Gadolinium-Enhanced MRI Classification of Sudden Sensorineural Hearing Loss and Its Clinical Correlates

Abstract

Importance

Sudden sensorineural hearing loss (SSNHL) has heterogeneous etiologies and prognoses. Early-imaging evaluation may help identify patients at higher risk of poor hearing recovery.

Objective

To investigate the relationship between labyrinthine gadolinium-enhanced MRI (Gd-MRI) findings and clinical characteristics in patients with SSNHL and to assess the clinical value of MRI-based subtyping for early evaluation and stratified treatment.

Design

Retrospective observational study.

Setting

Single tertiary referral center.

Participants

Ninety-six patients with unilateral SSNHL treated between January 2014 and May 2020 were included.

Exposure or Intervention

All patients underwent gadolinium-enhanced inner ear MRI. According to imaging features, they were categorized into an MRI-negative group, an endolymphatic hydrops (EH) group, and a labyrinthine signal abnormalities (LSA) group.

Main Outcome Measures

Clinical symptoms, audiological data, inner ear subunit involvement, and short-term hearing gain were compared among groups. Independent predictors of hearing improvement were analyzed using multivariate linear regression.

Results

Among 96 patients, 55 were MRI-negative, 20 were EH, and 21 were LSA. The LSA group had the highest rates of vertigo and tinnitus (100% and 95.2%, respectively; P = .001, P = .038) and more extensive inner ear involvement (≥2 subunits, 71.4%; P = .003). Pretreatment hearing thresholds were comparable between groups, but the LSA group had a significantly-poorer pure-tone average after treatment. Multivariate regression identified age and MRI classification as independent predictors of low-frequency hearing gain (P < .05).

Conclusion

Gadolinium-enhanced inner ear MRI effectively stratifies SSNHL by identifying distinct imaging phenotypes correlated with clinical severity and recovery potential. Labyrinthine signal abnormalities, indicating more extensive inner ear involvement, are associated with poorer hearing recovery.

Relevance

MRI-based subtyping can facilitate early evaluation and individualized management of SSNHL, supporting its integration into routine diagnostic assessment. These preliminary findings require confirmation through larger prospective studies with long-term follow-up.

Graphical Abstract

Keywords

sudden hearing loss magnetic resonance imaging labyrinth diseases endolymphatic hydrops prognosis

Key Messages

Gd-enhanced inner ear MRI stratifies sudden sensorineural hearing loss into MRI-negative, endolymphatic hydrops, and labyrinthine signal abnormalities (LSA).

LSA affects more inner ear subunits, suggesting greater structural damage.

Age and MRI classification independently predict low-frequency hearing gain, serving as valuable markers for early prognosis and personalized management.

Introduction

Sudden sensorineural hearing loss refers to hearing loss of ≥30 dB HL in at least 3 adjacent frequencies occurring within a 72 hour period. Among most patients, sudden sensorineural hearing loss is idiopathic (ISSNHL)¹ and may be multifactorial. The most common causes include viral infections,² vascular factors,³ autoimmunity,^4,5 and hydrops of the membranous labyrinth^6,7. Hearing outcomes are highly dependent on the early identification of the underlying cause, which helps assess the severity of the disease and provides a key basis for developing individualized intervention plans.^1,8,9

Conventional MR imaging in patients with SSNHL is primarily used to rule out brain lesions or to identify retrocochlear lesions such as vestibular schwannomas.¹⁰ In recent years, newer MR sequences, such as three-dimensional fluid-attenuated inversion recovery (3D-FLAIR) combined with gadolinium contrast enhancement, have been developed to assess abnormal signals in the inner ear labyrinth, including methemoglobin deposition or protein accumulation^11
-17 and to identify endolymphatic hydrops (EH) by volumetric measurement.¹⁸ Although endolymphatic hydrops is the primary pathological feature of Meniere’s disease, some patients may present with sudden deafness as the initial manifestation, necessitating follow-up for a definitive diagnosis.¹⁹

During the early evaluation of patients with SSNHL, gadolinium-enhanced inner ear MRI not only helps exclude organic causes but also reveals different inner ear pathological states. These imaging abnormalities may be associated with clinical symptoms and the potential for hearing recovery, but their prognostic value remains controversial.^13,16 The present study aimed to summarize and categorize 3D-FLAIR imaging sequences in patients with SSNHL to better understand the pathological mechanisms of sudden hearing loss and provide a basis for clinical stratification and treatment.

Materials and Methods

Patients

A total of 96 patients with unilateral sudden deafness were recruited during their initial visit to the Department of Otolaryngology Head and Neck Surgery between January 2014 to May 2020. Patients were screened according to strict inclusion and exclusion criteria, and clinical data on pretreatment and posttreatment audiological characteristics and gadolinium-based magnetic resonance imaging examinations were subsequently collected. Inclusion and exclusion criteria included: ① meeting the 2019 American Clinical Practice Guideline: Sudden Hearing Loss criteria for sudden deafness¹: ≥30 dB sensorineural hearing loss in at least 3 consecutive frequencies occurring within 72 hours; ② having complete clinical records and undergone pure-tone audiometry and gadolinium-based MRI examinations; ③ age ≥18 years, unilateral deafness, first onset; ④ excluding those with clear deafness-related factors, such as trauma, genetics, and other central and peripheral pathogenic factors; ⑤ excluding those with middle ear and inner ear space-occupying lesions as determined by specialist examinations and imaging examinations; ⑥ not receiving standard medication such as hormones or medications to improve inner ear microcirculation; and ⑦ excluding pregnant women and those with contraindications to the use of glucocorticoids.

