Nutritional Management of Low Birth Weight and Preterm Infants in Low- and Low Middle-Income Countries Open Access

Low birth weight (LBW) in infants may be due to preterm delivery, intrauterine growth restriction (IUGR), small for gestational age, or a combination of these factors [1]. Globally, more than 20 million infants are born LBW every year [2], and more than 13 million infants were born preterm in the year 2020 [3]; a significant majority being in low-income countries (LIC) or lower middle-income countries (LMICs) [2, 4] as 75% of LBW neonates were born in South Asia and Sub-Saharan Africa [1]. Although there have been advancements in quality of care and increased facility births, which have improved survival rates in recent years, yet they still remain at enhanced risks for morbidity, mortality, impaired growth, cognitive deficits, and chronic diseases later in life [5‒8].

LBW and preterm neonates face delays in feed initiation and experience feeding difficulties, particularly challenges to maintain exclusive breastfeeding (EBF) [9]. Optimal nutrition, achieved through better feeding practices, plays a crucial role in newborn survival, preventing adverse outcomes like sepsis, necrotizing enterocolitis (NEC), and shortening the duration of hospital stay [10‒13]. In 1960s, the United Kingdom improved feeding practices and these were amongst the initial interventions for vulnerable babies leading to reduced hospital mortality even prior to the availability of intensive care [14]. Studies show that early optimal nutrition influences the development of multiple organs [15], immune function [16], along with long-term effects on neurodevelopment [17, 18] and chronic conditions like diabetes and cardiovascular disease [19].

According to Convention on the Rights of the Child (CRC) every child is entitled to healthy food and adequate nutrition [20]. Numerous efforts to improve child feeding practices have been made in the past decades; some of these include “The Baby-Friendly Hospital Initiative” (BFHI) [21], “The Innocenti Declaration” [22], “International Code of Marketing of Breastmilk substitutes” [23], “Millennium Developmental Goals” (MDGs) [24], “Sustainable Developmental Goals” (SDGs) [25] and “Global Nutrition Targets 2025” (5). These initiatives are principally driven by the World Health Organization (WHO) and United Nations Children’s Funds (UNICEF) with the key objectives being protection, advocacy, and reinforcement of breastfeeding and optimal child nutrition.

Despite internationally unanimous guidelines published by the WHO in 2011 [26] for optimal feeding of vulnerable infants, consensus and delivery on their nutritional needs remains limited to high-income countries (HICs). Globally, there is a lack of comprehensive understanding regarding early feeding practices and the most effective and plausible feeding strategies for vulnerable infants, including LBW and preterm in LIC/LMIC settings. Therefore, we conducted this review to provide a comprehensive update describing the most recently available evidence for effective interventions to improve the feeding practices of preterm and/or LBW newborn infants, and the effects of supplementing with different micronutrients in LIC/LMIC settings.

The Lancet Every Newborn series, published in 2014 [27], was a series of papers that identified scalable interventions with specific targets to counter neonatal mortality and morbidity in LMICs by the year 2030. We updated evidence from the series by identifying new relevant systematic reviews; followed by extraction of LIC and LMIC data and creating new forest plots with revised estimates where applicable. Detailed methodology and rationale have been described elsewhere [28].

We conducted a search to identify the recently published systematic reviews that focused on optimal feeding practices and micronutrient supplementation interventions in preterm and LBW infants. The specific interventions were finalized after reviewing the previous Lancet Series and consultation with the technical advisory group. The aim was to update existing systematic reviews if the evidence was outdated, to use existing systematic reviews if the evidence was up-to-date, and to conduct de novo reviews if no systematic reviews were found relevant to the intervention of interest.

The eligible population included preterm (<37 weeks) and/or LBW (<2.5 kg) infants. However, when we did not find specific reviews on preterm and LBW infants, we synthesized the evidence available regardless of birthweight or gestational age. Relevant systematic reviews were identified and (1) updated if they were published prior 2020 and did not have an established intervention, (2) used as it is, if published after 2020 or if it had an established intervention in case it was published prior 2020. We used the same search strategy as described in the original review; where search strategy was not available, we used the given MeSH terms and developed a new strategy and which was run on the databases used in the original review including PubMed, Cochrane Library, CINAHL and Embase.

