Efficacy of Combined Hyperbaric Oxygen, per Os Steroid, and Prostaglandin E1 Therapy for Idiopathic Sudden Sensorineural Hearing Loss and Prognostic Factors for Recovery

Abstract

Background: Idiopathic sudden sensorineural hearing loss (ISSNHL) is an abrupt unilateral hearing loss of unknown origin. Combination therapy with hyperbaric oxygen (HBO), systemic steroids (SS), and prostaglandin E1 (PGE1) has been used in Japan; however, its prognostic factors remain unclear. Objective: To evaluate the efficacy of HBO combined with SS and PGE1 and to identify prognostic factors for hearing recovery in patients with ISSNHL. Methods: This retrospective study included 116 patients treated within 14 days of ISSNHL onset. Sixty patients received HBO, SS, and PGE1 (HBO group), and 56 received SS and PGE1 alone (No-HBO group). Hearing outcomes were assessed using PTA (arithmetic mean hearing at 250–4000 Hz) and graded by Siegel’s criteria. Prognostic factors were analyzed by multivariate logistic regression. Results: The HBO group showed significantly better hearing grade outcomes (p = 0.007) and greater PTA improvement (p = 0.003) than the No-HBO group. Vertigo and higher initial PTA were identified as independent predictors of poor hearing outcomes. Patients without vertigo showed significantly greater improvement at 2000 Hz (p = 0.009). Receiver operating characteristic analysis revealed an optimal initial PTA cutoff of ≥90.5 dB for predicting poor hearing outcome. Conclusions: HBO combined with SS and PGE1 significantly improves hearing outcomes in ISSNHL. However, the presence of vertigo and severe initial hearing loss remain poor prognostic indicators. These findings suggest that while the addition of HBO may enhance hearing outcomes, prognosis remains limited in severe cases. Further prospective studies are needed to confirm these results.

1. Introduction

Idiopathic sudden sensorineural hearing loss (ISSNHL) is defined as sudden unilateral hearing loss for which no cause can be identified even after a comprehensive examination [1]. Several etiologies of ISSNHL have been proposed, including viral infections, vascular disorders, autoimmune diseases, inner ear diseases, and central nervous system abnormalities [2,3]. The traditional initial treatment approach for ISSNHL involves the administration of hyperbaric oxygen therapy (HBO), systemic steroids (SS), systemic vasodilators, such as prostaglandin E1 (PGE1), or combinations of two of these therapies [4,5].

Although HBO therapy is not a global standard treatment for ISSNHL, HBO is mentioned as an “Option,” which represents the highest level of evidence for initial treatment, in both American and Japanese guidelines. The AAO-HNS 2019 guideline lists HBO as an Option for initial therapy within 2 weeks or as salvage within 1 month [4]. The Japanese clinical practice guideline likewise places HBO as an optional therapy, generally within 2 weeks of onset [6]. HBO increases arterial oxygen content and has been shown to raise the partial pressure of oxygen in the cochlear perilymph during treatment, which may enhance cochlear metabolic support. This increase has been directly measured in animal models; however, evidence for endolymphal oxygen changes is more limited than that for perilymph measurements [7].

Current guidelines recommend considering the administration of HBO alongside steroid treatment within two weeks after the onset of ISSNHL, or HBO alone within one month after onset [4]. Notably, treatment aimed at improving ischemia using PGE1 has also been attempted. Moreover, a systematic review showed potential benefits of vasodilators combined with steroids for the treatment of ISSNHL [4,8].

The prognosis of ISSNHL depends on several factors, including patient age, the presence of vertigo at onset, the degree of hearing loss, and the interval between onset and initiation of treatment [9]. It has been reported that vertigo negatively affects hearing recovery in patients with ISSNHL treated with HBO and SS or other pharmacological agents [1,10]. Previous work has suggested that cochlear and vestibular functions can be partly dissociated; therefore, the presence of vertigo in ISSNHL may indicate a more widespread or severe inner-ear injury [3]. Although the additional effect of HBO in combination with SS and PGE1 has been reported, prognostic factors for hearing recovery in patients with ISSNHL treated with this combination remain unclear [11].

