CONTEXT:

Human and bovine colostrum (HBC) administration has been linked to beneficial effects on morbidity and mortality associated with necrotizing enterocolitis (NEC).

OBJECTIVES:

To determine the effectiveness and safety of HBC for reducing NEC, mortality, sepsis, time to full-feed and feeding intolerance in preterm infants.

DATA SOURCES:

We conducted searches through Medline, Embase, Cumulative Index of Nursing and Allied Health Literature, Cochrane Central Register of Controlled Trials, and gray literature.

STUDY SELECTION:

Randomized controlled trials comparing human or bovine colostrum to placebo.

DATA EXTRACTION:

Two reviewers independently did screening, review, and extraction.

RESULTS:

Eight studies (385 infants) proved eligible. In comparison with placebo, HBC revealed no effect on the incidence of severe NEC (relative risk [RR]: 0.99; 95% confidence interval [CI] 0.48 to 2.02, I2 = 2.2%; moderate certainty of evidence), all-cause mortality (RR: 0.88; 95% CI 0.39 to 1.82, I2 = 0%; moderate certainty), culture-proven sepsis (RR: 0.78; 95% CI 0.53 to 1.14, I2 = 0%; moderate certainty), and feed intolerance (RR: 0.97; 95% CI 0.37 to 2.56, I2 = 55%; low certainty). HBC revealed a significant effect on reducing the mean days to reach full enteral feed (mean difference: −3.55; 95% CI 0.33 to 6.77, I2 = 41.1%; moderate certainty). The indirect comparison of bovine versus human colostrum revealed no difference in any outcome.

LIMITATIONS:

The number of patients was modest, whereas the number of NEC-related events was low.

CONCLUSIONS:

Bovine or human colostrum has no effect on severe NEC, mortality, culture-proven sepsis, feed intolerance, or length of stay. Additional research focused on the impact on enteral feeding may be needed to confirm the findings on this outcome.

Preterm birth is defined by the World Health Organization (WHO) as live births before 37 weeks of pregnancy. Preterm birth complications are the leading cause of death among children <5 years of age and were responsible for ∼1 million deaths in 2015.1 Extremely premature (birth weight <1250 g) newborns have substantial mortality and morbidity, often resulting from infectious morbidities including late-onset sepsis and necrotizing enterocolitis (NEC).2 

NEC is a multifactorial and life-threatening inflammation of the gastrointestinal tract and the most frequent surgical emergency in neonates. The mechanism of NEC is poorly understood, but researchers suggest that factors such as bowel hypoperfusion, use of antibiotics, and the delay to start enteral feeding seem to promote intestinal atrophy and abnormal bacterial intestinal colonization, which are crucial features of the disease.3 Despite the significant advances in neonatal care, morbidity and mortality related to NEC have remained unchanged for decades. The rate of NEC-related mortality is reported to be 20% to 30%, whereas in infants in need of surgery, the rate could be up to 50%.4,6 Furthermore, infants recovering from NEC are at increased risk for microcephaly, short-bowel syndrome, serious neurodevelopmental delays, and functional disabilities.7,8 

Mother’s milk has many immune and trophic factors (such as growth factors, cytokines, lactoferrin, lysozymes, and immunoglobulins)9,10 that may protect newborns from infection and might have an effect on the gastrointestinal tract maturation. Mother’s milk feedings have been linked with a reduced incidence of several prematurity-specific morbidities including NEC, bacteremia, and enteral feed intolerance for premature infants.11 Colostrum is the first milk produced by the mammary when the tight junctions in the mammary epithelium are open.12 It has been found that the immune protective factors are more highly concentrated in the colostrum of mothers delivering premature infants than in those who give birth at term.13,14 This, in turn, suggests that immune components in colostrum may provide infants with protection against infection.15 

The method of colostrum administration has been studied in 2 different ways: oropharyngeal and enteral. Human colostrum has been administered in small volumes directly into the buccal cavity of intubated premature infants.16 Likewise, commercially available bovine colostrum has been administered via enteral along with the enteral feeding. Bovine colostrum also contains protective factors, which have substantial homology to their human counterparts.17 

Randomized trials of both bovine and human colostrum in comparison with placebo have been performed in preterm infants to assess their potentially protective effects. To date, this evidence has not been systematically summarized. We have therefore conducted a systematic review to determine the effectiveness and safety of human and bovine colostrum in preterm infants for decreasing NEC-related outcomes, including mortality and morbidities.

