Findings from a recent study published in JAMA Pediatrics tested whether dexamethasone eyedrops can keep prethreshold retinopathy of prematurity (ROP) from progressing to the severe form that requires laser or intravitreal injection.
Infants on the drops progressed at roughly half the rate of those on placebo, though the trial missed its statistical threshold.
Editor’s note: We recently covered this very topic from a different clinical study. Check it out here.
Let’s start with some background.
Some context: ROP is one of the leading causes of preventable childhood blindness, and it blinds an estimated 35,000 children worldwide each year. It starts when retinal vessel growth stalls in very preterm infants, and the ischemia and inflammation that follow push up vascular endothelial growth factor (VEGF), which can drive neovascularization and retinal detachment.
But here’s the issue: Roughly 5% to 10% of screened infants end up needing laser photocoagulation or anti-VEGF injections, and both carry their own adverse effects.
- A global shortage of pediatric ophthalmologists also limits timely treatment.
Now, talk about the study.
DROPROP was the first randomized clinical trial (RCT) of topical dexamethasone in ROP, run double-masked at 6 university hospitals and 8 county hospitals in Sweden from September 2022 through October 2025.
The protocol: Infants received dexamethasone 1 mg/mL or preservative-free saline in identical single-dose containers. Dosing was 1 drop per eye every other day for stage 1 or 2 disease and daily for stage 3, for up to 12 weeks. Median duration was 5.3 weeks.
Who was included in the study?
Of 116 eligible infants, parents consented for 100. All were born before 30 weeks’ gestational age (GA) with prethreshold ROP, confirmed on retinal images by a steering committee of pediatric ophthalmologists.
To qualify, infants needed one of the following:
- Stage 1 or 2 without plus disease in zone I
- Stage 2 or 3 in posterior zone II without plus disease
Demographics: Mean GA at birth was 25.1 weeks and mean birth weight (BW) was 712.9 g. 42% were female, and most were already at stage 3 on entry (74% and 76%).
Findings?
In the intention-to-treat analysis, type 1 ROP requiring invasive treatment occurred in 10 of 50 infants (20.0%) on dexamethasone versus 19 of 50 (38.0%) on placebo.
Adjusted for GA and site, that came to an odds ratio (OR) of 0.44 (95% CI, 0.17 to 1.12; P = .08), a relative risk reduction of 47%.
- The confidence interval crossed 1, so the primary hypothesis was not confirmed.
- Per-protocol numbers ran the same direction, at 18.8% versus 38.8%.
Tell me more.
On safety: Adverse events tracked closely between groups, affecting 70.6% of infants on dexamethasone and 73.5% on placebo, with no treatment-related serious events in either group.
What to watch: 13 infants (25.5%) on dexamethasone and 10 (20.4%) on placebo had at least one intraocular pressure (IOP) reading above 20 mm Hg, 8 of those attributed to eyelid pressure during measurement. Salivary cortisol dipped during treatment and normalized afterward, with no clinical evidence of adrenal suppression in any infant.
Limitations?
The trial was underpowered for the effect it found. Sample size assumed a 60% treatment effect against a 50% placebo event rate. The observed effect was 47% and the placebo rate landed at 38%, which the authors attributed partly to shifts in neonatal care over the enrollment period.
Worth noting: Seven infants were treated invasively without the steering committee confirming type 1 ROP, 5 on dexamethasone and 2 on placebo, mostly for logistical reasons. Counting them dilutes the treatment effect. Every site was Swedish, so the findings reflect high-income neonatal care.
Expert opinion?
The authors argued that an effect of this size—delivered by a low-cost drop with no invasive step—is worth pursuing. They were direct that larger, adequately powered RCTs are needed before topical dexamethasone belongs in national guidelines.
Anything else?
Subgroup signals: The effect looked stronger in heavier infants (BW > 660 g; OR 0.23; 95% CI, 0.06 to 0.94) and in those on parenteral nutrition for fewer than 14 days. Both were exploratory analyses with wide intervals.
To note: Participants will be followed at ages 2.5 and 6.5 years for visual development and late effects.
- The authors also flagged low- and middle-income settings as where the drops could matter most, since screening and treatment capacity there is thin.
Take home.
Topical dexamethasone did not clear the statistical bar, but a 20% versus 38% split in progression to treatment-requiring ROP is hard to dismiss, and safety monitoring turned up nothing disqualifying.
Though it’s too early to change practice … if dexamethasone drops get used off-label in prethreshold ROP, IOP and growth are the parameters to watch.