Audiological Indicators

Classification of hearing curves in sudden deafness. The extent of hearing loss was calculated using the pure-tone average (PTA) of hearing thresholds at 500 Hz, 1 kHz, 2 kHz, and 4 kHz. Patients were subsequently classified into 4 subtypes^20
-22: ① Low-frequency decline subtype: Hearing loss at frequencies of 1000 Hz (inclusive) and below, with at least a 20 dB HL hearing loss at 250 and 500 Hz. ② High-frequency decline subtype: Hearing loss at frequencies of 2000 Hz (inclusive) and above, with at least a 20 dB HL hearing loss at 4000 and 8000 Hz. ③ Flat decline subtype: Hearing loss at all frequencies, with an average hearing threshold of ≤80 dB HL. ④ Total deafness subtype: Hearing loss at all frequencies, with an average hearing threshold of ≥81 dB HL at 250, 500, 1000, 2000, 3000, 4000, and 8000 Hz.

Early-efficacy assessment. All patients received standard treatment according to the Clinical Practice Guideline,²³ including uniform intravenous steroid therapy and improved circulatory drug therapy. No intratympanic steroid injections were administered. The treatment duration lasted until hearing recovery or until pure-tone audiometry retesting 10 days after the initiation of treatment. Responses to therapy were classified according to Siegel’s criteria²⁴: Complete recovery: final hearing threshold ≤25 dB HL; partial improvement: hearing gain >15 dB HL with a final threshold between 25 and 45 dB HL; slight improvement: hearing gain >15 dB HL with a final threshold >45 dB HL; no response: hearing gain <15 dB HL and final threshold >75 dB HL. Both “Complete recovery” and “Partial improvement” were considered “effective.”

Imaging Acquisition

All 96 patients provided written informed consent prior to the procedure. Before the intervention, patients were fully informed of the options for unilateral or bilateral intratympanic injection and were allowed to choose voluntarily. All patients ultimately included in this study elected to undergo bilateral intratympanic administration. Gadopentetate dimeglumine (Gd-DTPA) was diluted eightfold with normal saline prior to use. The contrast agent was injected into the middle ear cavity through the anteroinferior quadrant of the tympanic membrane, with an administered volume of approximately 0.4 to 0.5 mL per ear. Bilateral injections were performed simultaneously during the same session. After injection, patients remained seated with the head tilted backward for 1 hour and were instructed to minimize speaking, swallowing, and drinking to reduce variability in contrast distribution. After a 24 hour delay, 3D-FLAIR MRI was acquired on a 3.0 T Verio scanner (Siemens, Erlangen, Germany) equipped with a 16-channel head coil. Imaging parameters included repetition time (TR) of 6000 ms, echo time (TE) of 132 ms, and isotropic voxel resolution of 0.5 × 0.5 × 0.5 mm, with a total acquisition time of 4 minutes 16 seconds. In addition, 3D sampling perfection with application-optimized contrasts using different flip angle evolution (3D-SPACE) inversion recovery was applied for fluid signal suppression (TR = 6000 ms; TE = 388 ms; inversion time [TI] = 2100 ms; voxel size = 0.7 × 0.7 × 0.7 mm; scan time = 5 minutes 32 seconds).²⁵

MRI-Based Classification

Based on MRI findings, patients were categorized into 3 groups: MRI-negative, EH, and labyrinthine signal abnormality (LSA). EH was defined as visible enlargement of the endolymphatic space relative to the perilymphatic space on delayed contrast-enhanced 3D-FLAIR images.²⁵ LSA was defined as abnormally-increased signal intensity involving the cochlea, vestibule, and/or semicircular canals on MRI, including either hyperintensity on pre-contrast images or increased signal intensity on post-contrast images compared with the contralateral side.^13,26 The MRI-negative group was defined as the absence of abnormal MRI findings, with no evidence of EH or LSA. In this cohort, no patients demonstrated concomitant EH and LSA; therefore, the 3 groups were mutually exclusive. All MR images were independently reviewed by one radiologist and one senior otolaryngologist, each with over 5 years of experience in image interpretation. Both reviewers were blinded to the clinical data and the affected side.²⁵ Any discrepancies were resolved by consensus discussion. Figure 1 shows example images of inner ear gadolinium imaging results for the 3 groups.

Figure 1.

