Eye drops containing the steroid dexamethasone did not clearly reduce the number of very premature babies whose eye disease worsened to the point where invasive treatment was needed, a randomised trial in Sweden has found. Fewer babies who received the drops reached that stage than babies who received dummy drops, but the difference may be down to chance. The trial, called DROPROP, included 100 babies and was published in JAMA Pediatrics.
The study concerns retina disease in premature babies, known in medicine as retinopathy of prematurity. The retina is the thin layer at the back of the eye that senses light. According to the researchers, the severe form is treated with a laser or with injections into the eye. These treatments are invasive but necessary to prevent blindness. Laboratory work and limited experience in hospitals had suggested that steroid drops placed on the eye might slow the disease and reduce the need for such treatment.
How the trial was run
The trial took place at 6 university hospitals and 8 county hospitals in Sweden. It included babies born before 30 weeks of pregnancy between 2022 and 2025 whose disease was severe and close to the stage that needs treatment. On average, the babies were born at 25.1 weeks and weighed 712.9 grams; 42 of them were girls.
Chance decided which babies received dexamethasone drops and which received a placebo, drops of salt water with no medicine in them. The trial was double-blind: neither the families nor the staff knew which drops a baby was getting. One drop was given every day or every other day for up to 12 weeks. The main question was how many babies would progress to the stage that requires invasive treatment.
Fewer babies in the drops group, but chance cannot be ruled out
In the dexamethasone group, 10 of 50 babies (20.0%) reached the stage that requires invasive treatment. In the placebo group, 19 of 50 babies (38.0%) did. The difference did not reach statistical significance. This means it may be due to chance, so the trial shows no clear difference between the two groups.
The researchers also looked only at the babies who received the drops as planned. Among them, 9 of 48 babies (18.8%) in the dexamethasone group and 19 of 49 babies (38.8%) in the placebo group reached that stage. Here too, the range of likely results included the possibility of no difference at all.
The team monitored the babies for side effects and found no clinically important differences between the groups.
What the researchers conclude
The researchers write that giving dexamethasone drops in time lowered the risk in numerical terms, while noting that their analysis did not reach statistical significance. In their view, the findings suggest that the drops may be a way to reduce the need for invasive treatment without an invasive procedure.