Once the most recent systematic review was identified, we extracted all the relevant information and outcomes with estimates in duplicate. To conduct a subgroup analysis for all the outcomes of interest for LIC/LMIC-specific estimates, we cross-referenced the countries in which the primary studies were conducted with the World Bank list of countries [29]. Based on the year and the country where the study was conducted, we classified the study as conducted in a high-income country (HIC), upper-middle-income country, LMIC, or LIC. The reanalysis process was done by two authors, and we followed the same statistical procedure described in the original review to get the pooled effect estimates.

For reviews that we updated, two reviewers conducted the search. Following the inclusion criteria, we included all eligible studies and performed data extraction. We updated all the outcomes where new studies reported those outcomes. All the new studies were entered in the Review Manager 5.4.4 and the existing forest plots were updated and new estimates reported as risk ratio (RR) with 95% confidence interval (CI) for dichotomous and mean difference (MD) with 95% CI for continuous outcomes.

The existing systematic reviews include: early versus late initiation of enteral feeding [30], responsive versus scheduled feeding [31], mother’s own milk versus formula milk [32], duration of EBF [33], impact of infant and young child feeding (IYCF) nutrition interventions on breastfeeding practices, growth, and mortality [34], fast versus slow feed advancement [35], enteral calcium or phosphorous supplementation [36], low-dose vitamin A supplementation [37], enteral vitamin D supplementation [38], enteral iron supplementation [39], enteral multiple micronutrient (MMN) supplementation [40], and zinc supplementation for treatment and prevention of sepsis [41]. The systematic reviews that were updated include: formula versus maternal breast milk [42], formula versus donor breast milk [43], multi-nutrient fortification of human milk [44] and nutrient-enriched formula versus standard formula [45].

A total of 15 reviews focusing on high-risk infant feeding practices and micronutrient supplementation were included and these were sub-grouped under three broad categories as described below. Tables 1-3 provide the effect estimates of various interventions and Table 4 provides the details of country classifications. Additional information can be found in online supplementary File 1 (for all online suppl. material, see https://doi.org/10.1159/000542154).

• Timing of initiation of enteral feeding

• IYCF nutrition interventions

• Duration of EBF

• Formula and maternal breast milk

• Donor breast milk

• Types of feeding (responsive and scheduled)

• Rate of feeding advancement

• Multi-nutrient fortification of human milk

• Nutrient-enriched formula

• Iron

• Zinc

• Calcium and/or phosphorous

• Vitamin A

• Vitamin D

• MMN supplements

A review by Chitale et al. [30] assessed the effectiveness of early enteral feeding (<72 h), compared with delayed enteral feeding (≥72 h) in preterm and LBW infants. There were 14 studies included in the meta-analysis with one study from LMIC (India); the overall risk of bias was moderate to low risk. We found that early enteral feeding significantly improved the neonatal mortality (RR: 0.69, 95% CI: 0.48 to 0.99, n = 1,292, 12 studies) by discharge or 28 days of life in overall settings. The effect was nonsignificant in LMIC setting (RR: 0.60, 95% CI: 0.16 to 2.30, n = 62, one study). The subgroup analysis by day in overall and LMIC settings showed that the effect was nonsignificant on day 1 (<24 h), day 2 (<48 h), and day 3 (<72 h). In both overall and LMIC settings, the effect for NEC and sepsis was nonsignificant.

A recent systematic review by Lassi et al. [34] assessed the effectiveness of IYCF nutrition interventions on breastfeeding practices, growth and mortality. It also assessed the effectiveness of interventions in context of food security. According to 1996 World Food Summit, food security is achieved when every individual has physical and economic access to safe and nutritious food for leading a healthy and active life. The review had mothers or nursing mothers of children under 2 years of age living in LMICs of reproductive age irrespective or birthweight or gestational age. A total of 75 studies were included in the meta-analysis. Breastfeeding promotion interventions had a significant effect on early initiation of breastfeeding (RR: 1.20, 95% CI: 1.12 to 1.28, n = 84,092, 14 studies) causing a 20% increase in its prevalence. The subgroup analysis by setting showed that both facility-based (RR: 1.18, 95% CI: 1.03 to 1.36, 8 studies) and community-based studies (RR: 1.17, 95% CI: 1.07 to 1.28, 5 studies) had a significant effect. Postnatal and prenatal/postnatal interventions had a significant effect; however, prenatal (only) interventions were nonsignificant (RR: 1.06, 95% CI: 0.98 to 1.14, 3 studies). Interventions lasting more than 6 months duration had a significant effect (RR: 1.28, 95% CI: 1.18 to 1.38, 7 studies). The subgroup analysis by moderators of intervention showed that community health workers/volunteers as moderators had significant effects. Breastfeeding interventions had a significant effect on EBF at three (RR: 2.02, 95% CI: 1.88–2.17, n = 4,063, 6 studies) and 6 months of age (RR: 1.53, 95% CI: 1.47 to 1.58, n = 13,926, 19 studies) whereas there was a nonsignificant effect on neonatal mortality, infant mortality, weight-for-age z-score (WAZ) and weight-for-height z-score (WHZ). However, the rate of diarrheal disease decreased by 24% (RR: 0.76, 95% CI: 0.67 to 0.85, n = 4,585, 8 studies).