The protocol for this systematic review is registered with PROSPERO: CRD42018085566.

We searched Medline, Embase, the Cumulative Index to Nursing and Allied Health Literature, and the Cochrane Central Register of Controlled Trials for relevant published randomized controlled trials (RCTs) (search strategy is provided in the Supplemental Information). We did not apply language or publication status restrictions. We reviewed reference lists from eligible trials and related reviews for additional eligible RCTs and searched ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform for ongoing or unpublished trials.

Reviewers (B.S., I.D.F., and R.L.M.) independently screened the titles and abstracts of all identified studies by using a priori selection criteria. Subsequently, reviewers independently assessed eligibility of the full texts of potentially eligible studies. Reviewers resolved discrepancies through discussion.

We included RCTs in which researchers compared oropharyngeal or enteral administration of human or bovine colostrum to preterm infants (gestational age <37 weeks) within the first week of life irrespective of when enteral feeding was initiated, the type of milk used for enteral feeding or the feed advancement regimen with placebo, standard clinical care, or standard clinical care plus placebo. Standard clinical care typically includes parenteral nutrition or feeding of the infant’s own mother’s milk, donor’s milk, or preterm formula milk.

Our outcomes of interest were as follows: (1) NEC: stage II or more based on Bell’s criteria18,19; (2) NEC-related mortality; (3) all-cause mortality; (4) culture-proven sepsis; (5) patent ductus arteriosus; (6) intraventricular hemorrhage; (7) duration of hospitalization; (8) weight gain; and (9) incidence of adverse events (as reported by authors).

Reviewers (B.S., I.D.F., R.L.M., A.M.Z., and D.Z.) extracted the following data, independently and in duplicate: (1) general study information (author’s name, publication year, study design, and number of arms), (2) population-related information (birth weight, gestational age, Apgar score at 1 and 5 minutes, percentage of cesarean deliveries, and percentage of infants small for gestational age), (3) feeding details (feeding protocol and percentage of infants receiving mother’s milk or formula milk), (4) details on the intervention and comparison (type of colostrum, time of initiation, dose, duration of therapy, and type of control group), and (5) outcomes as listed above.

Two reviewers independently assessed risk of bias by using a modified Cochrane risk of bias instrument for RCTs20,21 that addresses the following issues: random sequence generation, allocation concealment, blinding of study participants, health care providers, and outcome assessors and/or adjudicators, incomplete outcome data, and other potential sources of bias.

To assess the certainty of evidence (CoE), we used the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach for evidence assessment that classifies evidence as high, moderate, low, or very low quality on the basis of considerations of risk of bias, inconsistency, indirectness, imprecision, and publication bias.22 We resolved disagreements between reviewers in data extraction, and assessments of risk of bias or CoE by discussion and, if needed, by third party adjudication. We used the GRADE profiler (GRADEpro GDT; https://gradepro.org/) to generate the GRADE summary of findings table.

For dichotomous outcomes, we calculated the relative risk (RR) and its corresponding 95% confidence interval (CI) and calculated the absolute effect by multiplying the RR and its CI with the estimated baseline risk. The median of the placebo group of included RCTs provided the baseline risk. For continuous outcomes, we calculated the mean difference and its corresponding 95% CI.

