Diagnostic audiology: audiometry, immittance, OAE and ABR

In short

Diagnostic audiology is the ordered battery of tests that establishes whether hearing loss exists, how much, in which ear, at which site in the auditory pathway, and whether the findings are internally consistent. It runs from pure tone and speech audiometry (behavioural) through immittance (middle ear), otoacoustic emissions (cochlear outer hair cells) and auditory brainstem response (eighth nerve and brainstem) — the cross-check principle. Every BASLP university paper carries it, and it is the core of AIISH/AYJNISHD entrance and PG tests, the RCI-registered audiologist's daily work, and the technical round for jobs in ENT hospitals, hearing-aid retail chains and cochlear implant companies. It is also the legal basis of disability certification under the Rights of Persons with Disabilities Act, 2016 and of noise-induced hearing loss claims in industry.

Explained

Pure tone audiometry (PTA)

• Purpose: find the threshold — the lowest intensity at which a tone is heard 50% of the time — at each frequency, by air conduction (AC) and bone conduction (BC).
• Units: dB HL (hearing level, referenced to normal hearing per the audiometric zero in the relevant audiometer calibration standard, implemented in India through BIS-adopted IEC/ISO references); dB SPL is referenced to 20 µPa; dB SL is above the individual's threshold.
• Frequencies: AC at 250, 500, 1000, 2000, 4000, 8000 Hz (inter-octaves 750, 1500, 3000, 6000 Hz when adjacent octaves differ by ≥20 dB). BC at 250–4000 Hz only — BC output limits and vibrotactile response restrict it.
• Transducers: supra-aural (TDH-39/49) or insert earphones (ER-3A, better interaural attenuation, avoid collapsing canals); bone vibrator (B-71) on mastoid or forehead; sound-field speakers for infants and aided testing.
• Method (modified Hughson–Westlake, ASHA-style): start 30 dB HL in the better ear at 1000 Hz → if heard, down in 10 dB steps till no response → up in 5 dB steps → threshold = lowest level with ≥2 of 3 (or 3 of 5) ascending responses. Order: 1000, 2000, 4000, 8000, then re-check 1000, then 500, 250.
• Audiogram convention: right ear O (AC, red), left X (AC, blue); BC unmasked < (right) and > (left); masked BC [ and ]; arrow added for "no response". Frequency on x-axis (log), dB HL on y-axis increasing downwards.
• Degree (Clark/Goodman-type grading used widely in India, on the better-ear average): normal ≤25 dB; mild 26–40; moderate 41–55; moderately severe 56–70; severe 71–90; profound >90 dB HL. WHO's 2021 revision uses ≥20 dB as the cut-off and renames the bands — exam answers should state which scheme is being used.
• Configuration: flat, rising, gradually/steeply sloping, trough (saucer), notched (4 kHz notch in noise-induced loss).
• Type from AC–BC relationship: conductive (BC normal, AC poorer, air–bone gap >10–15 dB), sensorineural (AC and BC both depressed, no significant gap), mixed (both depressed plus a gap).

Masking, and the test-room environment

• Why: sound presented to the test ear can cross to the non-test ear. Interaural attenuation (IA): ~40 dB for supra-aural AC (frequency-dependent, as low as 35–40 dB at 250 Hz), ~60 dB for inserts, and 0 dB for BC — so BC always risks crossover.
• When to mask (AC): AC of test ear − BC of non-test ear ≥ 40 dB (supra-aural). When to mask (BC): whenever an air–bone gap ≥10 dB exists in the test ear.
• Masker: narrow-band noise for tones; speech noise for speech tests. Effective masking (EM) level is expressed in dB EM.
• Plateau (Hood) method: raise masking in 5–10 dB steps; test-ear threshold stays unchanged over a plateau of ≥15–20 dB → true threshold. Overmasking when masker in non-test ear crosses back and shifts the test ear; undermasking when masker is too low.
• Masking dilemma: large bilateral air–bone gaps leave no masking level that is both adequate and non-overmasking — resolved with insert earphones (higher IA) or by ABR/immittance.
• Occlusion effect: covering the ear raises low-frequency BC sensitivity (about 20 dB at 250 Hz, 15 at 500, 10 at 1000, none ≥2000) in normal/SN ears — add it to the initial masking level for BC.
• Room: testing requires an audiometric room meeting permissible ambient noise limits; booths should be calibrated and annual audiometer calibration (electroacoustic, with biological daily check and listening check) documented.

