2Air-Bone Gap as the Core Outcome Measure

FWhat the air-bone gap is, and why we measure it

Every operation needs a yardstick, and for ossiculoplasty that yardstick is the air–bone gap (ABG). The reasoning starts from what the operation can and cannot do. Ossiculoplasty rebuilds the conductiveapparatus — the mechanical chain that carries sound from the eardrum to the oval window. It does nothing for the cochlea. So the right measure of success is not “how well does the patient hear” in absolute terms, but how much of the recoverable, mechanical loss did we recover.

A pure-tone audiogram gives us two lines for that judgement. The air-conductionline is what the ear hears through the whole pathway — canal, drum, ossicles and cochlea together. The bone-conductionline, delivered by a vibrator on the mastoid, bypasses the conductive chain and reports the cochlea directly. The vertical distance between them is the air–bone gap, and it is preciselythe part of the hearing loss attributable to the mechanics — the part an ossiculoplasty can hope to close. The bone line is therefore both the target the air line is driven down toward and the ceiling it can never cross: a patient with a 30 dB sensorineural component will never hear normally, however perfect the reconstruction, and the gap, not the absolute threshold, keeps that distinction honest [1991].

This is why the gap is the natural currency of the field. It strips away the cochlear reserve that surgery cannot influence and reports only the mechanical work the reconstruction did. Two ears with identical operations but different cochleae should earn the same credit, and the gap — unlike a raw threshold — gives it to them.

FThe four-frequency gap and the 20 dB success line

To compare one surgeon’s results with another’s, the gap had to be standardised, and in 1995 the American Academy of Otolaryngology—Head and Neck Surgery did exactly that. Its Committee on Hearing and Equilibrium guidelines define the gap from the four-frequency pure-tone average at 0.5, 1, 2 and 3 kHz, computed for air and for bone, with the difference reported as the air–bone gap [1995]. The choice of 3 kHz over the older 4 kHz, and the insistence on a consistent four-frequency set, exist so that headline numbers from different series actually mean the same thing — a series that quietly averages only 0.5, 1 and 2 kHz will flatter itself, because the highest, hardest-to-close frequency has been dropped.

By long convention an ossiculoplasty is called a success when the postoperative gap closes to within 20 dB. The guidelines also recommend reporting the proportion closed to within 10 dB— an “excellent” result — and the mean postoperative gap itself, so that a single threshold does not do all the work [1995]. The 20 dB line is a pragmatic, not a physiological, boundary: it marks roughly the point at which the residual conductive loss stops being a meaningful handicap for everyday listening. It is also, deliberately, a generousbar. A 19 dB result and a 4 dB result both count as “success,” yet the patient experiences them very differently, which is why the honest reader looks past the binary to the mean gap and the within-10 dB rate.

TWhy the gap, and not air-conduction change

A common error — in counselling and in the older literature — is to judge the operation by how much the air-conduction threshold improved, comparing the postoperative air line to the preoperative one. This sounds intuitive but is a trap, and seeing why sharpens the whole concept. Air-conduction change conflates two things the gap deliberately separates: the mechanical repair and any drift in the cochlea. Worse, it rewards a bad starting point. An ear that began at 55 dB and ends at 30 dB posts an impressive 25 dB “gain,” yet if its bone line sits at 10 dB it still carries a 20 dB residual gap; another ear that began at 35 dB and ends at 15 dB posts a smaller 20 dB “gain” but has closed its gap almost completely. The second is the better reconstruction, and only the residual gap says so.

Referencing the result to the patient’s ownbone line, rather than to their previous air line, is what makes the gap fair across ears and comparable across surgeons. It also protects against a subtler illusion: an apparent air-conduction “improvement” that is really a postoperative drop in bone conduction at 2 kHz, the so-called Carhart effect, which can shrink the gap without the ear hearing any better in absolute terms [1950]. Because the gap is anchored to the bone line, it tracks the bone line too — and a worsening bone threshold is read not as a benefit but as the warning sign of a sensorineural complication.

