The Role of Maternal Height in Pregnancy Outcomes and Complications
MetadataShow full item record
Background: Adult height reflects not only an individual’s genetic make-up but also their early life environmental exposures, socioeconomic status, infectious disease history, nutrition, and growth and development. Increasing evidence suggests an inverse association of height with the risk of type 2 diabetes and cardiovascular disease among men and non-pregnant women. Studies of the association of maternal height with pregnancy complications and outcomes have shown inconsistent results. Further, understanding the role of maternal height in pregnancy outcomes and complications, independent of pre-pregnancy weight, represents a major gap in the literature. Objectives: The specific aims of this dissertation were to 1) examine overall and infant-sex specific associations of maternal height, independent of maternal pre-pregnancy weight, with infant birth size (birth weight [BW], birth length [BL], and head circumference [HC]); 2) examine associations of maternal height, independent of maternal pre-pregnancy weight, with pregnancy complications (gestational diabetes [GDM], preeclampsia, and preterm birth); 3) examine the joint association of maternal height and GDM with neonatal outcomes: BW, BL, HC, and infant macrosomia; and 4) examine the causal effect of maternal height, independent of pre-pregnancy weight, on glucose tolerance during pregnancy using a Mendelian randomization approach. Methods: The specific aims were addressed using data collected as part of three prospective cohort studies of pregnant women: Omega (Aims 1–3), Hyperglycemia and Adverse Pregnancy Outcome (HAPO) (Aims 1–4) and Growing Up in Singapore Towards healthy Outcomes (GUSTO) (Aim 1). Socio-demographic and historical data for the participants were collected using a questionnaire at approximately 15 weeks (Omega) or 28 weeks (HAPO and GUSTO) of gestation whereas information on pregnancy outcomes, infant birth size, and other relevant information was abstracted from medical records. We defined cohort-specific percentiles of height (cm) using self-reported height (Omega) or the measured height collected during the oral glucose tolerance test visit (HAPO and GUSTO). We categorized height as short, average, or tall based on the <20th, 20th–80th, and ≥ 80th percentile values, respectively. We estimated the average difference in the mean of the infant birth size (BW, BL, HC) or the relative risk (RR) of the pregnancy complications (GDM, preterm birth, preeclampsia) and the corresponding 95% confidence interval (CI) using unadjusted and adjusted linear regression models or Poisson regression models (with robust standard errors), respectively. We repeated the analysis for associations of each height category (short, average, and tall) with infant birth size, including macrosomia (≥ 4000 g). We evaluated the p-value of the cross-product term between overall height and 1) GDM or 2) infant sex to determine whether maternal height-infant birth size association varied by GDM or infant sex, respectively. Using height-related single nucleotide polymorphisms (SNPs) (N = 36) and genotype data, we calculated two weighted genetic risk scores (GRS) for maternal height for each individual in HAPO—one using the effect size estimates (the increase in height Z-score per one additional risk allele) from a previous GWAS (external weight) and another from similar estimates derived from our analytical cohort (internal weight). We determined whether genetically determined height was associated with the risk of GDM using the Mendelian randomization approach. Results: Mean height and mean pre-pregnancy weight of the participants were 168 (Standard Deviation [SD]: 7.67) cm and 65.9 (SD: 13.6) kg in Omega, 161 (SD: 7.51) cm and 64.8 (13.6) kg in HAPO, and 158 (SD: 5.64) cm and 56.6 (11.3) kg in GUSTO, respectively. A 5 cm greater height of the mother was associated with an average of 23.82 g to 46.26 g greater birth weight in the three cohorts, and 1% and 18% lower the risk of gestational diabetes in Omega and HAPO, respectively. We found associations of maternal height (continuous as well as short and tall categories) with birth weight, birth length, and gestational diabetes, but not head circumference. We did not find any evidence of a non-linear relationship between maternal height and infant birth size. We observed evidence of infant sex-specific association of maternal height with BW in one cohort (GUSTO) such that associations for short height were significant only among female infants. There was a significant positive association between maternal height and preterm birth and a significant inverse association with preeclampsia in the Omega study, although the association with preeclampsia was not seen in HAPO. Results from the Omega study showed no significant association of gestational diabetes with birth weight, birth length, head circumference, and macrosomia within categories of short, average, or tall height women. In HAPO, gestational diabetes was associated with a significantly increased risk of macrosomia as well as a greater birth length in infants among tall height mothers or short height mothers, but not mothers of average height. There was no significant interaction between gestational diabetes and maternal height on birth weight, birth length, head circumference, or macrosomia in either cohort studied (interaction p > 0.05). The internal and external weighted height GRS were strongly associated with maternal height. Variants that predicted adult height were inversely associated with fasting, one-hour, and two-hour OGTT glucose levels. We did not find statistically significant associations between the externally weighted or internally weighted height GRS and gestational diabetes. Conclusion: We found that maternal height was associated with infant BW, BL, and HC, and short women, compared to average-height women, were more likely to deliver babies with lower BW and BL. Our findings indicate that the association of maternal height on infant BW varies by infant sex and the association of GDM on infant BL and macrosomia may vary by maternal height. Our results showed that maternal height is causally related with blood glucose levels. Future studies should include replicating the findings from this study and explore the effect of maternal height on other maternal health outcomes and long-term health in the post-natal period. Such studies can provide additional evidence for risk stratification based on maternal height to improve infant growth and pregnancy outcomes across generations.
- Epidemiology