Protein supplementation of human milk for promoting growth in preterm infants

Emma A Amissah, Julie Brown, Jane E Harding

Preterm infants require high protein intake to achieve adequate growth and development. Although breast milk feeding has many benefits for this population, the protein content is highly variable, and inadequate to support rapid infant growth. This is a 2018 update of a Cochrane Review first published in 1999. To determine whether protein-supplemented human milk compared with unsupplemented human milk, fed to preterm infants, improves growth, body composition, cardio-metabolic, and neurodevelopmental outcomes, without significant adverse effects. We used the standard search strategy of Cochrane Neonatal to search CENTRAL, MEDLINE via PubMed, Embase, and CINAHL (February 2018). We also searched clinical trials databases, conference proceedings and the reference lists of retrieved articles for randomised controlled trials (RCT) and quasi-randomised trials. Published and unpublished RCTs were eligible if they used random or quasi-random methods to allocate hospitalised preterm infants who were being fed human milk, to additional protein supplementation or no supplementation. Two review authors independently abstracted data, assessed risk of bias and the quality of evidence at the outcome level, using GRADE methodology. We performed meta-analyses, using risk ratio (RR) for dichotomous data, and mean difference (MD) for continuous data, with their respective 95% confidence intervals (CIs). We used a fixed-effect model and had planned to explore potential causes of heterogeneity via subgroup or sensitivity analyses. We included six RCTs, involving 204 preterm infants. Low-quality evidence showed that protein supplementation of human milk increased in-hospital rates of growth in weight (MD 3.82 g/kg/day, 95% CI 2.94 to 4.7; five RCTs, 101 infants; I² = 73%), length (MD 0.12 cm/wk, 95% CI 0.07 to 0.17; four RCTs, 68 infants; I² = 89%), and head circumference (MD 0.06 cm/wk, 95% CI 0.01 to 0.12; four RCTs, 68 infants; I² = 84%). There was no evidence of a clear difference in rate of growth of skin fold thickness between the supplemented and unsupplemented groups (triceps MD 0.06 mm/wk, 95% CI -0.09 to 0.21; one RCT, 20 infants; or subscapular MD 0.00 mm/wk, 95% CI -0.17 to 0.17; one RCT, 20 infants). Protein supplementation led to longer hospital stays (MD 18.5 days, 95% CI 4.39 to 32.61; one RCT, 20 infants; very low-quality evidence), and higher blood urea nitrogen concentrations compared to the unsupplemented group (MD 0.95 mmol/L, 95% CI 0.81 to 1.09; four RCTs, 81 infants; I² = 56%). Very low-quality evidence did not show that protein supplementation clearly increased the risk of feeding intolerance (RR 2.70, 95% CI 0.13 to 58.24; one RCT, 17 infants), or necrotizing enterocolitis (RR 1.11, 95% CI 0.07 to 17.12; one RCT, 76 infants), or clearly altered serum albumin concentrations (MD 2.5 g/L, 95% CI -5.66 to 10.66; one RCT, 11 infants), compared with the unsupplemented groups. No data were available about the effects of protein supplementation on long-term growth, body mass index, body composition, neurodevelopmental, or cardio-metabolic outcomes. Low-quality evidence showed that protein supplementation of human milk, fed to preterm infants, increased short-term growth. However, the small sample sizes, low precision, and very low-quality evidence regarding duration of hospital stay, feeding intolerance, and necrotising enterocolitis precluded any conclusions about these outcomes. There were no data on outcomes after hospital discharge. Our findings may not be generalisable to low-resource settings, as none of the included studies were conducted in these settings.Since protein supplementation of human milk is now usually done as a component of multi-nutrient fortifiers, future studies should compare different amounts of protein in multi-component fortifiers, and be designed to determine the effects on duration of hospital stay and safety, as well as on long-term growth, body composition, cardio-metabolic, and neurodevelopmental outcomes.