Prenatal Exposure to Maternal Diabetes May Increase Severe ROP Risk

Premature infants exposed to maternal diabetes in utero may face a higher risk of developing severe ROP, according to a large retrospective study published in Ophthalmology Science. Researchers analyzing more than 3,100 infants found that maternal Type 1 and Type 2 diabetes were significantly associated with Stage 3 to 5 ROP, even after controlling for gestational age, birthweight and other clinical factors, suggesting prenatal metabolic exposure may influence retinal vascular development. Photo: Don Lyon, OD, and Danielle Warren, OD. Click image to enlarge. Retinal vascularization isn’t complete until about 36 to 40 weeks of gestation, making premature infants susceptible to retinopathy of prematurity (ROP), a leading cause of childhood blindness, due to disrupted normal vascular growth ex-utero. ROP shares characteristics with proliferative diabetic retinopathy and, given that hyperglycemia can cross the placenta, it’s possible that maternal diabetes could predispose infants to retinal damage. This was the hypothesis posed by researchers at Vanderbilt University in Nashville, who recently published their findings in Ophthalmology Science, to which they concluded that both type 1 and type 2 diabetes in mothers increases an infant’s risk of developing vision-threatening ROP. The study population included infants of ≤30 weeks and six days gestational age (GA) at birth (<31 weeks GA) or with birthweight ≤1,500g who survived at least 30 days of life or 40 weeks corrected GA. The final cohort comprised 3,139 infants from 2004 to 2021, with those who were diagnosed with ROP Stage 3 through 5 (n=311, 10%) and controls defined as ROP Stage 0 (n=2,121, 69%). According to the study results, lower birthweight (BW) and gestational age were associated with more severe ROP. Maternal diabetes prevalence was 9% and resulted in higher BW and GA in premature infants; however, Stage 3 to 5 ROP was significantly associated with maternal diabetes (OR: 2.87).Most of the infants in the Stage 3 to 5 group were born outside of Vanderbilt (72%), while most controls (69%) were inborn. Comorbidities were found to be higher in the ROP group than controls, including necrotizing enterocolitis (NEC), bronchopulmonary dysplasia (BPD) and intraventricular hemorrhage (IVH) Grade 4. However, children with exposure to maternal diabetes had lower rates of these comorbidities than those from mothers without diabetes, and were more likely to be inborn. ROP progression is known to be associated with decreased BW and GA. On average, infants with higher stages of ROP (as well as those with comorbidities such as IVH, BPD and NEC) tended to have lower BW and GA, while infants exposed to maternal diabetes tended to have higher BW and GA than those without.“We used a multivariate logistic regression model to test for an association between prenatal exposure to maternal diabetes and ROP severity,” the authors wrote. “A statistically significant positive association was found using exposure to maternal diabetes as the independent variable and the risk of Stage 3-5 ROP as a dependent variable, controlling for potential confounding effects of sex, race, GA, BW, year of birth (YOB), birth center location, NEC, IVH and BPD (OR: 3.04). Similar associations were found when maternal diabetes was stratified by subtype but were only statistically significant for T1DM (OR: 6.36) and T2DM (OR: 5.82). Low GA, low BW, earlier year of birth, male sex, being born outside [Vanderbilt], and the presence of IVH were all also significantly associated with increased odds of Stage 3-5 ROP in the multivariate model.”The data also showed a trend towards association with gestational diabetes mellitus, but it was not found to be statistically significant. The authors stated, however, the median GA of infants with advanced stages of ROP (25 weeks GA) is in the early range of the window for maternal gestational diabetes testing (24 to 28 weeks GA), potentially leading to underdiagnoses.Among the limitations noted by the researchers, the first is the focus on the rare outcome of Stage 3 to 5 ROP. They mention that outborn infants with Stage 3 to 5 ROP may not have complete maternal medical data available, therefore information on maternal hemoglobin A1c (HbA1c) levels was not accessible. “Obtaining HbA1c for future studies would likely require better electronic health records (EHR) integration across medical systems,” they noted, adding, “This study was limited to infants with ROP screening and maternal diabetes data, which can introduce selection bias against infants with missing data. Finally, the database is generated from manually curated EHR, which can be subject to coding errors; however, this limitation was mitigated with additional manual chart review for the outcomes of interest and missing variables.”Ultimately, the authors did identify a positive association suggesting that neonates screened within typical BW and GA criteria and who were exposed to maternal diabetes in utero have significantly higher odds of developing Stage 3 to 5 ROP compared to those not exposed, assuming all other controlled variables remain constant. “Given the increasing number of premature infants that can survive despite low GA and BW and the dearth of ophthalmologists available to screen for ROP, future inclusion of maternal diabetes in decisions regarding ROP screening intervals or participation outside of current screening guidelines may be beneficial to prevent disease progression and preserve vision in at-risk infants,” they concluded in their paper.Click here for the journal source. Lewis, Adam et al.  Exposure to Type 1 and Type 2 maternal diabetes is associated with stage 3-5 retinopathy of prematurity. Ophthalmology Science. March 4, 2026. [Epub ahead of print.] This article was developed by the editorial staff in conjunction with experts in the field. In the process, AI may have been among the editorial tools used to meet the goals of human editors, who approved all content.