11Sensorineural Hearing Loss After Ossiculoplasty

FThe complication that cannot be revised

Almost everything that goes wrong after ossiculoplasty is, in principle, recoverable. A prosthesis that extrudes can be replaced; one that slips off the stapes can be reseated; an air–bone gap that reopens can be revised. These are conductivefailures — the mechanical link between drum and inner ear has broken again, but the cochlea is intact and a second operation can offer the patient another chance at hearing.

Sensorineural hearing loss is the exception, and that is why it is feared. Here the injury is to the inner ear itself— the hair cells of the cochlea or the perilymph that bathes them. The harm shows not as a persisting gap but as a fall in the bone-conduction threshold, the very ceiling on what any reconstruction could achieve. At its mildest it is a few decibels of high-frequency loss the patient may never notice. At its worst it is a dead ear: total, irreversible loss of hearing and often balance in that ear, from a single operative event. Unlike a conductive failure, it cannot be repaired by reseating a prosthesis or revising a graft. This asymmetry — one family of complications recoverable, the other not — shapes how a careful surgeon weighs every decision, accepting a more modest conductive result rather than gamble the cochlea [1992].

FHow often does it happen?

The reassuring news for counselling is that significant, permanent sensorineural loss is uncommon. In pooled titanium ossiculoplasty data, new sensorineural loss occurs on the order of 3–4% of ears [2023], and a prospective cohort of type I tympanoplasties recorded a 5.1%incidence — with fifteen of sixteen affected ears having been operated by trainees rather than the consultant, identifying surgeon experience as the dominant modifiable factor [2016]. A true dead ear is rarer still, a fraction of one percent, but its consequences are so grave that it must be discussed before every procedure.

These headline figures hide an important subtlety about timing. When hearing is measured in the first hours after surgery, far more ears show a bone-conduction dip — in one series, significant high-frequency loss appeared in 23% of tympanoplasty-only ears and 48% of ears that also underwent mastoidectomy at 24 hours. Most of this recovers: by six months the persistent figure had fallen to about 16% in the mastoid group and to nothing measurable in the tympanoplasty-only group [2015]. The early dip is largely reversible acoustic trauma; the small residue is the permanent injury that matters for prognosis. The chart below separates these.

The lesson is twofold. First, a temporary postoperative bone-conduction dip is common and usually settles, so an audiogram on day one should be interpreted with caution. Second, the fact that drilling and mastoid work multiply the early-trauma rate points straight at the mechanisms — and at where the surgeon’s habits can change the odds.

THow injury reaches the cochlea

Cochlear injury during ossiculoplasty arrives by four broad routes, and they are worth keeping mentally separate because each has its own defence. The first is acoustic trauma. A high-speed burr is an intense sound source, and when it touches a mobile, sound-conducting ossicle it couples that energy directly into the perilymph. Intracochlear pressure measurements in temporal bones show that incidental drilling on the ossicular chain generates pressures equivalent to ear-canal sound levels well above 100 dB SPL, with peaks reaching the range of damaging impulse noise [2017]. Suction-irrigation adds its own insult, exposing the ipsilateral ear to averaged levels up to about 107 dB(A) [1980]. The result is a classic high-frequency noise-induced loss, worst around 2–4 kHz [1989].

The second route is mechanical manipulation of the chain. Every forceful touch on the stapes or footplate sends an impulse into the cochlear fluid; cumulative levering, repositioning and crimping can disturb hair cells even without a visible breach. The third is the over-tight or over-long prosthesis, which pre-loads the annular ligament and footplate, immobilises the stapes and forces sustained pressure into the vestibule — dampening transmission and risking subluxation. The fourth and most catastrophic is a frank footplate fracture or annular-ligament tear, which opens the perilymphatic space to leakage and contamination. The widget below pairs each mechanism with its intraoperative countermeasure.