Statistical heterogeneity was determined by using the Q statistic and I2. We used the DerSimonian–Laird random-effects model for the meta-analysis of all outcomes. Regardless of the observed statistical heterogeneity, we conducted the following prespecified subgroup analyses: birth weight, assuming larger effects for infants with larger birth weights; gestational age, assuming larger effects for infants with higher gestational age; type of colostrum, assuming a larger effect for infants receiving human colostrum; and risk of bias, assuming larger effects for studies at high risk of bias. For subgroup analysis, we tested for interaction by using a χ2 significance test, when each subgroup was represented by at least 2 studies.23 We performed univariate and multivariate metaregression to assess the effects of birth weight, gestational age, duration of therapy, Apgar score at 1 and 5 minutes, percentages of cesarean deliveries, and publication year on the treatment effect. We planned to examine publication bias by using funnel plots for outcomes in which 10 or more studies were available.24 We performed indirect meta-analysis using the frequentist approach to compare the effect of human versus bovine colostrum. A conventional meta-analysis combines effect estimates from direct comparisons of interventions (ie, evidence from trials with head-to-head comparison of interventions). Indirect comparisons are made by looking at the impact of the interventions of interest versus a third intervention, a common comparator (in this case, inferring the effect of bovine versus human colostrum through trials of bovine colostrum versus placebo, and human colostrum versus placebo). Indirect meta-analysis is a relatively new technique and is intended for situations in which there is no direct evidence and comparisons are made pairwise. More details on the statistical methods can be found in Miladinovic et al.25 Data were analyzed by using Stata software version 14.2 (Stata Corp, College Station, TX).

We identified 1075 titles and abstracts through our literature search, of which 26 proved potentially eligible for full-text evaluations, and 18 were excluded for the following reasons: (1) not randomized trials (n = 8), (2) colostrum was not used as the intervention (n = 5), (3) not preterm infants (n = 2), and no relevant outcome was reported (n = 3). Figure 1 provides the details of the study selection process.

FIGURE 1

Study selection flowchart.

FIGURE 1

Study selection flowchart.

We included 8 RCTs that proved eligible, enrolling 394 individuals. The intervention in 6 studies13,14,16,26,28 was human colostrum, and in 2 studies the intervention was bovine colostrum.29,30 In 2 studies, researchers enrolled preterm infants with a birth weight of ≤1.0 kg or gestational age <28 weeks.13,14 In 5 studies, researchers enrolled preterm infants with a birth weight of ≤1.5 kg or gestational age <32 weeks.16,26,29 In 1 study, researchers enrolled infants with a birth weight between 1.0 to 1.8 kg and gestational age ≥28 weeks.30Table 1 presents details of included trials.

TABLE 1

Characteristics of Studies Included in the Systematic Review

StudyMean BW, gMean GA, wkNo. Randomized (Intervention, Control)Type of ColostrumDuration of Therapy or DoseTime of InitiationMethod of Colostrum Administration
Rodriguez et al13  842.0 26.3 9, 6 Human 0.2 mL every 2 h for 2 d Within 48 h of life Oropharyngeal 
Lee et al14  815.0 26.8 24, 24 Human 0.2 mL every 3 h for 3 d 48–96 h after birth Oropharyngeal 
Sohn et al16  1053.5 27.0 6, 6 Human 0.2 mL every 2 h for 46 h Median age of 39 h (range 32–87) Oropharyngeal 
Balachandran et al29  1202.9 29.9 43, 43 Bovine 1.2–2.0 g 4 times per d until discharge or death or d 21 of life In the first 96 h of life Orogastric tube 
Romano-Keeler et al27  1219.5 25.5 48, 51 Human 0.2 mL every 6 h for 5 d In the first 48 h of life Oropharyngeal 
Glass et al26  1109.0 28.4 17, 13 Human 0.2 mL every 3 h for 7 d In the first 48 h of life Oropharyngeal 
Zhang et al28  1244.5 30.2 32, 32 Human 0.2 mL every 4 h for 7 d Between d 2–4 of life Oropharyngeal 
Juhl et al30  1487.5 30.5 21, 19 Bovine Volume limited by a pre-set total protein intake of 4.5 g/kg per d for 10–14 d In the first 48 h of life Enteral 
StudyMean BW, gMean GA, wkNo. Randomized (Intervention, Control)Type of ColostrumDuration of Therapy or DoseTime of InitiationMethod of Colostrum Administration
Rodriguez et al13  842.0 26.3 9, 6 Human 0.2 mL every 2 h for 2 d Within 48 h of life Oropharyngeal 
Lee et al14  815.0 26.8 24, 24 Human 0.2 mL every 3 h for 3 d 48–96 h after birth Oropharyngeal 
Sohn et al16  1053.5 27.0 6, 6 Human 0.2 mL every 2 h for 46 h Median age of 39 h (range 32–87) Oropharyngeal 
Balachandran et al29  1202.9 29.9 43, 43 Bovine 1.2–2.0 g 4 times per d until discharge or death or d 21 of life In the first 96 h of life Orogastric tube 
Romano-Keeler et al27  1219.5 25.5 48, 51 Human 0.2 mL every 6 h for 5 d In the first 48 h of life Oropharyngeal 
Glass et al26  1109.0 28.4 17, 13 Human 0.2 mL every 3 h for 7 d In the first 48 h of life Oropharyngeal 
Zhang et al28  1244.5 30.2 32, 32 Human 0.2 mL every 4 h for 7 d Between d 2–4 of life Oropharyngeal 
Juhl et al30  1487.5 30.5 21, 19 Bovine Volume limited by a pre-set total protein intake of 4.5 g/kg per d for 10–14 d In the first 48 h of life Enteral 