Complementary feeding education interventions had a significant effect on WAZ in both food secure (MD: 0.41, 95% CI: 0.07 to 0.75, n = 1,562, 4 studies) and food insecure (MD: 0.47, 95% CI: 0.35 to 0.59, n = 572, 1 study) settings. Similarly, the effect was significant for height-for-age z-score (HAZ) in both food secure and food insecure settings. Complementary feeding education interventions did not have a significant effect on WHZ in food secure settings, but the effect was significant in food insecure settings (MD: 0.50, 95% CI: 0.35 to 0.65, n = 572, 1 study). The effect of the intervention on stunting and wasting in both food secure and food insecure settings was nonsignificant.

Supplementary feeding interventions did not have a significant effect on stunting or wasting. The interventions, however, had a significant impact on WHZ (MD: 0.15, 95% CI: 0.08 to 0.22, n = 3,664, 6 studies) and infant mortality (RR: 0.61, 95% CI: 0.38 to 0.97, n = 4,757, 2 studies).

The most recent review by Yang et al. [33] found only two studies, both in LMICs (Honduras, India), for optimal duration of EBF in preterm and LBW infants (less than 6 months vs. 6 months). The overall risk of bias showed some concerns. The studies could not be pooled because the data were measured and reported in a different manner. The effect on days with diarrhea (MD: −2.6, 95% CI: −5.2 to 0.0, n = 119, 1 study), fever (MD: −0.7, 95% CI: −3.4 to 2.0, n = 119, 1 study), cough (MD: 3.1, 95% CI: −4.6 to 10.8, n = 119, 1 study), congestion (MD: 3.6, 95% CI: −3.5 to 10.7, n = 119, one study), and nasal discharge (MD: 4.2, 95% CI: −1.2 to 9.6, n = 119, 1 study) was nonsignificant. Growth was similar between the two EBF groups for the following outcomes: WAZ at corrected age 12 months, absolute weight gain (gm) from 16 to 26 weeks of age and linear growth gain (cm) from 16 to 26 weeks of age. There were no trials reported on mortality.

A review by Brown et al. [42] assessing the effectiveness of formula milk compared to maternal breast milk, found no eligible randomized controlled trials (RCTs). We carried out an update; however, no new trials were reported (online suppl. Fig. 1).

A review by Strobel et al. [32] assessed the effectiveness of formula milk compared with mother’s own milk for feeding preterm or LBW infants. This review included RCTs, cohort, cross-sectional as well as case-control studies. A total of 36 studies were included in the meta-analysis; four studies were from LMICs (Ghana, Nepal, India) and most of the studies had high risk of bias. We found that formula milk did not have a significant effect on neonatal mortality in overall settings at the latest follow up (mean 116 days) (odds ratio (OR) 1.26, 95% CI: 0.91 to 1.76, n = 9,673, 5 studies) whereas studies from LMICs showed that formula milk increased the risk (OR: 1.51, 95% CI: 1.02 to 2.22, n = 6,711, 2 studies). The overall effect for sepsis was nonsignificant whereas the effect for NEC was significant (OR: 2.99, 95% CI: 1.75 to 5.11, n = 3,013, 15 studies) at the latest follow up. There was no evidence for an effect on sepsis or NEC in LMICs.

