6ABR and Electrocochleography

FReading the brainstem’s reply: ABR waves and latencies

The auditory brainstem response (ABR) is a series of far-field potentials, recorded from scalp electrodes, generated as a click travels up the early auditory pathway in the first ~10 milliseconds. Five waves are labelled I–V: wave I from the distal auditory nerve, wave III from the cochlear-nucleus/superior-olive region, and wave V — the most robust and clinically central — from the lateral lemniscus near the inferior colliculus.

The diagnostic currency is timing. The interval between waves (interpeak latencies I–III and I–V) and the difference in wave V latency between the two ears report on conduction along the nerve and brainstem. A prolonged I–V interval or an interaural wave V difference beyond about 0.2 ms is the classic flag for a lesion slowing transmission, such as a tumour [2013].

TABR for retrocochlear screening — and why MRI won

For decades the ABR was the screening test for vestibular schwannoma in patients with unilateral tinnitus or asymmetric loss, because it was cheap and non-invasive. Its fatal weakness is sensitivity: small, intracanalicular tumours frequently leave the ABR normal, so a clean test cannot confidently exclude a lesion. Direct comparisons found that contrast-enhanced MRI detects tumours the ABR misses, particularly the small ones now found earlier [2004].

The consequence is a clear hierarchy: where retrocochlear disease is genuinely suspected, MRI of the internal auditory canals is the definitive test, and modern tinnitus pathways reflect that [2025]. ABR retains niche value — in patients who cannot have MRI, for intra-operative monitoring, and for characterising neural synchrony — but it is no longer the gatekeeper it once was [2014].

CABR signatures in tinnitus itself

Beyond tumour-hunting, the ABR offers a window on the tinnitus mechanism. Several studies of tinnitus patients with normal audiograms report a reduced wave I amplitude — consistent with cochlear synaptopathy, a loss of auditory-nerve synapses that the audiogram cannot see — sometimes accompanied by a relatively preserved or enhanced wave V.

That pattern fits the central-gain model: diminished peripheral output (small wave I) is met by amplified central responses (maintained wave V), the same compensatory up-regulation thought to generate the phantom sound [2013]. These are group-level research findings rather than an individual diagnostic test, but they explain why a structurally ‘normal’ ear can still ring.

TElectrocochleography and the SP/AP ratio

Electrocochleography (ECochG) records potentials from very close to the cochlea — using a transtympanic, ear-drum or canal electrode — isolating the cochlear summating potential (SP) and the auditory-nerve compound action potential (AP). The headline measure is the SP/AP amplitude ratio. In endolymphatic hydrops the basilar membrane is biased by the distended duct, enlarging the SP and so raising the ratio.

An elevated SP/AP ratio (commonly cited above ~0.4, though laboratory thresholds vary) supports a diagnosis of Meniere disease when the clinical picture — episodic vertigo, fluctuating low-frequency loss, aural fullness and roaring tinnitus — fits [2003]. Area-ratio and tone-burst refinements improve on the simple amplitude ratio, but the test’s overall sensitivity is modest, so a normal ECochG never excludes Meniere [2017].

CChoosing the right electrophysiology

The two tests answer different questions, and matching test to suspicion keeps the work-up lean. ECochG belongs to the fluctuating, vertiginous, low-frequency, aural-fullness tinnitus where Meniere or a third-window lesion (such as superior canal dehiscence) is on the differential. ABR belongs to the patient who cannot have MRI but needs retrocochlear reassurance, or to research characterisation of neural synchrony and synaptopathy.

For the common scenario — unilateral tinnitus with asymmetric sensorineural loss and red-flag features — neither evoked-potential test is the answer; MRI is. Used selectively, ABR and ECochG add genuine information; used reflexively, they delay the scan that actually changes management [2014].