Sudden Hearing Loss (SHL) is defined as an abrupt reduction in hearing, occurring unilaterally or bilaterally within 72 hs, and is often accompanied by tinnitus and/or vertigo.1,2 It may present as sensorineural, mixed, or conductive, depending on the site and nature of auditory dysfunction. The most common form, Sudden Sensorineural Hearing Loss (SSNHL), is classically defined as a ≥30 dB loss across three consecutive audiometric frequencies within 72-hs, according to the National Institute on Deafness and Other Communication Disorders (NIDCD).3 Its incidence ranges from 5 to 27 cases per 100,000 individuals annually, with approximately 66,000 new cases reported each year in the United States. Bilateral association remains rare, occurring in only 0.4%–4.9% of cases.3–5

Despite being considered an otological emergency, the etiology of SSNHL remains unknown in nearly 70%–90% of patients and is thus classified as idiopathic.3,6 Several mechanisms have been proposed, including vascular, autoimmune, and infectious pathways.6,7 Among infectious agents, members of the Herpesviridae family ‒ such as HSV-1, HSV-2, VZV, EBV, and CMV ‒ have been extensively investigated due to their ability to establish latency and reactivate, potentially affecting inner ear structures including the cochlear nerve and the organ of Corti.7–10

More recently, SARS-CoV-2, the causative agent of COVID-19, has been implicated as a possible trigger of SSNHL. Reports of otoneurological manifestations associated with COVID-19 have highlighted sudden hearing loss as a potential complication, although the underlying mechanisms remain unclear.11

Direct identification of viral involvement in the inner ear is limited due to the invasiveness of cochlear biopsy, making indirect methods such as PCR and serological testing essential for etiological investigation.7–9 Given the high proportion of idiopathic cases, viral infections may be underdiagnosed in clinical practice.

Therefore, this study aimed to investigate the presence and clinical impact of Herpesviridae and SARS-CoV-2 association, in patients with sudden sensorineural hearing loss, characterizing their audiometric and clinical profiles to advance etiological understanding and guide management strategies.

A total of 67 patients were initially screened for eligibility. Fourteen patients were excluded based on predefined criteria, resulting in a final sample of 53 patients. Among the included patients, 56.6% were male, with a mean age of 46-years (Table 1). Most cases were unilateral (90.6%). Associated symptoms were frequent: tinnitus was reported by 90.6% of patients, while vertigo was present in 52.8%.

Hypertension (39.62%) and diabetes mellitus (18.87%) were the most prevalent comorbidities in the cohort. A considerable proportion of patients presented both conditions simultaneously (16.98%). Other less frequent comorbidities included dyslipidemia, asthma, hypothyroidism, autoimmune diseases, anxiety disorders, and chronic kidney disease.

Initial hearing loss severity was classified as normal in 9.4%, mild in 13.2%, moderate in 20.8%, severe in 30.2%, and profound in 26.4% of cases (Table 2). Overall, 66% of patients experienced some degree of clinical improvement during follow-up.

Serological testing revealed evidence of acute or recent viral exposure in 43.4% of patients. IgM positivity was observed for SARS-CoV-2 in 26.4% and for HSV in 22.6%. CMV IgM positivity was present in 5.6% (Table 3). All patients underwent PCR testing, which was negative for HSV, EBV, VZV, and SARS-CoV-2; only one patient tested positive for CMV (1.89%). Notably, IgG levels were elevated for VZV in 37.7% of patients (titers >1:100) and for EBV in 34% (titers >1:160), indicating past exposure or reactivation. Coinfection with two viruses occurred in 11.3% of patients, all of whom showed no clinical improvement (p = 0.0032, Table 4), highlighting the negative impact of viral overlap on prognosis.

Brain and temporal bone MRI were performed in 39 of the 53 patients. The main indication was to exclude retrocochlear pathology, particularly cerebellopontine angle tumors. Fourteen patients did not undergo MRI due to clinical or logistical constraints.

The MRI protocol did not include dedicated sequences for the evaluation of endolymphatic hydrops due to limited availability.

Abnormal findings were observed in 35.8% of patients. The predominant findings consisted of intralabyrinthine signal alterations, typically hyperintense on T2-weighted and FLAIR sequences, occasionally with contrast enhancement on T1-weighted imaging, as illustrated in Fig. 1.

Fig. 1.

Temporal bone magnetic resonance imaging in axial sections. Source: Authors’ personal archive.

