High protein intake on later outcomes in preterm children: a systematic review and meta-analysis.

BACKGROUND: Appropriate protein intake is crucial for growth and development in children born preterm. We assessed the effects of high (HP) versus low protein (LP) intake on neurodevelopment, growth, and biochemical anomalies in these children. METHODS: Randomised and quasi-randomised trials providing protein to children born preterm (<37 completed weeks of gestation) were searched following PRISMA guideline in three databases and four registers (PROSPERO registration CRD42022325659). Random-effects model was used for assessing the effects of HP (≥3.5 g/kg/d) vs. LP (<3.5 g/kg/d). RESULTS: Data from forty-four studies (n = 5338) showed HP might slightly reduce the chance of survival without neurodisability at ≥12 months (four studies, 1109 children, relative risk [RR] 0.95 [95% CI 0.90, 1.01]; P = 0.13; low certainty evidence) and might increase risk of cognitive impairment at toddler age (two studies; 436 children; RR 1.36 [0.89, 2.09]; P = 0.16; low certainty evidence). At discharge or 36 weeks, HP intake might result in higher weight and greater head circumference z-scores. HP intake probably increased the risk of hypophosphatemia, hypercalcemia, refeeding syndrome and high blood urea, but reduced risk of hyperglycaemia. CONCLUSIONS: HP intake for children born preterm may be harmful for neonatal metabolism and later neurodisability and has few short-term benefits for growth. IMPACT STATEMENT: Planned high protein intake after birth for infants born preterm might be harmful for survival, neurodisability and metabolism during infancy and did not improve growth after the neonatal period. Protein intake ≥3.5 g/kg/d should not be recommended for children born preterm.

Postnatal growth and neurodevelopment at 2 years' corrected age in extremely low birthweight infants.

BACKGROUND: Faltering postnatal growth in preterm babies is associated with adverse neurodevelopment. However, which growth reference is most helpful for predicting neurodevelopment is unknown. We examined associations between faltering growth and developmental delay in extremely low birthweight (ELBW) infants. METHODS: We categorized faltering growth (z-score decrease ≥0.8 for weight/length, >1 for head circumference) between birth, 4 weeks, 36 weeks' postmenstrual age and 2 years' corrected age using fetal (Fenton, UK-WHO and Olsen) and healthy preterm (INTERGROWTH-21st) references. Associations between faltering growth and developmental delay were examined using binary logistic regression and area under the receiver operating curve (AUC). RESULTS: In 327 infants, Olsen charts identified the highest prevalence of faltering growth (weight 37%, length 63%, head 45%). Agreement in classification was higher amongst fetal references (kappa coefficient, ĸ = 0.46-0.94) than between INTERGROWTH-21st and fetal references (ĸ = 0.10-0.81). Faltering growth in all measures between 4-36 weeks (odds ratio, OR 2.0-4.7) compared with other time intervals (OR 1.7-2.7) were more strongly associated with developmental delay, particularly motor delay (OR 2.0-4.7). All growth references were poorly predictive of developmental delay at 2 years (AUC ≤ 0.62). CONCLUSIONS: Faltering postnatal growth in ELBW infants is associated with, but is poorly predictive of, developmental delay at 2 years. IMPACT: In babies born preterm, different growth references result in wide variation in categorization of faltering postnatal growth. Faltering growth in weight, length, and head circumference from 4 weeks to 36 weeks' postmenstrual age are associated with developmental delay at 2 years' corrected age, particularly motor delay. However, postnatal growth is a poor predictor of later developmental delay in extremely low birthweight infants irrespective of the growth reference used.

Nutrition guidelines for preterm infants: A systematic review.

BACKGROUND: There is no consensus on optimal nutrition for preterm infants, leading to substantial practice variation. We aimed to assess the quality of nutrition guidelines for preterm infants, the consistency of recommendations, and the gaps in recommendations. METHODS: We searched databases and websites for nutrition guidelines for preterm infants before first hospital discharge, which were endorsed, prepared, or authorized by a regional, national, or international body, written in English, and published between 2012 and 2023. Two reviewers independently screened articles and extracted the recommendations. Four reviewers appraised the included guidelines using Appraisal of Guidelines, Research, and Evaluation II. RESULTS: A total of 7051 were identified, with 27 guidelines included, 26% of which were high in quality. Most guidelines lacked stakeholder involvement and rigor of development. We found considerable variation in recommendations, many of which lacked details on certainty of evidence and strength of recommendation. Recommendations for type of feed and breastmilk fortification were consistent among high-quality guidelines, but recommendations varied for intakes of almost all nutrients and monitoring of nutrition adequacy. Different guidelines gave different certainty of evidence for the same recommendations. Most gaps in recommendations were due to very low certainty of evidence. CONCLUSION: Future development of nutrition guidelines for preterm infants should follow the standard guideline development method and ensure the rigorous process, including stakeholders' involvement, to improve the reporting of strength of recommendation, certainty of evidence, and gaps in recommendation. Evidence is needed to support recommendations about macro and micronutrient intakes, breastmilk fortification, and markers on adequacy of intake of different nutrients.

