Effects of prenatal small-quantity lipid-based nutrient supplements on pregnancy, birth and infant outcomes: A systematic review and meta-analysis of individual participant data from randomized controlled trials in low- and middle-income countries

Dewey, Kathryn G.; Wessells, K. Ryan; Arnold, Charles D.; Adu-Afarwuah, Seth; Adu-Afarwuah, Seth; Arnold, Benjamin F.; et al. 2024. Effects of prenatal small-quantity lipid-based nutrient supplements on pregnancy, birth and infant outcomes: A systematic review and meta-analysis of individual participant data from randomized controlled trials in low- and middle-income countries. American Journal of Clinical Nutrition 120(4): 814-835. https://doi.org/10.1016/j.ajcnut.2024.08.008

Background Undernutrition during pregnancy increases the risk of giving birth to a small vulnerable newborn. Small-quantity lipid-based nutrient supplements (SQ-LNS) contain both macro- and micronutrients and can help prevent multiple nutritional deficiencies.

Objectives We examined effects of SQ-LNS provided during pregnancy, compared to a) iron and folic acid or standard of care (IFA/SOC) or b) multiple micronutrient supplements (MMS), and identified characteristics that modified the estimates of effects of SQ-LNS on birth outcomes.

Methods We conducted a 2-stage meta-analysis of individual participant data from 4 randomized controlled trials of SQ-LNS provided during pregnancy (n = 5,273). We generated study-specific and subgroup estimates of SQ-LNS compared with IFA/SOC or MMS and pooled the estimates. In sensitivity analyses, we examined whether results differed depending on methods for gestational age dating, birth anthropometry, or study design.

Results SQ-LNS (vs IFA/SOC) increased birth weight (mean difference: +49g; 95% CI: 26, 71g) and all birth anthropometric z-scores (+0.10-0.13 SD); it reduced risk of low birthweight by 11%, newborn stunting by 17%, newborn wasting by 11%, and small head size by 15%. Only 2 trials compared SQ-LNS and MMS; p-values for birth outcomes were >0.10 except for head circumference (e.g., z-score for gestational age +0.11; 95% CI: -0.01, 0.23). Effect estimates for SQ-LNS vs IFA/SOC were greater among female infants and, for certain outcomes, among mothers with body mass index < 20 kg/m2, inflammation, malaria, or household food insecurity. Effect estimates for SQ-LNS vs MMS were greater for certain outcomes among female infants, first-born infants, and mothers < 25 y.

Conclusions SQ-LNS had positive impacts on multiple outcomes compared to IFA/SOC, but further research directly comparing SQ-LNS and MMS is needed. Targeting SQ-LNS to vulnerable subgroups may be worth considering. Analysis registered at www.crd.york.ac.uk/PROSPERO (CRD42021283391).