In the present study, we therefore evaluated the impact of these clinical variables, including age, the presence of vertigo, initial pure-tone average (PTA), time from onset to treatment, and comorbidities such as diabetes mellitus (DM), hypertension (HT), and dyslipidemia (DL), on hearing recovery in patients with ISSNHL treated with or without HBO. Among these factors, particular attention was paid to vertigo, because previous studies have reported that ISSNHL accompanied by vertigo tends to present with more severe hearing loss and a poorer prognosis.

2. Materials and Methods

This study was approved by the Ethical Review Board of Juntendo University Faculty of Medicine (identification E24-0386-S01), which waived the requirement for informed consent; therefore, an opt-out approach was employed.

2.1. Patients

Inclusion criteria:

• Age ≥ 18 years.
• A diagnosis of unilateral idiopathic sudden sensorineural hearing loss, defined as ≥30 dB sensorineural hearing loss occurring within 72 h in at least three consecutive frequencies, according to our institutional protocol.
• Initiation of treatment at our hospital within 14 days of symptom onset.
• Availability of complete audiometric data at baseline and at least one month post-treatment.
• Brain MRI demonstrating no retrocochlear pathology (e.g., vestibular schwannoma, stroke, demyelinating disease).
• Availability of relevant medical records.

Exclusion criteria:

• Receipt of intratympanic corticosteroid therapy.
• History of previous sensorineural hearing loss in the affected ear.
• Identifiable causes of hearing loss (e.g., Ménière’s disease, ototoxic drug exposure, noise-induced trauma).
• Comorbidities contraindicating hyperbaric oxygen therapy.
• Inability to complete the one-month follow-up assessment.

Between January 2013 and June 2023, 116 patients who met the above criteria were treated at our hospital, were followed for at least one month after initiating treatment. Among them, 60 patients were treated according to the treatment protocol using HBO, SS, and PGE1 (HBO group), whereas 56 patients were treated with SS and PGE1 alone (No-HBO group). Patients who received intratympanic corticosteroids were excluded to ensure homogeneity of steroid dosage and method of delivery in the study cohort.

A retrospective evaluation of 116 patients was performed. Patients were evaluated by sex, age, the interval from onset to treatment, the presence of vertigo, and the presence of DM, HT, and DL. Use of SS, PGE1, and HBO was recorded.

2.2. Treatment

At our institution, we use SS and PGE1 as the standard treatment for ISSNHL, and, in principle, recommend adjunctive HBO therapy for cases in which it is indicated. However, the final decision regarding whether to undergo HBO is made by the patient after receiving an explanation from the attending physician. Additionally, chest CT is performed for all patients prior to HBO therapy, and HBO is not administered to those who present with findings that constitute contraindications, such as pulmonary cysts or severe pulmonary emphysema.

Most patients received treatment on an outpatient basis; however, those who had difficulty attending daily visits or those with comorbid diabetes were admitted for inpatient management. Both intravenous PGE1 infusion and HBO were administered once daily, and, depending on each patient’s schedule, treatments were generally performed consecutively on weekdays.

SS consisted of oral prednisolone. The dose of prednisolone for SS was tapered over 7 days as follows: 60 mg/day (30 mg twice daily) for the first 3 days, followed by 30 mg/day (15 mg twice daily) for 2 days, and 15 mg/day (15 mg once daily) for the last 2 days. PGE1 (40 μg for 7 days) was dissolved in 200 mL of saline and administered intravenously by drip infusion. HBO was administered at a pressure of 2 atm absolute (ATA) for 60 min. Fifteen patients discontinued treatment because of headache or otitis media with effusion; however, they were included in the HBO group if they underwent the treatment more than once.