The images include: (A) T1-weighted sequence without contrast, showing no pathological hyperintensity in the inner ear structures; (B) T2 FIESTA/CISS sequence, highlighting the inner ear structures, semicircular canals, and vestibulocochlear nerve with high fluid resolution; (C) Post-contrast T1-weighted sequence with fat suppression, showing moderate enhancement, consistent with an inflammatory process (801,000 normalized) in the internal acoustic meatus, basal turn of the cochlea, and right vestibule (red arrows); (D) Post-contrast T1-weighted sequence with fat suppression, showing reduced signal intensity (651,000 normalized), indicating absence of lesion or low inflammatory activity on the left.

Regarding prognostic factors, bilateral hearing loss was significantly associated with poorer outcomes (p = 0.0368, Table 5). Initial severity remained the strongest predictor of non-recovery (p = 0.0173).

In the univariate analysis, COVID-19 (PR = 2.79; 95% CI 1.39–5.57; p = 0.0038), herpes simplex virus types 1 and 2 (PR = 2.17; 95% CI 1.08–4.36; p = 0.0287), time to auditory recovery (PR = 18.00; 95% CI 2.68–120.92; p < 0.0001), sex (PR = 1.99; 95% CI 0.83–4.79; p = 0.1231), and time from symptom onset to treatment (PR = 2.45; 95% CI 1.21–4.96; p = 0.0128) were associated with the outcome at p < 0.25 and were included in the multivariate model.

In the adjusted analysis, COVID-19 infection remained independently associated with a higher prevalence of no improvement of hearing, corresponding to a 111% increase compared to COVID-19-negative patients (p = 0.0129).

Time to auditory recovery, analyzed as a clinical course variable (<15 vs. ≥15 days), was also associated with the outcome, with a 10.74-fold higher prevalence of no improvement compared to patients who recovered within 15 days (p = 0.0158). No significant difference was observed between patients with recovery times longer than 15-days and those with recovery within 15-days (p = 0.5015, Table 6).

This study explores the potential role of viral association in the pathogenesis and prognosis of Sudden Sensorineural Hearing Loss (SSNHL), as supported by high seropositivity rates for herpesviruses and SARS-CoV-2. Coinfection appeared to be associated with poorer hearing outcomes, raising the possibility of a synergistic effect of multiple viral agents in aggravating auditory damage.15,16 Given the sample size, this finding should be interpreted cautiously.

Bilateral involvement, although uncommon,17,18 was consistently associated with a lack of hearing recovery, in line with previous reports. Similarly, the degree of initial hearing loss remained the most important prognostic factor, corroborating prior studies and underscoring the need for early recognition and prompt management.14

In the multivariate analysis, COVID-19 infection remained independently associated with poorer hearing outcomes, reinforcing the potential role of viral factors in disease progression.

Delayed auditory recovery was also associated with worse outcomes, likely representing a marker of disease severity and clinical course rather than an independent predictor. Other variables did not retain statistical significance after adjustment, suggesting that their effects may be influenced by confounding factors.

Our findings contribute to the current body of knowledge by documenting the frequency of viral serological and molecular markers in patients with SSNHL. However, serological findings alone are insufficient to establish active infection or reactivation.

We identified patients with positive IgM for herpes simplex virus and SARS-CoV-2. In addition, cases compatible with Cytomegalovirus (CMV) reactivation were observed serologically, and notably, one adult patient presented CMV positivity on PCR testing, a finding rarely reported in the literature. These observations suggest that both primary infections and viral reactivation may be associated with the onset and clinical course of SSNHL, although they do not establish a causal relationship.16,19

Despite the absence of confirmed active or reactivated infections in some patients, elevated IgG titers for VZV and EBV were consistently observed, indicating prior immune stimulation and a possible role in susceptibility or recovery dynamics.20

PCR testing yielded predominantly negative results, which is consistent with the known limitations of blood-based PCR for neurotropic viruses. HSV, VZV, EBV, and CMV exhibit tropism for cochlear and neural tissues, where viral replication may not be adequately reflected in peripheral blood. Additionally, the interval between symptom onset and sample collection, up to 30 days, may have reduced detectability. Therefore, PCR negativity should not be interpreted as exclusion of viral participation, but rather as a limitation of systemic sampling in inner ear disease.

MRI examinations were performed primarily to exclude retrocochlear pathology. Neuroimaging findings provided additional context for a possible infectious-inflammatory process,21 although several methodological limitations must be acknowledged. Imaging was performed using a conventional 1.5-Tesla protocol without dedicated endolymphatic hydrops sequences, reducing sensitivity for detecting subtle inner ear abnormalities. Moreover, in most cases, imaging occurred only after completion of systemic corticosteroid therapy and at delayed intervals following SSNHL onset. This timing may have reduced the ability to identify acute cochlear inflammation, as corticosteroids can suppress contrast enhancement, while enhancement observed later may reflect post-inflammatory changes rather than active disease.