Speech audiometry

• SRT / SAT: speech recognition threshold — lowest level at which 50% of spondees (two-syllable, equally stressed words) are repeated; speech awareness/detection threshold when words cannot be repeated. SRT should agree with the PTA average within ±6–10 dB; a gross mismatch suggests non-organic loss, auditory processing problem or wrong word list.
• SDS / WRS: speech discrimination (word recognition) score — percentage of phonetically balanced monosyllables correct, usually at 40 dB SL (re SRT) or at the PB-max level.
• Indian language material: standardised word lists in Hindi, Kannada, Tamil, Telugu, Malayalam, Marathi, Bengali and others, largely developed and validated at AIISH Mysuru and allied institutes — language-matched lists are mandatory in practice, and using an English list for a non-English speaker invalidates the score.
• PI-PB function (performance-intensity): plot SDS against intensity. Rollover — score falls markedly at high levels (rollover index > about 0.45) — points to retrocochlear pathology.
• Other measures: MCL (most comfortable level), UCL/LDL (uncomfortable level), dynamic range = UCL − SRT; narrow dynamic range indicates recruitment, important for hearing-aid fitting.
• SNR tests: speech-in-noise/SPIN, QuickSIN-type tests, dichotic tests for central function.

Immittance audiometry (tympanometry and acoustic reflex)

• Principle: a probe delivers a 226 Hz probe tone (1000 Hz for infants under about 6 months, whose canal and middle ear are more compliant/mass-dominated), a pump varies canal pressure from about +200 to −400 daPa, and a microphone measures sound reflected — expressed as admittance (mmho or ml equivalent volume) or its reciprocal, impedance.
• Measures: peak compensated static admittance (Ytm) normal roughly 0.3–1.6 ml in adults; tympanometric peak pressure (TPP) normal about +50 to −100 daPa; ear canal volume (ECV) normal roughly 0.6–1.5 ml adults, 0.3–0.9 ml children; tympanometric width/gradient.
• Jerger types: A normal; As shallow (stiff — otosclerosis, tympanosclerosis); Ad deep/hypercompliant (ossicular discontinuity, flaccid membrane); B flat (middle-ear effusion if ECV normal; perforation or patent grommet if ECV large; impacted wax or probe against wall if ECV very small); C negative pressure peak (eustachian tube dysfunction).
• Acoustic (stapedial) reflex: contraction of the stapedius to loud sound (≈70–100 dB HL in normals), bilateral and consensual. Ipsilateral and contralateral arcs; pathway = cochlea → VIII nerve → cochlear nucleus → superior olivary complex → facial (VII) motor nucleus → stapedius.
• Reflex uses: cross-check of threshold (reflex at normal levels argues against severe loss), site-of-lesion patterns (absent with middle-ear disease even in a mild loss; elevated/absent with VIII nerve lesion), topodiagnosis of facial nerve palsy, and reflex decay (≥50% drop within 5 s at 500/1000 Hz = retrocochlear sign).
• Eustachian tube function tests, wideband/multifrequency tympanometry and resonance frequency extend the battery; Carhart's notch (BC dip ~2000 Hz) in otosclerosis is an audiometric, not immittance, sign.

Otoacoustic emissions (OAE)

• Nature: low-level sounds generated by the motility of outer hair cells (the cochlear amplifier), recorded in the ear canal — evidence of pre-neural cochlear function, not of hearing.
• Types: spontaneous (SOAE) in ~50% of normal ears; transient evoked (TEOAE) to clicks, broadband, screens roughly 1–4 kHz; distortion product (DPOAE) to two tones f1 and f2 (f2/f1 ≈ 1.22), measured at 2f1 − f2, frequency-specific up to 6–8 kHz; stimulus frequency (SFOAE).
• Pass criteria (typical): SNR ≥6 dB in required frequency bands with adequate reproducibility (TEOAE often ≥70%) — exact criteria are protocol- and device-specific.
• Clinical value: present OAE with raised behavioural thresholds and absent/abnormal ABR is the signature of auditory neuropathy spectrum disorder (ANSD); OAE suppression by contralateral noise tests the medial olivocochlear efferent system; OAEs detect ototoxic and noise damage earlier than the audiogram.
• Limitations: abolished by any conductive problem (wax, fluid, perforation) and by ambient/patient noise; absent once cochlear loss exceeds roughly 30–40 dB HL; gives no threshold value and no information above the cochlea.