2.3. Patient Evaluation

Data extracted from the patients’ medical records were evaluated. The evaluated data included sex, age, the interval from onset to treatment, the presence of vertigo, the presence of DM (glycated hemoglobin level > 6.5%; National Glycohemoglobin Standardization Program), HT, DL (low-density lipoprotein cholesterol level ≥ 140 mg/dL, high-density lipoprotein cholesterol level < 40 mg/dL, or triglycerides level ≥ 150 mg/dL), number of HBO sessions. In addition, the use of SS, PGE1, and HBO was recorded.

The initial and final pure-tone audiogram of the patients were assessed to evaluate their initial and post-therapy hearing. This study used PTA, which is the arithmetic mean hearing at five frequencies (250, 500, 1000, 2000, and 4000 Hz), to evaluate hearing acuity. The hearing level one month after onset was considered fixed, and hearing recovery was graded according to these criteria. PTA improvement (dB) was calculated as (initial PTA) − (post-therapy PTA). The degree of hearing recovery was evaluated using Siegel’s criteria [12,13]. A post-therapy PTA of ≤25 dB one month after onset indicated complete recovery. PTA improvement of >15 dB and post-therapy PTA of >25 dB, but <45 dB, indicated a partial recovery. PTA improvement > 15 dB and post-therapy PTA > 45 dB, but <75 dB, indicated a slight recovery. PTA improvement of ≤15 dB or post-therapy PTA > 75 dB was considered indicative of no recovery. Based on Siegel’s criteria, a good outcome was defined as complete or partial recovery, whereas a poor outcome was defined as slight recovery or no improvement.

2.4. Statistical Analyses

All statistical analyses were performed using SPSS Statistics, version 25 (International Corp., Armonk, NY, USA). Normality of continuous variables was assessed using the Shapiro–Wilk test and visual inspection of histograms and Q–Q plots. Because the distributions of key continuous variables (e.g., initial PTA and PTA improvement) were non-normal, continuous variables were presented as the median and interquartile range (IQR), and between-group comparisons were performed using the Mann–Whitney U test. Correlations between continuous variables were evaluated using Spearman’s rank correlation coefficient. Categorical variables were summarized as numbers (%), and between-group comparisons were performed using the χ² test. Variables with p < 0.05 in the univariate analyses were entered into a multivariate logistic regression model using a backward stepwise selection method. To identify prognostic factors for hearing recovery in patients with ISSNHL treated with the combination of HBO, SS, and PGE1, multivariate logistic regression analysis was performed to assess the associations between hearing recovery and variables including sex, age, the interval from onset to treatment, the presence of vertigo, DM, HT, DL, initial hearing level, and number of HBO sessions. The optimal cutoff value for initial PTA was determined using the Youden index of the receiver operating characteristic (ROC) curve. p < 0.05 was considered statistically significant.

3. Results

3.1. Patient Profiles

Table 1. Profiles of the patients in the HBO and the No-HBO groups.

Characteristic	HBO Group (n = 60)	No-HBO Group (n = 56)	p-Value
Sex (male), n (%)	27 (45.0%)	31 (55.3%)	0.26 †
Age (years)	65(56.0–70.5)	67.5 (58.0–74.0)	0.09 ‡
Incidence of vertigo, n (%)	30 (50.0%)	11 (19.6%)	<0.01 ***,‡
Incidence of DM, n (%)	20 (33.3%)	20 (35.7%)	0.78 ‡
Incidence of HT, n (%)	18 (30.0%)	18 (32.1%)	0.80 ‡
Incidence of DL, n (%)	10 (16.7%)	6 (10.7%)	0.41 ‡
Interval between onset and treatment (days)	6(4.0–9.0)	3.0 (2.0–6.0)	0.13 ‡
Initial hearing level (dB)
PTA	90.5(66.7–105)	73 (52.0–106.0)	0.11 ‡
125 Hz	70(52.5–70.0)	67.5 (40.0–70.0)	0.29 ‡
250 Hz	90(60.0–90.0)	95 (55.0–90.0)	0.25 ‡
500 Hz	97.5(70.0–110)	80 (65.0–110)	0.23 ‡
1000 Hz	90(65.0–110.0)	77.5 (50.0–110)	0.14 ‡
2000 Hz	85(65.0–110.0)	70 (50.0–110.0)	0.065 ‡
4000 Hz	92.5(62.5–110)	72.5 (52.5–110.0)	0.13 ‡
8000 Hz	95(65.0–100.0)	85 (65.0–100.0)	0.21 ‡

