Assisted reproductive technology and major congenital malformations: nationwide sibling-comparison study

초록

Objectives: To evaluate the association between assisted reproductive technology (ART) and major congenital malformations, and assess whether observed risks are due to ART procedures or shared parental and infertility-related factors. Methods: This nationwide, population-based cohort study analyzed data from the Korea National Health Insurance Service on singleton live births from 2018 to 2021 (N = 1,463,212). Exposures to ART included intrauterine insemination (IUI), fresh embryo transfer (ET), and frozen ET. The primary outcome was Core Major Malformations (ICD-10 codes: Q00–Q89) diagnosed within one year of birth. Analyses compared the general population, a multiparous cohort, and a within-mother sibling-comparison cohort (N = 11,367). Stabilized inverse probability of treatment weighting (IPTW) balanced maternal and pregnancy-related covariates. Relative risks (RRs) were estimated using Poisson regression with robust error variances. Results: In the general population, infants conceived through ART had a higher risk of core major malformations (11.94% vs 6.54%; Adjusted RR, 1.47 [95% CI, 1.43–1.51]; P<.001). In the sibling-comparison analysis, this association was greatly reduced and not statistically significant (Weighted RR, 1.14 [95% CI, 0.99–1.30]; P = .064), with the absolute risk difference (ARD) decreasing from 38.9 to 12.1 per 1,000 births. No significant risk increase was found for any ART subtype compared to natural conception within the sibling cohort: IUI (Weighted RR, 0.97 [95% CI, 0.77–1.22]), fresh ET (0.98 [95% CI, 0.79–1.21]), and frozen ET (0.99 [95% CI, 0.82–1.21]). Organ-specific analyses showed no significant associations, except for a slight persistent increase in core musculoskeletal defects (Weighted RR, 1.26 [95% CI, 1.02–1.55]; P = .03). Sensitivity analyses confirmed these findings when multiple gestations were excluded. Conclusions: The increased risk of congenital malformations observed at the population level is largely explained by shared parental and subfertility-related factors rather than ART procedures.