314 - Risk factor of retinopathy of prematurity in very preterm infants, is small for gestational age enough?

Poster Number: 314
Publication Number: 314.2814

Background: Small for gestational age (SGA) was associated with increased odds of retinopathy of prematurity (ROP) in preterm infants. Birth weight (BW) percentile reflecting more comprehensive information about the relative growth status might also be a relevant predictor for ROP.

Objective: To determine the associations between BW percentile and ROP in very preterm infants (VPIs).

Design/Methods: This retrospective multicenter cohort study was conducted at hospitals participating in the Chinese Neonatal Network and used prospectively collected data on VPIs with gestational age (GA) ≤316/7 weeks admitted between January 1, 2019, and December 31, 2021. Exposure is GA- and gender-specific BW percentiles based on 2013 Fenton growth curve. The primary outcome was any stage of ROP. The secondary outcomes were severe ROP (stage 3 or above). Odds ratios (ORs) with 95% confidence intervals (CIs) were estimated, adjusted for potential confounders. Firstly, BW percentile was used as continuous variable to estimate the association with ROP by spline smoothing and logistical regression model. Secondly, an optimal discriminative value of BW percentile was generated by receiver operating characteristic (ROC). VPIs were also divided into eight groups to test the risks of ROP in different groups. Stratified analyses were conducted in subgroups with different gender, GA, number of births, and maternal hypertension.

Results: A total of 17 882 very preterm infants were enrolled (median GA, 29.86 weeks [interquartile range, 28.57 and 31.00 weeks]), of which 4831 (27.0%) and 660 (3.7%) were diagnosed as any stage of ROP and severe ROP respectively. The incidence of any stage of ROP from 21.2% (BW percentile ≤97%) to 39.0% (BW percentile < 3%). After adjustment for maternal and infant characteristics, each 10 percentile of BW decreased was associated with 15% higher odds of any stage of ROP and severe ROP (adjusted OR, 0.85 [95%CI, 0.83, 0.86] and 0.85 [95%CI, 0.82, 0.89] respectively). The optimal discriminative BW percentile value for any stage of ROP and severe ROP were determined as 26% and 19%. Low BW percentile under the cut-off showed OR 2.20 (95%CI, 1.97, 2.47) for any stage of ROP and OR 2.91 (95%CI, 2.22, 3.80) for severe ROP. The associations between different BW percentile and ROP were similar in subgroup analysis.

Conclusion(s): A negative association was observed between BW percentile and ROP in very preterm infants. Infants with BW percentile lower than 26% and 19%, not only SGA infants, were with higher risk for any stage of ROP and severe ROP respectively, who should be pay more attention.