Intervention Review

You have free access to this content

Multicomponent fortified human milk for promoting growth in preterm infants

  1. Carl A Kuschel1,*,
  2. Jane E Harding2

Editorial Group: Cochrane Neonatal Group

Published Online: 26 JAN 2004

Assessed as up-to-date: 28 AUG 2003

DOI: 10.1002/14651858.CD000343.pub2


How to Cite

Kuschel CA, Harding JE. Multicomponent fortified human milk for promoting growth in preterm infants. Cochrane Database of Systematic Reviews 2004, Issue 1. Art. No.: CD000343. DOI: 10.1002/14651858.CD000343.pub2.

Author Information

  1. 1

    The Royal Women's Hospital, Neonatal Services, Carlton, Victoria, Australia

  2. 2

    University of Auckland, Department of Paediatrics, Auckland, New Zealand

*Carl A Kuschel, Neonatal Services, The Royal Women's Hospital, 132 Grattan Street, Carlton, Victoria, 3053, Australia. carl.kuschel@rwh.org.au.

Publication History

  1. Publication Status: Edited (no change to conclusions)
  2. Published Online: 26 JAN 2004

SEARCH

 
Characteristics of included studies [ordered by study ID]
Carey 1987

MethodsRandomized study, single centre
Sample size estimate: No
Randomization method: Not stated
Blinding of randomization: Can't tell
Complete follow-up: No
Blinding of outcome measure: Can't tell


ParticipantsBirthweight <1500g
Exclusions: significant illness and congenital malformation
Enteral feeds of 150ml/kg/day for 48 hours
Number of Treatment Infants Randomized: Unknown
Number of Control Infants Randomized: Unknown


InterventionsMaternal milk supplemented with Protein, Calcium, and Phosphate (exact quantities not specified) vs. unsupplemented maternal milk.
Intervention ceased at approximately 4.5 weeks.
No information about supplemental vitamins.


OutcomesShort term growth
Biochemical indices of bone metabolism
Serum protein levels


NotesSome results expressed as means for the number of observations, rather than individuals.
Some infants excluded from results because of missing data.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?UnclearB - Unclear

Faerk 2000

MethodsRandomized study, two centres
Sample size estimate: Yes, based on BMC (at term corrected)
Method of randomization: Block randomization by sealed envelopes, stratified by birthweight
Blinding of randomization: Adequate
Blinding of intervention: Yes
Complete follow-up: No
Blinding of outcome measurement: Yes


ParticipantsGestational age <32 weeks
Exclusions: major congenital abnormalities


InterventionsMaternal or donor milk supplemented with Eoprotin (0.4g Protein, 35mg Calcium, and 17mg Phosphorus) per 100ml vs. maternal or donor milk supplemented with 10mg Phosphorus per 100ml
Target intake of 200ml/kg/day
All infants received Vitamin D 800iU per day
Intervention ceased when breast fed or 36 weeks corrected GA


OutcomesBMC (whole body)
Growth at term


Notes103 infants randomized to fortifier or phosphorus. 9 infants were withdrawn (6 fortifier) and DEXA scans were not considered technically suitable for a further 18 infants (9 fortifier). Only extractable data were BMC and rates of NEC. Further information about other outcomes requested from the authors.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Greer 1988

MethodsRandomized study, single centre
Sample size estimate: No
Method of randomization: Random number table
Blinding of randomization: Can't tell
Blinding of intervention: No
Complete follow-up: No
Blinding of outcome measurements: Can't tell


ParticipantsInfants <32 weeks GA or <1600g
Exclusions: major congenital abnormalities, congenital intrauterine infection, significant gastrointestinal disease, or seizures requiring anticonvulsant therapy.
Study commenced once infants tolerating full oral feeds.
Number of Treatment Infants Randomized: Unknown
Number of Control Infants Randomized: Unknown


InterventionsMaternal milk supplemented with 0.85g Protein, 90mg Calcium, and 45mg Phosphorus per 100ml (Ross Laboratories) vs. unsupplemented maternal milk
Enteral intake >120ml/kg/day and <200ml/kg/day - varied according to weight gain, appetite, and tolerance.
All infants received Vitamin D 400IU/day.


