Your Child Has Keratoconus: Is Waiting Always Safe?

I am writing this for families whose child has been diagnosed with keratoconus, or in whom it is suspected. My aim is not to push you towards a decision; it is to help you arrive at the examination knowing what to ask. My focus is a single question: when is waiting a reasonable decision, and when is it not?

Why keratoconus behaves differently in children and adolescents

What the literature says: often rapid in children, but not the same in everyone

The consensus statement published in 2026 by the European Reference Network for Rare Eye Diseases (ERN-EYE) describes paediatric keratoconus as a condition that often advances rapidly and significantly affects visual development; the same document points to the lack of standardised paediatric-specific guidelines and to the importance of patient and family education.¹

The direct evidence behind the word "faster" is narrower than many assume. A study from Taiwan retrospectively reviewed 152 patients (288 eyes); there were 20 patients (37 eyes) in the group aged 18 and under and 132 patients (251 eyes) in the adult group, with a mean follow-up of 2.98 years. Refractive error and keratometry changed more rapidly in the children; however, a statistically significant difference between the two groups was seen only in spherical refractive error and spherical equivalent.² So the study points in one direction, but the evidence rests on 20 children. I prefer to say that as it is.

Progression in numbers: most eyes progressed in one cohort

A cohort from New Zealand evaluated 148 eyes of 106 patients who were 18 or younger at diagnosis and had at least 6 months of follow-up. Mean age was 15.2±2.5 years and mean follow-up 2.9±2.2 years. The rate of topographic progression was 77.0% (114 of 148 eyes).³

Which eyes carry more risk: stage at diagnosis, a steep cornea, allergy

In the same cohort, the eyes that progressed were more advanced at presentation: maximum keratometry (Kmax) 55.4±6.3 versus 52.2±5.4 dioptres, posterior elevation 108.2±40.9 versus 86.3±35.6 µm, and central corneal thickness 442.1±56.7 versus 454.4±47.5 µm.³ In another study, from Brazil, which followed 305 eyes of 168 patients aged 9 to under 18 for at least 36 months, progression was defined as a 1.5-dioptre increase in Kmax; eyes with a Kmax of 55 dioptres or more had a shorter time to progression (significant in left eyes and in the better-eye and worse-eye analyses). Allergy was also a predictor of progression in right eyes.⁴

Another reality in the youngest age group is that the diagnosis can come late. In a series from India of 294 patients (586 eyes) aged 12 or younger, the disease was already at an advanced stage in 25% of the children.⁵ A child often does not describe blurred vision; that is the nature of the age, not a failing of the family.

Eye rubbing and allergy: mixed evidence, a harmless precaution

On eye rubbing, the honest answer is "the evidence is mixed". In a study from Germany comparing 97 children (10–18 years) with adults, eye rubbing was associated with progression in univariate analysis in all age groups.⁶ In the same study, anterior chamber volume and the keratoconus match index stood out in the paediatric group; the authors attributed this to the probably different anatomical and biomechanical characteristics of a child's eye.⁶ In the New Zealand cohort, on the other hand, documented eye rubbing and atopy were not significantly associated with progression in the multivariate analyses.³ My own conclusion: helping a child stop rubbing and treating the allergy are harmless, practical and recommended measures; but we have no evidence that entitles anyone to say "it would not have happened without the rubbing". Your child rubbing their eyes is not your fault.

When "let's wait" is reasonable, and when it is not

Monitoring is a decision: the child without documented progression

In a child who has been diagnosed but in whom progression has not yet been documented, monitoring is a defensible decision. The condition is that it is planned: at set intervals, with comparable measurements. "Let them grow, we will see" and "we will measure again on the agreed date, on the same device" are not the same sentence.

The cost of waiting: how often the waiting group progressed in a randomised trial

The study that answers this most directly is KERALINK, from the United Kingdom. Sixty participants aged 10 to 16 with progressive keratoconus were randomised to two groups: cross-linking plus standard care (spectacles or contact lenses), or standard care alone. At 18 months, keratoconus progression had occurred in 2 patients (7%) in the cross-linking group and in 12 patients (43%) in the standard care group. Patients in the cross-linking arm had 90% lower odds of progression (odds ratio 0.1; 95% confidence interval 0.02–0.48; P=0.004).⁷

Which findings are worth a conversation with the doctor

The following are not for making a diagnosis; they are reasonable grounds to ask for an appointment without waiting for the planned check-up:

• Your child's prescription or astigmatism is changing at short intervals,
• They say they cannot see clearly even with new glasses, or struggle to read the board at school,
• One eye has started to see noticeably worse than the other,
• They rub their eyes often and hard, and the allergic itching cannot be brought under control,
• Someone in the family has been diagnosed with keratoconus.

Follow-up interval: a framework for families

Intervals in the literature

There is no single, universally agreed interval that is specific to children. In a systematic review of 221 articles on how progression is defined, progression was assessed between 6- and 12-month follow-up in 64.7% of the articles; the most frequent criterion was Kmax, used in 85.5%.⁸ That review is not specific to children. The authors of the Taiwan study, for their part, conclude that early detection and frequent follow-up are necessary in paediatric patients.²

My approach

I see my child and adolescent patients more frequently than my adult patients. The exact interval is set not by the calendar but by your child's age, the measurements at diagnosis and the change we saw at the previous visit; we agree on it together at the end of each examination.

