4.2 The tympanogram and the Jerger types

A tympanogram is a plot of acoustic admittance against ear-canal pressure as the latter is swept from about +200 daPa to about −400 daPa. The audiologist reads three numbers off the curve — the peak height ($Y_\text{peak}$), the pressure at which the peak occurs ($\text{TPP}$, the tympanometric peak pressure), and the ear-canal volume ($\text{ECV}$) — and a fourth feature, the shape: notched, double-peaked, or flat. The combination places the tympanogram into one of five categorical types named by James Jerger in 1970 (see the historical note in Lesson 4.1).

Why the curve has a peak

The middle ear sits at a static pressure set by Eustachian-tube exchanges with the nasopharynx — normally close to atmospheric. The probe sweeps the ear-canal pressure through this static middle-ear pressure. At the moment when canal pressure equals middle-ear pressure, the eardrum has zero static deflection and is at its most compliant — small AC pressure variations move it easily. At any other canal pressure the eardrum is pushed in or pulled out and is stretched stiff against the asymmetry, so its admittance drops. The peak of the curve therefore locates the middle-ear pressure.

Mathematically, the static deflection of the membrane is a function of $\Delta p = p_\text{canal} - p_\text{ME}$. The acoustic admittance scales with the local compliance — the inverse slope of the static pressure-deflection curve at the operating point. For a thin elastic membrane this compliance is maximum at $\Delta p = 0$ and falls off in both directions, producing the characteristic bell-shaped peak.

The five Jerger types

Each type, with the ECV reading, often produces a near-instantaneous middle-ear diagnosis:

• Type A. Healthy middle ear. The peak sits within the shaded normal region at modest amplitude and near 0 daPa.
• Type As (“shallow A”). The bell shape is correct but the peak is too low — the eardrum is stiff. Causes: otosclerosis (stapes fixation), tympanosclerosis (drum scarring after recurrent otitis), or malleus fixation. The audiogram in such cases typically shows conductive loss with a low-frequency-dominant gap (stiffness dominates at low frequencies), the so-called Carhart notch at 2 kHz for otosclerosis being the classic combined finding.
• Type Ad (“deep A”). The peak is too high — the system is too compliant. Causes: ossicular discontinuity (a broken or eroded incus, the most common — e.g., from chronic infection), or a monomeric tympanic membrane (a healed perforation where only the thin epithelial layers remain). The audiogram shows conductive loss but with intact stapedial reflexes inverted: in discontinuity the reflex acoustically increases admittance because the stapedius can no longer mechanically couple to the cochlea.
• Type B. Flat trace, no identifiable peak. The middle ear is contributing essentially no admittance — either because it’s full of fluid (effusion / otitis media with effusion / “glue ear”), because the membrane is perforated, or because the probe is blocked. The ECV distinguishes the three:
  • Type B with normal ECV (0.65–1.75 mL adult) = effusion. The drum is mobile only over a very narrow pressure range; the canal-air-only-admittance dominates everywhere.
  • Type B with very high ECV (> 2 mL) = perforation or pressure-equalisation (PE) tube. The probe sees both ear canal and middle-ear cavity.
  • Type B with very low ECV (< 0.3 mL adult) = probe blocked.
• Type C. Normal-height peak shifted to negative pressure (commonly −150 to −300 daPa). Eustachian-tube dysfunction has prevented the middle ear from re-equilibrating after the cabin pressure dropped on a flight, after a recent cold, or after upper-respiratory inflammation. The drum is mobile; it just operates at a sub-atmospheric set-point. Persistent type C in children predicts development of otitis media with effusion.

Static admittance and the canal correction

The number quoted as $Y_\text{peak}$ is static admittance — the admittance of the eardrum and middle ear, with the ear canal’s contribution subtracted off. Because the ear canal is in parallel with the eardrum (see Lesson 4.1),

$Y_\text{tymp}(p_\text{canal}) = Y_\text{ear-canal} + Y_\text{eardrum}(p_\text{canal}).$

The ear-canal admittance is what the probe reads at the extreme pressure (when the eardrum is locked); subtracting that off everywhere yields the eardrum’s own admittance trace. This correction is built into modern tympanometers and is why the displayed trace usually has its baseline near zero.

Tympanometric width / gradient

The peak’s width matters too: a sharply peaked curve indicates a system with strong stiffness restoring forces; a broad, smeared curve indicates a system that is losing its stiffness component. The tympanometric width is the pressure interval over which the curve sits within half the peak amplitude (the equivalent of FWHM for the bell). Normal adult tympanometric width is 50–110 daPa; values above ~150 daPa are abnormal even in the absence of a clear type-B flattening, and often predict early effusion before the trace fully flattens.

Multi-frequency tympanometry

The 226-Hz probe assumes the middle ear at this frequency is stiffness-dominated: the peak in admittance reflects the peak compliance. In the infant ear this assumption fails — small cavities, less developed ossicles, and immature drum properties make the system mass-dominated at 226 Hz. The 226-Hz tympanogram in a healthy newborn can look type B (flat) or have spurious peaks.

The fix: use a higher probe frequency. At 1000 Hz the infant middle ear is stiffness-dominated and the tympanogram becomes interpretable. ANSI S3.39 specifies 1000 Hz for newborn screening. Multi-frequency tympanometers can also sweep probe frequency at fixed canal pressure to map the middle ear’s resonance frequency — the frequency at which mass reactance equals stiffness reactance and the admittance phase passes through zero. The resonance frequency shifts up with stiffness pathology (otosclerosis) and down with mass pathology (ossicular disarticulation), giving a quantitative complement to the static-admittance shape.

Next lesson: the same probe hardware, with one addition — a stimulus delivery system at clinically loud levels — measures the acoustic reflex, our next localising test.