PREMFOOD Version 1/010812
Research type
Research Study
Full title
A pilot randomised controlled trial of either fortified Human Donor Milk, unfortified Human Donor Milk, or Preterm Formula in preterm babies, to make up any shortfall in the volume of Mother’s Own Milk.
IRAS ID
101786
Contact name
Neena Modi
Contact email
Sponsor organisation
Imperial College London
Research summary
Summary of Results
1. Study name
1.1 Study title
Short title: PREMFOOD study, a randomised controlled feasibility trial of different early diets for very preterm infants.
Long title: PREMFOOD (PREterM FOrmula Or Donor milk for very preterm infants) study, a pilot randomised controlled trial of either fortified human donor milk, unfortified human donor milk or preterm formula in preterm babies to make up any shortfall in the volume of mother's own milk.
1.2 Trial registration number
NCT01686477
Unique Protocol ID: CRO2006
1.3 Other identifiers
Research Ethics Approval No: 12/LO/13911.4 Abstract
Purpose of the study: It is unknown which is the best milk to feed very premature babies (babies born more than 8 weeks before their due date) when there is not enough mother’s own breast milk. A large study is needed to detect differences in important health benefits from different early diets for very premature babies. The purpose of this smaller study was to determine whether or not such a large study could be done.
What was tested:
If there is not enough mother’s own breast milk to meet the needs of their premature baby, doctors and nurses often use donor breast milk, from a human milk bank, or a formula specially designed for premature babies (preterm formula) to make up the volume (supplement). Extra nutrients (fortifier) are also sometimes added to mother’s own breast milk and donor milk supplement.
We compared three feeding groups for very premature babies: unfortified mother’s own breast milk and donor milk supplement (Unfortified Human Milk, UHM), fortified mother’s own breast milk and donor milk supplement (Fortified Human Milk, FHM), and unfortified mother’s own breast milk and preterm formula supplement (PTF). These diets were from birth to 5 weeks before the baby’s due date. At the time of birth the study was discussed with the parents and opportunity offered for them to not take part in the study (opt out consent). If the parents did not opt out, their baby was placed into one of the three feeding groups by chance (randomisation).
We assessed how many parents agreed to take part in the study and how many babies were later withdrawn from the feeding group, either due to parental wishes, or due to slow growth concerns from the healthcare team. We assessed the practicality of using national database newborn healthcare data for the research study. We also assessed growth and amount of body fat measured by magnetic resonance imaging (MRI) at term (due date) and term plus 6 weeks of age.
People taking part: One hundred and three very premature babies from 3 hospitals in London.
Results: Sixty two percent of parents approached agreed to take part in the study. Six percent of parents overall withdrew from the study diet. Two babies had slow growth and needed the addition of fortifier or preterm formula. Data were recorded accurately and completely into the national database. There were no differences in growth or body fat between the three feeding groups.
Conclusions: Randomisation of very premature babies to UHM, FHM, and PTF feeding groups with opportunity to opt out is acceptable to parents and healthcare teams, appears safe, and did not impact on long term body fat. Data from the national newborn healthcare database can be used effectively for feeding studies. Our data demonstrates that a study large enough to detect differences in important health benefits can be done.2. Study sponsor
Imperial College London3. General information about the clinical trial
3.1 Where was the study done?
This trial took place in London, UK. Participants were recruited from 3 London NHS hospitals, Chelsea and Westminster Hospital, Northwick Park Hospital, and Croydon University Hospital.
3.2 When was this study done?
Recruitment commenced 02/04/2013 and ended in November 2017, with a final follow-up in January 2018. The end of study was 01/10/20203.3 What was the main objective of this study?
Mother’s own milk is recommended for premature babies. However, mother’s are often unable to produce all the milk that is needed for their baby. In this situation donor breast milk from a human milk bank, or formula specially designed for premature babies (preterm formula), is used in addition to mother’s milk to make up the full amount required (supplement). Premature babies often grow slowly. If this is the case and a baby is receiving mother’s own milk or donor breast milk, a special nutrient supplement called ‘fortifier’ is added. We do not know if this has any long term benefit, or side effects.
There is wide variation in the use of each option. The choice is dependent on the availability of donor milk and the preference of the doctor. Doctors and nurses who prefer donor breast milk top ups believe that this will help prevent infection and a particular type of bowel disease sometimes seen in premature babies. Doctors and nurses who prefer formula top ups believe that this benefits growth, including brain growth and development.
