Theses Doctoral

Preconception nutrition interventions and intrauterine growth: Exploring mechanism and identifying high-risk groups

Aziz, Sumera

Impaired intrauterine growth, inability of the fetus to achieve the required growth potential, contributes to a higher burden of neonatal morbidity and mortality. Intrauterine growth is an inferred process and small for gestational age is considered a rough estimate of impaired intrauterine growth that requires reliable gestational age data. Due to sparse data and measurement errors in gestational age, researchers rely on markers such as birth weight, birth length, and birth head circumference to infer fetal growth.

While the etiology of impaired fetal growth is multifactorial, maternal anemia and undernutrition contribute substantially to impaired fetal growth and are prioritized in 2030 global nutrition goals by the World Health Assembly. Nutrition supplements such as lipid-based nutrient supplements, multiple micronutrients, and iron or folic acid during pregnancy are presumed to improve maternal anemia. But the effect of the supplements during pregnancy on fetal growth appears to be small to negligible, which has shifted the researcher’s focus to the pre-conception period. However, prior reviews on the preconception period have either synthesized the evidence from observational studies, or have explored outcomes such as congenital anomalies (e.g., neural tube defects), neurodevelopment disorders, or only birth weight. Hence the evidence from existing randomized controlled trials (RCTs) evaluating the effect of preconception nutrition supplements on maternal anemia and all markers of fetal growth including birth weight, birth length, and birth head circumference has not been systematically summarized and synthesized.

We bridged this knowledge gap in the current dissertation (Aim 1). The Women First (WF) Preconception Nutrition Trial found that lipid-based nutrient supplementation started preconception or during pregnancy conferred greater benefits for birth weight and birth length among mothers who were anemic (Hemoglobin (Hb) < 12 g/dL) than among mothers who were non-anemic (Hb ≥ 12 g/dL) pre-pregnancy. However, by dichotomizing women into anemic and non-anemic women, we may miss high-risk women with specific Hb levels who may obtain greater benefits for intrauterine growth associated with the supplements. Assessing the effect of preconception nutrient supplements across a range of pre-pregnancy Hb levels will help us identify women with specific Hb cut-offs who may have the greatest potential to respond to the supplements (Aim 2). Additionally, there are major gaps in understanding the mechanisms of how nutrient supplements, consumed either before or during pregnancy, improve intrauterine growth. A better understanding of the underlying mechanisms would allow for fine-tuning of nutrition interventions for greater efficacy. Here, we examined whether Hb during pregnancy could be a potential mechanism through which nutrition supplements improve intrauterine growth (Aim 3).

For Aim 1, we undertook a systematic review and meta-analysis of the RCTs evaluating the effect of preconception nutrition supplements on maternal hemoglobin and markers of intrauterine growth including birth weight, birth length, birth head circumference, and small for gestational age. Additionally, we examined preterm birth as an important perinatal outcome. We searched electronic databases including PubMed, Web of Science, Embase, CINAHL, and Cochrane Central. We computed pooled mean differences and risk ratios (RR) with 95% confidence intervals (CIs) using random-effect models. We employed I2 and Cochran’s Q test statistics to assess heterogeneity. We used the GRADE (grading of recommendations, assessment, development, and evaluations) tool to assess the quality of evidence.

For Aim 2 and Aim 3, we leveraged the existing data from a large multi-country Women First (WF) Preconception Nutrition Trial conducted in Pakistan, India, Guatemala, and the Democratic Republic of Congo. Women in the WF trial were randomized to consume a lipid-based nutrient supplement (LNS) at least three months before and during pregnancy (Arm 1- preconception), only during pregnancy (Arm 2- during), or not at all (Arm 3 - control). The outcome was weight, length, and head circumference within 48 hours of birth expressed as Z-scores. For Aim 2, we analyzed the WF trial data on 2443 women-newborn dyads. For each site, we computed adjusted mean differences in these Z-scores between the randomized arms across six pre-pregnancy Hb categories (8-8.9, 9-9.9, 10-10.9, 11-11.9, 12-12.9, and ≥13g/dL) based on Hb distributions. We pooled site-specific effect measures using meta-analysis. For Aim 3, hemoglobin measured at 12 (n=2075) and 32 weeks of gestation (n=2157) was a mediator. We employed causal mediation analysis under a counterfactual approach to estimate direct and indirect effects.

