Early feeding type, nutrient intake, and longitudinal growth of preterm infants at a Tertiary Neonatal Unitin Kuala Lumpur, Malaysia: prospective COHORT study

Abstract

With improved neonatal care, preterm infant survival has increased; however, postnatal growth failure and nutritional deficits remain common. Uncertainty persists regarding whether predominantly fortified breast milk (PBM) or predominantly formula milk (PFM) feeding during neonatal intensive care unit (NICU) admission is associated with different nutrient delivery, growth trajectories, or body composition outcomes, particularly in Malaysian NICUs where local longitudinal data remain limited. This study examined nutritional intake during NICU admission, longitudinal growth trajectories to 6 months corrected age (CA), and body composition and bone mineralisation outcomes among preterm infants receiving PBM or PFM at a tertiary neonatal unit in Kuala Lumpur, Malaysia. A prospective longitudinal cohort study was conducted in the NICU of Hospital Canselor Tuanku Muhriz, Kuala Lumpur, between September 2024 and November 2025. Preterm infants born at ≤34 weeks’ gestation and/or ≤1800 g, without major congenital anomalies or medical conditions known to affect growth, and who remained in the NICU for at least 14 days were enrolled. Daily nutrient intake and feeding milestones were extracted from clinical records. Weightfor- age z-score (WAZ), length-for-age z-score (LAZ), and head circumference-for-age z-score (HCZ) were assessed at birth, discharge, and at 1, 3, and 6 months CA. Infants were classified as PBM-fed or PFM-fed according to the predominant source of enteral intake during NICU admission, defined as the milk type contributing the largest proportion of cumulative enteral intake. This classification was fixed at discharge regardless of subsequent post-discharge feeding transitions. Whole-body dual-energy X-ray absorptiometry (DXA) was performed at 6–9 months CA in an available subsample to assess bone mineral content, bone mineral density, lean mass, and body fat percentage. Seventy-six infants were classified as PBM-fed (n = 34) or PFM-fed (n = 42). PBM infants had lower gestational age (GA) (median 32 vs 33 weeks; p = 0.01), higher rates of resuscitation (58.8% vs 35.7%; p = 0.04), higher rates of respiratory distress syndrome (85.3% vs 61.9%; p = 0.02), and longer length of stay (median 30 vs 21 days; p = 0.01). Despite these baseline differences, mean nutrient delivery during Weeks 1–3 was comparable between groups for energy (101.8 vs 102.4 kcal/kg/day; p = 0.87) and protein (2.78 vs 2.87 g/kg/day; p = 0.51). Both groups remained below the lower ESPGHAN target ranges for energy and protein intake, achieving 88.5–89.0% of the minimum energy target and 79.4–82.0% of the minimum protein target. At discharge, ΔWAZ did not differ between PBM and PFM infants (−1.27 ± 0.35 vs −1.13 ± 0.55; p = 0.21), and postnatal growth failure occurred in 47.1% and 45.2%, respectively (p = 0.87). Feeding group and early nutrient intake were not independently associated with ΔWAZ at discharge in adjusted models. During post-discharge followup (n = 57), WAZ declined progressively from 1 to 6 months CA in both groups. At the first post-discharge clinic visit, 28.1% of infants (16/57) remained exclusively on breast milk, 33.3% (19/57) remained exclusively on formula milk, 8.8% (5/57) had transitioned from formula milk to breast milk, 10.5% (6/57) had transitioned from breast milk to formula milk, and 19.3% (11/57) were receiving mixed feeds. In adjusted models, PFM infants had higher WAZ across follow-up (estimate = 1.00, 95% CI 0.15 to 1.85; p = 0.03), although this finding was not consistent across sensitivity analyses and may reflect residual confounding by baseline clinical differences. Birth WAZ was the strongest predictor of post-discharge WAZ trajectory in all models (p = 0.003 to p < 0.001), suggesting that birth growth status was more strongly associated with post-discharge growth than NICU feeding type. DXA was performed at a mean CA of 6.8 ± 1.0 months in 12 infants. Mean bone mineral content was 169.2 ± 29.2 g and mean bone mineral density was 0.25 ± 0.02 g/cm², comparable to published preterm cohorts and suggesting generally preserved bone mineralisation within this small cohort. Mean lean mass was 4887 ± 786 g, and mean body fat percentage was 39.4 ± 6.7%, appearing higher than values reported in healthy term-born infants at comparable age. This suggests a relatively adipose-predominant body composition profile that warrants further post-discharge monitoring. Under a standardised unit protocol, PBM and PFM infants achieved broadly comparable early macronutrient delivery and similar growth outcomes at discharge despite baseline clinical differences between groups. These findings suggest that infants receiving PBM can achieve growth outcomes comparable to those receiving PFM when nutrient delivery is adequately supported through fortification and structured feeding protocols. Post-discharge growth trajectory appeared to be more closely associated with baseline growth status and gestational maturity, reflected by birth WAZ and GA, than with predominant NICU feeding type or achieved nutrient intake. However, the progressive decline in WAZ through 6 months CA in both groups indicates an ongoing post-discharge growth concern and supports the need for structured nutritional follow-up beyond hospital discharge. This study also contributes contemporary local DXA-derived data on body composition and bone mineralisation among preterm infants at 6–9 months CA. The exploratory DXA findings suggest that growth monitoring in preterm infants should consider body composition and bone health, not weight gain alone. Overall, these findings provide preliminary Malaysian data to guide future adequately powered studies on postdischarge nutrition, growth quality, and longer-term metabolic outcomes in this population.