Hypophosphataemia definitions, incidence and associated outcomes in paediatric intensive care: A retrospective cohort study in post-cardiac surgical patients <2 years of age.

AIM: Hypophosphataemia has been linked to higher morbidity and mortality in intensive care but there is inconsistency in the definition of hypophosphataemia for infants and children. We aimed to determine the incidence of hypophosphataemia in a group of at-risk children in paediatric intensive care unit (PICU) and associations with patient characteristics and clinical outcomes using three different hypophosphataemia thresholds. METHODS: Retrospective cohort study of 205 post-cardiac surgical patients <2 years of age admitted to Starship Child Health PICU, Auckland, New Zealand. Patient demographics and routine daily biochemistry for 14 days after PICU admission were collected. Rates of sepsis, mortality and length of mechanical ventilation were compared between groups with different serum phosphate concentrations. RESULTS: Out of 205 children, 6 (3%), 50 (24%) and 159 (78%) had hypophosphataemia at thresholds of <0.7, <1.0 and <1.4 mmol/L, respectively. There were no differences in gestational age at birth, sex, ethnicity or mortality in those with and without hypophosphataemia at any threshold. Children with a serum phosphate <1.4 mmol/L had more mean (SD) total hours of mechanical ventilation (85.2 (79.6) vs. 54.9 (36.2) h, P = 0.02) and those with mean serum phosphate <1.0 mmol/L had more mean hours of mechanical ventilation (119.4 (102.8) vs. 65.2 (54.8) h, P < 0.0001), episodes of sepsis (14% vs. 5%, P = 0.03) and longer length of stay (6.4 (4.8-20.7) vs. 4.9 (3.9-6.8) days, P = 0.02). CONCLUSIONS: Hypophosphataemia is common in this PICU cohort and serum phosphate <1.0 mmol/L is associated with increased morbidity and length of stay.

Early Amino Acids in Extremely Preterm Infants and Neurodisability at 2 Years.

BACKGROUND: Whether higher parenteral amino acid intake improves outcomes in infants with extremely low birth weight is unclear. METHODS: In this multicenter, parallel-group, double-blind, randomized, placebo-controlled trial, we assigned infants with birth weights of less than 1000 g at 8 neonatal intensive care units to receive amino acids at a dose of 1 g per day (intervention group) or placebo in addition to usual nutrition for the first 5 days after birth. The primary outcome was survival free from neurodisability as assessed with the Bayley Scales of Infant and Toddler Development and neurologic examination at 2 years, corrected for gestational age at birth. Secondary outcomes were the components of the primary outcome as well as the presence or absence of neonatal disorders, the rate of growth, and nutritional intake. RESULTS: We enrolled 434 infants (217 per group) in this trial. Survival free from neurodisability was observed in 97 of 203 children (47.8%) in the intervention group and in 102 of 205 (49.8%) in the placebo group (adjusted relative risk, 0.95; 95% confidence interval [CI], 0.79 to 1.14; P = 0.56). Death before the age of 2 years occurred in 39 of 217 children (18.0%) in the intervention group and 42 of 217 (19.4%) in the placebo group (adjusted relative risk, 0.93; 95% CI, 0.63 to 1.36); neurodisability occurred in 67 of 164 children (40.9%) in the intervention group and 61 of 163 (37.4%) in the placebo group (adjusted relative risk, 1.16; 95% CI, 0.90 to 1.50). Neurodisability was moderate to severe in 27 children (16.5%) in the intervention group and 14 (8.6%) in the placebo group (adjusted relative risk, 1.95; 95% CI, 1.09 to 3.48). More children in the intervention group than in the placebo group had patent ductus arteriosus (adjusted relative risk, 1.65; 95% CI, 1.11 to 2.46). In a post hoc analysis, refeeding syndrome occurred in 42 of 172 children in the intervention group and 26 of 166 in the placebo group (adjusted relative risk, 1.64; 95% CI, 1.09 to 2.47). Eight serious adverse events occurred. CONCLUSIONS: In infants with extremely low birth weight, extra parenteral amino acids at a dose of 1 g per day for 5 days after birth did not increase the number who survived free from neurodisability at 2 years. (Funded by the New Zealand Health Research Council and others; ProVIDe Australian New Zealand Clinical Trials Registry number, ACTRN12612001084875.).