OutcomesShort term growth
BMC
Biochemical indices of bone metabolism
Urea
Total protein


NotesThe fortified group received significantly less milk (152 vs. 180 ml/kg/day) than the unsupplemented group. There was no significant difference in caloric intake.
Of 176 eligible infants, only 10 in the HMF arm and 10 in the unsupplemented arm completed the study.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?UnclearB - Unclear

Gross 1987 (1)

MethodsRandomized study, single centre.
Sample size estimate: No
Randomization method: Sealed envelopes
Blinding of randomization: Can't tell
Complete follow-up: No
Blinding of outcome measures: Can't tell
Two phase trial, referred to as Gross 1987 (1) and Gross 1987 (2)


ParticipantsBirthweight <1600g, AGA
Free from congenital anomalies and major disease, no supplemental oxygen, enteral feeds begun within one week of birth.
Number of Treatment Infants Randomized to Phase 1: 10
Number of Control Infants Randomized to Phase 1: 10
Number of Treatment Infants Randomized to Phase 2: 10
Number of Control Infants Randomized to Phase 2: 9


InterventionsPhase 1: Maternal or donor milk mixed ml-for-ml with Similac Special Care (Ross Laboratories) vs. unsupplemented human milk.
Phase 2: Maternal or donor milk with powdered HMF (Ross Laboratories) providing 0.9g protein, 87mg Calcium, and 50mg Phosphorus and electrolytes per 100ml vs. unsupplemented human milk.
Enteral intake 180ml/kg/day, adjusted according to weight gains.
Intervention ceased at a weight of 1800-2000g.
All infants received supplemental vitamins (including Vitamin D 400IU/day).


OutcomesShort term growth
Growth at 44 weeks postconceptional age
BMC
Biochemical indices of bone metabolism


NotesPhase 2 compared HMF to unsupplemented human milk, as well as a group who received preterm formula as in Phase 1. Only the control group and the group receiving powdered fortifier have been included in this review.
Results of short-term weight gain could not be included as they were expressed in g/day. Information has been requested from the author.
4 infants in the phase 2 study were withdrawn post-randomization because of feed intolerance (2 each from the HMF and formula supplementation arms).


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?UnclearB - Unclear

Gross 1987 (2)

MethodsSee Gross (1)


ParticipantsSee Gross (1)


InterventionsSee Gross (1)


OutcomesSee Gross (1)


NotesSee Gross (1)


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?UnclearB - Unclear

Kashyap 1990

MethodsRandomized study, single centre
Sample size estimate: No
Randomization Method: Sealed envelope
Blinding of randomization: Adequate
Blinding of intervention: Can't tell
Complete follow-up: No
Blinding of outcome measurement: Can't tell


ParticipantsBirthweight 900-1750g
Number of Treatment Infants Randomized: 30
Number of Control Infants Randomized: 36


InterventionsIntervention commenced once infants enterally feeding.
Maternal milk supplemented with (per kg/day) with 1.1g Protein, 3.7mmol Calcium, 2.11mmol Phosphate, and Sodium (Ross Laboratories) vs. unsupplemented maternal milk.
Feeds maintained at 180ml/kg/day
Outcomes assessed at 2200g.
All infants received supplemental Vitamin D (400 iU/day)


OutcomesShort term growth
Biochemical indices of bone metabolism
Nitrogen retention
Urea


NotesHigh attrition from both unsupplemented (22 of 36) and fortified (17 of 30) groups.
Reasons included NEC, PDA, and insufficient maternal milk. Some of these infants were placed into a third, non-random arm of the study in which they received supplemented term human milk. This arm has not been included in this review.
Other outcomes included plasma amino acid levels and skinfold thickness.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Lucas 1996