Auditory brainstem response (ABR/BERA) and electrophysiology

• What it is: a far-field, scalp-recorded auditory evoked potential in the first 10–15 ms after a click or tone burst, averaged over 1000–2000 sweeps; not dependent on patient response or sleep state (sedation acceptable).
• Recording: electrodes at Cz/high forehead (non-inverting), ipsilateral mastoid/earlobe (inverting), contralateral mastoid or low forehead (ground); impedance <5 kΩ and interelectrode difference <2 kΩ; filters ~30–1500/3000 Hz; rate ~11–30/s for neurodiagnosis, up to 30–50/s for threshold work; insert phones (adds ~0.9 ms tube delay).
• Waves I–VII (Jewett): I = distal VIII nerve, II = proximal VIII nerve/cochlear nucleus, III = cochlear nucleus/trapezoid body, IV–V complex = lateral lemniscus/inferior colliculus region. Adult normative latencies at 80 dB nHL: I ≈ 1.5 ms, III ≈ 3.7 ms, V ≈ 5.7 ms; I–V interpeak ≈ 4.0 ms (<4.5 ms), I–III ≈ 2.1, III–V ≈ 1.9.
• Neurodiagnostic indices: absolute wave V latency, I–V interval, interaural wave V latency difference (IT5, abnormal >0.2–0.4 ms), V/I amplitude ratio (<0.5 suspicious), latency shift with high repetition rate, and absent waves beyond I.
• Threshold estimation: lower intensity in 10–20 dB steps and track wave V to the lowest level with a replicable response (ABR threshold in dB nHL); click ABR reflects 2–4 kHz, so use tone-burst ABR at 500, 1000, 2000, 4000 Hz for frequency-specific thresholds and bone-conduction ABR to separate conductive from sensorineural loss in infants. Correction factors (roughly 5–15 dB, frequency- and protocol-specific) convert dB nHL to estimated HL.
• Polarity: rarefaction usually gives clearer wave I; a cochlear microphonic that inverts with polarity reversal and disappears when the insert tubing is clamped (a stimulus artefact would persist) is the key ANSD sign.
• Factors affecting ABR: intensity (latency lengthens, amplitude falls as level drops), conductive loss (all waves delayed, intervals normal), high-frequency cochlear loss (wave V present, wave I poor), age (immature until ~18 months, myelination), body temperature, sex and head size (small latency differences), muscle and electrical artefact.
• Related tests: ASSR (auditory steady-state response) gives objective, frequency-specific, statistically detected thresholds usable to high intensities, valuable for severe-to-profound loss and implant candidacy; ECochG/electrocochleography (SP/AP ratio >0.4–0.5 in Ménière's); middle latency (MLR) and late/cortical P1-N1-P2 and P300/MMN for cortical and processing assessment; vestibular VEMP and vHIT sit alongside but are not part of the hearing battery.

