Keratometry: How Corneal Curvature Sets Your Base Curve and Toric Fit
The first number taken is not your prescription
Before anyone asks which line is clearer, an instrument measures the front surface of your cornea. That reading, given either as a radius in millimetres or as a power in dioptres, is keratometry, and it quietly constrains several decisions that follow: which base curve the lens is made on, whether a toric design is needed, and whether the cylinder in your refraction originates at the cornea or deeper inside the eye.
It is also the oldest measurement in the room. The ophthalmometer was described in the 1880s, and the modern autorefractor simply automates a calculation that was previously done by watching a reflected mire move across a graduated scale.
Why a radius becomes a power
The cornea does most of the eye's bending: roughly two thirds of the total power comes from the air-to-cornea interface, which is why the shape of that surface matters so much. Converting curvature to power uses the same relationship as any refracting surface, with one deliberate fiction built in.
F = (n - 1) / r
where r is the radius in metres and n is the keratometric index of refraction, conventionally 1.3375. The fiction is that this index is not the true refractive index of corneal tissue, which is nearer 1.376. It is a calibration value chosen so that the calculated power matches what clinicians have historically called corneal power, and it exists because the cornea is a thick shell with a back surface rather than a thin refracting surface. Using 1.3375 keeps the numbers consistent with a century of tables.
A worked example: a radius of 7.8 mm gives 0.3375 / 0.0078 = 43.27 dioptres. A steeper cornea, 7.6 mm, gives 44.41; a flatter one, 8.0 mm, gives 42.19. Two meridians 7.6 and 8.0 mm apart therefore differ by about 2.2 dioptres, and that difference is corneal astigmatism.
What the reading actually changes
- Base curve. The front curve is not arbitrary. Labs stock a limited set of base curves, and the one chosen is a compromise between the calculated best-form curve for your power, the curvature of your cornea as a reference point, and what the surfacing range allows. A cornea at 43 dioptres and one at 47 do not get identical treatment.
- Toric decision for contact lenses. A soft lens on a toroidal cornea drapes over the surface and masks part of the corneal cylinder. Keratometry shows how much of the refractive cylinder is corneal, which is the part a toric soft lens can address, and how much is internal, which it cannot.
- Axis cross-check. If keratometry puts the steep meridian at 180 degrees and the refraction puts the correcting axis at 90, one of the two numbers is wrong. The mismatch is worth resolving before a lens is ordered.
- Detecting irregularity. A cornea whose two meridians are not roughly ninety degrees apart, or whose mires are distorted, is not a regular astigmatic surface, and a spectacle lens will not correct it cleanly. That finding changes the referral, not the prescription.
Corneal versus lenticular astigmatism
Total refractive astigmatism is the sum of everything in the optical path. Keratometry isolates the corneal contribution, so subtraction gives the residual: the astigmatism produced by the crystalline lens and any misalignment of the eye's internal optics. This distinction is not academic. Corneal astigmatism can be addressed by a rigid lens that replaces the front surface; lenticular astigmatism cannot. A person with two dioptres of total cylinder but only half a dioptre at the cornea is a poor candidate for that approach, and the keratometry is what tells you so before money is spent.
Why the two numbers never match exactly
Keratometry measures a small central zone, a few millimetres across, and assumes that zone is spherical. Real corneas flatten toward the periphery, and the central few millimetres are exactly the region a spectacle lens uses, so the assumption is usually serviceable. It also assumes the visual axis passes through the centre of the measured zone, which is approximately rather than exactly true. Topography maps the whole surface and resolves these limits, which is why an irregular or post-surgical cornea is imaged rather than merely measured.
Misconceptions
- "K readings are my prescription." They are not. Keratometry describes the corneal surface in isolation; your prescription describes the whole eye's focus at the spectacle plane, including axial length and lens thickness. A cornea at 43 dioptres is entirely compatible with any number of prescriptions.
- "A steep cornea means a strong prescription." No. Corneal power and refractive error are related but independent; the eye compensates through axial length. A 47 dioptre cornea can belong to an emmetrope.
- "The keratometric index is the refractive index of the cornea." It is a calibration convention of 1.3375, deliberately lower than the tissue's true index, chosen so calculated powers agree with established clinical values.
- "Any cylinder can be fixed with a toric contact lens." Only the corneal component is addressable that way, and only if the cornea is regular and the lens stays put.
When keratometry deserves more than a routine glance
Ask for the numbers to be looked at again if your astigmatism has changed substantially between visits without a change in your health, if the two meridians are not close to ninety degrees apart, if you have had corneal surgery or injury, or if a rigid or toric lens has repeatedly failed to perform as expected. Keratoconus in particular announces itself first as an irregularity in these readings rather than as a poor prescription, and it is far easier to manage when it is noticed early.
None of this replaces an examination. It simply explains why the technician measures your cornea before asking a single question about which lens looks better.