BW, birth wt; GA, gestational age; NR, not reported.

Among the included studies, 5 out of 8 RCTs demonstrated concerns for high risk of bias due to allocation concealment, blinding, and outcome reporting.16,26,28,30 One study had issues in incomplete outcome reporting,28 4 studies had issues in blinding of participants and/or outcome assessors,16,26,27,30 and 2 studies had issues in concealing the treatment allocation (the risk of bias assessment is described in Supplemental Figs 6 and 7).27,28 

The meta-analysis from 7 studies in which researchers reported the incidence of NEC stage II or more13,14,16,26,29 revealed no difference among infants who received colostrum versus those who received a placebo or usual care group (RR 0.99, 95% CI: 0.48 to 2.02; I2 = 2.2%; moderate CoE; Fig 2, Table 2). In 1 study, researchers reporting this outcome used bovine colostrum. Tests of interaction revealed no evidence of any subgroup effect (Table 3). The univariate metaregression confirmed the results of subgroup analysis (Supplemental Tables 4 through 6). The indirect comparison of human and bovine colostrum revealed no difference (Supplemental Table 7).

FIGURE 2

Forest plot revealing RR for NEC stage II or more (based on Bell’s criteria) for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled RR was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis.

FIGURE 2

Forest plot revealing RR for NEC stage II or more (based on Bell’s criteria) for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled RR was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis.