PTA, arithmetic mean hearing at five frequencies (250, 500, 1000, 2000, and 4000 Hz); DM, diabetes mellitus; HT, hypertension; DL, dyslipidemia. ***, p < 0.001; †, χ² tests; ‡, Mann–Whitney U test.

shows the patient profiles of the HBO and No-HBO groups. The number of patients with vertigo was significantly higher (p < 0.05) in the HBO group (30/60 [50.0%]) than in the No-HBO group (11/56 [19.6%]). There were no significant differences in age, sex, affected side, the interval between onset and treatment, or the presence of DM, HT, or DL between the two groups. In the HBO group, the means ± SD of the number of HBO sessions and the interval between onset and HBO were 7.3 ± 4.9 sessions and 6.6 ± 4.4 days, respectively. A brain MRI was performed for all patients and revealed no cranial nerve involvement.

3.2. Therapeutic Outcomes

Table 2. Hearing outcomes for the two groups.

	HBO Group (n = 60)	No-HBO Group (n = 56)	p-Value
Post-therapy PTA	44 (26.1–68.8)	50 (30.5–78.0)	0.39
PTA improvement (dB)	36.5 (16.5–46.0)	19.5 (7.0–38.5)	0.003 **
Hearing improvement of 125 Hz (dB)	25 (7.5–40.0)	10 (0–27.5)	0.005 **
Hearing improvement of 250 Hz (dB)	35 (15.0–52.5)	15 (0–40.0)	0.002 **
Hearing improvement of 500 Hz (dB)	40 (25.0–60.0)	30 (10.0–47.5)	0.001 **
Hearing improvement of 1000 Hz (dB)	37.5 (20.0–50.0)	20 (5–40.0)	0.003 **
Hearing improvement of 2000 Hz (dB)	30 (10.0–45.0)	15 (0–35.0)	0.009 **
Hearing improvement of 4000 Hz (dB)	20 (5.0–42.5)	10 (0–30.0)	0.027
Hearing improvement of 8000 Hz (dB)	10 (0–20.0)	5 (0–15.0)	0.31
Grade of hearing recovery
Complete recovery	16 (27%)	10 (17.9%)	0.007 **
Partial recovery	12 (20%)	11 (19.6%)
Slight recovery	21 (35%)	5 (8.9%)
No recovery	11 (18%)	30 (53.5%)

PTA, arithmetic mean hearing at five frequencies (250, 500, 1000, 2000, and 4000 Hz); **, p < 0.01; p-values were calculated using the Mann–Whitney U test.

shows the hearing outcomes for the two groups. PTA improvement (dB) was significantly greater in the HBO group (median [IQR]: 36.5 [16.5–46.0]) than in the No-HBO group (19.5 [7.0–38.5]). Similarly, the grade of hearing recovery was significantly better in the HBO group than in the No-HBO group.

Separate analyses of each frequency showed that PTA improvement at the 125–4000 Hz frequencies was significantly better in the HBO group than in the No-HBO group. PTA improvement for the 8000-Hz frequency did not differ between the two groups.

A multivariate logistic regression analysis was performed to confirm the association between HBO and hearing outcome in

Table 3. Univariate and multivariate logistic regression analyses examining the association between HBO and hearing outcomes.