What to bring, and why the same device and the same centre matter

Progression is read from the difference between two measurements, and that difference only means something if the measurements are comparable. So bring previous topography printouts and spectacle prescriptions with you. Where possible, have the check-ups done on the same device: measurements from different devices are not interchangeable. If your child wears rigid contact lenses, ask how long the lenses should be left out before the measurement.

The "stabilise" step: what cross-linking (CXL) does in a child

What has been shown: the data on halting progression

KERALINK's conclusion is clear: cross-linking arrested keratoconus progression in the majority of young patients, and the authors recommended that it be considered as a first-line treatment in progressive disease. At 18 months, mean steep keratometry (K2) was 49.7 dioptres in the cross-linking group and 53.4 dioptres in the standard care group; the adjusted mean difference was −3.0 dioptres (95% confidence interval −4.9 to −1.1; P=0.002).⁷ In a retrospective series from Türkiye that followed 40 eyes of 40 paediatric patients treated with an accelerated protocol for at least 60 months, visual and keratometric stabilisation also continued over 60 months; the same series reported worsening in posterior elevation and thinning in pachymetric indices.⁹

On the question of when to decide on cross-linking, studies take two different routes: KERALINK enrolled children with documented progression;⁷ a study from India that followed 38 eyes of 30 children (9–18 years) for one year performed the procedure within four weeks of diagnosis and found it safe, with three patients developing transient sterile infiltrates that resolved with medical therapy.¹⁰ These two strategies are not the same thing. Which one is appropriate is decided at the examination, from your child's measurements and the change seen during follow-up, not from a general rule.

What CXL does not do: it does not correct vision or change the shape

This is a point that is often confused. Cross-linking aims to stiffen the cornea; it is not done to remove the need for glasses. Some series report a small improvement: in a study from Romania that followed 28 eyes of 23 patients aged 13–18 for 24 months, uncorrected vision went from 0.53±0.16 to 0.44±0.16 logMAR and corrected vision from 0.31±0.11 to 0.25±0.11 logMAR (in logMAR, a smaller number means better vision); the authors described this as a "modest" improvement and noted that the results should be interpreted cautiously because there was no control group.¹¹ Expectations should be set correctly: the gain is the preservation of vision that would otherwise be lost. Your child may continue to need glasses or contact lenses after the procedure.

Follow-up continues after CXL

In a systematic review of 37 studies (2078 eyes) of children after cross-linking, the pooled progression rate after epithelium-off treatment, using an increase of 1.0 dioptre or more in Kmax, Kmean or Ksteep as the criterion, was 9.9% (95% confidence interval 6.1%–14.6%; 1508 eyes); heterogeneity between studies was high.¹² In other words, the procedure does not end the check-ups. Topography follow-up continues in a child who has had cross-linking. Like every procedure, cross-linking carries risks (such as the transient infiltrates mentioned above); these, and whether the procedure suits your child, need to be discussed at the examination.

"Reshape" is a separate step: CAIRS does not halt progression

CAIRS is a treatment that changes the geometry of the cornea using segments prepared from donor corneal tissue. Its job is to improve the quality of vision; it does not halt progression. The treatment that aims to halt progression is cross-linking. This is why two separate questions are asked in a child: "is it progressing?" and "do they see well enough with glasses or lenses?". The answer to the first brings cross-linking into the conversation; the answer to the second brings reshaping.

An honest limit: there is as yet no published long-term series specific to CAIRS in childhood. The long-term data obtained with synthetic segments in patients under 18 are a separate matter; I covered the childhood data and the segment results in my review of the last 10 years.

What families can do at home and at school

• Reduce eye rubbing. Not by forbidding it, but by dealing with the cause (itching, dryness, tiredness).
• Have the allergy treated. In the Brazilian cohort 70% of the children were allergic;⁴ while the itching persists, stopping the rubbing is usually difficult.
• Support wearing glasses and lenses. Clear vision matters for a child's school life; but seeing well does not show that the disease has stopped.
• Do not skip the check-up. Having no complaints is not a reason to postpone it.

I am not listing restrictions on sport, screens or school here, because the sources I rely on in this article contain no data showing that these change the course of keratoconus.

Siblings and family: when screening makes sense

In a study from Türkiye of the asymptomatic offspring of patients with keratoconus, keratoconus was identified during screening in 8 of 56 offspring (14.3%).¹³ That study examined the children of patients, not siblings, and the abstract does not state the participants' ages; even so, where there is keratoconus in the family, I recommend that first-degree relatives be assessed with topography without waiting for symptoms. I also discuss the place of family screening in Keratoconus Treatment: The Last 10 Years.

For family physicians and optometrists: when to refer

Questions you may want to ask at the appointment

• Is there a difference between my child's previous measurement and this one, and in which value?
• Is that difference larger than the device's measurement variability?
• When is the next check-up, and on which device?
• Which finding would lead us to discuss cross-linking?
• Are both eyes in the same condition, or is one ahead of the other?
• What should we pay attention to at home until the next visit?

Frequently asked questions