Snoring, sleep apnoea and keratoconus: an association worth asking about

If you or someone in your family has keratoconus, you have probably been asked about eye rubbing, allergies and family history. Here is a question that may come up more often in future: do you snore, or has anyone ever mentioned you might have sleep apnoea? A large North American database study published in Cornea in 2026 asked whether obstructive sleep apnoea (OSA) and keratoconus travel together in young adults. This article is based on the published abstract; full-text details such as attrition, funding and the full set of sensitivity analyses were not available to us.

What the researchers did

Butt and colleagues ran a retrospective cohort study in the TriNetX Research Network, a large multicentre health record database, looking in two directions at once.

• Forward direction: adults aged 18 to 45 with a diagnosis of OSA and evidence of a sleep study (polysomnography) were compared with matched people without that diagnosis, and followed to see who later developed keratoconus.
• Reverse direction: people with keratoconus and evidence of corneal imaging (topography, anterior segment OCT or pachymetry) were compared with matched people without keratoconus, and followed to see who later developed OSA.

In each direction, the study group was compared with three different comparator groups, matched one-to-one on demographics, body mass index category, atopic and metabolic conditions, and how often they used health care. The authors also used a "negative control" outcome, acute appendicitis, which should have no plausible link to either condition; if the OSA group had shown more appendicitis too, that would have been a warning that the matching had not worked.

What they found

The forward analysis matched 48,954 people per arm. Among those with OSA, the hazard ratio for developing keratoconus over 5 years was 2.01 (95% confidence interval 1.21 to 3.32; P = 0.006). At 10 years it was 2.03 (95% CI 1.28 to 3.21). The two sensitivity analyses using the other comparator groups gave hazard ratios of 1.72 and 1.88, pointing the same way.

A quick plain-English guide to those numbers. A hazard ratio of 2.01 means that, over the follow-up period, people with OSA were being diagnosed with keratoconus at roughly twice the rate of matched people without OSA. The 95% confidence interval is the range within which the true value plausibly sits given the data; here it runs from 1.21 to 3.32, which is wide but does not include 1.0 (no difference). The P-value of 0.006 means a difference this large would be unlikely to appear by chance alone if there were truly no association.

The reverse analysis was smaller, with 4,350 people per arm. Among those with keratoconus, the 5-year hazard ratio for a later OSA diagnosis was 1.13 (95% CI 0.91 to 1.40), which was not statistically significant. At 10 years it was 1.29 (95% CI 1.07 to 1.55; P = 0.008), a weaker and more delayed signal than the forward direction.

The negative control outcome was null in all primary comparisons, which is reassuring. The authors also report an E-value of 3.45 for the forward result: an unmeasured factor would need to be linked to both OSA and keratoconus by about that much to fully explain away the finding.

Why might sleep apnoea and keratoconus be linked?

The abstract does not test any mechanism. Ideas already discussed in the keratoconus community include sleeping positions that press the eye into a pillow, floppy eyelids that can accompany OSA, and eye rubbing, which is a recognised keratoconus risk. None of this is proven by this study, and the authors say the reverse direction may partly reflect detection bias: people with keratoconus have more medical contact, so a sleep problem may simply be noticed sooner.

What this could mean for you in Hamilton

The authors' own conclusion is modest. They suggest that a sleep evaluation is worth considering in young adults with suspected corneal ectatic disease. They state clearly that the study does not tell us whether treating OSA changes the course of established keratoconus.

For a keratoconus patient in the Waikato, the practical takeaways are:

• If you have keratoconus and you snore heavily, wake unrefreshed or have been told you stop breathing during sleep, mention it to your GP. Sleep apnoea is worth treating for your general health regardless of your eyes.
• If you have OSA and notice frequent prescription changes or increasing glare and ghosting, ask for a keratoconus check rather than putting it down to tiredness.
• Tell your optometrist about your sleep, alongside allergy, eye rubbing and family history.

Where to get your sleep checked in Hamilton

If snoring or broken sleep rings a bell, a home sleep test is a straightforward first step and does not need a hospital visit. AwakeNZ (0800 AWAKEN) provides home sleep testing, sleep consultations and CPAP trials for people across New Zealand. Locally, Paul Hsu, the audiologist at Veritas Hearing at 37 Lake Road, Frankton, works alongside AwakeNZ and can point you in the right direction for a sleep assessment; while you are there, a hearing check is a sensible thing to fold in, because sleep apnoea, hearing and eye health are all part of the same general-health picture. We have no financial relationship with either service; we mention them because patients ask us who to call.

The right front door for all of this is a keratoconus-focused optometrist, not a surgeon. At Rose Optometry the optometrist team, including Jagrut Lallu and Jacqueline Rowe, diagnose keratoconus, document whether it is progressing with sequential corneal maps, refer to an ophthalmologist for corneal cross-linking at the right moment if progression is documented, and fit scleral lenses when glasses and soft lenses no longer give clear vision. Sleep history is a natural part of that first conversation. You can read more about the keratoconus service at Rose Optometry, or book a keratoconus assessment in Hamilton.

What the studies can't tell us

• Association is not cause. This is a retrospective, observational database study. It cannot show that sleep apnoea causes keratoconus, or the reverse.
• Diagnoses came from billing codes. Keratoconus was identified by an ICD-10 code plus evidence of imaging, not a standardised tomography review, so mild cases will have been missed.
• Only relative risks are given. The abstract reports hazard ratios, not how many extra people per thousand developed keratoconus. Keratoconus in adults aged 18 to 45 is uncommon in absolute terms, so a doubling of a small risk is still a small risk.
• Residual confounding and detection bias. Matching covered measured factors such as body mass index and atopy, not unrecorded ones such as eye rubbing or sleep position. People with either condition also see clinicians more, so the other condition is more likely to be found.
• Not a New Zealand population. The data come from a mostly North American network. Keratoconus is more common in Māori and Pasifika New Zealanders, and we do not know whether the same association holds here.
• Abstract only. We could not review attrition, funding, conflicts of interest or the full sensitivity analyses. No treatment effect was studied, so nobody should expect a CPAP machine to alter their keratoconus.

Our view is that this paper adds a sensible, low-cost question to the keratoconus conversation rather than a new treatment. Keratoconus risk factors in New Zealand are an area of active interest for the New Zealand Eye Research Centre.

Reference

• Butt FR, et al. Bidirectional Association Between Obstructive Sleep Apnea and Keratoconus in Young Adults: A Multicenter Retrospective Cohort Study. Cornea. 2026; published online 31 August 2026 (volume/issue/pages not yet assigned). PMID 42746737. DOI: https://doi.org/10.1097/ICO.0000000000004290