Findings from a recent study published in Ophthalmology suggest that introducing low-dose dexamethasone eye drops when type 2 retinopathy of prematurity (ROP) is first diagnosed can prevent progression to the stage where invasive treatment is needed.
The results come from a difference-in-differences analysis comparing one intervention site to three control sites across Sweden.
Give me some background first.
ROP is one of the most common causes of severe visual impairment and blindness in children worldwide. Preterm infants are born with immature retinal vasculature, and during the neonatal period, abnormal fibrovascular vessel growth can lead to retinal detachment if left unchecked.
The standard approach: When ROP progresses to type 1 (the stage requiring treatment), infants undergo laser ablation of the avascular retina or intravitreal injections of anti-vascular endothelial growth factor (VEGF) agents. Both procedures typically require general anesthesia, which carries its own risks for premature babies.
- Type 2 ROP is the precursor stage. It warrants close monitoring but, under current international guidelines, no active treatment. Clinicians wait until the disease either regresses on its own or progresses to type 1, at which point they intervene.
Now, talk about the study.
A team based in the southern Swedish health care region began using low-dose topical dexamethasone drops (1 mg/ml) starting at the type 2 ROP diagnosis in 2020.
- The idea grew out of an earlier observation: when postoperative dexamethasone drops were started before laser therapy to reduce anterior segment inflammation, some infants’ ROP actually regressed, making laser treatment unnecessary.
The question: Could starting dexamethasone earlier, at a lower dose, prevent type 2 ROP from progressing to type 1 altogether?
And how was this tested?
Investigators used data from the Swedish Quality Register for Retinopathy of Prematurity (SWEDROP) and compared treatment rates across two time periods (2015 to 2018 vs. 2020 to 2021) at the intervention site and three control sites (Stockholm, Uppsala, Örebro) that did not use dexamethasone drops.
Who was included in the study?
All screened infants registered in SWEDROP with a gestational age at birth of less than 30 weeks were included, totaling 2,017 infants across all four sites and both time periods.
Demographics:
- Median gestational age ranged from 24.4 to 27.7 weeks across groups
- The intervention site had 32 infants with severe ROP in each period
- The three control sites had 175 and 57 infants with severe ROP in the control and intervention periods, respectively
To note: Infants born in 2019 were excluded because the dexamethasone protocol was still being standardized that year, with considerable variation in dose and timing.
Findings?
At the intervention site, the proportion of infants with severe ROP needing traditional treatment dropped from 23 of 32 (72%) during the control years to 4 of 32 (13%) during the intervention years (P < 0.001).
For all screened infants at the intervention site, treatment rates fell from 23 of 409 (5.6%) to 4 of 217 (1.8%) (P = 0.03).
At the control sites, no such decline occurred. Treatment rates for severe ROP actually ticked upward, from 82 of 175 (47%) to 32 of 57 (56%) (P = 0.22).
Tell me more.
The difference-in-differences analysis confirmed the differential decline was specific to the intervention site. For infants with severe ROP, the adjusted interaction odds ratio (OR) was 0.05 (95% confidence interval [CI], 0.01 to 0.22; P < 0.001).
Translation: Roughly a 95% greater reduction in the odds of needing traditional treatment at the intervention site compared with the control sites, after adjusting for gestational age, birth weight deviation, sex, bronchopulmonary dysplasia, necrotizing enterocolitis, and respiratory support.
When all screened infants were included, the adjusted interaction OR was 0.42 (95% CI, 0.12 to 1.52; P = 0.18). Same direction, but without reaching statistical significance.
Talk treatment details.
During the intervention period, 28 of 32 infants diagnosed with type 2 ROP at the intervention site received topical dexamethasone.
- Of these, 24 did not progress to requiring laser or anti-VEGF treatment.
The median duration of dexamethasone therapy was 4 weeks (interquartile range, 2 to 16 weeks), and 19 of 24 infants received a maximum dose of 1 drop/day per eye.
And the safety profile?
A review of medical records up to 3 years after the last screening visit found no documented reactivation of ROP, and no cases of cataract or glaucoma.
One infant had a suspected elevated intraocular pressure (IOP) of 26 mmHg in the right eye and 30 mmHg in the left during the final screening visit. It normalized at the subsequent follow-up without intervention.
Any limitations?
This was an observational study, not a randomized controlled trial. The difference-in-differences design helps control for confounding, but it can’t fully establish causation.
Other considerations:
- It’s possible severe ROP was underdiagnosed at the intervention site during the control period, or that some degree of overdiagnosis occurred at control sites
- The intervention site relied more heavily on widefield digital imaging, which may have influenced diagnostic consistency
- Interobserver variability across the four sites is another concern, though the same core group of ophthalmologists performed screenings during both periods
See here for more.
Expert opinion?
No outside expert commentary was included in the study.
That said: The authors argued that the current standard of care, monitoring type 2 ROP without active intervention, leaves clinicians waiting for the disease to worsen before acting.
- As such: Low-dose dexamethasone eye drops may offer a way to treat earlier in the disease process, before invasive procedures become necessary.
Their position: Timely administration is critical, as animal studies have shown that dexamethasone reduces neovascularization only when given before peak disease activity, not after.
Anything else?
It’s worth noting that a prospective, randomized, multicenter, double-masked trial (DROPROP) is currently underway to evaluate dexamethasone eye drops’ effect on ROP and potential systemic effects.
Long-term follow-up data from that trial will be critical for confirming these findings.
And now for the take home.
For premature infants with type 2 ROP, low-dose topical dexamethasone may offer a simple, cost-effective alternative to waiting for disease progression and treating with laser or anti-VEGF under anesthesia.
While the data requires confirmation from that aforementioned ongoing randomized trial, the magnitude of the reduction—from 72% to 13% for infants with severe ROP—is hard to ignore.