TABLE 2

Summary of Findings

Outcome (Studies), No. ParticipantsRelative Effect (95% CI)Anticipated Absolute Effects (95% CI)Certainty
Risk With No ColostrumRisk With ColostrumDifference
NEC stage II or more (7 RCTs) Risk ratio: 0.99 (0.48 to 2.02) 9.4% 9.3% (4.5% to 18.9%) 0.1% fewer (4.9 fewer to 9.5 more) ⨁⨁⨁◯ 
 No. participants: 345 Moderatea,b 
Culture-proven sepsis (8 RCTs) Risk ratio: 0.78 (0.53 to 1.14) 21.1% 16.4% (11.2% to 24.0%) 4.6% fewer (9.9 fewer to 2.9 more) ⨁⨁⨁◯ 
 No. participants: 385 Moderatea,b 
All-cause mortality (7 RCTs) Risk ratio: 0.88 (0.39 to 1.82) 7.4% 6.5% (2.9% to 13.5%) 0.9% fewer (4.5 fewer to 6.1 more) ⨁⨁⨁◯ 
 No. participants: 330 Moderatea,b 
Feed intolerance (2 RCTs) Risk ratio: 0.97 (0.37 to 2.56) 43.8% 42.4% (16.2% to 100.0%) 1.3% fewer (27.6 fewer to 68.3 more) ⨁⨁◯◯ 
 No. participants: 70 Lowa,b,c 
NEC-related mortality (4 RCTs) Not estimable 0.0% 0.0% (0.0% to 0.0%) 0.0% fewer (0 fewer to 0 fewer) ⨁⨁⨁◯ 
 No. participants: 105 Moderatea,d 
Duration of hospital stay (3 RCTs) — The mean duration of hospital stay was 79.0 d — MD 1.26 d more (13.73 fewer to 16.26 more) ⨁⨁◯◯ 
 No. participants: 160 Lowa,b,e 
Time to reach full enteral feed (6 RCTs) — The mean time to reach full enteral feed was 22.1 d — MD 3.55 d fewer (0.33 fewer to 6.77 fewer) ⨁⨁⨁◯ 
 No. participants: 285 Moderatea,f 
Outcome (Studies), No. ParticipantsRelative Effect (95% CI)Anticipated Absolute Effects (95% CI)Certainty
Risk With No ColostrumRisk With ColostrumDifference
NEC stage II or more (7 RCTs) Risk ratio: 0.99 (0.48 to 2.02) 9.4% 9.3% (4.5% to 18.9%) 0.1% fewer (4.9 fewer to 9.5 more) ⨁⨁⨁◯ 
 No. participants: 345 Moderatea,b 
Culture-proven sepsis (8 RCTs) Risk ratio: 0.78 (0.53 to 1.14) 21.1% 16.4% (11.2% to 24.0%) 4.6% fewer (9.9 fewer to 2.9 more) ⨁⨁⨁◯ 
 No. participants: 385 Moderatea,b 
All-cause mortality (7 RCTs) Risk ratio: 0.88 (0.39 to 1.82) 7.4% 6.5% (2.9% to 13.5%) 0.9% fewer (4.5 fewer to 6.1 more) ⨁⨁⨁◯ 
 No. participants: 330 Moderatea,b 
Feed intolerance (2 RCTs) Risk ratio: 0.97 (0.37 to 2.56) 43.8% 42.4% (16.2% to 100.0%) 1.3% fewer (27.6 fewer to 68.3 more) ⨁⨁◯◯ 
 No. participants: 70 Lowa,b,c 
NEC-related mortality (4 RCTs) Not estimable 0.0% 0.0% (0.0% to 0.0%) 0.0% fewer (0 fewer to 0 fewer) ⨁⨁⨁◯ 
 No. participants: 105 Moderatea,d 
Duration of hospital stay (3 RCTs) — The mean duration of hospital stay was 79.0 d — MD 1.26 d more (13.73 fewer to 16.26 more) ⨁⨁◯◯ 
 No. participants: 160 Lowa,b,e 
Time to reach full enteral feed (6 RCTs) — The mean time to reach full enteral feed was 22.1 d — MD 3.55 d fewer (0.33 fewer to 6.77 fewer) ⨁⨁⨁◯ 
 No. participants: 285 Moderatea,f 

The table contains human or bovine colostrum compared with placebo or usual care for prevention of morbidity and mortality in preterm infants. Population: preterm infants (gestational age <37 wk); intervention: colostrum (human or bovine); comparator: no colostrum. GRADE Working Group grades of evidence. High certainty: we are highly confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: We have little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect. The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). MD, mean difference; —, not applicable.

a

Although some studies were at risk for bias because of allocation concealment and blinding, subgroup analyses did not suggest that any heterogeneity due to risk of bias was introduced.

b

The 95% CI includes values suggesting substantial benefit and values suggesting substantial harm.

c

I2 value is 56%, suggesting some heterogeneity; however, exploratory analyses did not highlight the source.

d

No events were reported for either arm.

e

I2 value is 41%, demonstrating potential heterogeneity.

f

I2 value is 38%, potential heterogeneity; however, it is unlikely related to bias.

TABLE 3

Results of the Meta-analysis and Subgroup Analysis of RCTs Assessing the Effects of Colostrum