Inner ear signal abnormalities suggestive of inflammatory changes were identified in a subset of patients, more frequently among those with positive viral serology or PCR; however, given the observational nature of the study, these findings are not diagnostic and should be interpreted as indicative of a possible infectious-inflammatory process rather than establishing a causal relationship or defining the underlying etiology of SSNHL.

When considered alongside contemporary literature, including post-COVID SSNHL reports and the prior findings,16 our results are consistent with multiple mechanisms for virus-associated auditory dysfunction: (1) Viral reactivation from latency; (2) Endothelial injury and disruption of cochlear microvasculature; (3) Microthrombosis; (4) Cytokine-mediated inflammation; and (5) Direct or immune- mediated damage to the organ of Corti. The observation of coinfections further supports the possibility that combined viral and inflammatory pathways may contribute to cochlear injury and influence recovery.

These data support the concept that SSNHL is a heterogeneous condition in which viral exposure may act as a contributing or modulating factor rather than a direct etiological agent. Serological and imaging findings may therefore be considered complementary tools for clinical characterization and prognostic stratification, although their role in routine clinical decision-making remains to be established.

The present study has limitations that should be acknowledged. Its relatively small sample size, single-center design, and the absence of a control group restrict generalizability and limit causal inference regarding viral association. The absence of a control group also limits interpretation of whether seropositivity exceeds background population levels.

This study was designed as a prospective cohort with consecutive inclusion and longitudinal follow-up of patients with SSNHL, including repeated clinical, audiometric, serological, and imaging assessments during the acute and subacute phases. As such, the focus was on within-cohort temporal and prognostic associations rather than comparison with an external control group.

The interpretation of serological findings remains challenging, and blood-based PCR has inherently low sensitivity for neurotropic viruses.

Comorbidities such as hypertension and diabetes were actively managed during the study period, including monitoring blood pressure and glycemic levels, particularly during corticosteroid therapy. However, standardized longitudinal data allowing formal classification of disease control (e.g., well-controlled vs. uncontrolled) were not consistently available for all patients, which represents a limitation.

Furthermore, the 60–90 day follow-up period was intended to capture early hearing recovery after initial treatment, which occurs predominantly within the first month. Although longer follow-up is relevant for identifying recurrence and other etiologies, this study focused on acute-phase outcomes and infectious associations.

Despite these limitations, the study retains important methodological strengths, including its prospective design and the integration of audiometry, MRI, serology, and PCR. By evaluating both herpesviruses and SARS-CoV-2 during a unique epidemiological period, this study contributes to the growing body of literature exploring potential viral and immunological factors in SSNHL and highlights the need for future multicenter and mechanistic investigations.

Serological evidence of viral exposure was observed in a substantial proportion of patients with Sudden Sensorineural Hearing Loss (SSNHL). Positivity for herpesviruses and SARS-CoV-2, particularly in cases of coinfection, was associated with poorer clinical outcomes. Initial hearing loss severity and bilateral involvement remained the main prognostic determinants. Intralabyrinthine abnormalities on magnetic resonance imaging were more frequently observed among seropositive patients, suggesting a potential role of virus associated inflammatory mechanisms in SSNHL.

Tracy Lima Tavares Soeiro: 0000-0003-0665-2539

Marina Cançado Passarelli Scott: 0000-0002-1854-6509

Breno Lima de Almeida: 0000-0002-3276-617X

Jônatas Bussador do Amaral: 0000-0001-5819-0474

Elizabeth Cristina Perez: 0000-0003-3038-9902

Sujana S. Chandrasekhar: 0000-0002-8523-1769

Norma de Oliveira Penido: 0000-0003-3496-410X

Tracy Lima Tavares Soeiro: Study design; data acquisition; drafting the manuscript; approval of final version; agree to be held accountable for all information described in the manuscript. Marina Cançado Passarelli Scott and Breno Lima de Almeida: Data acquisition; statistical analysis; approval of final version; critical revision of the final version; agree to be held accountable for all information described in the manuscript. Jônatas Bussador do Amaral and Elizabeth Cristina Perez : Data acquisition; statistical analysis; drafting of the manuscript; critical revision and approval of final version; agree to be held accountable for all information described in the manuscript. Sujana S. Chandrasekhar: Study design; critical revision and approval of final version; agree to be held accountable for all information described in the manuscript. Norma de Oliveira Penido: Study design; data acquisition; drafting the manuscript; critical revision, and approval of final version; agree to be held accountable for all information described in the manuscript.

This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; process number 2021/07346-6). Tracy Lima Tavares Soeiro received funding from the Coordination for the Improvement of Higher Education Personnel (CAPES). Norma de Oliveira Penido received funding from the National Council for Scientific and Technological Development (CNPq; process number 402014/2025-1). For the remaining authors, none were declared.

The authors declare that all data are available in repository.