MethodsRandomized trial, two centres
Sample size estimate: Yes - neurodevelopmental outcome
Method of randomization: Sealed opaque envelope, stratified by birthweight (<1200 and >1200g)
Blinding of randomization: Yes
Blinding of intervention: No
Complete follow-up: Yes
Blinding of outcome measurement: Yes (neurodevelopmental outcome)


Participants<37 weeks and <1850g
Survival to 48-72 hours
Exclusions: major congenital abnormalities affecting neurodevelopmental outcomes, non-resident in UK
Number of Treatment Infants Randomized: 137
Number of Control Infants Randomized: 138


InterventionsMaternal milk supplemented with (per 100ml) 0.7g Protein (bovine), 2.73g Carbohydrate, 0.05g Fat, 90mg Calcium, and 45mg Phosphate, and electrolytes (Enfamil, Mead Johnson), vs. maternal milk supplemented with 15mg/100ml Phosphate.
Enteral intake 180ml/kg/day.
Intervention ceased at discharge or 2000g.
All infants received vitamins (including Vitamin D 260 IU/100ml).
Infants whose mothers could not provide sufficient milk were supplemented with a premature formula and not excluded from the analysis.


OutcomesNeurodevelopmental outcome at 9 and 18 months.
Short term growth.
Growth to 9 and 18 months.
Serum indices of bone metabolism
Urea


NotesThe authors felt it was not ethical to provide totally unsupplemented human milk so added 15mg Phosphate/100ml in control infants.
A large proportion of infants in both groups were supplemented with a premature formula when there was insufficient maternal milk.
Linear and head growth have been converted from mm/day to cm/week.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Modanlou 1986

MethodsRandomized study, single centre
Sample size estimate: No
Randomization Method: Sealed envelopes
Blinding of randomization: Adequate
Blinding of intervention: No
Complete follow-up: Yes
Blinding of outcome measurement: No


Participants1000-1500g AGA infants.
Exclusions: Ventilatory assistance >1 week, oxygen >10 days, diuretic therapy >3 days, not enterally fed by 14 days
Number of Treatment Infants Randomized: 20
Number of Control Infants Randomized: 19


InterventionsMaternal milk supplemented with (per 100ml) 0.7g Protein, 2.7g Carbohydrate, "trace" Fat, 60mg Calcium, and 33mg Phosphate, trace minerals, and electrolytes (preparation not specified), vs. unsupplemented maternal milk.
Enteral intake approximately 135-140ml/kg/day.
Intervention ceased at discharge or 1800g.
Vitamin D supplementation unknown.


OutcomesShort term growth
Indices of bone metabolism
BMC
Urea
Serum proteins


NotesA third arm of the study evaluated infants receiving a premature formula (not analysed in this review).
9 infants in the control arm and 10 in the HMF arm were withdrawn for insufficient maternal milk. 2 infants in the HMF arm were withdrawn for suspected NEC.
Infants whose mothers could not provide sufficient milk were supplemented with a term infant formula and not excluded from the analysis unless formula intake was >10% of weekly total.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Nicholl 1999

MethodsRandomized study, single centre
Sample size estimate: Yes, based on leg length by knemometry
Method of randomization: Sealed envelopes
Blinding of randomization: Yes
Blinding of intervention: No
Complete follow-up: Yes
Blinding of outcome: No


Participants<1500g
Enteral feeds at least 150ml/kg/day
Exclusions: Fluid restriction, diuretics, postnatal systemic steroid use, significant congenital abnormality


InterventionsMaternal (or pasteurised pooled donor milk) supplemented with (per 100ml) 0.7g Protein, 2.0g Carbohydrate, 30mg Calcium, 40mg Phosphorus, trace minerals and vitamins vs. unsupplemented maternal or donor milk
Intervention ceased when infants no longer required nasogastric feeds


OutcomesShort term growth (including knemometry)
Indices of bone metabolism (ALP)


NotesOne infant whose mother declined fortifier was included in the results of the non-fortified infants. A third group of infants received a preterm formula.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Pettifor 1989

