) reported that H₂ was effective for hearing loss induced by cochlear ischemia in an animal model, which is thought to be the main cause of ISSNHL.

In previous animal studies, transient cochlear ischemia induced hearing loss via the depletion of the adenosine triphosphate (ATP) supply, glutamate excitotoxicity, and cell damage by the generation of free radicals (Hakuba et al., 1997, 2000; Maetani et al., 2003; Morizane et al., 2005). As H₂ can selectively scavenge peroxynitrite and hydroxyl radicals produced after transient ischemia (Ohsawa et al., 2007), it has potential effectiveness for ISSNHL treatment. However, the effectiveness of H₂ in sensorineural hearing loss has not yet been reported in humans.

In previous clinical studies using H₂ inhalation, 1.3–3% of H₂ was used for various diseases (Ishibashi et al., 2014; Katsumata et al., 2017; Ono et al., 2017; Tamura et al., 2020; Botek et al., 2022). A randomized controlled study reported that 3% H₂ gas (1 h twice a day) was effective in patients with acute cerebral infarction (Ono et al., 2017). In addition, no adverse effects due to H₂ gas have been reported in the previous studies (Ishibashi et al., 2014; Katsumata et al., 2017; Ono et al., 2017; Tamura et al., 2020; Botek et al., 2022) which is in contrast to the conventional treatments, including glucocorticoids, which have side effects. Our hypothesis is that H₂ gas inhalation is a safe and effective treatment for ISSNHL. Thus, we investigated the effects of 3% H₂ gas inhalation in patients with ISSNHL.

Overall, 82 patients consented to participate in the study (Figure 1) and were randomly assigned to either the H₂ or control groups between June 2019 and December 2020. One patient in the H₂ group and one in the control group withdrew consent, while 39 and 41 patients in the H₂ and control groups, respectively, completed the intervention. In addition, eight patients in the H₂ group were excluded from the analysis due to the missed 3-month follow-up (n = 3), incomplete data (n = 4), and diagnosis of Meniere’s disease (n = 1). In contrast, seven patients in the control group were also excluded because of the missed 3-month follow-up (n = 6) and incomplete data (n = 1). Ultimately, 65 patients (31 and 34 in the H₂ and control groups, respectively) were included in the analysis.

Table 1 shows the baseline characteristics of the two groups. Days from onset to treatment were 6.2 (95% CI: 5.1–7.4) and 4.7 (95% CI: 3.7–5.7) days in the H₂ and control groups, respectively, which show statistical difference between the two groups (P = 0.049). Other parameters such as age, affected side, diabetes, hypertension, hyperlipidemia, vertigo/dizziness, PTA threshold of the affected and contralateral sides, and intratympanic steroid injection as salvage therapy were not significantly different between the two groups.

This study proposes that inhalation of H₂ gas and treatment with systemic glucocorticoids and PGE1 effectively improve hearing outcomes in patients with ISSNHL. To the best of our knowledge, this is the first study to examine the effectiveness of H₂ gas therapy in ISSNHL treatment.

However, the pathogenesis of ISSNHL remains unknown. Although various causes have been proposed, circulatory disturbance is recognized as the most plausible cause of the disease (Kim et al., 1999). Steroids and vasoactive agents are often used for ISSNHL treatment (Kitoh et al., 2017; Okada et al., 2017). Nevertheless, approximately 40% of patients with ISSNHL have a poor hearing prognosis (Kitoh et al., 2017; Okada et al., 2017). Interestingly, intratympanic steroid (ITS) therapy has recently been used as initial or salvage therapy for ISSNHL; however, the treatment effect of ITS is inadequate (Chandrasekhar et al., 2019).

Inhalation of H₂ gas was effective in changes in hearing threshold, although the duration from onset to the initial visit was significantly longer in the H₂ group than in the control group. In animal studies, it has been reported that transient cochlear ischemia causes inner hair cell loss through the depletion of the ATP supply, glutamate excitotoxicity, and cell damage by the generation of free radicals (Hakuba et al., 1997, 2000; Maetani et al., 2003; Morizane et al., 2005). In addition, cochlear blood flow disturbance causes a significant increase in nitric oxide (NO) production in the perilymph, which is attributed to the inducible nitric oxide synthase (iNOS) pathway (Kim et al., 1999). NO production through iNOS is believed to cause injury, which follows ischemia and contributes to neuronal damage through peroxynitrite production (Niwa et al., 1999). Furthermore, our previous study showed that immunostaining for iNOS was strongly expressed on days 1 and 4 after transient cochlear ischemia (Morizane et al., 2005). Notably, H₂ can selectively scavenge peroxynitrite and hydroxyl radicals produced following transient cochlear ischemia (Ohsawa et al., 2007). In our previous animal study, we also reported that H₂ is effective for acute hearing loss prevention due to transient cochlear ischemia (Ogawa et al., 2018). Moreover, the therapeutic effects of H₂ on the inner ear have been reported in animal studies related to noise-induced hearing loss (Lin et al., 2011; Zhou et al., 2012; Chen et al., 2014; Kurioka et al., 2014; Fransson et al., 2021) and ototoxic agents (Qu et al., 2012; Kikkawa et al., 2014; Fransson et al., 2017) through its antioxidative or anti-inflammatory effects (Lin et al., 2011; Qu et al., 2012; Chen et al., 2017). Therefore, H₂ therapy may be effective for ISSNHL through its antioxidative or anti-inflammatory effects, which include those of the iNOS pathway.

