When to Cross-Link: What the New Global Keratoconus Consensus Says

One of the panelists on the international expert group for non-invasive treatment of keratoconus
By Adel Barbara, MD — Medical Director, IVISION Refractive Surgery and Keratoconus Treatment Center, Haifa

Of all the questions our panel spent the past year debating, one mattered more to patients than any other: when, exactly, should a keratoconic eye be cross-linked? The answer just appeared in print, as part of the Global Consensus on Keratoconus and Ectatic Diseases — Edition 2 (Gomes et al., Cornea 2026;45:888–908), the update to the 2015 consensus that first established corneal cross-linking (CXL) as standard care. I served on the panel responsible for non-invasive treatment of keratoconus — one of seven expert groups that together comprised 128 ophthalmologists from 12 societies across six continents. Here is what changed specifically around cross-linking, and why it matters if you or a family member has keratoconus.

CXL once progression is confirmed: still the one point of full agreement

The single strongest statement to come out of the whole process, reached with complete agreement across the panel, is unchanged in substance but sharper in practice: cross-linking is the standard of care as soon as progression is confirmed, and it remains the only treatment proven to halt the disease. Nothing else — not glasses, not contact lenses, not diet or eye exercises — has been shown to stop a cornea from continuing to steepen and thin. If your eye is progressing, cross-linking is the conversation to have.

Progression itself is now defined more broadly

This is the technical shift that matters most. Ten years ago, progression was judged mainly by a single number — the steepest point on the cornea, Kmax. The new consensus moves away from relying on that one figure. Progression is now read as a pattern across several measurements together: regional changes in curvature (not just the single steepest point), changes on the back surface of the cornea, thinning of the cornea over time, and real changes in how well the patient sees. A cornea can be worsening in ways a single Kmax reading would miss — and the panel wanted that reflected in how we decide who needs treatment.

Children and teenagers: treat early, don't wait for proof

This is where the panel pushed hardest for a change in practice. Keratoconus in children and teenagers tends to progress faster than in adults, and a young patient who loses vision to advanced ectasia has decades left to live with the consequence. The panel supported treating young patients early — in the most rapidly progressing cases, even before progression has been formally documented over two visits, rather than waiting for proof while the cornea continues to change. For a parent, the practical takeaway is this: a teenager newly diagnosed with keratoconus should not be told to "come back in a year and we'll see."

Cross-linking is increasingly not a stand-alone decision

The 2015 consensus was about stabilization — stopping the disease from getting worse. This edition reflects a broader ambition: using cross-linking together with other procedures to improve vision, not just preserve it. In appropriately selected eyes, cross-linking is now often planned alongside corneal ring segments, or combined with surface laser treatment, specifically to flatten and regularize the cornea at the same time it is being stabilized. Which combination makes sense, and in what order, depends heavily on the individual eye — but the panel's discussion signals that "cross-linking alone, and nothing else" is no longer the default plan for every eye that needs treatment.

Thinner corneas are no longer automatically excluded

Corneal thickness has always been the main safety limit on cross-linking — a cornea that is too thin risks damage to the deeper structures during treatment. Refinements in protocol design discussed by the panel now allow some corneas that would previously have been turned away for being too thin to be treated safely. This doesn't mean every thin cornea is now a candidate — the decision still needs an experienced corneal specialist — but the exclusion is narrower than it used to be.

What's still genuinely unresolved

I want to be straightforward about this, because it matters. A Delphi consensus is built from expert votes, not from a clinical trial, and not every question reached comfortable agreement. Exactly when a previously treated eye should be cross-linked a second time, and which biomechanical measurements (beyond standard imaging) should be enough on their own to trigger treatment, remained open questions after all four rounds of voting. That's not a weakness of the process — it's an honest signal of where the evidence genuinely runs out today, and where your own ophthalmologist's judgment about your specific eye still carries real weight.

The practical message

If you have keratoconus, the question is no longer really "should I ever consider cross-linking" — for a progressing eye, that has been settled for a decade. The real question, sharpened by this update, is timing: are you being followed closely enough to catch progression early, especially if you're young, and is your specialist looking at the whole pattern of change rather than one number in isolation? Ask for a full corneal tomography, not just a quick check, and ask specifically whether your cornea has been stable or changing since your last scan.

This article is for general information only and is not a substitute for an in-person examination and consultation with an ophthalmologist. Any decision about your treatment should be made together with a specialist who has examined your eyes directly.

Reference: Gomes JAP, Hafezi F, Ambrósio R Jr, et al., for the Global Consensus Panelists. Global Consensus on Keratoconus and Ectatic Diseases—Edition 2. Cornea. 2026;45(7):888–908.