MethodsQuasi-randomized trial, single centre.
Sample size estimate: No
Method of randomization: Allocation by maternal hospital number
Blinding of randomization: No
Blinding of intervention: Can't tell
Complete follow-up: No
Blinding of outcome measurement: Can't tell


Participants1000-1500g birthweight
No major congenital abnormalities, no ventilator requirement, no serious infection, no major metabolic disturbance, and enteral intake at least 45ml/kg/day on Day 4
Number of Treatment Infants Randomized: 53
Number of Control Infants Randomized: 47


InterventionsMaternal milk supplemented with (per 100ml) 0.05g Protein, 1.1g Carbohydrate, 0.26g Fat, 72.3mg Calcium, and 34mg Phosphate, and electrolytes and vitamins (HMF, Ross Laboratories) vs. unsupplemented maternal milk.
Maximal enteral intake 200ml/kg/day
Intervention ceased at approximately 1800g.
All infants received additional Vitamin D 750IU/day.


OutcomesShort term growth
BMC
Serum indices of bone metabolism
Serum albumin


Notes41 of 100 infants enrolled were withdrawn - 13 for insufficient maternal milk, 16 for significant illness [2 NEC, 11 respiratory causes, and 3 others], and 7 died. 5 others were excluded for incomplete data.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?NoC - Inadequate

Polberger 1989

MethodsRandomized trial. Single centre
Method of randomization: Sealed envelopes
Blinding of randomization: Adequate
Blinding of intervention: Adequate
Complete follow-up: No


ParticipantsAGA preterm infants <1500g
Enteral intake 170ml/kg/day
Exclusions: major illness, requirement for supplemental oxygen
Number of Treatment Infants Randomized: 9
Number of Control Infants Randomized: 8


InterventionsMaternal or donor milk supplemented with (per 100ml) 1.0g human milk protein and 1.0g human milk fat vs. unsupplemented human milk.
Intervention ceased at 2200g or when breast fed.
All infants were supplemented with vitamin E, folic acid, multivitamins, and extra Vitamin D (to a total of 1200 IU/day).
All infants received Calcium (30mg/kg/day) and Phosphate (20mg/kg/day) from day 5.
2mg elemental iron per kg per day given from 4 weeks.


OutcomesShort term growth
Plasma protein levels


NotesThis study is included in the multicomponent review because of the two nutritional supplements. This study had four arms - unsupplemented, vs. supplemented with protein, vs. supplemented with fat, vs. supplemented with fat and protein. Supplementation with fat alone and protein alone is discussed in separate reviews. 34 infants were enrolled in all four study groups, but six infants were excluded following randomization (reasons discussed in separate reviews). Results were reported for only 7 infants in each group.
Ultrafiltration of human milk to obtain protein may result in additional calcium being provided to the fortifier group.
There were large fluctuations in the energy intake for all four groups across the study.
Other outcomes, not included in the review, included plasma and urine amino acid levels.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Wauben 1998

MethodsRandomized trial. Single centre
Method of randomization: Block randomization, random number tables
Blinding of randomization: Yes
Blinding of intervention: No
Complete follow-up: No
Sample size estimate: Yes (BMC at term corrected)


ParticipantsAGA preterm infants <1800g, greater than 1 week old
Enteral intake 160ml/kg/day
Exclusions: Gastrointestinal disease, major congenital anomalies
Number of Treatment Infants Randomized: 15
Number of Control Infants Randomized: 16


InterventionsCommenced when maternal milk providing >80% of enteral intake
Maternal supplemented with (per 100ml) 0.37g human milk protein, 3.47g carbohydrate, 61mg Calcium, 44mg Phosphorus, as well as electrolytes, other minerals, and vitamins (including Vit D 472IU/day) (Wyeth-Ayerst, Toronto) vs. unsupplemented human milk.
Mean fluid intakes significantly greater in the control group (177 vs. 164ml/kg/day).
Intervention ceased at discharge or 38 weeks corrected GA, whichever occurred later.
Control infants supplemented with vitamin D (600IU/day) whilst receiving supplements. All infants received "standard" vitamin supplementation following discharge.