Univariate Logistic Regression Analysis
Variable	Coefficient (β)	SE	Wald χ2	p-Value	Odds Ratio	95% CI
Age	0.008	0.013	0.360	0.548	1.008	0.982–1.034
Sex (male vs. female)	−0.071	0.376	0.035	0.851	0.932	0.446–1.947
HBO (yes vs. no)	0.377	0.378	0.994	0.319	1.458	0.695–3.061
Vertigo	−1.030	0.421	5.974	0.015 *	0.357	0.156–0.815
Diabetes mellitus	0.172	0.394	0.190	0.663	1.188	0.548–2.572
Hypertension	0.622	0.406	2.348	0.125	1.863	0.841–4.128
Dyslipidemia	0.018	0.544	0.001	0.973	1.019	0.351–2.958
Interval between onset and treatment (days)	−0.018	0.025	0.539	0.463	0.982	0.935–1.031
Initial PTA (dB)	−0.035	0.009	15.926	<0.001 ***	0.965	0.949–0.982
Multivariate Logistic Regression Analysis
Variable	Coefficient (β)	SE	Wald χ2	p-Value	Odds Ratio	95% CI
Intercept	2.271	0.729
HBO (yes vs. no)	1.124	0.477	5.549	0.018 *	3.076	1.208–7.834
Vertigo	−1.111	0.500	4.934	0.026 *	0.329	0.124–0.877
Initial PTA (dB)	−0.036	0.010	14.46263	<0.001 ***	0.960	0.939–0.981

HBO, hyperbaric oxygen therapy; PTA, arithmetic mean hearing at five frequencies (250, 500, 1000, 2000, and 4000 Hz); *, p < 0.05; ***, p < 0.001.

. After adjustment for the presence of vertigo and initial hearing level identified as significant in univariate analyses, HBO was independently associated with a good hearing outcome (odds ratio [OR] = 3.076, 95% confidence interval [CI] 1.208–7.834, p = 0.018).

3.3. Prognostic Factors for Hearing Recovery After ISSNHL Treated with a Combination of HBO, SS, and PGE1

The association between hearing recovery and the prognostic factors in the HBO group is shown in

Table 4. Association between patient characteristics and the therapeutic effect of the combination HBO, SS, and PGE1.

Characteristics		Outcomes				p-Value
		No Recovery (n = 11)	Slight Recovery (n = 21)	Partial Recovery (n = 12)	Complete Recovery (n = 16)
Sex	Male (n = 27)	5 (19%)	8 (30%)	5 (19%)	9 (33%)	0.463 †
	Female (n = 33)	6 (18%)	13 (39%)	7 (21%)	7 (21%)
Age		66 (45.5–75.5)	59 (42.0–75.0)	66 (61.0–69.5)	64 (61.5–67.5)	0.884 ‡
	≥60 years (n = 19)	7 (17%)	10 (24%)	11 (27%)	13 (32%)	0.052 †
	<60 years (n = 41)	4 (21%)	11 (58%)	1 (15)	3 (16%)
Interval between onset and treatment (days)		4.0 (2.1–6.0)	4.0 (2.7–8.3)	4.5 (2.4–7.1)	2.5 (2.1–5.6)	0.588 ‡
Incidence of vertigo, n (%)	Yes (n = 30)	8 (27%)	14 (47%)	5 (17%)	3 (10%)	0.001 ** †
	No (n = 30)	3 (10%)	7 (23%)	7 (23%)	13 (43%)
Incidence of DM, n (%)	Yes (n = 20)	3 (15%)	6 (30%)	7 (35%)	4 (20%)	0.750 †
	No (n = 40)	8 (20%)	15 (38%)	5 (13%)	12 (30%)
Incidence of HT, n (%)	Yes (n = 18)	4 (22%)	4 (22%)	7 (39%)	3 (10%)	0.867 †
	No (n = 42)	7 (17%)	17 (41%)	5 (12%)	13 (31%)
Incidence of DL, n (%)	Yes (n = 10)	2 (20%)	2 (20%)	3 (30%)	3 (30%)	0.620 †
	No (n = 50)	9 (18%)	19 (38%)	9 (18%)	13 (26%)
Initial PTA (dB)		96 (65.0–106.0)	102 (97.0–106)	74.5 (56.5–3.5)	68.5 (49.0–84.5)	<0.001 ***,‡
Number of HBO sessions		7 (6.0–9.5)	6 (4.0–8.0)	6.5 (4.5–9.5)	5 (3.5–7.5)	0.173 ‡

PTA, arithmetic mean hearing at five frequencies (250, 500, 1000, 2000, and 4000 Hz); HBO, hyperbaric oxygen therapy; SS, systemic steroid; PGE1, prostaglandin E₁; **, p < 0.01; ***, p < 0.001; †, Mann–Whitney U tests; ‡, Spearman correlation coefficients.