Representative MRI findings of SSNHL subtypes. (A) MRI-negative group: 3D-FLAIR sequence shows no abnormal signal, with symmetric appearance between the affected and contralateral sides. (B) Endolymphatic hydrops group: hydrops in left cochlea (white solid-line arrow) and vestibule (white dotted arrow); the right side is normal. (C) LSA group (pre-contrast): slight signal increase in left cochlea and vestibule (white arrowhead) compared with the right side. (D) LSA group (post-contrast): marked enhancement in left cochlea and vestibule (white arrowhead), with no enhancement on the right side. LSA, labyrinthine signal abnormality; SSNHL, sudden sensorineural hearing loss; 3D-FLAIR, three-dimensional fluid-attenuated inversion recovery.

Statistical Analysis

Statistical analyses were performed using R version 4.2.1. Normally-distributed data are expressed as mean ± standard deviation (SD) and were compared using one-way ANOVA; non-normally-distributed data are presented as median with interquartile range (IQR) and were compared using the Kruskal-Wallis test. Categorical variables were compared using the chi-squared test or Fisher’s exact test. When a significant overall difference was observed, post hoc pairwise comparisons were performed with the Bonferroni correction. After univariate analysis, significant variables were included in the multivariate linear regression model. All tests were 2-sided, and differences were considered statistically significant when P < .05.

Results

Demographics

A total of 96 patients were enrolled and classified into 3 groups based on MRI findings: MRI-negative (n = 55), EH (n = 20), and LSA (n = 21). The mean ages of the MRI-negative, EH, and LSA groups were 43.2 ± 12.7, 47.3 ± 14.2, and 48.1 ± 13.9 years, respectively (P = .262). The incidence of vertigo was 52.7%, 50.0%, and 100.0% (P = .001), with a higher incidence in the LSA group than in the MRI-negative group and the EH group (Bonferroni correction, adj. P < .05). Aural fullness was reported in 54.5%, 40.0%, and 28.6% of the patients (P = .107), and tinnitus was reported in 94.5%, 75.0%, and 95.2% of the patients (P = .038). The incidence of underlying diseases (hypertension, hyperlipidemia, or diabetes) was 21.8%, 25.0%, and 19.0%, respectively (P = .891). Baseline data are shown in Table 1.

Table 1.

Demographic and Clinical Characteristics of Patients by MRI Findings.

Characteristic	MRI-negative (n = 55)	Endolymphatic hydrops (n = 20)	Labyrinthine signal abnormality (n = 21)	P value
Gender, female, n (%)	38 (69.1)	8 (40.0)	14 (66.7)	.064
Ear affected, left, n (%)	31 (54.4)	12 (60.0)	10 (47.6)	.703
Age, mean ± SD	43.2 ± 12.7	47.3 ± 14.2	48.1 ± 13.9	.262
Vertigo, n (%)	29 (52.7)	10 (50.0)	21 (100.0)	.001
Aural fullness, n (%)	30 (54.5)	8 (40.0)	6 (28.6)	.107
Tinnitus, n (%)	52 (94.5)	15 (75.0)	20 (95.2)	.038
Comorbidities,^a n (%)	12 (21.8)	5 (25.0)	4 (19.0)	.891

Bold indicates statistical significance (P < .05).

Comorbidities include hypertension, hyperlipidemia, and diabetes mellitus.

MRI

Table 2 shows the distribution of inner ear subunit involvement in the EH and LSA groups. The cochlear involvement rates were 30% and 9.5% in the EH and LSA groups, respectively (P > .05); the vestibular involvement rates were 45% and 19% (P > .05). Involvement of two or more subunits was observed in 25% of the EH group and 71.4% of the LSA group (χ² = 8.838, P = .003), indicating that the LSA group exhibited more extensive inner ear involvement.

Table 2.

Inner Ear Subunit Involvement in Patients With Endolymphatic Hydrops and Labyrinthine Signal Abnormality.

Subsite	Endolymphatic hydrops (n = 20)	Labyrinthine signal abnormality (n = 21)	P
Cochlea, n (%)	6 (30.0)	2 (9.5)	.13
Vestibule, n (%)	9 (45.0)	4 (19.0)	.07
Semicircular canal, n (%)	0 (0.0)	0 (0.0)	—*
≥2 Subsites Involved, n (%)	5 (25.0)	15 (71.4)	.003

Bold indicates statistical significance (P < .05).

P-value could not be calculated due to zero events in both groups.

Clinical Symptoms

Audiological outcomes were compared among the 3 groups (Table 3, Figure 2A). Pretreatment PTA across the 3 groups was 71.3, 68.1, and 90.0 dB HL, respectively (Kruskal-Wallis test, P = .076). After treatment, these values decreased to 47.5, 68.1, and 81.3 dB HL, respectively (P = .013). Posttreatment low-frequency PTA was 35.0, 60.0, and 77.3 dB HL, respectively (P = .006), mid-frequency PTA was 45.0, 66.7, and 77.3 dB HL, respectively (P = .025), and high-frequency PTA was 62.5, 77.5, and 85.0 dB HL, respectively (P = .010). Multiple comparisons indicated that the LSA group had higher posttreatment PTA and subgroup frequency PTA than the MRI-negative group (Figure 2A; adj. P = .013, .008, .024, and .013, respectively). Analysis of low-frequency ΔPTA revealed significant differences among the 3 groups (Kruskal-Wallis test, P = .028), with post hoc tests confirming significantly poorer improvement in the LSA group than in the MRI-negative group (adj. P < .05). The distribution of the 4 hearing curve types (low-frequency decline, high-frequency decline, flat, and total deafness) was not statistically different among the groups (P = .119). Paired analysis (Wilcoxon signed-rank test, Figure 2B-D) indicated that the MRI-negative group observed significant hearing improvement after treatment (P < .001), while the EH group and the LSA group showed significantly less improvement. Overall, the MRI-negative group showed the greatest hearing improvement after treatment, particularly in PTA and low-frequency, while the LSA group showed the worst recovery.

