10Coupling and Stability at the Prosthesis Interfaces

FWhy coupling and stability decide the result

An ossicular prosthesis is only as good as the points where it touches living tissue. The strut itself — titanium, hydroxyapatite or sculpted bone — is inert; the interfaces are where sound is handed across and where reconstructions succeed or fail. Coupling describes how efficiently vibration crosses each contact; stabilitydescribes whether that contact holds over months and years against the relentless small forces of pressure swings, mucosal adhesion and scar contracture. The two are bound together: a strut that slips loses contact and stops conducting, and a strut that is wedged too tightly to move stops conducting too. From Wullstein’s first columellar reconstructions onward, the surgeon’s task has been to build a bridge that conducts sound and stays put [1956].

Most prosthesis failures are not material failures — they are interface failures. Inadequate contact at either end produces micromotion, the prosthesis works loose, the air–bone gap re-opens, and a bare alloplastic head may erode through the drum and extrude. The corollary is reassuring: when the interfaces are right, designs that differ severalfold in weight and stiffness give broadly comparable hearing, because the prosthesis–tissue interface and surgical technique dominate the outcome over the precise properties of the strut. This module is about getting those contacts right.

FThe three interfaces: drum, stapes head, footplate

A reconstruction couples at up to three places, and which ones are in play depends on what the disease has left behind:

• The lateral interface (drum or malleus handle).This is where vibration enters the strut. A head plate spreads load against the undersurface of the drum, or — far better — engages a retained malleus handle, which preserves part of the native lever and keeps the strut aligned with the drum’s vibratory axis. In a large reconstructed series the residual gap was 11.6 dB when the malleus handle was present versus 16.9 dB when it was absent [2001].
• The medial interface at the stapes head.A partial prosthesis (PORP) rests on the capitulum of a present, mobile stapes superstructure — a stable, well-aligned platform held by the crura. Loose seating lets the strut tilt off the arch, so modern designs grip it: the titanium clip PORP snaps onto the stapes head and resists displacement even under the forces of middle-ear pressure change [2004].
• The medial interface at the footplate. When the superstructure is gone, a total prosthesis (TORP) must balance on the footplateitself — a small, mobile, far less forgiving platform. Without an arch to capture it, a TORP is more prone to tilting and displacement and depends on a stabilising cartilage shoe and meticulous tension.

The clinical signature of these differences is consistent. In a titanium series the mean residual air–bone gap was 14.3 dB for PORPs versus 25.2 dB for TORPs, with closure to within 20 dB in 77% versus 52% of ears, and a present, mobile stapes superstructure was a main predictor of success [2006]. The lesson is that the stapes head is the most stable coupling platform an ear can offer, and preserving a usable arch preserves stability.

TCentral placement and broad contact

Two geometric principles govern stable coupling. The first is where the head sits. The tympanic membrane does not vibrate uniformly: its excursion is greatest centrally, around the umbo and malleus handle, and falls toward the fixed annulus. A head positioned near the centre therefore captures the most acoustic energy and is mechanically more stable, because it loads the strut symmetrically; an eccentric head out near the annulus couples poorly and pivots, inviting slippage. Where the malleus is medialised and pulls the head off-centre, gentle lateralisation of the malleus or bending the shaft toward the promontory can restore both a central contact and an efficient line of force.

The second principle is broad rather than point contact. Each interface should present a spread, conforming surface — a head plate under the drum, a cup or clip on the stapes head — not a narrow tip. Point contact concentrates load, increases micromotion and is a leading driver of slippage, extrusion and the local inflammatory response that erodes tissue around the strut. Cadaveric vibrometry confirms the functional payoff of good lateral coupling: a prosthetic malleus-to-stapes-head assembly transmitted vibration better than a drum-to-stapes-head assembly, because engaging the manubrium restores a more physiological, better-aligned input [2004].

Length is what ties placement and contact together. A strut that is too short loses contact at one end as the ear heals and the construct settles; one that is too long over-tensions the system. The target is a length that gives continuous, light contact at every interface in play, with the head sitting just beneath a centrally placed drum.