OutcomesShort term growth
Biochemical indices of bone metabolism
BMC (whole body)
Nitrogen retention


Notes6 infants (3 in each group) were withdrawn - 2 in the HMF group for feed intolerance (defined as abdominal distension - personal communication, Dr S Atkinson), and 1 for insufficient maternal milk. 2 infants in the control group developed chronic lung disease, and 1 developed a metabolic acidosis. These infants' results are not included in the paper.
A third arm of the study (not analysed in this review) evaluated infants who received a preterm formula.
Infants in the control arm were significantly lighter at birth, and significantly lighter and shorter at study entry and exit than the group receiving HMF.
Nutrient intakes were also measured.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?YesA - Adequate

Zuckerman 1994

MethodsQuasi-randomized trial. Single Centre
Method of randomization: by hospital number (even/odd)
Blinding of randomization: No
Blinding of intervention: No
Complete follow-up: No
Blinding of outcome measure: Can't tell
Sample size estimate: No


ParticipantsInfants <1200g, greater than 2 weeks old
Enteral intake maximum 200ml/kg/day
Exclusions: congenital abnormalities, infections, any diseases causing bone disease


InterventionsMaternal milk mixed in equal proportions with premature infant formula (Alprem, Nestle) to give supplements (per 100ml) of calcium 14.5mg, phosphorus 7mg, and protein 0.6g vs. unsupplemented human milk.
Intervention ceased at 1800g.
All infants received 800iU/day Vitamin D


OutcomesShort term growth
Serum indices of bone metabolism
Radiographic changes of metabolic bone disease
Serum vitamin D levels


NotesMainly SGA infants in each group.
3 infants in the control group were excluded because of incorrect feeding. 16 of 27 control infants completed follow-up. All 29 infants in the supplemented group completed the hospital study but only 14 completed post-discharge follow-up. Energy intakes were similar between both groups, despite the increased caloric density in the supplemented group.


Risk of bias

ItemAuthors' judgementDescription

Allocation concealment?NoC - Inadequate

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Boehm 1991No unsupplemented control group. The interventions contrasted were multicomponent fortifier (EOPROTIN) versus supplementation with human albumin, minerals, and sodium.

dos Santos 1997Comparison of multicomponent fortifiers.

Ewer 1996Did not include any prespecified outcomes of this review.

Gupta (unpublished)Randomized study (method not stated) comparing a commercial multicomponent fortifer vs. individualised supplements (fat, minerals, vitamins) vs. low-birth-weight formula +/- human milk. Concerns about randomization method and imbalance in characteristics of the study groups.

Lucas 1984Infants received a preterm formula or donor term human milk as a substitute for maternal milk. The supplements were given in varying amounts, with the median intake 42%. Over a third of mothers provided less than 20% of their infant's intake.

McClure 1996Did not include any prespecified outcomes of this review.

Metcalf 1994Comparison of multicomponent fortifiers.

Moyer-Mileur 1992Comparison of multicomponent fortifiers.

Plath 1988Did not include any prespecified outcomes.

Porcelli 2000Comparison of multicomponent fortifiers.

Reis 2000Comparison of multicomponent fortifiers.

Ronnholm 1982Unable to abstract data for those infants supplemented with fat and protein together vs. those who were unsupplemented.

Sankaran 1996Comparison of multicomponent fortifiers.

Schanler 1995Comparision of two multicomponent fortifiers in two cohorts over two separate time periods.

Venkataraman 1988Unable to abstract data from published report.