Table 3.

Audiometric Thresholds and Curve Types Stratified by MRI Findings.

Variable	MRI-negative (n = 55)	Endolymphatic hydrops (n = 20)	Labyrinthine signal abnormality (n = 21)	P
Full-frequency PTA (dB HL)
Pretreatment	71.3 (63.7)	68.1 (42.5)	90.0 (37.5)	.076
Posttreatment	47.5 (61.3)	68.1 (40.0)	81.3 (39.4)	.013
ΔPTA	6.3 (16.3)	0.0 (11.9)	0.0 (17.5)	.051
Low-frequency PTA (dB HL)
Pretreatment	70.0 (25.8)	60.0 (18.5)	80.0 (33.0)	.140
Posttreatment	35.0 (55.0)	60.0 (43.1)	77.3 (55.0)	.006
ΔPTA	10.0 (27.5)	1.3 (25.9)	0.0 (11.4)	.028
Mid-frequency PTA (dB HL)
Pretreatment	73.3 (63.3)	65.8 (44.6)	90.0 (37.5)	.104
Posttreatment	45.0 (61.7)	66.7 (46.7)	77.3 (45.0)	.025
ΔPTA	7.5 (15.0)	0.0 (12.5)	0.0 (16.3)	.142
High-frequency PTA (dB HL)
Pretreatment	67.5 (52.5)	86.3 (26.9)	90.0 (20.0)	.078
Posttreatment	62.5 (62.5)	77.5 (25.6)	85.0 (27.5)	.010
ΔPTA	5.0 (12.5)	3.8 (13.1)	0.0 (18.8)	.274
Hearing curve types, n (%)
Low-frequency drop	11 (20.0)	2 (10.0)	0 (0.0)	.119
High-frequency drop	9 (16.4)	4 (20.0)	3 (14.3)
Flat pattern	13 (23.6)	9 (45.0)	6 (28.6)
Total deafness	22 (40.0)	5 (25.0)	12 (57.1)

Data are presented as median (IQR) for PTA. ΔPTA represents posttreatment minus pretreatment thresholds. Bold indicates statistical significance (P < .05).

Abbreviation: PTA, pure-tone average.

Figure 2.

Changes in hearing thresholds before and after treatment among MRI-defined groups. (A) Posttreatment full-frequency and frequency-specific PTA in MRI-negative, endolymphatic hydrops, and labyrinthine enhancement groups. Group comparisons were performed using the Kruskal-Wallis test with the Bonferroni correction. LF, low frequency; MF, mid-frequency; HF, high frequency. (B-D) Pretreatment and posttreatment full-frequency PTA in each group. Within-group differences were evaluated using the Wilcoxon signed-rank test. PTA, pure-tone average.

Significant differences were observed among the 3 groups in Siegel’s criteria distribution (Fisher’s exact test, P = .032, Table 4) and recovery rate (χ² = 8.19, P = .017). The MRI-negative group demonstrated higher rates of complete recovery and overall recovery than the LSA group (adj. P < .05).

Table 4.

Hearing Recovery Based on Siegel’s Criteria.

Outcome	MRI-negative (n = 55)	Endolymphatic hydrops (n = 20)	Labyrinthine signal abnormality (n = 21)	P
Siegel’s criteria
Complete recovery, n (%)	15 (27.3)	2 (10.0)	0 (0.0)	.032
Partial recovery, n (%)	6 (10.9)	1 (5.0)	2 (9.5)
Slight improvement, n (%)	2 (3.6)	0 (0.0)	2 (9.5)
No improvement, n (%)	32 (58.2)	17 (85.0)	17 (81.0)
Effective recovery, n (%)	21 (38.2)	3 (15.0)	2 (9.5)	.017

Bold indicates statistical significance (P < .05).

Regression Analysis

In the univariate analysis of full-frequency ΔPTA, age (P = .004), MRI classification (P = .028), vertigo (P = .097), and tinnitus (P = .070) met the predefined inclusion criterion (P < .10). Based on baseline comparisons shown in Table 1, sex (P = .064), vertigo (P = .001), and tinnitus (P = .038) differed across MRI groups and were therefore considered potential confounding factors. Accordingly, sex, age, vertigo, tinnitus, and MRI classification were included in the multivariable linear regression model. In the multivariable analysis, only age remained independently associated with full-frequency ΔPTA (B = −0.312, SE = 0.107, β = −.288, P = .004), indicating that older age was associated with smaller hearing gain. MRI classification yielded a P value of .067 in the multivariable model.