TTension: the loosest stable seat

If placement decides where the strut couples, tension decides how hard. This is the most nuanced variable in the whole reconstruction, because the same force that holds a strut steady can choke off the motion it is meant to transmit. The native middle ear’s compliance is set overwhelmingly by the annular ligamentthat suspends the footplate — it accounts for the great majority of the system’s stiffness. A prosthesis seated too tightlydistends and stiffens that ligament, raises the cochlear input impedance and restricts pistonic stapes motion, hurting transmission — and the penalty falls hardest on the low frequencies. Over-tension can also bow or sublux the stapes and, at the extreme, traumatise the footplate.

A strut seated too loosely fails the other way: intermittent contact, acoustic leakage, and eventual displacement, which patients experience as fluctuating or recurrent conductive loss. Temporal-bone vibrometry shows that tension has a very significant effect on transmission, with both extremes degrading function and the low frequencies most sensitive [2004]. The same group quantified the optimum directly and found it is the loosest configuration that still sits stably [2004]. Differences of only 0.1–0.2 mm in length can swing a construct between too tight and too loose, which is why intraoperative sizing is so demanding.

The practical rule that flows from this is the classical ideal: a prosthesis that is firmly contacting yet freely mobile— stable, but not over-pressurised. And because postoperative fibrosis and scar can quietly tighten or tether a construct weeks later, the surgeon aims for that loosest-stable seat at the time of surgery, leaving a little mechanical headroom for the healing ear.

CProtecting the lateral end against extrusion

The lateral interface carries a particular hazard the medial one does not: a bare alloplastic head sitting against the delicate, mobile drum tends, over time, to erode through the membrane and extrude. Extrusion is the most visible interface failure — it ends the reconstruction and re-opens the gap — and the standard prophylaxis is cartilage interposition: a thin shield of tragal or conchal cartilage placed between the head and the drum.

Cartilage does three things at once. It protects the drum from direct contact with a hard alloplastic surface; it broadens and softens the interface, giving a forgiving, conforming contact that resists micromotion; and it provides a stable platform that helps maintain length and angle. Series using universal cartilage interposition under titanium heads report low extrusion and good tolerance [2004], and large alloplastic experiences show that cartilage between the head and the drum is a key determinant of staying seated rather than extruding [2001]. The biomaterial of the head matters too: hydroxyapatite, which mimics bone mineral, is unusually well tolerated in direct drum contact, and hybrid HA-head designs reached low extrusion rates that helped establish the material [1992].

There is a trade-off, and it is worth naming. A cartilage shield adds a little mass and stiffness and shaves a small amount off acoustic transfer, and over-large or peripherally placed cartilage can blunt the central vibratory input the head is meant to capture. The accepted bargain is to use a thin, centrally trimmed shield— accepting a minor acoustic penalty for a large gain in durability — rather than to leave a bare head exposed to a slow extrusion.

CSecuring coupling at the microscope

The biomechanics above become a short list of operative decisions once the ear is open:

• Capture both ends, broadly.Seat the medial end on the stapes head with a cup or clip that grips the capitulum, and centre a head plate under the drum — ideally engaging a retained malleus handle [2004, 2001].
• Centre, don’t corner. Place the head near the umbo where drum excursion is greatest, not out at the annulus; re-centre by lateralising a medialised malleus if needed.
• Size for loosest-stable. Trim the shaft to give continuous light contact, not a wedge; remember that a fraction of a millimetre separates too tight from too loose, and that tension hits the low frequencies [2004, 2004].
• Shield a bare head. Interpose a thin cartilage cap between an alloplastic head and the drum to lower extrusion, accepting the small acoustic cost [2004, 2001].
• Prefer the arch.Keep a mobile stapes superstructure whenever you can — it is the most stable coupling platform available, and it is why PORPs out-perform footplate TORPs [2006].

The final humility mirrors the rest of the chapter: the middle-ear environment usually matters more than the strut. Aeration, mucosal health and eustachian-tube function govern whether any interface stays healthy, and no coupling technique rescues a fibrotic, non-aerated ear. Good coupling and stable seating give a reconstruction its best chance; the biology has the final word.