The effect of H₂ was remarkable in patients with diabetes. Generally, diabetes and hypertension have been reported to contribute to the causes of ISSNHL through microcirculatory disturbances and cellular stress (Teranishi et al., 2007). Both diabetes and hypertension are associated with endothelial dysfunction and oxidative stress (Petrie et al., 2018). Yet, H₂ was effective only in patients with diabetes whereas ineffective in those with hypertension and hyperlipidemia. Since the mechanisms by which diabetes, hypertension, and hyperlipidemia affect the inner ear pathologies in ISSNHL remain unknown, we cannot explain these results. In addition, further large-scale study in human and animal studies are warranted because of the small number of patients.

Furthermore, the ratio of complete recovery was significantly higher in the H₂ group among patients with severe hearing loss, although this outcome was not included in the secondary outcomes. Since the severe hearing loss at the initial visit is believed to be a poor prognostic factor, inhalation of H₂ gas may improve the ISSNHL prognosis. Notably, in this study, there were no adverse effects due to the inhalation of H₂ gas. In addition, no side effects have been reported in previous studies that administered H₂ gas to humans (Ishibashi et al., 2014; Katsumata et al., 2017; Ono et al., 2017; Tamura et al., 2020; Botek et al., 2022). This study’s results suggest that the inhalation of H₂ gas is a safe and effective treatment for severe ISSNHL. However, large-scale studies are required to verify the efficacy of H₂ gas in treating ISSNHL because of the small number of patients.

This study has several limitations. First, since the number of patients was small, the study population may be insufficient for analysis. In the power analysis of our primary outcome (1-ß), was 0.51, suggesting that large-scale studies are required. In addition, multivariate analysis is more suitable for analyzing the effect of complications on hearing outcomes. However, the number of patients was insufficient for the analysis. Further large-scale studies are needed to confirm the effect of complications. Second, randomization was performed according to the hearing thresholds at the initial visit and age. Although other factors that might influence the hearing outcomes were not significantly different between the H₂ and control groups, excluding the duration from onset to initial visit, further studies that randomize by including these factors will be required. Third, the optimal concentration of H₂ and duration of inhalation of H₂ gas are still unknown for treating ISSNHL. Therefore, further studies are needed to verify these findings. Fourth, we did not perform a speech perception test, as it is not included in the Japanese criteria for hearing outcomes.

This is the first clinical report to analyze the efficacy of H₂ for ISSNHL. In animal studies, there have been reports that H₂ alleviates hearing loss due to various causes (Lin et al., 2011; Qu et al., 2012; Zhou et al., 2012; Chen et al., 2014; Kikkawa et al., 2014; Kurioka et al., 2014; Fransson et al., 2017, 2021; Ogawa et al., 2018). However, prior to the present study, no other study had evaluated the efficacy of H₂ in humans. Further clinical trials to confirm the efficacy of H₂ in hearing loss by other causes, such as noise-induced or drug-induced, are required. The route of H₂ administration is roughly classified into three types: inhalation of H₂ gas, drinking H₂-dissolved water, and injection of H₂-dissolved saline (Iketani and Ohsawa, 2017). Inhalation of H₂ gas is an easy, safe, and reliable way to administer H₂. No side effects of H₂ inhalation were reported in this study or in other clinical trials (Ishibashi et al., 2014; Katsumata et al., 2017; Ono et al., 2017; Tamura et al., 2020; Botek et al., 2022). Since there are few disadvantages to inhalation of H₂ gas, it is expected to be applied to various types of hearing loss.

In conclusion, inhalation of H₂ gas was effective for treating ISSNHL, and the changes in hearing threshold were significantly better in the H₂ group than in the control group. The effect of H₂ gas inhalation was remarkable in patients with diabetes and in those with severe hearing loss. However, H₂ gas inhalation was ineffective in patients with hypertension but was effective in those without hypertension. Furthermore, no side effects were observed after the inhalation of H₂ gas. Therefore, these results suggest that the inhalation of H₂ gas may be safe and effective in ISSNHL treatment.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Certified Review Board of Ehime University. The patients/participants provided their written informed consent to participate in this study.

MO and NHa designed the study. MO, HO, TT, EN, TY, JH, YS, NHo, MT, and HY collected the data. MO and TF analyzed the data. S-IH prepared the equipment. MO wrote the draft of this manuscript. NHa revised the manuscript. All authors contributed to the manuscript and approved its submission.