Further analyses were performed for frequency-specific ΔPTA. In the univariable linear regression analysis of low-frequency ΔPTA, age (B = −0.387, SE = 0.135, P = .005) and MRI classification (B = −5.958, SE = 2.205, P = .008) were significantly associated with low-frequency hearing gain, while vertigo (P = .079) met the predefined inclusion criterion (P < .10). Other variables showed no significant associations (Table S1). Based on the univariable results and baseline differences, relevant variables were entered into the multivariable linear regression model. In the multivariable analysis (Table 5), both age (B = −0.337, SE = 0.134, P = .014) and MRI classification (B = −5.074, SE = 2.174, P = .022) remained independently associated with low-frequency ΔPTA, with older age and LSA phenotype predicting smaller hearing gain. No significant associations between MRI classification and mid- or high-frequency ΔPTA were observed in the regression analyses.

Table 5.

Multivariable Linear Regression Analysis of Factors Associated With Low-Frequency ΔPTA.

Variable	B	SE	P value	95% CI
Constant	28.396	6.240	<.001	16.006 to 40.787
Age	−0.337	0.134	.014	−0.603 to −0.070
MRI classification	−5.074	2.174	.022	−9.391 to −0.758

Bold indicates statistical significance (P < .05).

Abbreviation: PTA, pure-tone average.

Discussion

SSNHL is commonly manifested as the first symptom of various inner ear diseases. The etiology of SSNHL is complex, and its clinical manifestations are highly heterogeneous. Early-objective evaluation and classification have been challenging to achieve in clinical practice. In recent years, gadolinium-enhanced inner ear MRI of the inner ear has been integrated for the study of inner ear pathology. MRI can be used to evaluate inner ear fluid dynamics, blood-labyrinth barrier function, and lesion extent.^27
-29 In contrast to the limitations of clinical symptoms, MRI can provide direct pathophysiological information. Building on this advancement, we classified SSNHL into 3 imaging-based subtypes: MRI-negative, endolymphatic hydrops, and labyrinthine signal abnormality. Subsequently, the association between imaging classification and clinical phenotype and prognosis was evaluated.

Previous studies have shown that the unaffected ear of SSNHL patients typically lacks hyperintensity on 3D-FLAIR sequences,^27,30 a feature also consistently observed in healthy individuals.¹³ Therefore, 3D-FLAIR inner ear hyperintensity is considered an abnormal or even pathological phenomenon, which we collectively term LSA. Theoretically, these signal abnormalities can be further subdivided based on underlying mechanisms, such as intralabyrinthine hemorrhage, inflammatory response, or blood-labyrinthine barrier disruption.¹³ However, due to the limited sample size in this study, reliable subcategorization at this level was not feasible. Consequently, all cases exhibiting 3D-FLAIR hyperintensity were analyzed collectively as LSA.

The results of this study indicate that there are significant differences in the clinical phenotypes among patients with different MRI classifications. Patients with LSA exhibited distinct clinical features: 71.4% had involvement of ≥2 inner ear subunits, with vertigo in 100% and tinnitus in 95.2% of cases. Posttreatment hearing thresholds were significantly higher in this group, and functional recovery rates were the lowest, consistent with previous reports.^31,32 We speculate that inner ear hyperintensity reflects severe blood-labyrinth barrier disruption and irreversible outer hair cell damage,³³ which may impede drug delivery and explain the poor prognosis. Combined with standard clinical treatments, more aggressive early interventions may be warranted for these patients, such as intratympanic steroid injections, while the use of defibrination or anticoagulant drugs should be approached with caution to improve therapeutic outcomes.

EH represents a distinct imaging subtype with important prognostic implications. Longitudinal studies have shown that more than half of SSNHL patients with EH eventually develop Menière’s disease.³⁴ This progression suggests that EH, which is the primary pathological manifestation of MD,³⁵ may not be a coincidental imaging finding in SSNHL but rather identifies an early-stage or a shared pathological substrate between the 2 conditions. In our cohort, the EH group exhibited moderate clinical involvement: vertigo in 50% and tinnitus in 75% of cases, with most patients (75%) having a single affected inner ear subunit (cochlear: 30%; vestibular: 45%). Audiologically, the full-frequency, low-frequency, and mid-frequency hearing thresholds of patients in the EH group were at the intermediate level among the 3 groups before and after treatment. The ΔPTA gains were limited, and the hearing curves were predominantly flat or showed low-frequency decline. The recovery rate of the EH group was also in the middle among all 3 groups (15.0%). Although the initial hearing impairment in EH-related SSNHL may not be the most severe, the prognosis for recovery warrants caution as it may represent a chronic, evolving pathology rather than a monophasic event.

