Long-term administration of vitamin B12 and adenosine triphosphate for idiopathic sudden sensorineural hearing loss: a retrospective study

Study design and ethical considerations

This retrospective study (B20-056) was approved by the Institutional Review Board of Kitasato University Hospital. The need for informed consent was waived owing to the retrospective nature of the study.

Patients

The subjects were 117 patients with ISSNHL who underwent pure-tone audiometry at Kitasato University Hospital between 2015 and 2018. The inclusion criteria were as follows: (1) sudden sensorineural HL of 30 dB or greater in at least three consecutive frequencies; (2) underwent early therapeutic management, that is, treatment was started within 2 weeks after onset; (3) age over 18 years; and (4) follow-up duration longer than 6 months. The exclusion criteria were as follows: (1) age younger than 17 years; (2) middle ear or retro-cochlear pathology; (3) history of Meniere’s disease or autoimmune HL; (4) history of genetic or fluctuating HL; (5) history of hearing aid wearing or otologic surgery; and (6) history of intratympanic steroid injection for salvage treatment after systemic steroid administration.

Assessment of hearing function

Pure-tone audiometry was performed using a conventional device (AA-78; Rion, Tokyo, Japan) in a soundproof room. First, the hearing thresholds were obtained through air conduction (AC) and bone conduction (BC) in frequencies of 0.25, 0.5, 1, 2, and 4 kHz for both ears. To prevent cross-hearing phenomenon causing an erroneous measurement, masking noise was used to occupy the non-test ear while the other ear was tested, as necessary. Briefly, the necessity of masking during the measurement of the AC thresholds was identified based on the minimum interaural attenuation level of 40 dB for retesting through AC. Meanwhile, in the measurement of BC thresholds, a masking process was applied using ABC methods (Kurioka et al., 2020b). Thresholds were obtained across all the frequency octaves from 0.25 kHz to 4 kHz, and the arithmetic average AC and BC thresholds were calculated from the thresholds at 0.25, 0.5, 1, 2, and 4 kHz. The grade of HL, defined according to the Japanese Ministry of Health and Welfare guidelines (Nakashima et al., 2014), was then determined using the initial audiogram data. Hearing recovery was calculated as the difference between average hearing thresholds at different time points, including initial day of treatment and 2, 4, and 6 months after initial treatment. Siegel’s criteria were employed to assess treatment results 6 months after treatment (Table 1), and the patients were accordingly classified into two groups as the recovery group (i.e., complete and partial recovery) and the no recovery group (i.e., slight recovery and no improvement). These two groups were treated with same drug regimen, except for the duration of vit B12 and ATP administration.

Treatment

All patients were treated with a 10-day course of systemic corticosteroids (betamethasone 8 mg via intramuscular injection for the first day followed by betamethasone 4 mg via oral administration for the first 3 days, tapered to 2 mg for the second 3 days, and finally to 1 mg for the last 3 days). Oral administration of vit B12 (1.5 mg daily) and ATP (300 mg daily) was started from initial day of treatment, concurrent with the corticosteroids. To investigate the therapeutic effects of long-term administration of vit B12 and ATP on hearing outcomes in ISSNHL, the patients were divided into three groups according to the administration duration as the short-term (<8 weeks), middle-term (≥8 weeks, <16 weeks) and long-term (≥16 weeks) administration groups.

Statistical analyses

The Chi-squared test was used to evaluate clinical characteristics and possible prognostic factors. Student’s t-tests and non-parametric Mann–Whitney U tests were used to investigate continuous prognostic factors. For comparisons between more than two groups, a one-way analysis of variance was used followed by Dunn’s multiple comparisons for the post hoc test. The parameters that were statistically significant in the univariate analysis were entered into a binary logistic regression model for multivariate analysis. All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software Inc., La Jolla, CA) or JMP 14.2 (SAS Institute Japan Inc., Tokyo, Japan). A p value of <0.05 was considered statistically significant.

