# Visceral and subcutaneous adipose tissue in children born after ART with frozen and fresh embryo transfers

## TL;DR

Children born via frozen embryo transfer had slightly lower visceral to subcutaneous fat ratios compared to naturally conceived children, but no major differences were found in metabolic risk factors.

## Contribution

This study is the first to report visceral to subcutaneous adipose tissue ratios in children born after assisted reproductive technology.

## Key findings

- Children conceived via frozen embryo transfer had significantly lower VAT/SAT ratios compared to naturally conceived children.
- No significant difference in VAT/SAT ratio was found between children born via fresh embryo transfer and natural conception.
- The lower VAT/SAT ratio in frozen embryo transfer children remained significant after adjusting for multiple confounders.

## Abstract

Is the ratio of visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) comparable between children following ART and natural conception (NC)?

Children conceived by frozen embryo transfer (FET) had slightly lower VAT/SAT ratios than children following NC; no difference in VAT/SAT ratio was observed in children born following fresh embryo transfer (Fresh-ET) as compared to those born from NC.

The VAT/SAT ratio is closely related to the metabolic profile, with a high ratio increasing the risk of cardiometabolic diseases. To our knowledge, no studies have reported the VAT/SAT ratio in children following ART.

This prospective exploratory observational cohort study included 150 singletons aged 7–10 years. All children were born in eastern Denmark. The study was conducted between November 2018 and August 2020.

This is a sub-study of the ‘Health in Childhood following Assisted Reproductive Technology’ (HiCART) study. The children were conceived after FET (n = 50), Fresh-ET (n = 50), and NC (n = 50), and children conceived by NC were matched to ART children by sex and birth year. The children underwent abdominal MRI for the quantification of abdominal adipose tissues along with measurements of blood pressure, fasting blood samples, anthropometric measurements, and dual-energy X-ray absorptiometry scans. The volumes of VAT and SAT were semi-automatically quantified, blinded for the mode of conception. The level of statistical significance was set to a P-level below 0.05. Multivariable linear regression analysis of the VAT/SAT ratio was performed to adjust for confounders in a five-step approach: Model 1: Adjusted for child age and sex; Model 2: Model 1 plus maternal age at delivery and maternal BMI at pregnancy; Model 3: Model 2 plus birth weight and child BMI; Model 4a: Model 3 plus maternal educational level; Model 4b: Model 3 plus pubertal status. The confounders were selected based on their association with metabolic risk factors according to previous studies.

As previously reported in the HiCART studies, there were no differences between the groups in anthropometric measurements including BMI, lean body mass, blood pressure, or triglycerides. The crude VAT/SAT ratio differed significantly between the three groups (mean (SD); FET 0.26 (0.08), Fresh-ET 0.29 (0.07), NC 0.30 (0.08), ANOVA—P = 0.014). Pairwise comparison revealed that children conceived after FET had lower crude VAT/SAT ratio than children conceived after NC (P = 0.007) with a mean difference of −0.04, 95% CI (−0.07; −0.01), and a tendency for a lower VAT/SAT ratio as compared to the Fresh-ET group (P = 0.059) with a mean difference of −0.03, 95% CI (−0.06; 0.00). Lower VAT/SAT ratio in FET as compared to NC remained after adjustment for child age and sex (Model 1: −0.04 (−0.07; −0.01)), maternal age at delivery and maternal BMI at pregnancy (Model 2: −0.04 (−0.07; −0.01)), birth weight and child BMI (Model 3: −0.04 (−0.07; −0.01)), maternal educational level (Model 4a: −0.05 (−0.08; −0.01)), and puberty (Model 4b: −0.04 (−0.08; −0.01)) in a five-step approach. Repeated analysis of twenty MRI scans showed good intra-rater repeatability of VAT and SAT volume quantifications.

The sample size was relatively small and selection bias due to differences in intrinsic factors between the three groups may affect the results. Well-described confounders from the literature were included in the multivariable regression analysis, but the observational nature of this cohort study hinders the establishment of causality.

Reassuringly, this study found no clinically important difference in VAT/SAT ratio between children following ART (both FET and Fresh-ET) and NC, although a small but significantly lower VAT/SAT ratio was found in children born after FET compared with NC children.

A.R.O was supported by a scholarship from Herlev-Gentofte Copenhagen University Hospital. The study was funded by grants from Novo Nordisk Foundation (NNF18OC0034092, NFF19OC0054340) and The Research Foundations at Rigshospitalet and Herlev-Gentofte Copenhagen University Hospital (unrestricted grant). A.P. has received grants (via her institution), honoraria, and consulting fees from Gedeon Richter, Ferring Pharmaceuticals, and Merck A/S, as well as consulting fees from Novo Nordisk A/S and Cryos, honoraria from Organon and support for attending meetings (via her institution) from Gedeon Richter. K.M.M. has received royalties from Gyldendal and consulting fees from The National Board of Wealth and Welfare in Sweden, in addition to honoraria from Novo Nordisk A/S and Lundbeck A/S, and serves as a medical expert for the Ministry of Justice, Department of Civil Affairs. All other authors declare no conflicts of interest.

ClinicalTrials.gov identifier: NCT03719703.