Multivariable regression analysis identified age as an independent predictor of hearing gain, with older patients showing less improvement, consistent with previous reports.^34
-37 This may reflect age-related declines in outer hair cell plasticity and inner ear microcirculation, limiting recovery capacity. For low-frequency hearing gain, both age and MRI classification emerged as independent predictors, indicating that the prognostic value of imaging subtypes is frequency-specific. Specifically, the LSA group showed poorer low-frequency recovery, whereas the MRI-negative group had relatively-better outcomes at low frequencies; the EH group showed intermediate recovery. This may be explained by several factors. First, the LSA group exhibited more extensive inner ear involvement, which is associated with more severe impairment of auditory and vestibular function.³⁸ Second, EH is characterized by fluctuating low-frequency hearing loss,³⁹ and recovery in SSNHL cases with low-frequency involvement or ascending audiograms is generally better than in those with high-frequency involvement or descending audiograms.^37,40
-42 Therefore, these observations collectively explain the association between MRI classification and low-frequency hearing gain. However, given the retrospective design and limited sample size, these findings warrant validation in larger prospective studies.

This study has several limitations. The relatively-small sample size, particularly in the EH and LSA subgroups, may have limited statistical power to detect group differences, necessitating validation in larger prospective cohorts. In addition, while intratympanic gadolinium administration was chosen for its superior inner ear enhancement via direct diffusion across the round and oval window membranes,^43,44 this technique has inherent methodological limitations, including variability in round window permeability, potential contrast loss, and the logistical burden of delayed imaging. Well-designed comparative studies are needed to clarify the relative advantages of intratympanic versus intravenous approaches and to inform optimal imaging strategies for SSNHL. Finally, as this study focused primarily on early-treatment response during hospitalization, delayed recovery cannot be excluded. Therefore, follow-up evaluations at the 3rd and 6th month, when hearing thresholds typically stabilize, are warranted. Overall, the present findings are more applicable to the prediction of early-treatment sensitivity, and the predictive value of imaging classification for long-term prognosis requires validation in larger prospective cohorts.

Conclusion

This study demonstrates that gadolinium-enhanced inner ear MRI can identify distinct imaging features in patients with SSNHL, which are associated with differences in clinical presentation and hearing outcomes. Labyrinthine signal abnormalities were associated with more extensive inner ear involvement and poorer hearing recovery, while older age was independently associated with reduced hearing gain. These findings suggest that MRI-based classification may provide additional information for early assessment of SSNHL. Given the retrospective design, limited subgroup sample size, and short-term follow-up, these findings should be interpreted cautiously. Larger prospective studies with long-term follow-up are needed to confirm the clinical applicability of MRI-based classification in SSNHL.

Supplemental Material

sj-docx-1-ohn-10.1177_19160216261440784 – Supplemental material for Gadolinium-Enhanced MRI Classification of Sudden Sensorineural Hearing Loss and Its Clinical Correlates

Supplemental material, sj-docx-1-ohn-10.1177_19160216261440784 for Gadolinium-Enhanced MRI Classification of Sudden Sensorineural Hearing Loss and Its Clinical Correlates by Xiaojing Guo, Xiaoying Ke, Shengnan Ye and Heng Xiao in Journal of Otolaryngology - Head & Neck Surgery

Footnotes

Acknowledgements

The authors thank all the patients who participated in this study.

Author Contributions

Xiaojing Guo and Xiaoying Ke designed and coordinated the study. Xiaoying Ke conducted data collection. Xiaojing Guo analyzed the data and drafted the manuscript. Heng Xiao and Shengnan Ye interpreted the findings and critically revised the manuscript. All authors contributed to the study and approved the final version for submission.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Natural Science Foundation of Fujian Province (No. 2024J01557), Joint Funds for the Innovation of Science and Technology, Fujian Province (No. 2024Y9190) and Startup Fund for Scientific Research of Fujian Medical University (No. 2024QH1085).

ORCID iDs

Xiaojing Guo

Heng Xiao

Supplemental Material

Additional supporting information is available in the online version of the article.

References

Chandrasekhar

Tsai Do

Schwartz

, et al. Clinical practice guideline: sudden hearing loss (update). Otolaryngol Head Neck Surg. 2019;161(1_suppl):S1-S45. doi:10.1177/0194599819859885

Berrettini

Bianchi

Segnini

Sellari-Franceschini

Bruschini

Montanaro

Herpes zoster oticus: correlations between clinical and MRI findings. Eur Neurol. 1998;39(1):26-31. doi:10.1159/000007900

Capaccio

Cuccarini

Ottaviani

Fracchiolla

Bossi

Pignataro

Prothrombotic gene mutations in patients with sudden sensorineural hearing loss and cardiovascular thrombotic disease. Ann Otol Rhinol Laryngol. 2009;118(3):205-210. doi:10.1177/000348940911800308

Schreiber

Agrup

Haskard

Luxon

LM.

Sudden sensorineural hearing loss. Lancet. 2010;375(9721):1203-1211. doi:10.1016/S0140-6736(09)62071-7

Kim

Lee

Kim

Choi

JY.

