A Phase 3 interventional study of Donor Milk and Preterm Formula in Infant, Newborn, Infant, Small for Gestational Age and Infant, Extremely Low Birth Weight, sponsored by NICHD Neonatal Research Network. Completed at 17 sites in United States. Open to participants aged Up to 21 Days. Per ClinicalTrials.gov, last updated 2026-07-16.
Sponsored by NICHD Neonatal Research Network · Phase 3, Interventional, and Prevention
The Milk Trial seeks to determine the effect on neurodevelopmental outcomes at age 22-26 months of donor human milk as compared to preterm infant formula as the in-hospital diet for infants whose mothers choose not to provide breast milk or are able to provide only a minimal amount. Infants will be randomized to receive donor breast milk or formula during their hospital stay. Infant's will be followed until they reach 22-26 months of age.
There is strong evidence that maternal breast milk feedings in infancy confer multiple health benefits in the extremely preterm population (extremely low birth weight, ELBW, \<1000 g). Studies suggest an IQ advantage of up to 8 points conferred by maternal milk feeding in this population. Rates of sepsis and necrotizing enterocolitis are also lower in human milk fed ELBW infants, and they experience shorter hospital stays and fewer re-hospitalizations in the first year of life. When mothers choose not to or are unable to provide milk, preterm formula is usually used. Recently, pasteurized donor human milk is available in some NICUs in the US as an alternative to preterm formula. Donor milk has not been well studied with regard to its safety and efficacy. It is unknown if donor human milk confers the same benefits as maternal milk with regard to neurodevelopmental and health outcomes. The proposed study will be the first US multicenter randomized trial of the health and developmental effects of donor milk as compared to preterm formula in ELBW infants receiving little or no maternal milk. Our long-term goal is to optimize neurodevelopmental and health outcomes for ELBW infants, maximizing their quality of life and societal functionality throughout their lives. If donor human milk has similar effects to maternal milk, the public health benefit of donor milk feedings in ELBW infants unable to receive maternal milk would be considerable.
Exclusion Criteria:
Feeding Group Eligibility:
Donor milk provided by the Human Milk Banking Association of North America
Biological: Donor Milk
Preterm formula determined by center practice
Dietary Supplement: Preterm Formula
Donor milk provided by the Human Milk Banking Association of North America
Preterm Formula determined by center practice.
Bayley Scales of Infant Development (BSID) Cognitive Composite Score
Mean cognitive composite score (standardized mean 100, SD 15, range 54-145). Subjects who died prior to follow-up assigned the score of 54. (lower scores indicating greater impairment)
Time frame: At 22-26 months corrected age
Total Deaths Before Discharge
Infant died before discharge home.
Time frame: From day of randomization to Neonatal Research Network NRN infant status i.e., the first occurring of: discharge home, death, transfer, or 1 year following birth
Late Onset Sepsis (LOS)
Number of infants diagnosed with LOS
Time frame: From birth to Neonatal Research Network NRN infant status i.e., the first occurring of: discharge home, death, transfer, or 120 days following birth
Necrotizing Enterocolitis (NEC)
Number of infants diagnosed with NEC
Time frame: From birth to Neonatal Research Network NRN infant status i.e., the first occurring of: discharge home, death, transfer, or 120 days following birth
Death or Necrotizing Enterocolitis (NEC)
A composite outcome that measures the occurrence of death or NEC
Time frame: From birth to Neonatal Research Network NRN infant status i.e., the first occurring of: discharge home, death, transfer, or 120 days following birth
Change in Weight-for-age Z-score During Study
Weight-for-age Z-scores were calculated at both baseline (study initiation) and study end (within one week of last study data collected) based on Fenton growth curves (2013). This outcome represents the change in weight-for-age Z-score during the course of the study (i.e., the Z-score at baseline was subtracted from the Z-score at study end). A value of 0 represents that the infant's weight-for-age Z-score is the same at the beginning and the end of the study. Positive values indicate the increase in the infant's weight-for-age Z-score during the study; negative values indicate the decrease in the infant's weight-for-age Z-score during the study.
Time frame: During Study Intervention, the time between study randomization and discontinuation of study protocol. Infants exited from the study protocol 1-2 weeks prior to anticipated hospital discharge or 120 days, whichever is sooner
Bayley Scales of Infant Development (BSID) Motor Composite Score
Mean motor composite score (standardized mean 100, range 44-155). Subjects who died prior to follow-up assigned the score of 44. (lower scores indicating greater impairment)
Time frame: At 22-26 months corrected age
Bayley Scales of Infant Development (BSID) Language Composite Score
Mean language composite score (standardized mean 100, range 46-155). Subjects who died prior to follow-up assigned the score of 46. (lower scores indicating greater impairment)
Time frame: At 22-26 months corrected age
Moderate to Severe Cerebral Palsy
Number of infants with moderate or severe grade of cerebral palsy
Time frame: At 22-26 months corrected age
Neurodevelopmental Impairment (NDI).
