IMPE Abstracts

Background: Cardiometabolic (CM) risk is linked to being born small for gestational age (SGA, birthweight <-2SDS). Using data from the Avon Longitudinal Study of Parents and Children (ALSPAC), an ‘omic signature in SGA catch-up predicts pre-hypertension in adolescence (Garner et al JES 2021). Fetal growth restriction (FGR) alone may be linked with greater CM risk. Therefore, we focused on CM risk in children born following pregnancies at higher risk for FGR, irrespective of birthweight.

Aims:  1.To identify associations between fetal and childhood weight trajectories, CM risk markers, and molecular pathways. 2. To validate the predictive ability of an ‘omic signature for pre-hypertension within this cohort.

Methods:  We recruited 81 children aged 3-6 years, following pregnancies >34 weeks at increased risk of FGR based on adverse maternal antenatal serology (PAPP-A, α-fetoprotein or inhibin A). Body mass index (BMI) SDS, abdominal circumference (AC), mid-upper arm circumference (MUAC), %fat, systolic blood pressure (SBP) and brachial augmentation index (AI) were recorded. Fasting blood samples were collected for CM markers and ‘omic analyses (n=31). ∆fetalwt ([birthweight centile minus 23-week estimated fetal weight centile]/days) and ∆childwt ([weight centile minus birthweight centile]/years) were divided into quartiles and differences in CM markers compared between Q1 and Q4. Differentially expressed genes (DEGs) and metabolites (DEMs) were established using EdgeR® and MetaboAnalyst®. Gene set enrichment analysis (GSEA) identified pathways. Random forest was used to determine predictive ability of the ALSPAC-derived ‘omic signature.

Results:  69% (56/81) had ∆fetalwt <0, but only 12% (10/81) were born SGA. SBP was higher and HDL lower in ∆fetal Q1 (lowest intrauterine weight gain) vs Q4 (P<0.05). SBP, BMI SDS, AC, MUAC, AI and %fat were higher in ∆childwt Q4 (highest childhood weight gain) vs Q1 (P<0.05). GSEA based on DEGs between ∆child quartiles highlighted a pathway including ARG1. Ornithine was a DEM between ∆fetalwt and ∆childwt quartiles. Transcriptomic data were available for 33/47 ALSPAC genes predictive of pre-hypertension. Random forest was able to accurately predict child SBP Q4 (area under the curve 0.98, error rate 3.9%).

Conclusions:  Low ∆fetalwt and high ∆childwt were associated with CM risk, with a less favourable CM profile after pregnancies with a majority showing FGR but a minority being SGA. ‘Omic analyses uncovered the arginine-nitric-oxide pathway, which has previously been associated with hypertension. Pre-hypertension genes in SGA catch-up predicted higher SBP within this cohort enriched for FGR. Therefore, fetal growth restriction without SGA could predict higher offspring SBP.