Autoimmunity as a candidate for the etiopathogenesis of meniere’s disease: detection of autoimmune reactions and diagnostic biomarker candidate. PLoS One. 2014;9(10):e111039. doi:10.1371/journal.pone.0111039

Sone

Kobayashi

Yoshida

Naganawa

Pathophysiological analysis of idiopathic sudden sensorineural hearing loss by magnetic resonance imaging: a mini scoping review. Front Neurol. 2023;14:1193104. doi:10.3389/fneur.2023.1193104

Qin

, et al. Visualization of endolymphatic hydrops in patients with unilateral idiopathic sudden sensorineural hearing loss with four types according to Chinese criterion. Front Surg. 2021;8:682245. doi:10.3389/fsurg.2021.682245

Chen

Wen

Zong

Liu

Sun

Audiological outcomes in sudden sensorineural hearing loss with presumed inner ear hemorrhage. Am J Otolaryngol. 2019;40(2):274-278. doi:10.1016/j.amjoto.2018.09.017

Ballesteros

Tassies

Reverter

Alobid

Bernal-Sprekelsen

Idiopathic sudden sensorineural hearing loss: classic cardiovascular and new genetic risk factors. Audiol Neurootol. 2012;17(6):400-408. doi:10.1159/000341989

10.

Aarnisalo

Suoranta

Ylikoski

Magnetic resonance imaging findings in the auditory pathway of patients with sudden deafness. Otol Neurotol. 2004;25(3):245-249. doi:10.1097/00129492-200405000-00008

11.

Weissman

Curtin

Hirsch

WL.

High signal from the otic labyrinth on unenhanced magnetic resonance imaging. AJNR Am J Neuroradiol. 1992;13(4):1183-1187.

12.

Cervantes

Barrs

DM.

Sudden sensorineural hearing loss associated with intralabyrinthine hemorrhage. Otol Neurotol. 2015;36(8):e134-e135. doi:10.1097/MAO.0000000000000474

13.

Berrettini

Seccia

Fortunato

, et al. Analysis of the 3-dimensional fluid-attenuated inversion-recovery (3D-FLAIR) sequence in idiopathic sudden sensorineural hearing loss. JAMA Otolaryngol Head Neck Surg. 2013;139(5):456-464. doi:10.1001/jamaoto.2013.2659

14.

Yoshida

Sugiura

Naganawa

Teranishi

Nakata

Nakashima

Three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging findings and prognosis in sudden sensorineural hearing loss. Laryngoscope. 2008;118(8):1433-1437. doi:10.1097/MLG.0b013e318172ef85

15.

Sugiura

Naganawa

Teranishi

Nakashima

Three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging findings in patients with sudden sensorineural hearing loss. Laryngoscope. 2006;116(8):1451-1454. doi:10.1097/01.mlg.0000228005.78187.23

16.

Lee

Jung

Park

Yeo

Lee

SK.

Feasibility of three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging as a prognostic factor in patients with sudden hearing loss. Eur Arch Otorhinolaryngol. 2012;269(8):1885-1891. doi:10.1007/s00405-011-1834-1

17.

Kim

Lee

Goh

, et al. Clinical significance of an increased cochlear 3D fluid-attenuated inversion recovery signal intensity on an MR imaging examination in patients with acoustic neuroma. AJNR Am J Neuroradiol. 2014;35(9):1825-1829. doi:10.3174/ajnr.A3936

18.

Gürkov

Berman

Dietrich

, et al. MR volumetric assessment of endolymphatic hydrops. Eur Radiol. 2015;25(2):585-595. doi:10.1007/s00330-014-3414-4

19.

Rodriguez-Martín

González-Aguado

Sotos

Morales-Angulo

Ménière’s disease with a prior diagnosis of idiopathic sudden hearing loss. Otol Neurotol. 2025;46(6):680-685. doi:10.1097/MAO.0000000000004508

20.

Guo

Xiao

Huang

, et al. Differentiating definite and probable ménière disease: a comprehensive evaluation of audio-vestibular function testing combined with inner ear MRI. Otol Neurotol. 2024;45(8):925-931. doi:10.1097/MAO.0000000000004282

21.

Zhang

Jing

Association between meteorological factors and audiogram configurations in patients with sudden sensorineural hearing loss: a cross-sectional study. BMJ Open. 2021;11(12):e045768. doi:10.1136/bmjopen-2020-045768

22.

Chen

, et al. Effectiveness of salvage intratympanic dexamethasone treatment for refractory sudden sensorineural hearing loss classified by audiogram patterns. Acta Otolaryngol. 2021;141(5):449-453. doi:10.1080/00016489.2021.1889029

23.

Editorial Board of Chinese Journal of Otorhinolaryngology Head and Neck Surgery; Society of Otorhinolaryngology Head and Neck Surgery, Chinese Medical Association. [Guideline of diagnosis and treatment of sudden deafness (2015)]. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2015;50(6):443-447.

24.

Siegel

LG.

The treatment of idiopathic sudden sensorineural hearing loss. Otolaryngol Clin North Am. 1975;8(2):467-473.

25.

Xiao

Guo

Cai

, et al. Magnetic resonance imaging of endolymphatic hydrops in Ménière’s disease: a comparison of the diagnostic value of multiple scoring methods. Front Neurol. 2022;13:967323. doi:10.3389/fneur.2022.967323

26.