Variation in Neonatal Nutrition Practice and Implications: A Survey of Australia and New Zealand Neonatal Units.

Background: Significant global variation exists in neonatal nutrition practice, including in assigned milk composition values, donor milk usage, fortification regimens, probiotic choice and in methods used to calculate and report nutrition and growth outcomes, making it difficult to synthesize data to inform evidence-based, standardized nutritional care that has potential to improve neonatal outcomes. The Australasian Neonatal Dietitians' Network (ANDiN) conducted a survey to determine the degree to which neonatal nutritional care varies across Australia and New Zealand (A&NZ) and to highlight potential implications. Materials and Methods: A two-part electronic neonatal nutritional survey was emailed to each ANDiN member (n = 50). Part-One was designed to examine individual dietetic practice; Part-Two examined site-specific nutrition policies and practices. Descriptive statistics were used to examine the distribution of responses. Results: Survey response rate: 88%. Across 24 NICU sites, maximum fluid targets varied (150-180 mL.kg.d-1); macronutrient composition estimates for mothers' own(MOM) and donor (DM) milk varied (Energy (kcal.dL-1) MOM: 65-72; DM 69-72: Protein (g.dL-1): MOM: 1.0-1.5; DM: 0.8-1.3); pasteurized DM or unpasteurized peer-to-peer DM was not available in all units; milk fortification commenced at different rates and volumes; a range of energy values (kcal.g-1) for protein (3.8-4.0), fat (9.0-10.0), and carbohydrate (3.8-4.0) were used to calculate parenteral and enteral intakes; probiotic choice differed; and at least seven different preterm growth charts were employed to monitor growth. Discussion: Our survey identifies variation in preterm nutrition practice across A&NZ of sufficient magnitude to impact nutrition interventions and neonatal outcomes. This presents an opportunity to use the unique skillset of neonatal dietitians to standardize practice, reduce uncertainty of neonatal care and improve the quality of neonatal research.

Strategies in Neonatal Care to Promote Growth and Neurodevelopment of the Preterm Infant.

Recommendations for nutrition of very preterm and very low birth weight infants have developed over time with our understanding of the requirements of preterm babies and the awareness of widespread poor postnatal growth. In general, the trend has been towards enhancing nutrition, but more recent recommendations have begun to raise questions with respect to the potential for high and early nutritional intakes, particularly of protein, to carry risks such as refeeding syndrome. However, large gaps in our knowledge remain for both macro- and micronutrient requirements to support optimal growth and how nutrition and growth relate to important long-term outcomes. Closing these knowledge gaps has been hampered by inconsistent reporting of nutrition intakes and growth parameters, small trials with short-term outcomes and the use of a variety of different methods of monitoring growth. The challenge now is for future research to address these issues through consensus building around the important questions that need to be answered, how to report data from neonatal nutritional trials and whether large trials answering important questions can take place through development of consortia that undertake similar trials in multiple jurisdictions with agreements to share data.

Neonatal Refeeding Syndrome and Clinical Outcome in Extremely Low-Birth-Weight Babies: Secondary Cohort Analysis From the ProVIDe Trial.