Battery logic, special tests and Indian practice

• Cross-check principle (Jerger & Hayes): no single test result is accepted without an independent confirmation — PTA against SRT, immittance and OAE; behavioural results against ABR in children.
• Site-of-lesion pattern:
  • Conductive: air–bone gap, type B/C/As tympanogram, absent reflexes, absent OAE, ABR waves delayed with normal intervals.
  • Cochlear: no gap, type A, reflexes present at reduced sensation level (recruitment), OAE absent in affected frequencies, ABR with normal I–V.
  • Retrocochlear: reflex decay/elevation, rollover on PI-PB, prolonged I–V or IT5, OAE often present.
  • ANSD: OAE/CM present, ABR absent or grossly abnormal, reflexes absent, poor speech scores out of proportion to the audiogram.
• Non-organic (functional) hearing loss: SRT–PTA disagreement, inconsistent thresholds, Stenger test, Lombard/Doerfler-Stewart (historic), and objective confirmation by ABR/ASSR — important in medico-legal and employment screening.
• Older site-of-lesion tests (still examined): SISI, tone decay (Carhart, Rosenberg), ABLB for recruitment, Bekesy tracing types I–V — largely replaced by immittance, OAE and ABR.
• Indian regulatory and practice context: audiologists must be registered with the Rehabilitation Council of India (RCI) under the RCI Act, 1992 to practise; the National Commission for Allied and Healthcare Professions Act, 2021 also covers the sector. Disability certification for hearing impairment follows the RPwD Act, 2016 and the Government of India's gazette guidelines for assessing hearing disability, which use a frequency-weighted better-ear calculation together with speech discrimination; benchmark disability is 40%. Hearing devices are listed under ADIP; newborn screening under RBSK/NPPCD uses OAE and ABR. Industrial hearing conservation now sits under the Occupational Safety, Health and Working Conditions Code, 2020 (in force 21 November 2025, replacing the former Factories Act provisions), with ambient-noise limits under the Noise Pollution (Regulation and Control) Rules, 2000; the audiometric-zero and audiometer performance references are the BIS-adopted IEC 60645/ISO 389 series — quote the clause only if the current edition is at hand.

Key points

• Threshold - lowest level detected 50% of the time; measured in dB HL, plotted on the audiogram.
• Air–bone gap - AC threshold minus BC threshold at the same frequency; >10–15 dB means a conductive component.
• Interaural attenuation - ~40 dB supra-aural, ~60 dB insert, 0 dB bone — the whole basis of masking rules.
• Plateau method - masking raised in steps; an unchanged test-ear threshold across ≥15 dB of masking confirms the true threshold.
• Occlusion effect - 20/15/10 dB at 250/500/1000 Hz; added to initial masking for bone conduction.
• Carhart's notch - BC dip of about 15 dB at 2000 Hz, classically in otosclerosis; an artefact of middle-ear mechanics, not a cochlear loss.
• SRT–PTA agreement - within about ±6–10 dB; disagreement flags non-organic loss or wrong word list.
• Rollover - fall in word score at high intensity; retrocochlear indicator from the PI-PB function.
• Tympanogram types - A normal, As stiff, Ad hypercompliant, B flat, C negative pressure.
• Ear canal volume with a type B curve - large = perforation, small = wax or blocked probe; it decides the interpretation.
• Acoustic reflex arc - cochlea → VIII → cochlear nucleus → superior olive → VII nucleus → stapedius; absent in middle-ear disease.
• Reflex decay - ≥50% amplitude loss in 5 s at 500/1000 Hz = retrocochlear sign.
• OAE - outer hair cell product; needs a clear conductive path; absent beyond roughly 30–40 dB cochlear loss; no threshold information.
• DPOAE vs TEOAE - DPOAE is frequency-specific to higher frequencies; TEOAE is broadband, best 1–4 kHz.
• ABR wave V - most robust wave, used for threshold tracking; I–V interval under ~4.5 ms is the key neural index.
• ANSD - OAE and cochlear microphonic present, ABR absent/abnormal, reflexes absent, speech disproportionately poor.
• ASSR - objective, frequency-specific, statistically detected thresholds; useful in severe-profound loss and implant work-up.
• Cross-check principle - no test stands alone; every result must be confirmed by an independent measure.

Formulas and examples

Example 1 — PTA, type and degree. Right ear AC: 250 Hz 45, 500 Hz 50, 1 k 55, 2 k 55, 4 k 60 dB HL. BC: 500 Hz 20, 1 k 25, 2 k 30 dB HL.

