01% Atropine Did Not Produce a Corneal Ectasia Signal in Children Treated for Myopia
Low-dose atropine remains one of the most widely discussed medical strategies for slowing childhood myopia, but corneal safety has not always been a background issue. Corneal curvature changes have been reported during treatment, raising a practical question for clinicians who monitor children over years rather than weeks: does the drug alter the cornea itself, or do the observed refractive shifts simply reflect changes elsewhere in the eye?
A randomized controlled trial in children with progressive myopia adds an important layer of reassurance. Among 153 participants aged 6 to 16 years, children were assigned in a 2:1 ratio to receive 0.01% atropine eyedrops or placebo for 2 years, followed by a 1-year washout. Scheimpflug imaging was obtained at baseline, 24 months, and 36 months, allowing investigators to track corneal biometry over time rather than relying on a single snapshot. One child in the atropine group developed keratoconus during treatment.
Corneal Tomography Stayed Broadly Stable, Except for Targeted Metrics
The most clinically important finding is what did not happen: the trial did not identify a treatment-related rise in corneal ectatic change. Across the full cohort, all other corneal parameters and cylindrical power remained similar between the atropine and placebo groups. For a therapy now used in children who may remain under ophthalmic follow-up for years, that absence of a broad corneal safety signal matters more than any isolated fluctuation in a single index.
Two corneal indices did move in the atropine arm. At 24 and 36 months, BAD-D values were significantly lower than placebo, with reductions from baseline of −0.075 at 24 months and −0.115 at 36 months. These differences reached statistical significance at both time points. At 36 months, the atropine group also showed a significant steepening of Kmax by +0.301D. Neither change, by itself, establishes clinically meaningful harm, but both deserve attention because corneal shape metrics are the same measurements ophthalmologists use when screening for early ectasia.
The biological interpretation is not straightforward. Atropine acts through muscarinic receptor blockade and is being used to modify ocular growth, not to reshape the corneal stroma directly. The cornea and the posterior segment do not respond to every pharmacologic intervention in lockstep, so a modest shift in a corneal index can emerge even when the overall ectasia risk remains unchanged. In other words, a statistical movement in tomography does not automatically translate into a dangerous structural weakening of the cornea.
One Incident of Keratoconus in the Treatment Arm Needs Context, Not Panic
The trial’s single case of incidental keratoconus occurred in the atropine group, but the frequency remained compatible with population prevalence rates. That is the key epidemiologic anchor. One case in 153 children is not enough to imply causation, particularly in a condition that can appear during adolescence regardless of eye-drop exposure.
Still, the bedside implication is immediate. Children receiving low-dose atropine who develop suspicious corneal findings should not be reassured away from further workup simply because their treatment is considered low concentration. Early keratoconus often presents subtly, and pediatric myopia clinics are exactly where those first irregularities may be noticed. The study supports continued vigilance, not alarm.
Clinicians should also remember that corneal screening is usually driven by risk pattern recognition rather than by a single number. A modest rise in Kmax or an unusual tomography profile deserves the same follow-through it would have received without atropine exposure, especially if symptoms, refraction, or corneal asymmetry are changing. The trial does not justify a new rule to stop atropine reflexively when a corneal metric moves.
What the 1-Year Washout Adds to the Interpretation
The 1-year washout strengthens the analysis by showing whether corneal changes persisted after treatment ended. By 36 months, the atropine group still showed the lower BAD-D values and the steeper Kmax, yet the overall pattern remained reassuring because the broader corneal profile did not diverge in a way that suggested progressive ectasia across the cohort. Longitudinal follow-up is essential here. Pediatric myopia management is cumulative, and corneal safety cannot be judged from a brief exposure window.
This is also why the trial design matters. Scheimpflug imaging at baseline, after two years of treatment, and after a washout period gives a more credible view of corneal behavior than cross-sectional screening alone. It captures both on-treatment effects and whether any signal lingers once atropine is withdrawn. For a drug used over long horizons, that distinction is clinically useful.
The study does have boundaries. The cohort was modest, and a rare adverse corneal event would be difficult to quantify with precision in a sample of this size. Even so, the data are enough to reject a large treatment-related increase in ectatic risk within this population. For everyday practice, that means 0.01% atropine can continue to be viewed primarily as a myopia-control therapy, with corneal monitoring reserved for the same kinds of suspicious findings that would trigger concern in any child.
No new corneal surveillance algorithm emerges from this trial, but one clear message does: low-dose atropine did not produce a broad corneal harm signal, and the lone keratoconus case should be treated as a rare clinical event rather than proof of drug toxicity.