TReading the bone line

Detecting a sensorineural complication depends on one principle that every post-operative audiogram must be read against: air conduction tells you the result, but only bone conduction tells you about the cochlea.The air–conduction line reflects the whole pathway — canal, drum, ossicles and inner ear together — so an air threshold that is worse after surgery is non-specific, equally explained by effusion, packing, a displaced prosthesis or cochlear injury. Bone conduction bypasses the middle ear and probes the inner ear directly. A genuine fall in the bone-conduction threshold is the unambiguous signature of sensorineural injury and cannot be produced by any conductive cause [1995].

This is precisely why the AAO-HNS reporting standard insists on tracking the bone line and reporting any change in it, not merely the air–conduction gain [1995]. A particularly treacherous trap is the audiogram in which the air–bone gap appears to have closed— but only because the bone line has sunk to meet a static air line. That is not a success; it is a sensorineural loss masquerading as one. The diagram contrasts the two outcomes that can share an identical air–conduction curve.

In practice, the high frequencies fall first, so masked bone-conduction thresholds at 2, 4 and ideally higher frequencies are the sensitive sentinels [1989]. A worsening bone line, especially when paired with vertigo, is never a benign finding to be reassured away — it is the warning that the inner ear has been touched.

CMinimising the risk at the microscope

Because the injury is largely unrecoverable, prevention is the whole game, and it is overwhelmingly a matter of technique and discipline rather than instrumentation. The single highest-yield habit follows directly from the acoustic data: never let a running burr contact an intact, mobile ossicular chain.If drilling is needed near a sound-conducting chain — for instance when widening an attic or removing disease — the chain should be disarticulated or otherwise protected first, the lowest effective drill speed used, and the burr kept off the ossicles. Suction should be gentle and kept well away from an open oval or round window [2017, 1980].

Around the reconstruction itself, a handful of principles protect the cochlea:

• Plan and measure, then place once. Use a depth gauge, decide the construct in advance, and seat the prosthesis in as few deliberate movements as possible. Each extra manipulation of the stapes is another impulse into the perilymph.
• Size for contact, not compression. An over-long prosthesis loads the footplate continuously; tension should be just enough to couple, with the stapes still visibly mobile after seating.
• Respect the vector.Orient the shaft roughly 45–90° to the footplate so force is not driven axially into the vestibule, which also improves stability.
• Steady, never lever. Stabilise the stapes during placement; levering against it risks both subluxation and direct cochlear trauma.
• Operate on a dry, controlled field.Active infection and a bloody field both compound the risk, particularly where the inner ear may be entered — the same reason combined stapedotomy work demands an infection-free ear.

None of these requires special equipment; all require restraint. The recurring theme of the outcome literature — that the state of the ear and the judgement of the surgeon matter more than the brand of prosthesis — applies with full force to cochlear safety [2023].

CFootplate disasters and consent

The worst intraoperative moment in ossiculoplasty is a footplate fracture or annular-ligament tear with perilymph welling up in the oval window. The instinct to push on and finish the reconstruction is exactly wrong. Driving a prosthesis into a breached footplate forces it toward the vestibule and worsens both the leak and the contamination, courting a dead ear. The correct response is to stop: abandon the ossicular reconstruction, seal the oval window immediately with fascia, fat or perichondrium to prevent a persistent perilymph fistula, and stage the ossiculoplasty to a later, dry sitting [1992]. A postoperative fistula — declaring itself as fluctuating vertigo, nausea and a dropping bone line — warrants bed rest and head elevation initially, but persistent symptoms demand early re-exploration and repair rather than watchful waiting.

Finally, the rarity of sensorineural injury does not excuse omitting it from consent. The conversation has a precise structure that mirrors the biology: the common risks (a gap that fails to close or reopens) are recoverableby revision, whereas the uncommon risk — a partial or, at worst, total and permanentsensorineural loss — is not. Quoting an honest figure of a few percent for some inner-ear deterioration, and a much smaller but real chance of a dead ear, lets the patient weigh an elective hearing operation against the small possibility of leaving with worse hearing than they came in with. That is the bargain at the heart of ossiculoplasty, and stating it plainly is part of operating well.