Screening-detected refractive errors in infants born after assisted reproductive technology: a prospective cohort study.

PURPOSE: Infants born after assisted reproductive technology (ART) have more adverse health outcomes than spontaneously conceived infants. Evidence on the impact of ART on the offspring's visual system, particularly refractive errors (REs), remains limited and inconsistent.

METHODS

The present study included 6847 infants from the Jiangsu Birth Cohort (JBC), which enrolled couples who conceived either through ART or spontaneously between 2015 and 2022 in Jiangsu Province, China. ART procedures and parental characteristics were collected prospectively, and REs were assessed at one year of age using the Spot Vision Screener. Poisson regression models were used to examine the association between ART and screening-detected REs.

RESULTS

Screening-detected REs were identified in 369 infants born after ART (11.4%) and in 405 infants spontaneously conceived (11.2%). No overall association was observed between ART and the risk of REs (RR, 0.91; 95% CI 0.77-1.07). Infants born to parents with different infertility diagnoses or following specific ART procedures had similar risks of REs compared with spontaneously conceived infants. In the analyses restricted to ART-conceived infants, infants conceived using the gonadotropin-releasing hormone antagonist (GnRH-antagonist) protocol had a lower risk of REs (RR, 0.56; 95% CI 0.44-0.71) compared with those using the gonadotropin-releasing hormone agonist (GnRH-agonist) protocol.

CONCLUSION

We observed no increased risk of screening-detected REs at one year of age in ART-conceived infants compared with spontaneously conceived infants, independent of parental infertility diagnoses or specific ART procedures. Among the ART-conceived infants, the use of the GnRH-antagonist protocol may be associated with a lower risk of screening-detected REs; however, this finding should be interpreted cautiously because protocol selection was clinically guided and residual confounding cannot be excluded.