• PTA₃ (AC) = (50 + 55 + 55)/3 = 160/3 = 53.3 dB HL → moderate loss on the 41–55 dB band.
• PTA₃ (BC) = (20 + 25 + 30)/3 = 75/3 = 25 dB HL.
• ABG = 30, 30, 25 dB at 500/1 k/2 k → average 28.3 dB → mixed hearing loss (both AC and BC raised, with a large gap).
• Masking check (supra-aural, IA 40 dB): is AC(right) − BC(left) ≥ 40? With left BC at, say, 10 dB, 50 − 10 = 40 → masking of the left ear is required for the right AC test; and since ABG ≥10 dB, right BC must also be masked.

Example 2 — initial masking and overmasking. Testing right BC at 500 Hz. Left ear AC threshold at 500 Hz = 15 dB HL; right BC apparent threshold = 20 dB HL; IA for BC = 0 dB; OE at 500 Hz = 15 dB.

• Initial masking M = AC(NTE) + OE + 10 = 15 + 15 + 10 = 40 dB EM in the left ear.
• Overmasking limit: masker crosses to the test ear when M − IA(AC of masker through headphone, 40 dB) ≥ BC(test ear) → M ≥ 20 + 40 = 60 dB EM.
• Usable plateau = 40 to <60 dB EM, i.e. about 20 dB wide — adequate; if the right BC were 55 dB, the limit would be 95 dB and the plateau wider still.

Example 3 — SRT cross-check and dynamic range. PTA₃ = 48 dB HL; SRT = 50 dB HL; UCL = 100 dB HL; SDS measured at 40 dB SL.

• SRT − PTA = 50 − 48 = 2 dB → good agreement, results reliable.
• Presentation level for SDS = SRT + 40 = 90 dB HL, which is within the UCL of 100 dB — acceptable.
• Dynamic range = 100 − 50 = 50 dB — comfortable for a conventional hearing aid; a DR of 25 dB or less would demand strong compression.

Example 4 — rollover index. Word scores: 50% at 60 dB HL, 70% at 70 dB, 80% at 80 dB (PB-max), 60% at 90 dB, 40% at 100 dB (PB-min).

• RI = (80 − 40)/80 = 40/80 = 0.50 → above ≈0.45 → rollover present, so retrocochlear involvement must be ruled out with ABR and imaging.

Example 5 — ABR interpretation. Right ear at 80 dB nHL: I = 1.6, III = 3.9, V = 6.4 ms. Left ear: I = 1.5, III = 3.7, V = 5.7 ms.

• Right I–V = 6.4 − 1.6 = 4.8 ms (normal <4.5) → prolonged. Right III–V = 6.4 − 3.9 = 2.5 ms (normal ≈1.9) → prolonged.
• Left I–V = 5.7 − 1.5 = 4.2 ms → normal.
• IT5 = |6.4 − 5.7| = 0.7 ms → well above 0.2–0.4 ms.
• Conclusion: right retrocochlear (brainstem-side) abnormality, delay arising above wave III; refer for MRI with contrast of the internal auditory canal. Tympanometry and OAE were normal, which excludes a conductive cause of the delay.

Where it is used

• Hospital and ENT audiology: AIIMS, PGIMER, CMC Vellore, government medical college ENT departments and district hospitals — diagnostic batteries, pre-operative assessment for tympanoplasty and stapedotomy, ABR for acoustic neuroma work-up.
• National institutes and teaching: AIISH Mysuru, AYJNISHD Mumbai, NISH Thiruvananthapuram, SVNIRTAR — clinical services, standardised test material development, RCI-approved training.
• Hearing-aid retail and manufacturer roles: audiologists at Hearing-aid chains and OEM/distributor networks (Phonak/Sonova, Oticon, Signia/WS Audiology, ReSound, Starkey, Widex and Indian firms such as Arphi and Alps) — diagnostic PTA plus real-ear measures as the basis of fitting.
• Cochlear implant and device programmes: candidacy evaluation (aided audiogram, ASSR, imaging correlation) and intra-operative/post-operative objective measures under ADIP and state implant schemes.
• Newborn and school screening programmes: OAE-first and ABR-confirm protocols in NPPCD and RBSK, hospital-based universal screening in corporate chains.
• Occupational and medico-legal audiology: industrial hearing conservation audiometry and noise surveys under the OSH&WC Code, 2020; disability certification boards under the RPwD Act, 2016; fitness and compensation assessments for railways, mining-sector employers, defence and aviation.