. This table summarizes the clinical characteristics of the 60 patients who received HBO. Regarding the patients’ age, Spearman correlation coefficient analysis did not show any significant correlation between age and hearing recovery. However, a subgroup analysis in which patients were classified into two age groups using a cutoff age of 60 years revealed that patients <60 years tended to have better hearing recovery than those ≥60 years. Nevertheless, this result was not statistically significant (p = 0.052).

Within the HBO treatment group, univariate and multivariate logistic regression analyses were performed to identify prognostic factors associated with a good hearing outcome (

Table 5. Univariate and multivariate logistic regression analyses of factors associated with good hearing outcomes in the HBO group.

Univariate Logistic Regression Analysis
Variable	Coefficient (β)	SE	Wald χ2	p-Value	Odds Ratio	95% CI
Age	0.030	0.020	2.238	0.135	1.030	0.991–1.071
Sex (male vs. female)	−0.379	0.522	0.529	0.467	0.684	0.246–1.903
Vertigo	−1.705	0.566	9.069	0.003 *	0.182	0.060–0.551
Diabetes mellitus	0.503	0.552	0.831	0.362	1.654	0.561–4.875
Hypertension	0.511	0.568	0.810	0.368	1.667	0.548–5.070
Dyslipidemia	0.647	0.706	0.840	0.359	1.909	0.479–7.610
Interval between onset and treatment (days)	−0.041	0.063	0.414	0.520	0.960	0.849–1.086
Initial PTA (dB)	−0.062	0.017	13.461	<0.001 ***	0.940	0.909–0.972
Number of HBO sessions	−0.040	0.055	0.520	0.471	0.961	0.863–1.071
Multivariate Logistic Regression Analysis
Variable	Coefficient (β)	SE	Wald χ2	p-Value	Odds Ratio	95% CI
Intercept	5.353	1.520
Vertigo	−1.536	0.662	5.389	0.020 *	0.215	0.059–0.787
Initial PTA (dB)	−0.057	0.017	11.350	<0.001 ***	0.944	0.913–0.976

HBO, hyperbaric oxygen therapy; PTA, arithmetic mean hearing at five frequencies (250, 500, 1000, 2000, and 4000 Hz); *, p < 0.05; ***, p < 0.001.

). In an univariate analysis, the presence of vertigo and initial PTA were significantly associated with hearing outcome, therefore, these variables were entered into the multivariate logistic regression analysis. No significant associations were observed for age, sex, comorbidities, interval between onset and treatment, or the number of HBO sessions.

In the multivariate logistic regression analysis, both the presence of vertigo (OR = 0.22, 95% CI 0.06–0.79, p = 0.020) and initial PTA (OR = 0.94 per 1-dB increase, 95% CI 0.91–0.98, p < 0.001) remained independently associated with hearing outcome.

Patients without vertigo showed significantly better hearing recovery than those with vertigo (p = 0.001, Figure 1A). Analysis of hearing recovery at different frequencies revealed that patients without vertigo showed significantly better hearing improvement at 2000 Hz than those with vertigo (p = 0.009, Figure 1B).

Specifically, patients with a lower initial PTA showed better hearing recovery. Moreover, the patients who showed complete hearing recovery had the lowest mean initial PTA. A ROC analysis was performed to determine the optimal cut-off value of the initial PTA for the prediction of hearing recovery. The area under the curve was 0.849, which indicated excellent predictive ability (Figure 2B). The analysis revealed that the optimal cutoff for the initial PTA for predicting poor hearing recovery was ≥90.5 dB.

