Background Nutritional analysis of donated human milk has been suggested as a means of optimising its use.

Methods We analysed pooled, single donor milk samples using the MIRIS Human Milk Analyser to obtain values for fat, protein, lactose and calculated energy content. These values were compared with those of formula milks and then extrapolated to demonstrate whether donated human expressed breast milk (DEBM)±fortification would meet preterm nutritional requirements. We demonstrated the potential variability in macronutrient intake from unselected DEBM use by comparing energy intake from milk samples with energy values on the 3rd percentile (nutrient poor) with intake from milk with values on the 97th percentile (nutrient rich).

Results 179 sequential samples from 42 unique donors were analysed. Mean nutritional values, excluding protein, were comparable with those of term formula; energy 66±12 kcal/100 mLs, fat 4.0±1.4 g/100 mLs, protein 0.9±0.4 g/100 mLs and lactose 6.6±0.7 g/100 mLs. At volumes of 180 mL/kg/day 119 samples (66%) would have provided minimum preterm energy requirements without fortification however protein levels were low and even with fortification 39% of samples would fail to meet the recommended preterm requirements. The potential difference in energy intake between nutrient-poor and nutrient-rich milk was 88 kcal/kg/day.

Interpretation In summary our data confirm the nutritional variability of DEBM and demonstrate the potential of nutritional analysis to target the use of donor milk according to energy content.