Formula versus donor breast milk for feeding preterm or low birth weight infants

Review question

When a mother's own breast milk is not available, does feeding preterm or low birth weight infants with formula rather than donor breast milk affect digestion, growth and the risk of severe bowel problems?

Background

Preterm infants often find artificial formula more difficult to digest than human milk, and concerns exist that formula could increase the risk of severe bowel problems. If preterm infants are fed with donor breast milk (when a mother's own breast milk is insufficient or unavailable), rather than an artificial formula, this might reduce the risk of these problems. Donor breast milk, however, is more expensive than many formulas, and may not contain sufficient amounts of key nutrients to ensure optimal growth for preterm or low birth weight infants. Given these concerns, we have reviewed all of the available evidence from clinical trials that compared formula versus donor breast milk for feeding preterm or low birth weight infants.

Study characteristics

We found 12 completed trials (involving 1871 infants). Most trials, particularly those trials conducted more recently, used reliable methods. Evidence is up to date as of 3 May 2019.

Key results

The combined analysis of data from these trials shows that feeding with formula increases rates of growth during the hospital stay, but is associated with a higher risk of developing the severe gut disorder called 'necrotising enterocolitis'. There is no evidence of an effect on survival or longer-term growth and development.

Conclusions

The currently available evidence suggests that feeding preterm infants with artificial formula (rather than donor breast milk when mother's own breast milk is not available) is associated with faster rates of growth, but with a near-doubling of the risk of developing necrotising enterocolitis. Further, larger trials could provide stronger and more precise evidence to help clinicians and families make informed choices about this issue. Currently, four such trials (involving more than 1100 infants) are ongoing internationally, and we plan to include the data from these trials in this review when these become available.

Authors' conclusions: 

In preterm and LBW infants, moderate-certainty evidence indicates that feeding with formula compared with donor breast milk, either as a supplement to maternal expressed breast milk or as a sole diet, results in higher rates of weight gain, linear growth, and head growth and a higher risk of developing necrotising enterocolitis. The trial data do not show an effect on all-cause mortality, or on long-term growth or neurodevelopment.

Read the full abstract...
Background: 

When sufficient maternal breast milk is not available, alternative forms of enteral nutrition for preterm or low birth weight (LBW) infants are donor breast milk or artificial formula. Donor breast milk may retain some of the non-nutritive benefits of maternal breast milk for preterm or LBW infants. However, feeding with artificial formula may ensure more consistent delivery of greater amounts of nutrients. Uncertainty exists about the balance of risks and benefits of feeding formula versus donor breast milk for preterm or LBW infants.

Objectives: 

To determine the effect of feeding with formula compared with donor breast milk on growth and development in preterm or low birth weight (LBW) infants.

Search strategy: 

We used the Cochrane Neonatal search strategy, including electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL; 2019, Issue 5), Ovid MEDLINE, Embase, and the Cumulative Index to Nursing and Allied Health Literature (3 May 2019), as well as conference proceedings, previous reviews, and clinical trials.

Selection criteria: 

Randomised or quasi-randomised controlled trials (RCTs) comparing feeding with formula versus donor breast milk in preterm or LBW infants.

Data collection and analysis: 

Two review authors assessed trial eligibility and risk of bias and extracted data independently. We analysed treatment effects as described in the individual trials and reported risk ratios (RRs) and risk differences (RDs) for dichotomous data, and mean differences (MDs) for continuous data, with respective 95% confidence intervals (CIs). We used a fixed-effect model in meta-analyses and explored potential causes of heterogeneity in subgroup analyses. We assessed the certainty of evidence for the main comparison at the outcome level using GRADE methods.

Main results: 

Twelve trials with a total of 1879 infants fulfilled the inclusion criteria. Four trials compared standard term formula versus donor breast milk and eight compared nutrient-enriched preterm formula versus donor breast milk. Only the five most recent trials used nutrient-fortified donor breast milk. The trials contain various weaknesses in methodological quality, specifically concerns about allocation concealment in four trials and lack of blinding in most of the trials. Most of the included trials were funded by companies that made the study formula.

Formula-fed infants had higher in-hospital rates of weight gain (mean difference (MD) 2.51, 95% confidence interval (CI) 1.93 to 3.08 g/kg/day), linear growth (MD 1.21, 95% CI 0.77 to 1.65 mm/week) and head growth (MD 0.85, 95% CI 0.47 to 1.23 mm/week). These meta-analyses contained high levels of heterogeneity. We did not find evidence of an effect on long-term growth or neurodevelopment. Formula feeding increased the risk of necrotising enterocolitis (typical risk ratio (RR) 1.87, 95% CI 1.23 to 2.85; risk difference (RD) 0.03, 95% CI 0.01 to 0.05; number needed to treat for an additional harmful outcome (NNTH) 33, 95% CI 20 to 100; 9 studies, 1675 infants).

The GRADE certainty of evidence was moderate for rates of weight gain, linear growth, and head growth (downgraded for high levels of heterogeneity) and was moderate for neurodevelopmental disability, all-cause mortality, and necrotising enterocolitis (downgraded for imprecision).