Early enteral feeding is a potentially modifiable risk factor for necrotising enterocolitis (NEC) and late onset sepsis (LOS), however enteral feeding practices for preterm infants are highly variable. High-quality evidence is increasingly available to guide early feeding in preterm infants. Meta-analyses of randomised trials indicate that early trophic feeding within 48 h after birth and introduction of progressive enteral feeding before 4 days of life at an advancement rate above 24 ml/kg/day can be achieved in clinically stable very preterm and very low birthweight (VLBW) infants, without higher mortality or incidence of NEC. This finding may not be generalisable to high risk infants such as those born small for gestational age (SGA) or following absent/reversed end diastolic flow velocity (AREDFV) detected antenatally on placental Doppler studies, due to the small number of such infants in existing trials. Trials targeting such high-risk preterm infants have demonstrated that progressive enteral feeding started in the first 4 days is safe and does not lead to higher NEC or mortality, however there is a paucity of data to guide feeding advancement in such infants. There is little trial evidence to support bolus or continuous gavage feeding as being superior in clinically stable preterm infants. Trials that examine enteral feeding are commonly unblinded for technical and practical reasons, which increases the risk of bias in such trials, specifically when considering potentially subjective outcome such as NEC and LOS; future clinical trials should focus on objective, primary outcome measures such as all-cause mortality, long term growth and neurodevelopment. Alternatively, important short-term outcomes such as NEC could be used with blinded assessment.