BACKGROUND: Refeeding syndrome (RS) following preterm birth has been linked to high intravenous (IV) protein intake in the presence of low electrolyte supply. In extremely low-birth-weight (ELBW) babies, we aimed to determine the incidence of RS and associations with birth characteristics and clinical outcomes. METHOD: Prospective cohort study of ELBW ProVIDe Trial participants in 6 New Zealand neonatal intensive care units. RS was defined as serum phosphate < 1.4 mmol.L-1 and total calcium > 2.8 mmol.L-1 . Relationships between RS and other factors were explored using 2-sample tests and logistic regression adjusted for sex, gestation, and birth-weight z-score. RESULTS: Of 338 babies (mean [SD] birth-weight, 780 (134) g, gestational age, 25.9 [1.7] weeks), 68 (20%) had RS. Mortality was greater in babies with RS (32% vs 11%; P < .0001). More small- than appropriate-for-gestational-age babies developed RS (22% vs 8%; P = .001). Growth from birth to 36 weeks' corrected age was not different between babies who did and did not have RS. In logistic regression, the odds of RS decreased by 70% for each 1 mmol per kg-1 .d-1 IV phosphate intake (odds ratio [OR], 0.3; CI, 0.1-0.6; P = .002) and increased by 80% for each 1 g.kg-1 .d-1 IV protein intake (OR, 1.8; CI, 1.3-2.7; P = .002). CONCLUSIONS: Neonatal RS is common in this cohort of ELBW babies and is associated with increased morbidity and mortality. Optimizing phosphate and calcium intakes in IV nutrition solutions may reduce RS and its consequences.

Do preterm girls need different nutrition to preterm boys? Sex-specific nutrition for the preterm infant.

Boys born preterm are recognised to be at higher risk of adverse outcomes than girls born preterm. Despite advances in neonatal intensive care and overall improvements in neonatal morbidity and mortality, boys born preterm continue to show worse short- and long-term outcomes than girls. Preterm birth presents a nutritional crisis during a critical developmental period, with postnatal undernutrition and growth-faltering common complications of neonatal intensive care. Furthermore, this preterm period corresponds to that of rapid in utero brain growth and development, and the developmental window relating to foetal programming of adult non-communicable diseases, the prevalence of which are associated both with preterm birth and sex. There is increasing evidence to show that from foetal life, boys and girls have different responses to maternal nutrition, that maternal breastmilk composition differs based on foetal sex and that early neonatal nutritional interventions affect boys and girls differently. This narrative review examines the evidence that sex is an important moderator of the outcomes of preterm nutrition intervention, and describes what further knowledge is required before providing nutrition intervention for infants born preterm based on their sex. IMPACT: This review examines the increasing evidence that boys and girls respond differently to nutritional stressors before birth, that maternal breastmilk composition differs by foetal sex and that nutritional interventions have different responses based on infant sex. Boys and girls born preterm are given standard nutritional support which does not take infant sex into account, and few studies of neonatal nutrition consider infant sex as a potential mediator of outcomes. By optimising early nutrition for boys and girls born preterm, we may improve outcomes for both sexes. We propose future studies of neonatal nutritional interventions should consider infant sex.

Early Use of Antibiotics Is Associated with a Lower Incidence of Necrotizing Enterocolitis in Preterm, Very Low Birth Weight Infants: The NEOMUNE-NeoNutriNet Cohort Study.

OBJECTIVE: To determine whether commencement of antibiotics within 3 postnatal days in preterm, very low birth weight (VLBW; ≤1500 g) infants is associated with the development of necrotizing enterocolitis (NEC). STUDY DESIGN: Preplanned statistical analyses were done to study the association between early antibiotic treatment and later NEC development, using the NEOMUNE-NeoNutriNet cohort of VLBW infants from 13 neonatal intensive care units (NICUs) in 5 continents (n = 2831). NEC incidence was compared between infants who received early antibiotics and those who did not, with statistical adjustments for NICU, gestational age, birth weight, sex, delivery mode, antenatal steroid use, Apgar score, and type and initiation of enteral nutrition. RESULTS: The incidence of NEC was 9.0% in the group of infants who did not receive early antibiotics (n = 269), compared with 3.9% in those who did receive early antibiotics (n = 2562). The incidence remained lower in the early antibiotic group after stepwise statistical adjustments for NICU (OR, 0.57; 95% CI, 0.35-0.94, P < .05) and other potential confounders (OR, 0.25; 95% CI, 0.12-0.47; P < .0001). CONCLUSIONS: In this large international cohort of preterm VLBW infants, a small proportion of infants did not receive antibiotics just after birth, and these infants had a higher incidence of NEC. It is important to better understand the role of such variables as time, type, and duration of antibiotic treatment on NEC incidence, immune development, gut colonization, and antibiotic resistance in the NICU.

Relationships between Neonatal Nutrition and Growth to 36 Weeks' Corrected Age in ELBW Babies-Secondary Cohort Analysis from the Provide Trial.