We do not know which approach is better for babies. We can only find this out by comparing a large number of babies. The purpose of this small study was to test all the aspects of a proposed larger study in order to determine the likelihood of it succeeding.3.4 What was tested?
We compared three diets for very premature babies: unfortified mother’s own breast milk and donor milk supplement (Unfortified Human Milk, UHM), fortified mother’s own breast milk and donor milk supplement (Fortified Human Milk, FHM), and unfortified mother’s own breast milk and preterm formula supplement (PTF). These diets are widely used in newborn care. They were received from birth to 5 weeks before the baby’s due date.
We designed the study with involvement from parent focus groups, doctors and researchers. The study was in two stages. We used an opt out approach for parental agreement (consent) to one of three diets for their baby. This means that the study was discussed with the parents within 48hrs of birth and opportunity offered for them to not take part or opt out of the study. If this happened, type of milk feed would be decided by the doctor looking after the baby. We used a later opt in approach for parental consent for research related investigations for their baby. This means that the research related investigations were discussed with the parents at a convenient time, and opportunity offered for them to take part or opt in to the study investigations.
Babies were put into one of three feeding groups by chance (randomised) to reduce differences between the groups. Randomisation helps to make the 3 groups equal. Reducing differences between the groups in this way, makes the comparison fairer.
We assessed how many parents agreed to enter their baby into the study and how many babies were withdrawn from the study diet, either due to parental wishes or for slow growth concerns from the healthcare team. We assessed the practicality of using routinely collected healthcare data from a national newborn database for the research study. We also assessed growth and body fat measured by magnetic resonance imaging (MRI) at term (due date) and term plus 6 weeks of age.
All scans were undertaken at either Chelsea and Westminster Hospital or Queen Charlottes and Chelsea Hospital with scanning costs supported through unrestricted research funds held by Professor Neena Modi.4. What patients were included in this study?
4.1 Inclusion and exclusion critera
This study included very premature babies (those born more than 8 weeks before their due date), whose mother received care during her pregnancy at the hospital of birth, and who were unlikely to be transferred to another hospital. Infants with life-threatening or major birth abnormalities that prevented early milk feeding, and those unable to be randomised within 48 hours of age were excluded.
4.2 Characteristics of participants
Overall there were 103 participants born at an average of 11 weeks before their due date. Fifty eight percent were boys and 43% were twins. Birthweight, gender, how premature the babies were, and other characteristics were similar between the three feeding groups.5. What were the results of the study?
Thirty five infants were randomised to UHM, 34 to FHM, and 34 to PTF groups, of which 21, 19, and 24 infants completed imaging at term, and 14, 5, and 19 respectively, at term plus 6 weeks.
Sixty two percent of parents agreed to take part in the study, above the predicted 50% target; 6% of parents withdrew from the study diet, all from the FHM arm.
Two infants had slow growth and needed the addition of fortifier or preterm formula. No other safety concerns emerged. There was 100% agreement between national database newborn healthcare data and individual patient notes. There was also less than 10% of missing data recorded onto the national database.
There were no between group differences in the primary endpoint, total body fat at term. There was also no between group differences in growth, or regional body fat at term or term plus 6 weeks.6. How has this study helped patients and researchers?
Randomisation of very premature babies to UHM, FHM, and PTF feeding with opportunity to opt out is acceptable to parents and healthcare teams, appears safe, and it was possible to obtain the majority of data for the study from a national database of newborn healthcare data. Feeding group did not affect long term growth, or body fat. Our data demonstrates that a study large enough to detect differences in important health benefits can be done. This is an important first step in reducing uncertainty in supplemental milk feeding, and optimising health both in the short and long term for very premature babies.7. Are there plans for further studies?
Follow up studies during childhood addressing body fat and other outcomes are planned for these babies.8. Where can I find more information about this study?
To learm more about this study, you can find more detailed information on this website:
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.clinicaltrials.gov%2F&data=05%7C01%7Capprovals%40hra.nhs.uk%7C250266718d83490ddcd108daa2fa80ec%7C8e1f0acad87d4f20939e36243d574267%7C0%7C0%7C638001494031039642%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=NeX0Z40sapzHMlIUCRocS8eJ4YUCuqZ1X7GGGht90tU%3D&reserved=0REC name
London - Fulham Research Ethics Committee
REC reference
12/LO/1391
Date of REC Opinion
10 Dec 2012
REC opinion
Further Information Favourable Opinion