We carried out update for the review by Quigley et al. [43] that compared formula milk with donor breast milk for preterm and LBW infants but no new trials were reported (online suppl. Fig. 2). The original review identified 12 trials, all from HICs, of which most studies had unclear or low risk of bias. There were no studies from LMICs. We found that infants who were fed formula milk regained birthweight more rapidly (MD: −3.08, 95% CI: −4.38 to −1.77, n = 236, 3 studies). The overall rate of weight gain (gm/kg/day) was also faster in the formula fed group (MD: 2.51, 95% CI: 1.93 to 3.08, n = 1,028, 9 studies), with higher rate of increase in linear growth (crown-heel length) (MD: 1.21, 95% CI: 0.77 to 1.65, n = 820, 8 studies) and occipitofrontal head circumference (MD: 0.85, 95% CI: 0.47 to 1.23, n = 894, 8 studies, grade = moderate). Formula fed infants also showed a higher risk for NEC (RR: 1.87, 95% CI: 1.23 to 2.85, n = 1,675, 9 studies). The effect was nonsignificant for all-cause mortality.

The most recent review by Talej et al. [31] assessed the effectiveness of the type of feeding approaches for preterm and LBW infants. Eight studies were included in the meta-analysis; there were no studies from LMICs. Most studies had high risk of bias related to the randomization process. The overall weight (in gm) at discharge (MD: −22.21, 95% CI: −130.63 to 86.21, n = 183, three studies) was nonsignificant. The overall weight gain (g/day) showed that infants with scheduled feeding (MD: −2.80, 95% CI: −3.39 to −2.22, n = 213, 4 studies) had greater weight gain. The overall duration of hospitalization days till discharge between the two groups was nonsignificant (MD: −1.42, 95% CI: −5.43 to 2.59, n = 342, 5 studies). However, the effect for time to establish full oral feeds (after trial entry) was significant in the scheduled feeding group (MD: −5.54, 95% CI: −6.87 to −4.20, n = 167, 2 studies). There were no trials that reported on neonatal mortality, morbidity, length, head circumference, or neurodevelopment.

A recent review by Yang et al. [35] assessed the effects of fast feed advancement (≥30 mL/kg per day) compared with slow feed advancement (<30 mL/kg per day). There were 12 trials included in the meta-analysis; five studies were from LMICs (India, Bangladesh). The overall risk of bias had some concerns as none of the trials had blinding. The effect was nonsignificant for sepsis in overall settings whereas there was a 36% decrease in the risk of sepsis in LMICs in infants who were fed through fast advancement (RR: 0.64, 95% CI: 0.43–0.96, n = 367, 4 studies). The overall (MD: −3.69, 95% CI: −4.44 to −2.95, n = 933, seven studies) and LMIC (MD: −4.39, 95% CI: −5.09 to −3.69, n = 453, 4 studies) evidence showed that infants who were fed through fast advancement regained birthweight more quickly. The overall and LMIC effect was nonsignificant for all-cause mortality, NEC, and feed intolerance.

We carried out an update of a recent review by Brown et al. [44] that compared fortified breast milk with unfortified breast milk. We did not find any new trials (online suppl. Fig. 3). There were 18 trials included in the original review of which eight studies were from LMICs (Egypt, India, South Africa, Brazil) and most studies had high or unclear risk of bias. The overall (MD: 1.76, 95% CI: 1.30 to 2.22, n = 951, 14 studies) and LMIC (MD: 1.88, 95% CI: 1.23 to 2.53, n = 492, 5 studies) evidence showed that infants who were fed fortified breast milk gained weight more quickly. There was a higher rate of length gain (cm/week) and head growth (cm/week) in the fortification group in both overall and LMIC settings. The effect for feed intolerance, NEC and length of hospital stay was nonsignificant in both overall and LMIC settings. There were no trials from LMICs for neurodevelopmental outcomes.

A review by Walsh et al. [45] compared nutrient-enriched formula with standard formula for growth and development of preterm infants. We carried out an update; however, no new trials were reported (online suppl. Fig. 4). There were seven trials included in the meta-analysis; with three studies from LMICs (Thailand, South Africa, Turkey) and four from HICs. The overall risk of bias had some concerns due to insufficient reporting. The review showed that there was no significant difference in the time to regain birthweight in both overall (MD: 95% CI: −1.48, 95% CI: −4.73 to 1.77, n = 74, 3 studies) and LMIC settings (MD: −4.5, 95% CI: −9.22 to 0.22, n = 25, 1 study). Infants who were fed nutrient-enriched formula gained weight more rapidly in both overall (MD: 2.43, 95% CI: 1.60 to 3.26, n = 440, 6 studies) and LMIC settings (MD: 4.15, 95% CI: 0.24 to 8.05, n = 88, 3 studies). Nutrient-enriched formula showed a significant effect on the rate of head circumference gain (mm/week) in both overall (MD: 1.04, 95% CI: 0.18 to 1.89, n = 399, 5 studies) and LMIC settings (MD: 2.25, 95% CI: 0.35 to 4.14, n = 47, 2 studies). There was no significant difference for NEC. There were no LMIC-specific trials reported for all-cause mortality and neurodevelopmental outcomes.