Patient characteristics and treatment outcome

The mean age was 58.2 ± 16.6 years, and there were 66 and 51 male and female patients, respectively. The patients’ clinicodemographic characteristics are presented in Table 2. The mean interval between symptom onset and initial treatment was 4.2 ± 3.8 days. As accompanying symptoms and complications, 56 patients (47.9%) had vertigo and 32 patients (27.4%) had diabetes. The mean hearing threshold at the initial examination was 80.7 ±  24.0 dB. With respect to initial HL grade, 8 patients had grade 1; 18 patients, grade 2; 45 patients, grade 3; and 46 patients, grade 4. The average duration of vit B12 and ATP administration was 4.2 ± 2.4 months. With regard to final recovery according to Siegel’s criteria, the overall recovery rate (complete + partial recovery) was 32.5%. Specifically, 19.7%, 12.8%, 39.3%, and 28.2% of the patients achieved complete recovery, partial recovery, slight recovery, and no improvement, respectively (Table 1).

Prognostic factors

In univariate analysis, the initial hearing threshold and initial grade of HL was significantly lower in the recovery group than that in the no recovery group (p <  0.0001). With regard to the associated symptoms, there was a significantly higher incidence of vertigo in the no recovery group than in the recovery group (p = 0.04). The other variables, including age, sex, days to initiation of treatment, and diabetes were not significantly different between the two groups. The duration of vit B12 and ATP administration was slightly longer in the no recovery group than that in the recovery group, but the difference was not significant. The significant variables in the univariate analysis were included in the multivariate analysis. The results showed that higher initial hearing threshold and higher initial grade of HL were associated with a poor prognosis in ISSNHL patients (Table 3).

Administration duration of vit B12 and ATP

The short-term, middle-term, and long-term administration groups involved 15, 41, and 38 patients, respectively. The patients with complete recovery within 24 weeks after onset (n = 23) were excluded because the vit B12 and ATP administration was terminated following the confirmation of complete recovery. Specifically, the duration of drug administration was shortened due to complete recovery, but the shortened administration did not contribute to complete recovery. Therefore, it was not appropriate to include the patients with complete recovery within 24 weeks after onset for the analysis of the effect of drug administration duration. In this study, the duration of administration was determined by the patients after consultation with their physician, even though they did not achieve complete recovery. The clinical features of each group are presented in Table 4. The mean duration of drug administration in the short, middle, and long-term groups was 1.9, 3.3, and 6.3 months, respectively. In univariate analysis, the duration of drug administration was significantly different among the groups (p < 0.0001), but the clinical characteristics and hearing results were comparable (p > 0.05). This indicated that the duration of drug administration was not influenced by the patient’s clinical characteristics and did not affect the overall hearing outcome. Ultimately, these results show that long-term administration of vit B12 and ATP had no significant impact on the hearing prognosis within our observation period.

We also investigated the time course of hearing recovery in each group. As shown in Fig. 1A, the thresholds (dB) of total hearing recovery within 24 months after symptom onset in the short-, middle-, and long-term administration groups were 18.9 ± 5.8, 22.4 ± 2.3, and 22.7 ± 3.0, respectively, with no significant differences (p = 0.76). The time course of hearing recovery in the overall population was 18.9 ± 1.7, 3.1 ± 0.7, and 0.04 ± 0.4 within 0–2, 2–4, and 4–6 months after symptom onset, respectively (p <  0.0001; Fig. 1B). Hearing recovery within the initial 2 months after symptom onset was significantly higher than that within 2–4 months (p <  0.0001) and 4–6 months (p <  0.0001). Moreover, hearing recovery within 2–4 months after symptom onset was significantly higher than that within 4–6 months (p = 0.04). This result indicated that hearing recovery in the initial 2 months after onset is higher and that the probability of recovery decreases as the duration after symptom onset increases. The time course of hearing recovery of both the affected and contralateral ears in the three groups are shown in Fig. 2. There was no significant difference in hearing recovery among the three groups until 16 weeks after onset. However, hearing recovery of the affected ears was significantly lower in the short-term administration group than that in the long-term administration group at 16–24 weeks after symptom onset (p = 0.03). Meanwhile, there was no significant difference in hearing recovery of the contralateral ears among the three groups throughout the study period. This indicated a slow progression of HL in the affected ear and that long-term administration of vit B12 and ATP might prevent progressive HL in the affected ear in ISSNHL. We performed an additional analysis of hearing recovery at each frequency. As shown in Fig. 3, hearing recovery appeared to be greater at 0.5 and 1.0 kHz than that at 2.0 and 4.0 kHz until 8 weeks after onset. In addition, significant differences between long-term and short-term drug administration were observed at only 2.0 kHz during the 4–6 months after the onset of ISSNHL.