Number of infants with NDI. NDI is defined as any of the following: Gross Motor Function Classification System score greater than or equal to 2, Bayley III cognitive or motor score less than 85 (1 standard deviation), Vision Impairment or Hearing impairment
Time frame: At 22-26 months corrected age
Profound Impairment
Number of infants with profound impairment.
Time frame: At 22-26 months corrected age
Death or Neurodevelopmental Impairment (NDI)
A composite outcome that measures the occurrence of death through 22-26 months or NDI.
Time frame: At 22-26 months corrected age
| Milestone | Donor Milk | Formula |
|---|---|---|
| Started | 239 | 244 |
| Complete data for bsid iii cognitive score at 22-26 months | 175 | 192 |
| Died after discharge | 5 | 7 |
| Died before discharge | 24 | 18 |
| Survived to discharge | 215 | 226 |
| Completed | 204 | 217 |
| Not completed | 35 | 27 |
| Withdrew: Lost to follow-up | 29 | 24 |
| Withdrew: Incomplete follow-up or no study data | 6 | 3 |
Mean cognitive composite score (standardized mean 100, SD 15, range 54-145). Subjects who died prior to follow-up assigned the score of 54. (lower scores indicating greater impairment)
| Score on a scale | Donor Milk | Formula |
|---|---|---|
| Bayley Scales of Infant Development (BSID) Cognitive Composite Score | 80.7 ± 17.4 | 81.1 ± 16.7 |
Infant died before discharge home.
| Participants | Donor Milk | Formula |
|---|---|---|
| Death before discharge | 24 | 18 |
| Survival to discharge | 215 | 226 |
Number of infants diagnosed with LOS
| Participants | Donor Milk | Formula |
|---|---|---|
| Missing data | 1 | 0 |
| Late-onset sepsis | 47 | 37 |
| No late-onset sepsis | 191 | 207 |
Number of infants diagnosed with NEC
| Participants | Donor Milk | Formula |
|---|---|---|
| Necrotizing enterocolitis | 10 | 22 |
| No necrotizing enterocolitis | 229 | 222 |
A composite outcome that measures the occurrence of death or NEC
| Participants | Donor Milk | Formula |
|---|---|---|
| Death or NEC | 27 | 33 |
| Survival without NEC | 212 | 211 |
Weight-for-age Z-scores were calculated at both baseline (study initiation) and study end (within one week of last study data collected) based on Fenton growth curves (2013). This outcome represents the change in weight-for-age Z-score during the course of the study (i.e., the Z-score at baseline was subtracted from the Z-score at study end). A value of 0 represents that the infant's weight-for-age Z-score is the same at the beginning and the end of the study. Positive values indicate the increase in the infant's weight-for-age Z-score during the study; negative values indicate the decrease in the infant's weight-for-age Z-score during the study.
| Change in Z-score (difference) | Donor Milk | Formula |
|---|---|---|
| Change in Weight-for-age Z-score During Study | -0.4 ± 0.9 | -0.1 ± 0.9 |
Mean motor composite score (standardized mean 100, range 44-155). Subjects who died prior to follow-up assigned the score of 44. (lower scores indicating greater impairment)
| Score on a scale | Donor Milk | Formula |
|---|---|---|
| Bayley Scales of Infant Development (BSID) Motor Composite Score | 80.3 ± 21.6 | 80.1 ± 19.9 |
Mean language composite score (standardized mean 100, range 46-155). Subjects who died prior to follow-up assigned the score of 46. (lower scores indicating greater impairment)
| Score on a scale | Donor Milk | Formula |
|---|---|---|
| Bayley Scales of Infant Development (BSID) Language Composite Score | 76.7 ± 19.6 | 75.8 ± 18.6 |
Number of infants with moderate or severe grade of cerebral palsy
| Participants | Donor Milk | Formula |
|---|---|---|
| Missing data | 3 | 2 |
| Moderate-severe cerebral palsy | 14 | 20 |
| No moderate-severe cerebral palsy | 171 | 177 |
Number of infants with NDI. NDI is defined as any of the following: Gross Motor Function Classification System score greater than or equal to 2, Bayley III cognitive or motor score less than 85 (1 standard deviation), Vision Impairment or Hearing impairment
| Participants | Donor Milk | Formula |
|---|---|---|
| Missing data | 11 | 10 |
| Neurodevelopmental impairment | 89 | 98 |
| No neurodevelopmental impairment | 88 | 91 |
Number of infants with profound impairment.
| Participants | Donor Milk | Formula |
|---|---|---|
| Missing data | 4 | 5 |
| No profound NDI | 154 | 152 |
| Profound NDI | 30 | 42 |
A composite outcome that measures the occurrence of death through 22-26 months or NDI.