4. Discussion

4.1. Summary of Main Findings

This study focused on combined therapy with HBO, SS, and PGE1 for ISSNHL. The effectiveness of HBO combined with SS and PGE1 for treating ISSNHL was evaluated. The results showed that patients treated with HBO, SS, and PGE1 had significantly better hearing recovery than those treated with SS and PGE1 alone. We also investigated the prognostic factors for hearing recovery in patients treated with combined therapy with HBO, SS, and PGE1 for ISSNHL. Patients with vertigo and those with a higher initial PTA (cutoff ≥ 90.5 dB) treated with HBO, SS, and PGE1 showed significantly poorer hearing recovery. In addition, patients with vertigo exhibited significantly poorer hearing improvement at 2000 Hz. These findings suggest that the addition of HBO to SS and PGE1 may improve hearing outcomes in patients with ISSNHL. Furthermore, the presence of vertigo and a higher initial PTA (cutoff ≥ 90.5 dB) were identified as prognostic factors for hearing recovery in patients treated with the combined therapy.

4.2. Potential Mechanisms Underlying the Benefit of HBO

Although the underlying mechanism of efficacy of the combined therapy with HBO, SS, and PGE1 remains unclear, we hypothesized that this combination improves cochlear ischemia more effectively than SS and PGE1 alone. Vascular impairment of the cochlea may be a principal etiology of ISSNHL, because the cochlea is extremely sensitive to ischemia [14]. HBO facilitates the delivery of high partial pressures of oxygen to the cochlea and has been used to improve cochlear ischemia [15]. SS is also known to have effects at multiple sites in the inner ear and to be effective in hearing loss of vascular origin [3]. PGE1 has also been used to enhance cochlear blood flow in the treatment of ISSNHL [16].

4.3. Prognostic Factors Associated with Hearing Recovery: The Presence of Vertigo

Previous studies indicated that vertigo at the time of ISSNHL onset is a prognostic factor for hearing recovery [4,14,17,18,19,20]. The pathophysiological mechanism for this remains unclear, but may involve a more extensive insult to the inner ear. The finding that auditory and vestibular deficits can progress independently is not unique to ISSNHL. A recent study by Maihoub et al. found no significant correlation between the degree of hearing loss and the degree of caloric weakness in patients with definite Ménière’s disease, highlighting a dissociation between cochlear and vestibular function [3]. This supports our hypothesis that the presence of vertigo in ISSNHL signifies a more diffuse or severe injury pattern, potentially affecting vascular supply or neural structures serving both systems, ultimately limiting recovery potential despite aggressive combination therapy.

It has been reported that approximately 28–57% of patients with ISSNHL experience vertigo at the time of onset [18]. Similarly, 50% of the patients in the present study had vertigo at the time of ISSNHL onset. Patients with ISSNHL who present with vertigo are more likely to have severe hearing loss than those without vertigo. Patients with ISSNHL who exhibit profound hearing loss experience vertigo and abnormal caloric results more frequently than those with less severe hearing loss [18]. Therefore, vertigo has been suggested to be a poor prognostic factor for hearing recovery after extensive and severe inner ear damage [21].

Patients with ISSNHL who exhibit hearing loss at high frequencies (2000 Hz and above) tend to be more susceptible to vertigo because, anatomically, the basal cochlear turn is more proximal to the vestibular end organs than the apical cochlear turn [22,23]. Moreover, it has been proposed that the obstruction of the arteria vestibulocochlearis leads to reduced blood supply to the lower cochlea, saccule, and posterior ampulla, resulting in high-frequency hearing loss accompanied by vertigo [24]. This anatomical theory could explain our finding that patients without vertigo showed significantly better hearing improvement at 2000 Hz than those with vertigo.

