In the first part of this series, I wrote about a married couple with violet sclerae, and I ended on a sentence that has stayed with me since: whatever we put in, the next clinician has to see past. This is the other kind of ink – and this time, I was the next clinician.

My patient’s eye color had been changed – not with colored contact lenses, and not with an iris implant, but with keratopigmentation: pigment laid into the cornea as a ring, with a clear window left over the pupil. What brought the patient to me was the most ordinary thing in our specialty: a cataract. Anything that could identify the patient stays out of this piece; the lessons do not need it.

The color is in the wrong layer

The first thing to understand about cosmetic keratopigmentation is that the color is not where it appears to be. The iris is untouched. What the world sees is a ring of micronized mineral pigment in the corneal stroma, these days usually delivered into a tunnel cut by a femtosecond laser, with the center left clear. Unlike the scleral tattoos of part one, it is a medical procedure, performed by ophthalmologists, and, when done well, can be remarkably convincing.

It is also a procedure in an early stage of development. Jorge Alió’s group published the first purely cosmetic series in 2016, detailing the results of seven patients. Five years later, the same group reported 79 eyes of 40 patients with a mean age of 34 – excellent cosmetic results and high satisfaction, but more than a third of the eyes went back for further pigment, light sensitivity in the first month was a  common complaint at 30 percent, and one patient with a 4.5-mm window noticed a narrowed visual field. A 2025 series of 166 eyes found acuity, pressure and visual fields unchanged at a year, alongside a measurable drop in contrast sensitivity – mostly under mesopic conditions, and still within the normal range – and a small fall in endothelial cell density. Case reports have since added ectasia, a severe corneal ulcer and – tellingly – techniques for taking the pigment back out again.

In July, Cornea printed the argument as a debate: “Color at What Cost? The Case Against Cosmetic Keratopigmentation in Sighted Eyes” on one side, “Cosmetic Keratopigmentation Works” on the other. Even the case for the defense asks for long-term data, careful patient selection and controlled clinical evaluation. Richard Zegers, reviewing the long history of attempts to change eye color, named the problem that concerns us most directly: many of the people having these procedures are young, and so trouble may come later “when they will need cataract or other eye surgeries.” For my patient, "later" had now arrived.

Measuring the window

The reassuring news is that a well-made window leaves most of the eye open to examination. In 60 eyes examined six months after purely cosmetic femtosecond-assisted keratopigmentation with a 5-mm aperture, every routine test could be acquired and interpreted – with one exception, the endothelial cell count in the corneal periphery.

Two measurements matter more than usual. The first is the diameter of the clear window against the diameter of the dilated pupil, because the smaller of the two is the size of your operating field. The second is corneal shape. Ectasia has been reported after cosmetic keratopigmentation, in some patients years after the procedure, so topography belongs in the work-up before any IOL calculation is trusted.

Phaco through a keyhole

In theater, the pigment ring behaves like a pupil that will not dilate. Everything that normally happens across seven or eight millimeters has to happen inside the window. The capsulorhexis must sit within it, because its edge can only be seen there – and since the capsule lies a few millimeters behind the pigment and is seen magnified through the cornea, the stretch of capsule you can actually see is smaller than the window itself. A rhexis of ordinary size can end up partly under the ring unless the window is generous. The equatorial cortex lies under the ring, largely out of sight, and so has to be brought into the window rather than chased beneath the pigment. The haptics disappear under the ring as the lens unfolds, so their position is judged by the centration and stability of the optic rather than by looking. And on a toric lens, the axis marks near the edge of the optic can sit right under the pigment.

The operation is on video, and it makes the point better than any description: everything the surgeon can see happens inside that circle.

Not the first – and that is good news

The only earlier report I could find on the procedure describes a 62-year-old woman who underwent femtosecond laser-assisted cataract surgery with a trifocal lens in both eyes, six months after her own color change, and reached 20/20 uncorrected. It shows that an excellent result is possible. It also describes something close to a best -case scenario: a tattoo only six months old, in eyes that already saw 20/20 with glasses – the aim was freedom from spectacles. Most of us will meet these eyes differently: years later, with a real cataract, and without the operative record of the color change.

The card nobody carries

Every patient with an intraocular lens can be handed a card that says what is inside the eye. A patient with keratopigmentation usually leaves with nothing of the kind, yet almost every detail of the procedure matters to whoever operates next. The pigment matters: in one ex vivo study, iron-oxide-based inks produced MRI artefacts and pigment migration – subtle under normal conditions, pronounced under extreme exposure – while another study this year found no heating or movement. The depth and diameter of the tunnel matter for any future corneal or cataract surgery. The size of the clear zone matters for every examination and every lens calculation. Removal is possible – recent reports describe lamellar flaps and excimer ablation – but it means more surgery on a healthy cornea.

If we are going to put permanent pigment into healthy corneas, the least we owe the next surgeon is a record of what was done: the pigment, the depth, the inner and outer diameter, the date. It costs nothing, and it would turn the next operation from a reconstruction of someone else’s decisions into a more structured plan.

Two kinds of ink

Scleral pigment silences the surface and leaves the inside of the eye open. Corneal pigment does the opposite: the conjunctiva can still redden and tell us things, but the way in narrows to a circle. My patient made a personal choice, just as the couple in part one did, and deserved what they did – to be treated as a patient with a cataract, not as a curiosity.

Whatever we put in, the next clinician has to see past. With keratopigmentation, that is not just a metaphor. Every step of this operation happened inside a circle drawn for cosmetic reasons. The least we can do, as the ones holding the pen, is to write down its size for the ophthalmologist who will next encounter that patient.

Patient details have been anonymised.