A review by Manapurath et al. [39] assessed the effects of enteral iron supplementation compared with no supplementation in LBW or preterm infants. There were eight trials included in the meta-analysis, of which two studies were from LMIC (India), and most of the studies had high risk of bias. The effect for sepsis in both overall (0.98, 95% CI: 0.56 to 1.73, n = 401, 4 studies) and LMIC setting (RR: 1.23, 95% CI: 0.46 to 3.31, n = 70, 2 studies) was nonsignificant. Similarly, the effect was nonsignificant for NEC in both overall and LMIC-specific estimates. The effect for increase in hemoglobin was significant in both overall (MD: 4.79, 95% CI: 2.90 to 6.69, n = 506, 5 studies) and LMIC setting (MD: 8.44, 95% CI: 1.22 to 15.65, n = 66, 2 studies). The linear growth was increased in the intervention group in overall settings (MD: 0.69, 95% CI: 0.01–1.37, n = 384, 3 studies), but this effect was nonsignificant in LMIC. The overall effect was significant for anemia (RR: 0.25, 95% CI: 0.10 to 0.62, n = 381, two studies), for which there were no LMIC trials. There were no trials found reporting on mortality.

The most recent systematic review by Irfan et al. [41] assessed the effectiveness of zinc supplementation compared to no intervention for treating neonatal and young infant sepsis. The review included all infants regardless of birthweight or gestational age; there were nine trials included in the meta-analysis with six studies reporting therapeutic efficacy of zinc and three studies reporting preventive efficacy. Eight studies were from LMICs, and one study was from HIC (Italy); most studies had low risk of bias. We found that preventive zinc supplementation in infants had a lower risk of mortality in overall settings (RR: 0.28, 95% CI: 0.12–0.67, n = 265, 2 studies); the LMIC-specific evidence showed that the effect was nonsignificant (RR: 0.24, 95% CI: 0.03 to 2.01, n = 72, 1 study). The evidence showed that therapeutic zinc supplementation may significantly reduce treatment failure in infants (RR: 0.61, 95% CI: 0.44 to 0.85, n = 964, 3 studies) in overall settings for which there were no LMIC trials. There was no significant difference between the groups for neonatal bacterial sepsis in both overall and LMICs settings. Furthermore, the effect of zinc on infant mortality in LMICs was nonsignificant.

The most recent review by Kumar et al. [36] investigated the effects of calcium and phosphorous supplementation compared with no supplementation in preterm or LBW infants. There were three trials included in the meta-analysis with no trials from LMIC. Only one study had high risk of bias due to insufficient reporting of the randomization process. There was no significant difference in weight (gm) between the two groups at latest follow-up (MD: 138.50, 95% CI: −82.16 to 359.16, n = 40, 1 study). The effect for length (cm) and head circumference (cm) was nonsignificant. The risk for osteopenia/rickets was reduced by 32% in the supplemented group (RR: 0.68, 95% CI: 0.46 to 0.99, n = 159, 3 studies).

The most recent review by Manapurath et al. [37] assessed the effectiveness of enteral low-dose vitamin A supplementation compared with no supplementation in preterm and LBW infants. There were four trials included in the meta-analysis with only one study from LMIC (India); three studies had low risk of bias and one study had some concerns. The effect for neonatal mortality was nonsignificant in overall settings (RR 0.74, 95% CI: 0.53 to 1.02, n = 800, four studies) but was reduced by 56% in LMIC in the supplemented group (RR 0.44, 95% CI: 0.23 to 0.84, one study). There was no difference between the two groups for either sepsis or feed intolerance. The effect for NEC and Bronchopulmonary dysplasia was nonsignificant in overall settings; and for which there were no LMIC trials.

