Many people with keratoconus are told, at some point, that their cornea has become “too thin” for cross-linking – the one treatment proven to slow the disease. For someone whose keratoconus is still advancing, that can feel like a door quietly closing. A study from the ELZA Institute, published in the American Journal of Ophthalmology, asked a direct question: can a customised, higher-energy protocol called ELZA-sub400 cross-linking treat these very thin corneas while keeping the delicate cell layer at the back of the eye safe?

Key takeaways

  • In 76% of very thin, unstable corneas, ELZA-sub400 cross-linking showed no further steepening over 12 months.
  • The protocol is built for corneas thinner than 400 µm after riboflavin soaking – eyes usually turned away from standard cross-linking.
  • Every treatment “demarcation line” stayed within the corneal stroma, and no eye developed inner-cell-layer failure (endothelial decompensation).
  • A small, non-significant decline in glasses-corrected vision was seen in this severely affected group, and needs confirming in larger studies.

Why very thin corneas are usually turned away

Keratoconus makes the cornea – the clear front window of the eye – thinner and more cone-shaped over time. That irregular shape bends light unevenly, blurring and distorting vision. Corneal cross-linking (CXL) uses vitamin B2 drops (riboflavin) and ultraviolet-A light to stiffen the cornea so it is less likely to keep bulging, and it has become the standard way to halt progression.

The catch is thickness. The original cross-linking protocol required a cornea of at least about 400 µm, to keep a safe buffer above the endothelium – the single layer of cells that pumps fluid out of the cornea and keeps it clear. Below that thickness, many patients are simply told they are not candidates, even when their keratoconus is clearly getting worse. That is the gap ELZA-sub400 cross-linking was designed to close.

What makes ELZA-sub400 cross-linking different

Rather than fixing one recipe for every eye, the ELZA team individualised the treatment to each cornea. Using a published nomogram, the surgeon measures the cornea’s thinnest point during surgery and titrates the UV energy to leave roughly a 70 µm untreated safety margin above the endothelium. It helps to picture that margin as a buffer zone: the upper cornea is stiffened, while the deepest layer near the inner cells is left alone.

This second-generation version adds two refinements over the first-generation ELZA-sub400 protocol from 2021. It raises the maximum UV dose (fluence) to about 10 J/cm², and it allows a faster delivery rate (irradiance) of 9 mW/cm² as well as the original 3 mW/cm². Together, these shorten treatment to roughly 18 minutes, down from about 30. Importantly, the faster setting changes only how quickly the energy is delivered, not the total dose.

How the study was done

The research was a retrospective, single-centre case series: 29 eyes of 24 patients with progressive keratoconus or post-LASIK ectasia, each with a cornea thinner than 400 µm after soaking. Progression was judged mainly by maximum keratometry (Kmax), a measure of how steep the cornea is, using Scheimpflug tomography and anterior-segment OCT. The main outcome was the share of eyes without progression at 12 months, defined as less than 1 dioptre of steepening.

What the results showed

At 12 months, 22 of 29 eyes – about 76% – met the non-progression mark. On average, Kmax edged down slightly rather than up, though results varied widely between individual eyes. For someone living with keratoconus, that 76% is the figure that matters: most, but not all, of these very thin corneas appeared to stabilise for at least a year.

The safety signals were reassuring. The treatment boundary sat on average about 205 µm below the surface and about 64 µm above the endothelium, and every demarcation line stayed within the stroma. In roughly half the eyes it came within 70 µm of the endothelium – close, but still in safe tissue. No eye developed endothelial decompensation, none showed deep stromal haze, and corneal clarity (densitometry) held steady.

What happened to vision

Here the study is candid. Median glasses-corrected vision drifted from 20/25 to about 20/42 over the year – a numerical decline that did not reach statistical significance, but that the authors flag as clinically important. The point of ELZA-sub400 cross-linking in these eyes is structural preservation: keeping a patient’s own cornea and delaying or avoiding a transplant, rather than sharpening vision, which is usually restored afterwards with specialty contact lenses or other means. Corneal thickness itself barely changed, by about 4 µm.

Is this right for every thin cornea?

Not necessarily. This was a retrospective, single-centre series of 29 eyes followed for one year – valuable, but not the same as a large, long-term, randomised trial. The data do not show that the higher-energy approach outperforms the first-generation protocol; the clear practical gain is the shorter treatment time. Eyes that still progressed tended to have the steepest, thinnest corneas at the outset. Larger prospective studies are planned, and they will decide how widely this fits.

What this could mean for you

If you have very thin keratoconus or post-LASIK ectasia, being told your cornea is “too thin” may no longer be the end of the conversation.

  • Ask whether an individualised, thickness-based approach like ELZA-sub400 cross-linking suits your specific measurements.
  • Bear in mind that in this series about three in four ultra-thin corneas did not progress over 12 months – but some did.
  • Set expectations that the goal is to stabilise the cornea; sharpness of vision may not improve, and could decline somewhat.

Questions worth asking your ophthalmologist

  • Is my keratoconus or ectasia currently progressing, based on my Kmax and scans?
  • How thin is my cornea after soaking, and would an individualised sub400-style protocol be an option?
  • What are the specific risks to my endothelium with higher-fluence cross-linking?
  • If my glasses or contact-lens vision worsens afterwards, how would we manage it?
  • How often should I return for follow-up scans and vision checks?

Frequently asked questions

What is keratoconus?
Keratoconus is a condition in which the cornea gradually thins and bulges into a cone, distorting vision and often calling for specialty contact lenses or, in advanced cases, surgery.

How does cross-linking help?
Riboflavin drops and ultraviolet-A light create new bonds within the cornea, stiffening it so it is less likely to keep steepening. It aims to halt progression rather than reverse it.

Can very thin corneas be cross-linked safely?
Standard protocols usually avoid them. In this series, an individualised energy dose treated 29 corneas under 400 µm with no endothelial decompensation over 12 months, though longer, larger studies are still needed.

Will ELZA-sub400 cross-linking improve my vision?
Not usually. The aim is to stabilise the cornea; in this study glasses-corrected vision actually declined numerically, so vision is typically restored separately with specialty lenses.

How long do the effects last?
This analysis followed eyes for up to 12 months, so longer-term durability in ultra-thin corneas remains to be confirmed.

ELZA developed and continues to refine corneal cross-linking – including cross-linking for thin corneas – at our clinics and research spaces in Zurich, Switzerland.

References

  • Hafezi F, Akcan RE, Kling S, et al. Second-Generation ELZA-sub400 Protocol: Individualized High-Fluence Cross-Linking for Ultra-Thin Keratoconus Corneas. American Journal of Ophthalmology. 2026. DOI: 10.1016/j.ajo.2026.06.034
  • American Academy of Ophthalmology. What is keratoconus?