4.4. Prognostic Factors Associated with Hearing Recovery: Initial Hearing Level

Initial hearing level is a well-known prognostic factor for ISSNHL [4,13,14]. ISSNHL at higher thresholds is typically associated with poorer hearing recovery outcomes. Furthermore, patients with low initial hearing levels at the onset of symptoms show a slower rate of recovery than those with moderate initial hearing levels [13]. In the present study, ROC analysis demonstrated that the cutoff value of initial PTA for predicting a poor hearing recovery was ≥90.5 dB. This cutoff is close to 90 dB, which has traditionally been used as the cutoff for profound hearing loss [4,25,26,27]. Therefore, the present findings confirm that patients with severe ISSNHL treated with HBO, SS, and PGE1 are likely to have poor hearing outcomes.

With an increased number of cases, this study retrospectively analyzed 60 patients who received combined therapy with HBO, SS, and PGE1, focusing on prognostic factors [11]. The results revealed that vertigo and initial hearing level (cutoff ≥ 90.5 dB) were independent prognostic factors for poor hearing outcome. Furthermore, vertigo was significantly associated with hearing improvement at 2000 Hz. To maintain homogeneity of treatment delivery, only patients who received the full combination of HBO, SS, and PGE1 were included. Notably, these therapies are not used in combination in all patients but are individualized according to each case.

4.5. Limitations of the Present Study

This study has several limitations. First, it was a single-center retrospective study with a relatively small sample size, raising the possibility of selection bias. Although our findings suggest that adjunctive HBO is associated with improved hearing recovery in ISSNHL, the retrospective and non-randomized design precludes definitive conclusions regarding causality. However, unifying the treatment protocol allowed a clearer evaluation of the therapeutic effects of the combined regimen.

Although all patients received the same protocol, including HBO, SS, and PGE₁, the number of HBO sessions differed between patients. The attending otologist decided whether to continue or discontinue HBO based on hearing recovery, patient preference, and adverse effects. Such heterogeneity in HBO exposure may have influenced outcomes and potentially led to an underestimation of its therapeutic effect. However, Spearman’s correlation analysis revealed no significant association between the number of HBO sessions and PTA improvement. Specifically, there was no significant association between an increase in the number of HBO sessions and the degree of hearing improvement. This result suggests that an increase in the number of HBO sessions may not necessarily lead directly to hearing recovery, and other treatment factors and individual patient conditions may have influenced recovery.

Another limitation is that speech intelligibility was not evaluated; hearing outcomes were assessed solely using pure-tone audiometry. Although pure-tone audiometry is essential, speech intelligibility provides a more practical measure of functional hearing, and incorporating it into future studies may yield a more comprehensive understanding of hearing outcomes.

In this study, the incidence of vertigo was significantly higher in the HBO group than in the no-HBO group. Because HBO treatment requires frequent hospital visits, it tends to be added for more severe cases that are often accompanied by vertigo. This likely explains the higher number of vertigo cases in the HBO group. To minimize this bias, multivariate logistic regression analysis including vertigo was performed, and HBO treatment remained independently associated with better hearing recovery after adjustment.

Finally, although vertigo and initial hearing level have been reported previously as prognostic factors [4,6], the clinical significance of this study lies in confirming that these factors also influence prognosis under the unique treatment protocol of combined HBO, SS, and PGE1 therapy. Despite limited novelty, this study provides valuable evidence regarding prognostic factors in ISSNHL treated with this specific regimen.

5. Conclusions

These findings demonstrate that additional HBO in combination with SS and PGE1 improves hearing outcomes in patients with ISSNHL. The presence of vertigo and a higher initial PTA (≥90.5 dB) were independent prognostic factors for poor recovery, with vertigo showing a significant negative association with hearing improvement at 2000 Hz. Although age > 60 years tended to be associated with worse outcomes, this finding was not statistically significant. These results suggest that additional HBO in combination with SS and PGE1 may provide a therapeutic advantage; however, hearing prognosis remains limited in patients with severe hearing loss and vertigo. Further prospective studies with larger cohorts are warranted to validate these findings and to optimize treatment strategies for ISSNHL.