A key modifiable factor for improving neurodevelopment in extremely low birthweight (ELBW) babies may be improving growth, especially head growth, by optimising nutrition in the early neonatal period. We aimed to investigate relationships between nutrient intakes in the 4 weeks after birth, and growth from birth to 36 weeks' corrected age (CA) in ELBW babies. We undertook a prospective cohort study of 434 participants enrolled in a randomised controlled trial (ProVIDe) in eight New Zealand and Australian neonatal intensive care units. Macronutrient intakes from birth to 4 weeks and weight, length and head circumference measurements from birth to 36 weeks' CA were collected. From birth to 36 weeks' CA, the median (IQR) z-score changes were: weight -0.48 (-1.09, 0.05); length -1.16 (-1.86, -0.43), and head circumference -0.82 (-1.51, -0.19). Changes in z-score to 4 weeks and 36 weeks' CA were correlated with protein intake. Each 1 g·Kg-1·d-1 total protein intake in week 2 was associated with 0.26 z-score increase in head circumference at 36 weeks' CA. Both nutritional intake and change in z-scores to 36 weeks' CA differed widely amongst sites. Correlations between nutrition and growth, and differences in these amongst sites, indicate there may be potential to improve growth with enhanced nutrition practices.

Plasma ammonia concentrations in extremely low birthweight infants in the first week after birth: secondary analysis from the ProVIDe randomized clinical trial.

BACKGROUND: Little is known about normative ammonia concentrations in extremely low birthweight (ELBW) babies and whether these vary with birth characteristics. We aimed to determine ammonia concentrations in ELBW babies in the first week after birth and relationships with neonatal characteristics and protein intake. METHODS: Arterial blood samples for the measurement of plasma ammonia concentration were collected within 7 days of birth from ProVIDe trial participants in six New Zealand neonatal intensive care units. RESULTS: Three hundred and twenty-two babies were included. Median (range) gestational age was 25.7 (22.7-31.6) weeks. Median (interquartile range (IQR)) ammonia concentration was 102 (80-131) µg/dL. There were no statistically significant associations between ammonia concentrations and birthweight or sex. Ammonia concentrations were weakly correlated with mean total (Spearman's rs = 0.11, P = 0.047) and intravenous (rs = 0.13, P = 0.02) protein intake from birth, gestational age at birth (rs = -0.13, P = 0.02) and postnatal age (rs = -0.13, P = 0.02). CONCLUSIONS: Plasma ammonia concentrations in ELBW babies are similar to those of larger and more mature babies and only weakly correlated with protein intake. Currently, recommended thresholds for investigation of hyperammonaemia are appropriate for ELBW babies. Protein intake should not be limited by concerns about potential hyperammonaemia.

Growth and Clinical Outcome in Very Low-Birth-Weight Infants After the Introduction of a Multicomponent Intravenous Lipid Emulsion.

BACKGROUND: Very low-birth-weight (VLBW; birth weight < 1500 g) infants are often dependent on intravenous nutrition after birth. Conventional soy-based intravenous lipid emulsions (Soy LE) are associated with inflammatory and metabolic complications that may be harmful to preterm infants. Evidence to support any clinical benefit associated with newer multicomponent emulsions (Mixed LE), remains inconsistent and unsubstantiated in appropriate studies. This retrospective study aimed to determine whether growth and clinical outcomes differed between VLBW infants given Mixed LE vs Soy LE at Auckland City Hospital. METHODS: Data were collected on nutrition, growth, and neonatal morbidities for the first 4 weeks after birth. Outcomes were compared between 2 lipid cohorts, Soy LE (February 2013 - August 2014) and Mixed LE (August 2014 - December 2015), using univariate and multivariate analysis. RESULTS: 207 infants (Soy LE, 105 vs Mixed LE, 102) were included in the study. Significantly fewer infants in the Mixed LE cohort developed any stage retinopathy of prematurity (Soy LE 59% vs Mixed LE 39%, P = .005) or intraventricular hemorrhage (Soy LE 27% vs Mixed LE 15%, P = .03) during their admission. Mixed LE was also associated with significantly lower mean (P = .01), minimum (P = .03), and maximum (P = .04) total bilirubin concentrations across the first 4 weeks after birth. There was no difference in growth velocity or weight, length, and head circumference z-score change. CONCLUSION: SMOFlipid may represent a favorable alternative to conventional lipid emulsions in neonatal parenteral nutrition regimens; however, long-term effects should be further evaluated.

Sex-specific relationships between early nutrition and neurodevelopment in preterm infants.