A review by Kumar et al. [38] assessed the effect of enteral vitamin D supplementation compared with no supplementation in preterm or LBW infants. There were three trials included in the meta-analysis, with one trial from LMIC (India); the studies had low or unclear risk of bias. The overall and LMIC evidence showed that there was no significant difference between the two groups for mortality and any serious morbidity. Vitamin D deficiency was decreased in both overall (RR: 0.58, 95% CI: 0.49 to 0.68, n = 504, 2 studies) and LMIC settings (RR: 0.59, 95% CI: 0.50 to 0.70, 1 study). The effect for WAZ at 6 months was significant in the supplemented group (MD: 0.12, 95% CI: 0.01 to 0.23, n = 1,273, 1 study) for which the study was from LMIC. The effect was nonsignificant at 3–6 years of age. The effect for HAZ in LMIC was significant at 6 months (MD: 0.12, 95% CI: 0.03 to 0.21, n = 1,258, 1 study) and not at 3–6 years of age.

A review by Kumar et al. [40] assessed the effects of MMN supplementation compared to no supplementation in human-milk fed preterm or LBW infants. There were two trials (four reports) included in the meta-analysis with one study from LMIC (Tanzania); two studies had a high risk of bias and two had some concerns. The overall and LMIC evidence showed that there was no significant difference for wasting, stunting, and underweight at the latest follow-up. Furthermore, the effect was nonsignificant for WAZ, WHZ, and HAZ in both overall and LMIC estimates.

We found 15 reviews assessing the effect of feeding practices and micronutrient supplementation interventions in preterm and LBW infants. The evidence suggested that early initiation of enteral feeding reduced neonatal mortality in overall settings but the effect was nonsignificant in LMIC, and this evidence was limited to only one study in India with potential differences based on healthcare and economic contexts. The findings also suggested that breastfeeding interventions increased the prevalence of initiation of breastfeeding as well as EBF at 3 and 6 months of age in LMICs. The consistency of this effect across different settings, like facility and community, depicts that the benefits of such interventions are not limited by the context in which they are delivered. The interventions that extended beyond 6 months had a more significant effect suggesting that continued engagement and support is crucial for maximizing the benefits of breastfeeding programs. The demarcation in the significance of effect between prenatal/postnatal interventions, and prenatal-only interventions, emphasizes the importance of sustained support during and beyond the postpartum period. The considerable impact of community health workers and volunteers as intervention moderators highlights the crucial role they can play in low resource settings, acting as a bridge between healthcare system and the local population, advocating for community-based breastfeeding practices in a culturally sensitive manner, and subsequently improving neonatal outcomes. There was also a reduction in diarrheal diseases; and since diarrhea is the third leading cause of death in children under five, advocacy and promotion of breastfeeding programs may be employed as public health strategy for diarrhea incidence reduction. On the contrary, there was a lack of significant impact on neonatal mortality, WAZ, and WHZ; this suggests that while these interventions may improve breastfeeding practices and some outcomes, they may not solely reduce neonatal mortality and improve anthropometric measures. This lack of impact is likely due to the multifaceted nature of these outcomes which also depend on antenatal and postnatal care as well as socioeconomic conditions. Complementary feeding education interventions improved WAZ and HAZ in both food secure and food insecure settings, and WHZ in food insecure settings, highlighting the effectiveness in addressing nutritional gaps in regions prevalent with food insecurity. When breast milk is not available or insufficient, formula milk or donor milk [46] from a lactating mother could be valuable alternatives and our analysis showed that formula milk did not have a significant impact on neonatal mortality overall but accentuated its risk in LMICs, possibly due to shortage of clean water needed to prepare powdered formula milk [47]. Formula milk, while contributing to weight gain, also increased the risk of NEC in overall settings; this finding underscores the need for healthcare professionals to provide tailored strategies for vulnerable neonates or consider alternative feeding options and additional protective measures to mitigate NEC incidence.

Scheduled feeding led to greater weight gain and a longer period to full oral feedings in infants, indicating that structured feeding regimens may support a more gradual and stable growth; and those fed through fast feed advancement had a reduced risk of sepsis and regained their birthweight quicker indicating that rapid progression of feeds may optimize health outcomes, particularly weight gain.