 
Comparison 1. Multicomponent fortification vs control (all trials)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Weight gain (g/kg/day)8450Mean Difference (IV, Fixed, 95% CI)2.33 [1.73, 2.93]

 2 Weight gain (g/day)355Mean Difference (IV, Fixed, 95% CI)4.74 [2.78, 6.70]

 3 Length gain (cm/week)8416Mean Difference (IV, Fixed, 95% CI)0.12 [0.07, 0.18]

 4 Head growth (cm/week)9428Mean Difference (IV, Fixed, 95% CI)0.12 [0.07, 0.16]

 5 Weight at 12 months (kg)125Mean Difference (IV, Fixed, 95% CI)Not estimable

 6 Length at 12 months (cm)125Mean Difference (IV, Fixed, 95% CI)-1.0 [-3.50, 1.50]

 7 Head circumference at 12 months (cm)125Mean Difference (IV, Fixed, 95% CI)0.10 [-2.22, 2.42]

 8 Weight at 18 months (kg)1245Mean Difference (IV, Fixed, 95% CI)-0.04 [-0.35, 0.27]

 9 Length at 18 months (cm)1245Mean Difference (IV, Fixed, 95% CI)-0.10 [-0.93, 0.73]

 10 Head circumference at 18 months (cm)1245Mean Difference (IV, Fixed, 95% CI)Not estimable

 11 Serum alkaline phosphatase (IU/l)8469Mean Difference (IV, Fixed, 95% CI)0.22 [-33.99, 34.44]

 12 Bone mineral content (mg/cm)279Mean Difference (IV, Fixed, 95% CI)8.30 [3.84, 12.76]

 13 Whole body bone mineral content (g)2101Mean Difference (IV, Fixed, 95% CI)1.65 [-1.65, 4.95]

 14 Mental development index at 18 months1245Mean Difference (IV, Fixed, 95% CI)2.20 [-3.35, 7.75]

 15 Psychomotor development index at 18 months1245Mean Difference (IV, Fixed, 95% CI)2.40 [-1.90, 6.70]

 16 Nitrogen retention (mg/kg/day)252Mean Difference (IV, Fixed, 95% CI)66.05 [35.28, 96.82]

 17 Hypercalcemia2263Risk Ratio (M-H, Fixed, 95% CI)1.18 [0.76, 1.82]

 18 Feed intolerance367Risk Ratio (M-H, Fixed, 95% CI)2.85 [0.62, 13.08]

 19 Necrotizing enterocolitis7640Risk Ratio (M-H, Fixed, 95% CI)1.33 [0.69, 2.54]

 20 Blood pH1275Mean Difference (IV, Fixed, 95% CI)Not estimable

 21 Blood urea (mmol/l)5350Mean Difference (IV, Fixed, 95% CI)0.27 [0.14, 0.40]

 22 Death8603Risk Ratio (M-H, Fixed, 95% CI)1.48 [0.66, 3.34]

 
Comparison 2. Multicomponent fortification vs control (trials without mineral supplementation of the control group)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Weight gain (g/kg/day)5136Mean Difference (IV, Fixed, 95% CI)3.62 [2.69, 4.55]

 2 Weight gain (g/day)355Mean Difference (IV, Fixed, 95% CI)4.74 [2.78, 6.70]

 3 Length gain (cm/week)5102Mean Difference (IV, Fixed, 95% CI)0.18 [0.08, 0.28]

 4 Head growth (cm/week)6114Mean Difference (IV, Fixed, 95% CI)0.14 [0.09, 0.20]

 5 Serum alkaline phosphatase (IU/l)6169Mean Difference (IV, Fixed, 95% CI)-43.24 [-98.29, 11.81]

 6 Bone mineral content (mg/cm)279Mean Difference (IV, Fixed, 95% CI)8.30 [3.84, 12.76]

 7 Nitrogen retention (mg/kg/day)127Mean Difference (IV, Fixed, 95% CI)82.5 [32.53, 132.47]

 8 Feed intolerance119Risk Ratio (M-H, Fixed, 95% CI)4.55 [0.25, 83.70]

 9 Necrotizing enterocolitis4261Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.44, 2.20]

 10 Blood urea (mmol/l)361Mean Difference (IV, Fixed, 95% CI)0.96 [0.56, 1.36]

 11 Death6297Risk Ratio (M-H, Fixed, 95% CI)13.33 [0.78, 227.34]