Interview questions

• Q: Why is bone conduction tested only up to 4000 Hz? A: The bone vibrator's maximum output is limited and at high intensities low-frequency vibration is felt rather than heard (vibrotactile response), so 250–4000 Hz is the valid range.
• Q: When do you mask for air conduction? A: When the test ear's AC threshold minus the non-test ear's BC threshold equals or exceeds the interaural attenuation — about 40 dB with supra-aural phones, 60 dB with inserts.
• Q: What is the masking dilemma and how do you deal with it? A: With large bilateral air–bone gaps, any masker loud enough to mask is also loud enough to cross back and overmask; use insert earphones for their higher interaural attenuation, or rely on immittance and ABR.
• Q: A flat type B tympanogram with a large ear canal volume — what does it mean? A: The probe is measuring a volume beyond the drum, so there is a tympanic membrane perforation or a patent ventilation tube, not an effusion.
• Q: Can a patient have a normal audiogram and absent OAEs? A: Yes — a mild conductive problem such as wax, negative middle-ear pressure or fluid can abolish OAEs while thresholds stay within normal limits; conversely early outer hair cell damage can show up before the audiogram changes.
• Q: OAEs are present but ABR is absent. What is your diagnosis? A: Auditory neuropathy spectrum disorder — the cochlea's outer hair cells work but neural synchrony in the eighth nerve/brainstem fails; confirm with the cochlear microphonic polarity test and absent acoustic reflexes.
• Q: Is ABR a test of hearing? A: No. It measures neural synchrony to a transient stimulus; a response at a given level estimates threshold but a normal ABR does not guarantee normal hearing, nor does an absent one always mean deafness.
• Q: Why do we prefer tone-burst ABR in infants? A: Click ABR mainly reflects the 2–4 kHz region; tone bursts at 500, 1000, 2000 and 4000 Hz give frequency-specific thresholds needed to fit a hearing aid, with bone-conduction ABR added to identify a conductive component.
• Q: How does a conductive loss change the ABR? A: It attenuates the effective stimulus, so all waves are delayed and amplitudes reduced, but the interpeak intervals stay normal — unlike a retrocochlear lesion which stretches I–V.
• Q: What is Carhart's notch and why does it occur? A: A bone-conduction dip of about 15 dB near 2000 Hz in otosclerosis; it is a mechanical artefact of the fixed ossicular chain's effect on bone-conducted sound, and it typically closes after successful stapes surgery.
• Q: How do you detect non-organic hearing loss? A: Look for poor test–retest consistency and an SRT much better than the pure tone average, use the Stenger test for unilateral claims, and confirm objectively with ABR or ASSR.
• Q: What legal and professional requirements govern your practice in India? A: Practice requires registration with the Rehabilitation Council of India under the RCI Act, 1992 (with the National Commission for Allied and Healthcare Professions Act, 2021 also applying); disability certification follows the Government of India's hearing-disability assessment guidelines under the RPwD Act, 2016, and occupational audiometry falls under the OSH&WC Code, 2020.

Common mistakes

• Reporting an air–bone gap without masked bone conduction — unmasked BC can belong to the other ear, since interaural attenuation for bone is zero.
• Forgetting the occlusion effect when calculating the initial masking level for bone conduction, which starts the plateau too low.
• Calling a type B tympanogram "middle-ear fluid" without reading the ear canal volume.
• Treating absent OAEs as proof of sensorineural loss, when wax, fluid or a noisy baby explains it.
• Using wave V latency alone in neurodiagnosis instead of the I–V interval and interaural difference, or using ABR thresholds as dB HL without applying the correction factor.
• Running speech audiometry with an English or non-native word list, then interpreting a low discrimination score as pathology.
• Skipping the cross-check — accepting a behavioural audiogram in a young child without immittance, OAE and, where indicated, ABR.

Related topics

Anatomy and physiology of the auditory and vestibular system · Paediatric audiology and newborn hearing screening · Hearing aids: selection and fitting · Cochlear implants and aural rehabilitation · Vestibular assessment: VNG, VEMP and vHIT · Industrial audiology, noise measurement and hearing conservation.