Fortified breast milk had positive effects on both weight gain and anthropometric outcomes leading to greater length and head growth in both overall and LMIC settings. This highlights the efficacy of breast milk fortification in enhancing growth measures and maybe of clinical significance for LBW infants who require more nutrients than their term counterparts, particularly in developing regions. The findings for nutrient-enriched formula suggest that it was effective in accelerating weight and head circumference growth in both settings with limited effect on birthweight recovery and NEC.

Supplementing breastfed preterm and LBW infants with iron led to an increase in linear growth in overall settings but not in LMICs, indicating a possible variation in intervention impact by setting. It also significantly increased hemoglobin in both settings while reducing anemia in HIC settings; for which there was no LMIC data available. Zinc supplementation reduced mortality significantly in overall settings, but this effect was nonsignificant in LMIC settings. Calcium and/or phosphorous supplementation reduced the risk of osteopenia/rickets, highlighting its crucial role in enhancing bone health. However, it did not markedly affect other growth metrics like length or head circumference. Vitamin A supplementation reduced neonatal mortality in LMICs, while the effect was nonsignificant in overall settings. This discrepancy suggests that vitamin A maybe particularly effective in LMICs, where deficiencies of this vitamin are more prevalent, contributing significantly to neonatal mortality. The effect for NEC and bronchopulmonary dysplasia was nonsignificant, with no LMIC trials, leaving the gap in understanding its effect in LMIC context. Vitamin D supplementation did not significantly affect mortality or serious morbidity; however, it was effective in reducing vitamin D deficiency. It also improved WAZ and HAZ in LMICs at 6 months. MMN supplementation did not significantly affect key growth and nutritional outcomes. The lack of significant effect may be due to a multitude of factors, indicating that while MMN supplementation might help in correcting deficiencies, it might not be sufficient to produce profound changes in these critical outcomes. It is also important to assess dietary requirements by gender as studies show that growth rates and composition differ between boys and girls, leading to varying nutritional needs [48‒51]. Therefore, it is important to conduct an audit of nutritional intakes and growth, not only before but also after implementing any change in nutritional care practice for these high-risk infants, realizing that clinical outcomes are consistently poorer in boys than girls [52‒54]. A minor limitation of our review is the ambiguity in the primary studies regarding infants born with IUGR.

Prematurity and low birthweight carry a high risk of morbidity and mortality for the infant. Early adequate nutritional support of preterm is paramount to hindering adverse outcomes. Our analysis depicts the use of formula milk increased the risk of mortality in LMICs and NEC (overall); therefore, it is crucial to promote breastfeeding practices through advocacy and comprehensive IYCF education. The use of nutrient-enriched formula and breast milk fortification could potentially be crucial toward promoting optimal growth in LMICs where growth restriction and stunting is prevalent. Despite no international unanimity on supplementation of iron, zinc, calcium/phosphorous, vitamin D, in the LBW/preterm infant, the benefits seem to outweigh the risks since our review demonstrates little to no adverse effects deriving from their supplementation, particularly for a breastfed high-risk infant. Vitamin A supplementation was particularly effective in reducing neonatal mortality in LMICs. And while the risk factors for outcomes such as preterm birth, LBW, and IUGR, as well as the benefits of different nutritional interventions like enhanced human breast milk, fortifying breast milk to prevent postnatal growth issues and stunting, and using preterm formulas for better growth, are quite analogous in both low- and HICs, this paper underscores the importance of conducting more randomized trials at a larger scale in LMICs to generate a more robust evidence validating the effectiveness of these interventions.

We thank the Aga Khan University for providing internal resources.

Ethical exemption was taken from the Aga Khan University’s Ethics Review Committee (ERCs) and National Bioethics Committee (NBC). Ethical approval and consent were not required as this study was based on publicly available data.

The authors declare no conflict of interest.

This study is supported by the Bill and Melinda Gates Foundation Grant (#INV-042789). The funders played no role in the design, data collection, data analysis, and reporting of this study. Dr. Sawera Hanif is a Fogarty fellow with funding from list funding from the National Institutes of Health 5D43 TW011625.

M.A., R.Y., S.H., and S.A.B. carried out the literature review extracted data from the reviews, carried out the analysis and interpreted the results. M.A. drafted the paper. J.K.D. and Z.A.B. provided senior supervision for each step, conceptualization and contributed to the critical revision of the manuscript.

Data used in our systematic review are publicly available, as it was sourced from published papers. Further inquiries can be directed to the corresponding author.