The same logic explains why speech and word-recognition scores, valuable as they are, do not serve as the primary ossiculoplasty metric. A purely conductive loss usually leaves word recognition intact once the signal is made loud enough, so the score is often near-normal both before and after surgery and discriminates poorly between a closed and an unclosed gap. The conductive component — the very thing the operation changes — is exactly what pure-tone thresholds capture and speech scores do not.

TWhat success actually looks like in the data

It is worth setting realistic figures against the optimism the word “reconstruction” can imply. In House and Teufert’s large series of over 1200 ears the mean last postoperative gap was 19.2 dB, with closure to within 20 dB in about 63% [2001]. A pooled meta-analysis of titanium prostheses found a gap under 20 dB in roughly 70% of PORPs and 57% of TORPs, reflecting the prognostic value of a preserved stapes superstructure [2023]. A recent multicentre study of 1679 ears reported a mean postoperative four-frequency gap of about 21 dB— sitting right on the success line — and used the spread to build an ear-environment risk score [2025]. These are clinically worthwhile, typical results, not the near-universal cure the term can suggest.

Two patterns deserve emphasis. First, the stricter the threshold, the rarer the result: the proportion reaching the within-10 dB “excellent” bar is far smaller than the within-20 dB “success” bar, which is the whole reason both are reported. Second, the success figure depends heavily on the ear, not the implant: stapes status, mucosal health, aeration and prior surgery move the gap far more than the brand of prosthesis does, so a single success percentage is meaningful only alongside the case mix that produced it [2025].

CReading and reporting the gap honestly

A clinician who understands the gap can read a published series, or audit their own, without being misled. Several disciplines separate honest reporting from flattering reporting, and all of them follow from the definitions above:

• State the frequencies. Confirm the gap is the AAO-HNS four-frequency average (0.5, 1, 2, 3 kHz). A three-frequency average that omits 3 kHz is not comparable and tends to look better [1995].
• Use the postoperative bone line. The gap should be calculated against the postoperative bone conduction, because surgery itself can shift the bone threshold; using the preoperative bone line inflates apparent closure [1995].
• Report the distribution, not just the binary. Give the mean gap and the within-10 and within-20 dB proportions together. Two series with the same 63% success rate can have very different mean gaps.
• Track the bone line for harm. A rise in bone thresholds is a sensorineural complication, not a success; never let a shrinking gap hide a deteriorating cochlea.
• Fix the follow-up interval. A six-month gap and a five-year gap are different claims; success rates drift downward over years, so an undated figure means little.

Applied at the bedside, this turns a number into a conversation. The surgeon can tell a patient not “you will hear normally” but “in an ear like yours we expect to close the gap to within 20 dB in roughly two-thirds of cases, your cochlea sets the best we can do, and the good early result is not guaranteed to be permanent.” That sentence is built entirely from the air–bone gap and the bone-conduction ceiling.

CThe gap’s blind spot: the binaural patient

For all its virtues, the air–bone gap has one structural blind spot, and a thoughtful clinician keeps it in view. The gap judges the operated ear in isolation. It says nothing about the patient. Toner, Smyth and Kerr made the point sharply: a gap closed to 20 dB in an ear that still lags far behind a near-normal contralateral ear may give that patient very little real-world benefit, because they already hear well on the other side; meanwhile the same 20 dB closure in an only-hearing or worse-hearing ear may transform their life [1991]. A surgically “successful” gap is therefore not the same as a useful result.

This is why patient-oriented standards — reporting whether the operated ear reaches a usable level relative to the other ear, for instance within 15 dB of it — were proposed as a complement to gap reporting, not a replacement [1991]. The air–bone gap remains the irreplaceable measure of what the reconstruction achieved, because it is comparable, surgery- specific and fair across cochleae. But the measure of what the operation was worth to the personrequires also looking at the better ear, the binaural picture and, ideally, what the patient reports. Holding both in mind — the gap as the currency of technical success, and binaural benefit as the currency of human value — is the mark of a clinician who reports outcomes honestly rather than merely favourably [1991].