| Participants | Donor Milk | Formula |
|---|---|---|
| Missing data | 11 | 10 |
| Death or NDI | 118 | 123 |
| Survival without NDI | 88 | 91 |
Collected over From day of randomization to Neonatal Research Network NRN infant status i.e., the first occurring of: discharge home, death, transfer, or 1 year following birth. For All-Cause Mortality, the time frame extends to follow-up (22-26 months) since deaths after discharge are included.. Non-serious events are listed at a 0.05% frequency threshold.
| Group | Deaths | Serious | Other |
|---|---|---|---|
| Donor Milk | 29/239 (12.1%) | 48/239 (20.1%) | 19/239 (7.9%) |
| Formula | 25/244 (10.2%) | 47/244 (19.3%) | 17/244 (7%) |
| Event | Donor Milk | Formula |
|---|---|---|
| Necrotising enterocolitis neonatalGastrointestinal disorders | 13/239 | 20/244 |
| Sepsis neonatalInfections and infestations | 14/239 | 14/244 |
| Neonatal intestinal perforationGastrointestinal disorders | 4/239 | 1/244 |
| Neonatal respiratory distressRespiratory, thoracic and mediastinal disorders | 1/239 | 4/244 |
| Neonatal respiratory failureRespiratory, thoracic and mediastinal disorders | 1/239 | 3/244 |
| PneumoperitoneumGastrointestinal disorders | 2/239 | 0/244 |
| Cytomegalovirus infectionInfections and infestations | 2/239 | 2/244 |
| Neonatal hypoglycemiaMetabolism and nutrition disorders | 2/239 | 0/244 |
| Neonatal pneumoniaInfections and infestations | 0/239 | 2/244 |
| Urinary tract infection fungalInfections and infestations | 0/239 | 2/244 |
| Event | Donor Milk | Formula |
|---|---|---|
| Sepsis neonatalInfections and infestations | 13/239 | 6/244 |
| Necrotising enterocolitis neonatalGastrointestinal disorders | 2/239 | 9/244 |
| Cytomegalovirus infectionInfections and infestations | 1/239 | 3/244 |
| Patent ductus arteriosusCongenital, familial and genetic disorders | 2/239 | 0/244 |
| Retinopathy of prematurityEye disorders | 1/239 | 0/244 |
| Gastroesophageal reflux diseaseGastrointestinal disorders | 1/239 | 0/244 |
| Feeding intoleranceMetabolism and nutrition disorders | 1/239 | 0/244 |
| Metabolic acidosisMetabolism and nutrition disorders | 1/239 | 0/244 |
| Neonatal hypoglycemiaMetabolism and nutrition disorders | 1/239 | 0/244 |
| ApneaRespiratory, thoracic and mediastinal disorders | 1/239 | 0/244 |
| Age, Continuous(weeks) | DONORMILK | FORMULA | Total |
|---|---|---|---|
| Mean | 26 ± 1.8 | 26.1 ± 1.6 | 26 ± 1.7 |
| Sex: Female, Male(Participants) | DONORMILK | FORMULA | Total |
|---|---|---|---|
| Female | 133 | 116 | 249 |
| Male | 106 | 128 | 234 |
| Race/Ethnicity, Customized(Participants) | DONORMILK | FORMULA | Total |
|---|---|---|---|
| Black | 126 | 121 | 247 |
| Missing | 4 | 1 | 5 |
| Other | 11 | 14 | 25 |
| White | 98 | 108 | 206 |
| Race/Ethnicity, Customized(Participants) | DONORMILK | FORMULA | Total |
|---|---|---|---|
| Hispanic or Latino | 37 | 31 | 68 |
| Missing | 1 | 1 | 2 |
| Not Hispanic or Latino | 201 | 212 | 413 |
| Weight of Infant at Birth(grams) | DONORMILK | FORMULA | Total |
|---|---|---|---|
| Mean | 848.6 ± 232 | 842.5 ± 219.8 | 845.5 ± 225.8 |
| Maternal Education(Participants) | DONORMILK | FORMULA | Total |
|---|---|---|---|
| College degree/more | 17 | 23 | 40 |
| High school degree | 82 | 93 | 175 |
| Less than high school degree | 64 | 57 | 121 |
| Missing | 19 | 17 | 36 |
| Partial college | 57 | 54 | 111 |
Documents are hosted by the registry — open the source record to download them.
Plan to share: Yes — NIH has had a long-standing policy to share and make available to the public the results and accomplishments of the activities that it funds. The NRN plans to share de-identified data after final publication in an NIH supported data repository such as the NICHD Data and Specimen Hub (https://dash.nichd.nih.gov)
This study is completed, as verified in Jun 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
NICHD Neonatal Research Network