Abstract Study question Is infant mortality, i.e., mortality up to one year of age, higher after assisted reproductive technology (ART) than after natural conception (NC)? Summary answer Infant mortality was increased in ART-singletons vs NC-singletons, with similar increased mortality across ART fertilization methods. Infant mortality was not increased in ART-twins vs NC-twins. What is known already Children born after ART have an increased risk of poor perinatal outcomes such as preterm birth and low birth weight, mainly due to the higher twin birth rates in the ART population. ART singletons also carry an increased risk of poor perinatal outcomes as compared to their NC counterparts. Studies on consequences of being born after ART are mostly sparse and inconclusive which also applies to studies on infant mortality. Study design, size, duration A Nordic registry-based study including all children born alive in Denmark (1994–2014), Finland (1990–2014), Norway (1984–2015) and Sweden (1985–2015) resulting in a cohort of 171,836 ART-children and 7,775,739 NC-children. Among ART-children 127,248 (74%) were singletons and 42,569 (25%) twins, and among NC-children 7,576,396 (97%) were singletons and 194,775 (3%) twins. Multiples with >2 children were excluded (<0.1%). All singletons and twins were followed until death, emigration or one year of age. Participants/materials, setting, methods Data originated from national ART-, birth-, and causes of death registers. Infant mortality was compared for ART vs NC singletons and ART vs NC twins in separate analyses using Cox proportional hazards model yielding hazard ratios (HRs) with 95% confidence intervals (95% CIs). Infant mortality in singletons and twins was compared for IVF, ICSI, fresh and frozen embryo transfer vs NC counterparts. Cause-specific death analysis accounted for other causes of death as a competing risk. Main results and the role of chance Overall, 515 (4‰) ART- and 26,076 (3‰) NC-singletons and 615 (14‰) ART- and 3317 (17‰) NC-twins died during infancy. Median age at death was 3 (IQR 0–195) and 10 days (IQR 0–245) for ART- and NC-singletons, respectively, and 1 (IQR 0–151) and 2 days (IQR 0–161) for ART- and NC-twins, respectively. After adjusting for country, parity, maternal age, and birth year, the adjusted HR (aHR) for infant mortality was 1.75 (95% CI 1.60–1.91) for ART-singletons vs NC-singletons, and 0.94 (95% CI 0.86–1.03) for ART-twins vs NC-twins. A higher mortality was found consistently among ART-singletons born after IVF, ICSI, fresh embryo transfer and frozen embryo transfer vs NC-singletons. Infant mortality was consistently similar between ART-twins and NC-twins across treatment methods. Leading causes of death were conditions originating in the perinatal period (23% and 47% of deaths among singletons and twins) with aHR 1.87 (95% CI 1.65–2.12) for ART-singletons vs NC-singletons, and aHR 0.97 (95% CI 0.86–1.10) for ART-twins vs NC-twins, and congenital malformations (19% and 11% of deaths among singletons and twins) with aHR 1.31 (95% CI 1.09–1.58) for singletons and aHR 1.37 (95% CI 1.08–1.74) for ART-twins vs NC-twins. Limitations, reasons for caution Observational studies are prone to confounding. Despite adjustment for some major confounders, residual confounding from underlying parental conditions cannot be excluded. Causes of death may be more complex than the causes registered. Further, 44% of deaths among NC and 12% of deaths in ART were without an assigned cause. Wider implications of the findings Findings that infant mortality is increased among ART- vs NC-singletons corresponds with previous findings of poorer perinatal outcomes in this group. Risks were similar across ART fertilization methods and cryopreservation status. Trial registration number Not applicable
Read full abstract