For Aim 1, we identified 20 eligible RCTs (n=27,659 women). Preconception nutrition supplements (iron and folic acid, multiple micronutrients, and a lipid-based nutrient supplement) increased maternal hemoglobin by 0.30g/dL ((0.03, 0.57); I2=79%). However, we did not find a significant effect of the supplements on birth weight (12.47gm ((-33.14, 58.08); I2=58%)), birth length (0.15cm (-0.26, 0.56); I2=68%; n=5), birth head circumference (-0.23cm (-0.88, 0.43); I2=84%), small for gestational age (RR: 0.91 (0.80,1.04); I2=31%), or preterm birth (RR: 0.93 (0.69,1.25); I2=57%). Overall, the quality of evidence was assessed as moderate and very low for maternal hemoglobin and three markers of intrauterine growth including birth weight, birth length, and birth head circumference, respectively.

In Aim 2, we found that the effect of LNS on birth weight, length, and head circumference varied by pre-pregnancy Hb categories. Pooled mean differences in the Z-scores for birth length (0.60 (0.03, 1.23)), birth weight (0.50, (0.11, 0.89)), and birth head circumference ((0.26, (0.02, 0.51)) were greatest for Arm 1-preconception vs. Arm 3-control women with Hb 9-9.9g/dL. Women with Hb 10-10.9g/dL also benefited from preconception LNS. However, compared to controls, the effects of preconception LNS on birth weight, birth length, and birth head circumference attenuated for women with Hb 10-10.9g/dL. Compared to Arm 3-control, LNS during pregnancy (Arm 2) improved birth length, birth weight, and birth head circumference for women with Hb 8-10.9g/dL. Preconception LNS (Arm1) vs. LNS during pregnancy (Arm 2) improved the three markers of intrauterine growth only for women with Hb 9-9.9g/dL. Women with nearly normal (11-11.9 g/dL) and normal Hb (≥12g/dL) did not appreciably benefit from LNS, offered before and or during pregnancy.

For Aim 3, Hb at 12 or 32 weeks of gestation did not mediate the relationship between the LNS and intrauterine growth. Indirect effects of preconception LNS (Arm 1) vs. Arm 3, mediated by Hb at 12 weeks, were 0.02 (-0.02, 0.01), 0.01 (-0.01, 0.02), and 0.01 (-0.01, 0.02) for length, weight, and head circumference Z-scores, respectively. The corresponding direct effects, not mediated by Hb, were 0.18 (0.09, 0.33), 0.12 (0.03, 0.23), and 0.06 (-0.03, 0.20), respectively. Site-specific and gestational age-adjusted data analyses both at 12 and 32 weeks of gestation confirmed the findings of negligible mediation by Hb during pregnancy.

All types of preconception nutrition supplements studied to date appear to reduce maternal anemia. However, it is uncertain whether there are beneficial effects of preconception nutrition supplements on markers of intrauterine growth. Low quality of evidence from the RCTs examining the markers of intrauterine growth warrants future well-designed RCTs to produce solid scientific data, particularly on the benefits of a more comprehensive package of preconception nutrition supplements that include both macro- and micronutrients. The findings from the WF trial suggest that the benefits of preconception LNS on fetal growth are mainly confined to women with pre-pregnancy Hb 9-9.9g/dL. Compared to controls, women with Hb 10-10.9g/dL also benefited from preconception LNS, albeit the magnitude of effect on three markers of intrauterine growth was modest for women with Hb 10-10.9g/dL. However, LNS, started during pregnancy, appeared to improve markers of intrauterine growth for women with a wider range of Hb 8-10.9g/dL. Women with Hb ≥ 11g/dL did not benefit from LNS started either pre-conception or during pregnancy. These findings suggest that prioritizing women with specific pre-pregnancy Hb categories for targeted nutrition interventions may be advisable. This would target limited resources most efficiently in LMICs. Lastly, negligible mediation by Hb during pregnancy suggests that alternative pathways that potentially mediate the relationship between LNS and intrauterine growth need to be investigated.

Files

  • thumnail for Aziz_columbia_0054D_18338.pdf Aziz_columbia_0054D_18338.pdf application/pdf 4.75 MB Download File

More About This Work

Academic Units
Epidemiology
Thesis Advisors
Kuhn, Louise
Degree
Ph.D., Columbia University
Published Here
March 20, 2024