The SeaSHeL National Prospective Cohort Study

Each year, approximately 20 people per 100,000 experience sudden loss of hearing that is sensorineural in nature. 'Sensorineural' indicates an abnormality of the cochlea, the auditory nerve, or higher central auditory pathways. When the hearing loss is of 30 dB (decibel) or more, over at least 3 contiguous frequencies and within 3 days, the condition is termed sudden onset sensorineural hearing loss (SSNHL). SSNHL is predominantly unilateral and the hearing loss can range from mild to profound. In 71% to 90% of cases, the cause is unknown despite investigation, and these cases are termed idiopathic.

Associated symptoms include tinnitus, vertigo and aural fullness. Idiopathic SSNHL is a serious condition, adversely impacting people's lives with research indicating associations with emotional distress, depression, difficulties at work and impaired social integration. It is estimated that 32% to 65% of cases of idiopathic SSNHL recover spontaneously, although clinical experience suggests that this may be an overestimation, with further research required in this area.

Current care pathways for patients suffering from idiopathic SSNHL appear to vary considerably in terms of the type of service patients first present to, their subsequent referral, length of time between onset of symptoms, presentation and start of treatment, the treatment plan as well as follow-up. Treatment options for idiopathic SSNHL include systemic and intratympanic steroids, antiviral agents, rheological agents, diuretics, hyperbaric oxygen treatment, and observation alone. The lack of evidence regarding the comparative effectiveness of these treatments is recognised by clinical guidelines published by National Institute for Health and Care Excellence (NICE), the American Academy of Otolaryngology-Head and Neck Surgery (ORL-HNS), and the British Academy of Audiology.

Based upon recent discoveries in the molecular mechanisms that lead to sensorineural hearing loss, biotechnology and pharmaceutical companies are developing new treatments for patients with idiopathic SSNHL. These treatments require rigorous testing in clinical trials before they can become available for application in a clinical service setting. To allow for such trials to be run effectively reliable information is required on where patients with idiopathic SSNHL present and can be recruited from within the optimum timeframe from onset of symptoms to start of treatment based upon the preclinical profile of the drug. This means that there is an urgent need for information on patient numbers, geographical distribution, demographics, patient and treatment pathways, as well as outcomes.

This study proposes to collect these data through an ENT trainee and Audiologist led nationwide prospective cohort study of adult patients presenting with SSNHL within the NHS. Importantly, this study will not only provide key data to inform future trial design and delivery, but also a unique dataset to develop a prediction model to predict recovery for patients with idiopathic SSNHL.

OBJECTIVES

1. To map the patient pathway and identify the characteristics of adult patients presenting to NHS ENT and hearing services with SSNHL.
2. To develop a prediction model to predict recovery for patients with idiopathic SSNHL.
3. Establish the impact of idiopathic SSNHL on patients' quality of life (QoL)

METHODS

Study design National multicentre prospective observational cohort study.

Setting A multicentre study taking place at NHS centres providing ENT and Hearing Services.

Inclusion criteria

• Adult patients (male or female) aged over 16 years of age presenting to NHS ENT and hearing services with SSNHL AND
• Diagnosed with a hearing loss in one or both ears of 30 dB HL or more, over at least 3 contiguous frequencies, between 250, 500, 1000, 2000, 4000 and 8000 Hz.

AND -Willing and able to provide written informed consent.

Exclusion criteria

-Patients with mixed or conductive hearing loss (CHL). CHL will be defined as a 'true' air-bone gap of 15 dB HL or more in 3 or more contiguous frequencies between 500, 1000, 2000, 4000 Hz.

Primary outcome:

The change in auditory function in the affected ear from initial presentation to follow-up (at any one time between 6 and 16 weeks from onset of symptoms). Auditory function will be defined as the Pure Tone Average (PTA) of air conduction thresholds at 250, 500, 1000, 2000, 4000 and 8000 Hz. If multiple pure tone audiograms have been carried out between 6 and 16 weeks, the most recent pure tone audiogram will be used for the calculation of auditory function.

Change in auditory function classified as:

1. Full recovery: Final PTA in affected ear within 10dB of PTA of unaffected ear (≤10dB)
2. Partial to no recovery: Final PTA in affected ear ≥10dB of PTA of unaffected ear.

Secondary outcomes:

Degree of change in auditory function:

• Complete recovery: Final PTA in affected ear within 10dB of PTA of unaffected ear (≤10dB)
• Marked recovery: PTA improvement ≥30 dB (and final PTA in affected ear ≥ 10dB of PTA of unaffected ear)
• Slight recovery: PTA improvement ≥10dB and ≤30 dB (and final PTA in affected ear ≥10dB of PTA of unaffected ear)
• No improvement: PTA improvement ≤10 dB (and final PTA in affected ear ≥10dB of PTA of unaffected ear)

Quality of life (QoL):

Change in QoL score from initial presentation to follow-up at any one time between 6 and 16 weeks following treatment. QoL will be measured using the Hearing Handicap Inventory for Adults (HHIA) (for patients under 60 years of age) or Hearing Handicap Inventory for Elderly (HHIE) (for patients over 60 years of age) and the Health Utility Index Mark 3 (HUI3). QoL data will be only be collected in a selection of sites (see below).

Ranges for HHIA and HHIE are from 0 (no handicap) to 100 (total handicap) Range for HUI3 = 8 (no handicap) to 45 (total handicap).

Statistical analysis

Prognostic model:

The investigators will develop a multivariable prognostic model to predict recovery for patients with SSNHL in NHS ENT and Hearing services. Missing outcome data at study end will be imputed using multiple imputation by a chained equations procedure. Internal validation will be performed to quantify optimism in the predictive performance (calibration and discrimination) of the developed model using bootstrapping techniques. Bootstrapping techniques provide information on the performance of the model in comparable datasets and generate a shrinkage factor to adjust the regression coefficients.Statistical analysis will be carried out using R software (version 3.5.1).

QoL:

The mean change in HHIA, HHIE and HUI3 scores will be calculated from initial presentation to follow up (any one time between 6 and 16 weeks). The investigators will use the non-parametric Wilcoxon and McNemar-Bowker tests with a significance level of 5% to compare results at patients' initial presentation and at their final follow up. Statistical data analysis will be carried out using the SPSS program 19.0 (SPSS, Chicago, IL, USA). QoL data will be collected from a sub-set of sites (20%, n=20).