BACKGROUND: Although early nutrition is associated with neurodevelopmental outcome at 2 years' corrected age in children born very preterm, it is not clear if these associations are different in girls and boys. METHODS: Retrospective cohort study of infants born <30 weeks' gestational age or <1500 g birth weight in Auckland, NZ. Macronutrient, energy and fluid volumes per kg per day were calculated from daily nutritional intakes and averaged over days 1-7 (week 1) and 1-28 (month 1). Primary outcome was survival without neurodevelopmental impairment at 2 years corrected age. RESULTS: More girls (215/478) survived without neurodevelopmental impairment at 2 years (82% vs. 72%, P = 0.02). Overall, survival without neurodevelopmental impairment was positively associated with more energy, fat, and enteral feeds in week 1, and more energy and enteral feeds in month 1 (P = 0.005-0.03), but all with sex interactions (P = 0.008-0.02). In girls but not boys, survival without neurodevelopmental impairment was positively associated with week 1 total intakes of fat (OR(95% CI) for highest vs. lowest intake quartile 62.6(6.6-1618.1), P < 0.001), energy (22.9(2.6-542.0), P = 0.03) and enteral feeds (1.9 × 109(9.5-not estimable), P < 0.001). CONCLUSIONS: Higher early fat and enteral feed intakes are associated with improved outcome in girls, but not boys. Future research should determine sex-specific neonatal nutritional requirements.

The Influence of Early Nutrition on Brain Growth and Neurodevelopment in Extremely Preterm Babies: A Narrative Review.

Extremely preterm babies are at increased risk of less than optimal neurodevelopment compared with their term-born counterparts. Optimising nutrition is a promising avenue to mitigate the adverse neurodevelopmental consequences of preterm birth. In this narrative review, we summarize current knowledge on how nutrition, and in particular, protein intake, affects neurodevelopment in extremely preterm babies. Observational studies consistently report that higher intravenous and enteral protein intakes are associated with improved growth and possibly neurodevelopment, but differences in methodologies and combinations of intravenous and enteral nutrition strategies make it difficult to determine the effects of each intervention. Unfortunately, there are few randomized controlled trials of nutrition in this population conducted to determine neurodevelopmental outcomes. Substantial variation in reporting of trials, both of nutritional intakes and of outcomes, limits conclusions from meta-analyses. Future studies to determine the effects of nutritional intakes in extremely preterm babies need to be adequately powered to assess neurodevelopmental outcomes separately in boys and girls, and designed to address the many potential confounders which may have clouded research findings to date. The development of minimal reporting sets and core outcome sets for nutrition research will aid future meta-analyses.

Time to Full Enteral Feeding for Very Low-Birth-Weight Infants Varies Markedly Among Hospitals Worldwide But May Not Be Associated With Incidence of Necrotizing Enterocolitis: The NEOMUNE-NeoNutriNet Cohort Study.

BACKGROUND: Transition to enteral feeding is difficult for very low-birth-weight (VLBW; ≤1500 g) infants, and optimal nutrition is important for clinical outcomes. METHOD: Data on feeding practices and short-term clinical outcomes (growth, necrotizing enterocolitis [NEC], mortality) in VLBW infants were collected from 13 neonatal intensive care units (NICUs) in 5 continents (n = 2947). Specifically, 5 NICUs in Guangdong province in China (GD), mainly using formula feeding and slow feeding advancement (n = 1366), were compared with the remaining NICUs (non-GD, n = 1581, Oceania, Europe, United States, Taiwan, Africa) using mainly human milk with faster advancement rates. RESULTS: Across NICUs, large differences were observed for time to reach full enteral feeding (TFF; 8-33 days), weight gain (5.0-14.6 g/kg/day), ∆z-scores (-0.54 to -1.64), incidence of NEC (1%-13%), and mortality (1%-18%). Adjusted for gestational age, GD units had longer TFF (26 vs 11 days), lower weight gain (8.7 vs 10.9 g/kg/day), and more days on antibiotics (17 vs 11 days; all P < .001) than non-GD units, but NEC incidence and mortality were similar. CONCLUSION: Feeding practices for VLBW infants vary markedly around the world. Use of formula and long TFF in South China was associated with more use of antibiotics and slower weight gain, but apparently not with more NEC or higher mortality. Both infant- and hospital-related factors influence feeding practices for preterm infants. Multicenter, randomized controlled trials are required to identify the optimal feeding strategy during the first weeks of life.