Birthweight trends for babies born using Assisted Reproductive Treatments between 1991 and 2018: a national cohort...

Has the average birthweight (BW) for singletons born in the UK through ART changed between 1991 and 2018 relative to the general population?

The average gestation-adjusted BW for ART singletons increased steadily between 1991 and 2012 before becoming almost static until 2018; these changes were not observed in the general UK population.

Many older studies have shown that ART babies are low BW (LBW) compared to naturally conceived babies. Some subsequent evidence suggests that the average BW for fresh and frozen transfer ART babies has increased over recent years.

This cohort study utilized the UK Human Fertilisation and Embryology Authority (HFEA) dataset of 140 859 singleton ART babies born between 1991 and 2018 in the UK following both fresh and frozen embryo transfer (FET). Comparative aggregated data of 11 790 372 singleton babies born in the UK general population during the same period were obtained from NHS Digital.

The HFEA dataset was used to explore ART BW (raw or gestation-adjusted) and gestational age trends between 1991 and 2018. Appropriate regression models assessed the outcomes and adjusted for child sex, available treatment factors (fresh/frozen ET, embryo transfer stage (cleavage or blastocyst transfer), number of embryos transferred, ovarian stimulation, ICSI, treatment centre/clinic, and pre-implantation genetic screening of embryos) and patient factors (maternal age, previous IVF attempts, previous live births, previous pregnancies, ethnicity, country of treatment within the UK (England, Northern Ireland, Scotland, and Wales), funding source, total number of oocytes retrieved, and infertility causes (endometriosis, idiopathic, ovulatory, tubal diagnosis, and male factor)). BW, LBW, and macrosomia models were additionally adjusted for gestational age. Segmented trend analysis using linear splines was used to capture non-linear time effects for BW and gestation outcomes. For the general population data, mean BW and gestation outcomes were assessed as raw, unadjusted yearly averages.

Unadjusted raw BW trends show that ART babies were initially smaller than the general population, but over time, are becoming closer to it. Fully adjusted mean ART BW increased by ∼75 g between 1991 and 2012, and after that, it was almost static. A drop in ART raw BW starting in 2013 was concurrent with a drop in average gestation by an overall average of 2.5 days between 2012 and 2018. The mean raw BW in the general population remained almost static over the study period and consistently higher than the ART BW average, with a notable drop beginning in 2013, concurrent with a decline in average gestation. Between 1991 and 2000, and 2012 and 2018, average gestation trends were comparable in the ART and general population, with varying patterns between 2001 and 2012.

We could only explore factors available in the dataset, possibly missing some unobserved/unrecorded covariates such as socio-economic and lifestyle factors. The ART dataset used required patient consent, leading to the exclusion of ∼40% of all cycles conducted during the study period.

ART singletons getting bigger over time is reassuring, as fewer ART babies now have adverse birth outcomes such as LBW. This is desirable within the developmental origins of health and disease (DOHaD) framework, which links adverse child outcomes to later-life disease predisposition. These changes in ART singletons (not observed in the general population) might be the result of changes or improvements in ART technologies, clinic practices, and ART patient population characteristics. Therefore, there is a need to continue monitoring BW-related outcomes such as large for gestational age and macrosomia, which are more prevalent in frozen ET ART babies and, though directly associated with the observed increase in average ART BW, can be detrimental to child health.

This study was funded by the EU H2020 Marie Sklodowska-Curie Innovative Training Networks (ITN) grant Dohartnet (H2020-MSCA-ITN-2018-812660).

The authors have no competing interests to declare.

n/a.