Background: Observational data suggest that slowly advancing enteral feeds in preterm infants may reduce necrotising enterocolitis but increase late-onset sepsis. The Speed of Increasing milk Feeds Trial (SIFT) compared two rates of feed advancement.
Objective: To determine if faster (30 ml/kg/day) or slower (18 ml/kg/day) daily feed increments improve survival without moderate or severe disability and other morbidities in very preterm or very low-birthweight infants.
Design: This was a multicentre, two-arm, parallel-group, randomised controlled trial. Randomisation was via a web-hosted minimisation algorithm. It was not possible to safely and completely blind caregivers and parents.
Setting: The setting was 55 UK neonatal units, from May 2013 to June 2015.
Participants: The participants were infants born at < 32 weeks' gestation or a weight of < 1500 g, who were receiving < 30 ml/kg/day of milk at trial enrolment.
Interventions: When clinicians were ready to start advancing feed volumes, the infant was randomised to receive daily feed increments of either 30 ml/kg/day or 18 ml/kg/day. In total, 1400 infants were allocated to fast feeds and 1404 infants were allocated to slow feeds.
Main outcome measures: The primary outcome was survival without moderate or severe neurodevelopmental disability at 24 months of age, corrected for gestational age. The secondary outcomes were mortality; moderate or severe neurodevelopmental disability at 24 months corrected for gestational age; death before discharge home; microbiologically confirmed or clinically suspected late-onset sepsis; necrotising enterocolitis (Bell's stage 2 or 3); time taken to reach full milk feeds (tolerating 150 ml/kg/day for 3 consecutive days); growth from birth to discharge; duration of parenteral feeding; time in intensive care; duration of hospital stay; diagnosis of cerebral palsy by a doctor or other health professional; and individual components of the definition of moderate or severe neurodevelopmental disability.
Results: The results showed that survival without moderate or severe neurodevelopmental disability at 24 months occurred in 802 out of 1224 (65.5%) infants allocated to faster increments and 848 out of 1246 (68.1%) infants allocated to slower increments (adjusted risk ratio 0.96, 95% confidence interval 0.92 to 1.01). There was no significant difference between groups in the risk of the individual components of the primary outcome or in the important hospital outcomes: late-onset sepsis (adjusted risk ratio 0.96, 95% confidence interval 0.86 to 1.07) or necrotising enterocolitis (adjusted risk ratio 0.88, 95% confidence interval 0.68 to 1.16). Cost-consequence analysis showed that the faster feed increment rate was less costly but also less effective than the slower rate in terms of achieving the primary outcome, so was therefore found to not be cost-effective. Four unexpected serious adverse events were reported, two in each group. None was assessed as being causally related to the intervention.
Limitations: The study could not be blinded, so care may have been affected by knowledge of allocation. Although well powered for comparisons of all infants, subgroup comparisons were underpowered.
Conclusions: No clear advantage was identified for the important outcomes in very preterm or very low-birthweight infants when milk feeds were advanced in daily volume increments of 30 ml/kg/day or 18 ml/kg/day. In terms of future work, the interaction of different milk types with increments merits further examination, as may different increments in infants at the extremes of gestation or birthweight.
Trial registration: Current Controlled Trials ISRCTN76463425.
Funding: This project was funded by the National Institute for Health Research Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 24, No. 18. See the NIHR Journals Library website for further project information.
Keywords: DISABILITY; INFECTION; MILK FEEDING; NECROTISING ENTEROCOLITIS; PRETERM; RANDOMISED CONTROLLED TRIAL; SURVIVAL.
Some infants who are born early need to be fed through a tube into their stomach. A small volume of milk is given to begin with, which is gradually increased. To determine whether infants do better if they are fed faster or slower, this study compared increasing the milk feeds by 30 ml/kg/day with increasing the milk feeds by 18 ml/kg/day, aiming to get to full feeds (when other fluids are not needed) in 5 or 9 days. We compared results from the two groups at discharge from hospital and at 24 months of age, after correcting for prematurity. We also assessed the economic impact of the two daily feed increments, interviewed parents about taking part in multiple studies and tested methods for improving questionnaire returns. The faster-fed group reached full milk feeds sooner and needed less intravenous nutrition, and the proportion of infants developing bowel inflammation or bloodstream infection were similar. At 24 months of age, we found an unexpected increase in the risk of moderate or severe motor impairment in the faster-fed group, which is difficult to explain. We also saw that other types of disability were more frequent in the faster group, although this was not significantly different mathematically. This means that no clear advantage of increasing feeds at faster or slower rates was identified and health professionals will need to carefully consider how to increase feeds. After accepting the increased risk of disability, an economic evaluation showed that increasing milk feed volumes at a faster rate was not a cost-effective strategy. Interviews with parents showed that they valued opportunities for their infant to take part in studies, but this interaction is complex and difficult to remember at a stressful and confusing time and made worse by considering multiple studies. More questionnaires were returned when vouchers were given before rather than after receiving them.