Lammers

MJW

Young

Fenton

Lea

Westerberg

. The prognostic value and pathophysiologic significance of three-dimensional fluid-attenuated inversion recovery (3D-FLAIR) magnetic resonance imaging in idiopathic sudden sensorineural hearing loss: a systematic review and meta-analysis. Clin Otolaryngol. 2019;44(6):1017-1025. doi:10.1111/coa.13432

27.

Kim

Park

Lee

, et al. Comparison of inner ear contrast enhancement among patients with unilateral inner ear symptoms in MR images obtained 10 minutes and 4 hours after gadolinium injection. AJNR Am J Neuroradiol. 2015;36(12):2367-2372. doi:10.3174/ajnr.A4439

28.

Sousa

Lobo

Cadilha

Eça

Campos

Luis

Is there progression of endolymphatic hydrops in Ménière’s disease? Longitudinal magnetic resonance study. Eur Arch Otorhinolaryngol. 2023;280(5):2225-2235. doi:10.1007/s00405-022-07721-z

29.

Hornibrook

Mudry

Curthoys

Smith

CM.

Ductus reuniens and its possible role in Menière’s disease. Otol Neurotol. 2021;42(10):1585-1593. doi:10.1097/MAO.0000000000003352

30.

Lee

Yoon

Lee

, et al. Prognostic value of labyrinthine 3D-FLAIR abnormalities in idiopathic sudden sensorineural hearing loss. AJNR Am J Neuroradiol. 2016;37(12):2317-2322. doi:10.3174/ajnr.A4901

31.

Bossy

Gallois

Escudé

Marx

Diagnostic value of three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging sequences at 1.5 Tesla in management of sudden sensorineural hearing loss: our experience in 20 patients. Clin Otolaryngol. 2022;47(2):331-335. doi:10.1111/coa.13890

32.

Liao

, et al. Revisiting the relationship of three-dimensional fluid attenuation inversion recovery imaging and hearing outcomes in adults with idiopathic unilateral sudden sensorineural hearing loss. Eur J Radiol. 2016;85(12):2188-2194. doi:10.1016/j.ejrad.2016.10.005

33.

Kaya

Hizli

Schachern

Paparella

Cureoglu

Effects of intralabyrinthine hemorrhage on the cochlear elements: a human temporal bone study. Otol Neurotol. 2016;37(2):132-136. doi:10.1097/MAO.0000000000000927

34.

Cai

Xiao

Lin

, et al. The value of gadolinium-enhanced MRI in predicting the development of sudden hearing loss into Ménière’s disease. Clin Otolaryngol. 2024;49(1):117-123. doi:10.1111/coa.14116

35.

Hallpike

Cairns

Observations on the pathology of Ménière’s syndrome: (section of otology). Proc R Soc Med. 1938;31(11):1317-1336.

36.

Fukushima

Ueno

Kitayama

Akahani

Inohara

Takeda

Assessment of the progression of vertical semicircular canal dysfunction and increased vestibular endolymphatic hydrops in patients with early-stage Ménière disease. JAMA Otolaryngol Head Neck Surg. 2020;146(9):789. doi:10.1001/jamaoto.2020.1496

37.

Xiong

Dai

Wang

, et al. Idiopathic sudden sensorineural hearing loss in different ages: prognosis of patients with initial total hearing loss. Front Psychol. 2022;13:818967. doi:10.3389/fpsyg.2022.818967

38.

Chen

Sun

Huang

Liang

Liu

Labyrinthine lesions in presumed inner ear hemorrhage-related sudden deafness. Am J Otolaryngol. 2022;43(2):103331. doi:10.1016/j.amjoto.2021.103331

39.

Niu

Chen

Lin

Sha

Development and characteristics of hearing loss with the progression of endolymphatic hydrops. Ear Nose Throat J. 2025;104(5):NP302-NP307. doi:10.1177/01455613221101088

40.

Psillas

Rizou

Rachovitsas

Tsiropoulos

Constantinidis

Hearing outcome of low-tone compared to high-tone sudden sensorineural hearing loss. Int Arch Otorhinolaryngol. 2019;23(1):65-69. doi:10.1055/s-0038-1657789

41.

Wittig

Wittekindt

Kiehntopf

Guntinas-Lichius

Prognostic impact of standard laboratory values on outcome in patients with sudden sensorineural hearing loss. BMC Ear Nose Throat Disord. 2014;14(1):6. doi:10.1186/1472-6815-14-6

42.

Tseng

Hsu

Chen

Chu

Liao

WH.

A novel audiogram model for predicting the prognosis of sudden sensorineural hearing loss. J Chin Med Assoc. 2025;88(11):851-857. doi:10.1097/JCMA.0000000000001299

43.

Naganawa

Nakashima

Visualization of endolymphatic hydrops with MR imaging in patients with ménière’s disease and related pathologies: current status of its methods and clinical significance. Jpn J Radiol. 2014;32(4):191-204. doi:10.1007/s11604-014-0290-4

44.

Conte

Lo Russo

Calloni

, et al. MR imaging of endolymphatic hydrops in Ménière’s disease: not all that glitters is gold. Acta Otorhinolaryngol Ital. 2018;38(4):369-376. doi:10.14639/0392-100X-1986