15Choosing Autograft Versus Prosthesis by Defect Pattern

FThe defect decides before the material does

The endless “autograft versus prosthesis” debate is usually argued in the abstract, as though one family of material were simply better than the other. At the microscope it is never abstract: you are looking at a specific defect, and the defect itself dictates much of the answer. The most useful organising framework is Austin’s classification, which sorts ossicular defects by the presence or absence of just two structures — the malleus handle and the stapes superstructure [1971]. Those two yes/no answers produce four common patterns:

The crucial insight is that geometry comes first. A present stapes head means a partial reconstruction (a PORP-type span); an absent superstructure means a total reconstruction down to the footplate (a TORP-type span). Only once the geometry is fixed does the question of material— sculpted native ossicle or manufactured alloplast — arise, and the answer changes from pattern to pattern. This module walks the four patterns and asks, for each, when the patient’s own tissue wins and when the reverse holds.

FType A: where the autograft shines

The type A defect— an intact, mobile malleus handle and an intact, mobile stapes superstructure with a gap between them (classically an eroded incus long process or incudostapedial joint) — is the most favourable pattern in all of ossiculoplasty, and it is the home ground of the sculpted autograft. The geometry required is a short bridge from the stapes head up to the malleus, and that is almost exactly the shape of the patient’s own incus body: drill a well to accept the stapes head, notch it onto the malleus, and the native ossicle slots into the gap it was made for. Because the graft is the patient’s own tissue it is immunologically inert, provokes minimal foreign-body reaction, and historically extrudes far less than early synthetics, all at no cost [2023].

The hearing record for this defect is excellent. In a large series of sculpted incus interpositions — a pure type A reconstruction — the mean air-bone gap improved from 26.8 to 18.6 dB, with about two-thirds of ears closing the gap to within 20 dB and no graft extrusions [2005]. The autograft does not merely tie the alloplast here; in the best head-to-head evidence it edges ahead, which the next sections quantify. The one absolute caveat for type A is availability and disease: the advantage holds only when a clean, disease-free, sculptable ossicle is on hand. An ossicle destroyed by erosion, or one engulfed in cholesteatoma, must never be reharvested — for the same type A geometry you then reach for a fresh PORP instead.

TAbsent superstructure: where the prosthesis takes over

Cross to the patterns with an absent stapes superstructure— types B and D — and the balance swings decisively the other way. Now the reconstruction must span from the mobile footplate all the way up to the drum or malleus: a long columella, the least forgiving geometry in the middle ear, where small errors in length, angle and footplate contact translate directly into failure. A hand-sculpted ossicle rarely delivers a reliable, reproducible length for this span, nor a stable seat on the small footplate. The manufactured TORP answers precisely this need: it comes in adjustable or graded lengths with a footplate shoe and a broad head, giving consistent geometry that sculpting cannot match. Titanium TORPs and PORPs in mostly revision ears, used with a cartilage shield at the drum, are easy to insert, well tolerated, low-extruding, and close the gap to within 20 dB in a useful majority of cases [2004].

The type C defect— absent malleus handle but present stapes superstructure — sits between these poles. The present stapes head still calls for a partial reconstruction, but the missing malleus removes two things the autograft relied on: a natural notch to seat the graft, and the lever advantage of the handle. Indeed, an intact malleus handle is itself a strong predictor of hearing success: across 105 reconstructions the present handle was one of the few independent determinants of a good result [2010]. With the handle gone, most surgeons place a cartilage-capped PORP onto the drum, though a clean sculpted ossicle to the drum remains workable. The general rule that emerges is intuitive: the more native scaffolding survives (especially the malleus), the more the autograft is at home; the more is missing, the more the manufactured prosthesis earns its place.

TReading the per-defect evidence

Two randomised trials, both confined to the type A defect, anchor the comparison where it matters most. The first randomised a sculpted autologous incus against a titanium PORP for Austin type A defects and found the gap closed to within 20 dB in 65% of incus ears versus 35% of titanium ears, with fewer post-operative complications in the incus group (20% vs 45%) — the autograft was both better and more predictable for this pattern [2017]. The second randomised an incus autograft PORP against a titanium angular-clip PORP specifically for incudostapedial-joint erosion (a type A geometry) and found no significant difference — success in 66.7% of incus versus 61.5% of titanium ears [2021]. Together they say that for the short type A gap the autograft is, at worst, the equal of the alloplast and at best meaningfully ahead.

Read the chart with care. The type A bars compare like with like — native ossicle against titanium for the same short bridge — and the autograft holds its own or wins. The third pair is not a head-to-head: for the long footplate-to-drum span the practical comparator is a titanium TORP series closing the gap in 57% of ears [2004], because a sculpted ossicle is rarely the realistic option there. The figures do not crown a universal champion; they show a defect-dependent answer— autograft for the short, scaffolded type A gap when clean tissue exists, prosthesis for the long or denuded span.

CThe modifiers that override defect pattern

Defect pattern frames the choice, but it does not finish it. Layered on top are modifiersthat can override the pattern’s default and that the clinician must weigh in real time:

• The ear environment, above all. Statistical staging of ossiculoplasty (the OOPS index) shows that mucosal health, aeration, drainage, the ossicular remnant and prior surgery drive hearing and extrusion more than the material chosen [2001]. In a wet, atelectatic ear a bone or cartilage autograft is more likely to resorb and anyrigid prosthesis is more likely to extrude — so a favourable type A pattern in a hostile ear is no guarantee of success for either material.
• Disease.A cholesteatoma-involved ossicle is disqualified as an autograft regardless of how perfectly its geometry would fit, because reusing it risks re-implanting matrix. The defect may say “type A, sculpt the incus,” but the disease says “use a fresh prosthesis.”
• Revision and remnant integrity. Even when the superstructure is present, a scarred revision ear with no usable remnant pushes toward a manufactured prosthesis of standard geometry.
• The interface. Whatever the pattern or material, a cartilage shieldis interposed between a rigid head and the drum to curb extrusion — near-universal for alloplasts and prudent for sculpted bone [2023].

CA per-defect choosing routine

The whole module collapses into a short routine you can run for any ossicular defect:

• Step 1 — classify the defect.Is the malleus handle present? Is the stapes superstructure present? That fixes the geometry: present superstructure → partial bridge; absent → total columella to the footplate [1971].
• Step 2 — type A with a clean, sculptable ossicle → favour the autograft. The short malleus-to-stapes gap suits the native incus, which is biocompatible, low-extruding, geometry-matched, free, and in randomised data at least equal to and often better than a titanium PORP [2017, 2021].
• Step 3 — absent superstructure (type B/D), a long span, no usable remnant, or revision → favour the prosthesis. An adjustable-length titanium TORP/PORP gives the reproducible length and angle that sculpting cannot, with low extrusion under a cartilage shield [2004].
• Step 4 — type C (no handle) → either, leaning PORP.The lost malleus removes the handle’s lever advantage and natural seat, so a cartilage-capped PORP to the drum is usual; a clean ossicle still works [2010].
• Step 5 — apply the overrides. Never reuse a diseased ossicle; honour the ear environment above the material; cap every rigid head with cartilage [2001].

The mature position is not tribal. The sculpted autograft owns the short, well-scaffolded type A gap when clean tissue is available; the manufactured prosthesis owns the long footplate-to-drum span and the denuded or diseased ear. The skilled otologist keeps both in the tray, lets the defect set the geometry and the ear set the odds, and remembers that within any pattern the most important variable is the middle ear into which any reconstruction is placed [2001, 2010].