14 CHIPS-Child: Testing the developmental programming hypothesis in the offspring of the CHIPS trial

Abstract

Introduction CHIPS-Child is a follow-up to the international CHIPS trial that showed that ‘less tight’ (vs. ‘tight’) control of maternal blood pressure (BP) was associated with no decrease (or increase) in either the primary perinatal outcome (pregnancy loss or high level neonatal care for >48 h) or measures of fetal growth (in utero). Objectives In CHIPS-Child, we tested the developmental programming hypothesis by determining if children born of mothers in ‘less tight’ (vs. ‘tight’) BP control arms of CHIPS showed differences in postnatal growth and health. Methods Follow-up was extended to 12 ± 2 months corrected post-gestational age for anthropometric measurements. For CHIPS subjects who consented to a CHIPS-Child study visit, we collected biological samples from the child (hair samples, buccal swabs) to evaluate hypothalamic–pituitary–adrenal axis (HPA) function (hair cortisol levels) and epigenetic change (DNA methylation analysis of buccal cells). The primary outcome was ‘change in z -score for weight’ (95% CI) between birth and 12 ± 2 months, compared between groups using linear regression adjusted for maternal pre-randomisation factors (hypertension type and centre, antihypertensive type, and maternal body mass index, body mass index [BMI]) (two-tailed p > 0.05) and any between-group differences at baseline among babies followed-up. Secondary outcomes were genome-wide and targeted DNA methylation status (from buccal swabs, measured using the Illumina infinium HumanMethylation450 array platform) and hair cortisol (as a measure of hypothalamic–pituitary activation). Results Of 683 eligible CHIPS babies (59 sites), 500 (73.2%) were followed up in CHIPS-Child [244 (70.1%) in ‘less tight’ vs. 256 (75.7%) in ‘tight’ control] and 414 consented to participate [196, (80.3%) vs. 218 (85.2%)]. 272 (76.8%) of babies had anthropometry at 12 ± 2 months. Babies in ‘less tight’ (vs. ‘tight’) probably differed with regards to change in z -score for weight (primary outcome) [−0.27 (−0.54, 0.00); p = 0.05]; there was no difference in z -scores for length [0.14 (−0.20, 0.49); p = 0.42] or head circumference [0.19 ( - 0.89, 1.26); p = 0.71], or waist circumference (cm) [0.99 ( - 1.75, 3.72); p = 0.45]. Secondary outcomes were measurable only among the babies who were eligible for a study visit ( N = 92), consented ( N = 45), and supplied biological samples ( N = 41). 16 DNA samples (9 vs. 7) passed quality control testing; no difference in DNA methylation was seen based on a false discovery rate >20%. In the 35 adequate hair samples, cortisol levels were significantly lower in ‘less tight’ (vs. ‘tight’)[−496 (−892, −100); p = 0.02]. Conclusions The results of CHIPS-Child suggest that ‘tight’ BP control may be associated with both more rapid postnatal growth (birth to age 12 months) and developmental programming of the HPA axis. These results are limited by small sample size. Longer follow up would be necessary to determine if the findings are associated with important clinical outcomes. (CHIPS-Child Working Group: JP Chanoine, AM Cote, A Devlin, A Gafni, W Ganzevoort, A Gruslin, M Helewa, E Hutton, G Koren, SK Lee, D McArthur, E Rey, WP Robinson, T Roseboom, J Singer, S Wilson, JM Moutquin).