Nutritional and Physiological Factors in Relation to Birth Outcomes and the Role of a Lipid-Based Nutrient Supplement (LNS) in Pregnant Ghanaian Women

Preterm birth has become the leading cause of mortality in children under five years of age. This dissertation provides new insights from a nutritional perspective of two risk factors for preterm birth, high circulating maternal cortisol and abnormally low cholesterol. In pregnant Ghanaian women, we examined: 1) whether circulating cortisol at 36 wk gestation is reduced in women receiving multiple micronutrients (MMN) or a lipid-based nutrient supplement (LNS); 2) the direct and indirect associations of maternal cortisol, inflammation, hemoglobin, iron status and BMI with duration of gestation and infant birth weight-for-age (WAZ), length-for-age (LAZ), and head circumference z-score (HCZ); and 3) whether maternal blood lipid concentrations, at ≤ 20 wk or 36 wk gestation, are associated with duration of gestation.;The first study examined the effects of LNS supplementation during pregnancy on maternal cortisol at 36 wk gestation. This study was a sub-study within the International Lipid-Based Nutrient Supplements (iLiNS) -DYAD trial in Ghana, a randomized intervention trial on the effects of LNS on maternal and child outcomes, and was limited to the pregnancy portion of the trial. Enrolled women were assigned to receive daily LNS, MMN, or an iron and folic acid supplement (IFA). We measured morning cortisol in saliva at baseline (≤ 20 wk gestation), 28 wk, and 36 wk gestation. There were no significant differences in mean cortisol concentration at 36 wk gestation among the three supplementation groups (LNS = 7.15 +/- 0.2, MMN = 7.34 +/- 0.2, IFA = 7.44 +/- 0.2 nmol/L, p = 0.54). However, there was a significant interaction with age, and we found that in the younger age group (≤ 25 y), women receiving LNS had lower mean cortisol at 36 wk gestation than women receiving MMN or IFA (LNS = 6.63 +/- 0.04, MMN = 7.83 +/- 0.03, IFA = 7.69 +/- 0.04 nmol/L; p = 0.038 for LNS vs. MMN). This effect was not evident in the older age groups (26-34 y and ≥ 35 y).;The second study looked at the direct and indirect associations of maternal cortisol, inflammation, hemoglobin, iron status and BMI with birth outcomes in all women enrolled in iLiNS-DYAD in Ghana. This was analyzed as a cohort study and controlled for supplementation group assignment. In general, higher cortisol during pregnancy was associated with a shorter duration of gestation and higher risk of preterm birth, but when analyzed by infant gender, this association was evident only among women carrying male fetuses. Higher cortisol was also associated with a smaller birth size, but this was due to the association of higher cortisol with a shorter duration of gestation. Unexpectedly, we did not find evidence of cortisol acting as a mediator for inflammation, maternal BMI, iron status, or hemoglobin in relation to birth outcomes. Higher BMI was associated with lower concentrations of morning cortisol and higher concentrations of CRP, hemoglobin, and iron deficiency biomarkers. While higher hemoglobin in early pregnancy was associated with a longer duration of gestation and a decreased risk of low birth length, the associations of iron status with birth outcomes were generally not significant, and high ZPP, an indicator of iron deficiency, was actually associated with higher WAZ. Our results highlight the complexity of the relationship between birth outcomes and maternal hemoglobin or iron status.;The final study examined the association between blood lipid concentrations (high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), total cholesterol, and triglycerides) and duration of gestation in a nested cohort within the iLiNS-DYAD trial in Ghana. None of the blood lipid concentrations measured at baseline was associated with duration of gestation. At 36 wk, higher HDL-C (mmol/L) was associated with a longer duration of gestation. Women below the 10th percentile of HDL-C at 36 wk gestation had a gestational period that was 5.9 days shorter than the referent group (10th-90th percentile) and 8.5 days shorter than women above the 90th percentile. Among women below the 10th percentile of LDL-C at 36 wk gestation, duration of gestation was 4.9 days longer when compared with the 10th-90th percentile group. Total cholesterol and triglycerides were not significantly associated with duration of gestation. While a variety of mechanisms could possibly explain these results, we propose and discuss the mechanism centered around vasculopathy of the placenta.