Outcome and SubgroupsaNo. TrialsES95% CINo. ParticipantsI2P for Interaction
LowerUpperInterventionControl
NEC (stage ≥II)         
 Human 0.83 0.39 1.75 131 128 0.0 b 
 Bovine 4.00 0.47 34.34 43 43 — 
 BW <1000 g 0.67 0.22 2.07 33 30 — .488 
 BW ≥1000 g 1.26 0.49 3.27 141 141 9.4 
 Low risk of bias 1.13 0.51 2.49 99 92 0.0 .752 
 High (due to allocation concealment) 0.62 0.06 6.06 75 79 52.5 
 Low risk of bias 1.29 0.23 7.30 76 73 54.0 .871 
 High (due to blinding) 0.98 0.35 2.70 98 98 3.6 
 Total 0.99 0.48 2.02 174 141 2.2 — 
Mortality         
 Human 0.74 0.27 2.06 104 100 0.0 .765 
 Bovine 1.00 0.31 3.21 64 62 — 
 BW <1000 g 0.86 0.15 4.80 33 30 33.7 .922 
 BW ≥1000 g 0.99 0.36 2.78 135 132 0.0 
 Low risk of bias 0.78 0.35 1.72 120 111 0.0 b 
 High (due to allocation concealment) 3.18 0.13 76.31 48 51 — 
 Low risk of bias 0.83 0.36 1.88 76 73 0.0 .947 
 High (due to blinding) 0.98 0.11 8.91 92 89 1.9 
 Total 0.84 0.39 1.82 168 162 0.0 — 
Culture-proven sepsis         
 Human 0.79 0.51 1.23 131 128 0.0 .873 
 Bovine 0.73 0.33 1.61 64 62 0.0 
 BW <1000 g 1.22 0.24 6.20 33 30 43.8 .744 
 BW ≥1000 g 0.72 0.41 1.24 162 160 0.0 
 Low risk of bias 0.83 0.55 1.25 120 111 0.0 .331 
 High (due to allocation concealment) 0.47 0.16 1.43 75 79 0.0 
 Low risk of bias 0.80 0.50 1.26 76 73 0.0 .839 
 High (due to blinding) 0.73 0.36 1.48 119 117 0.0 
 Total 0.78 0.53 1.14 195 190 0.0 — 
Time to reach full feed, d         
 Human −2.87 −6.02 0.28 123 122 34.3 b 
 Bovine −9.60 −19.46 0.26 21 19 — 
 BW <1000 g −4.55 −14.33 5.23 31 30 72.5 .794 
 BW ≥1000 g −3.17 −6.65 0.31 113 111 29.8 
 Low risk of bias −4.19 −9.40 1.03 69 62 48.8 .854 
 High (due to allocation concealment) −3.47 −9.06 2.13 75 79 48.1 
 Low risk of bias −4.55 −14.33 5.23 31 30 72.5 .794 
 High (due to blinding) −3.17 −6.65 0.31 113 111 29.8 
 Total −3.55 −6.77 −0.33 144 141 38.3 — 
Outcome and SubgroupsaNo. TrialsES95% CINo. ParticipantsI2P for Interaction
LowerUpperInterventionControl
NEC (stage ≥II)         
 Human 0.83 0.39 1.75 131 128 0.0 b 
 Bovine 4.00 0.47 34.34 43 43 — 
 BW <1000 g 0.67 0.22 2.07 33 30 — .488 
 BW ≥1000 g 1.26 0.49 3.27 141 141 9.4 
 Low risk of bias 1.13 0.51 2.49 99 92 0.0 .752 
 High (due to allocation concealment) 0.62 0.06 6.06 75 79 52.5 
 Low risk of bias 1.29 0.23 7.30 76 73 54.0 .871 
 High (due to blinding) 0.98 0.35 2.70 98 98 3.6 
 Total 0.99 0.48 2.02 174 141 2.2 — 
Mortality         
 Human 0.74 0.27 2.06 104 100 0.0 .765 
 Bovine 1.00 0.31 3.21 64 62 — 
 BW <1000 g 0.86 0.15 4.80 33 30 33.7 .922 
 BW ≥1000 g 0.99 0.36 2.78 135 132 0.0 
 Low risk of bias 0.78 0.35 1.72 120 111 0.0 b 
 High (due to allocation concealment) 3.18 0.13 76.31 48 51 — 
 Low risk of bias 0.83 0.36 1.88 76 73 0.0 .947 
 High (due to blinding) 0.98 0.11 8.91 92 89 1.9 
 Total 0.84 0.39 1.82 168 162 0.0 — 
Culture-proven sepsis         
 Human 0.79 0.51 1.23 131 128 0.0 .873 
 Bovine 0.73 0.33 1.61 64 62 0.0 
 BW <1000 g 1.22 0.24 6.20 33 30 43.8 .744 
 BW ≥1000 g 0.72 0.41 1.24 162 160 0.0 
 Low risk of bias 0.83 0.55 1.25 120 111 0.0 .331 
 High (due to allocation concealment) 0.47 0.16 1.43 75 79 0.0 
 Low risk of bias 0.80 0.50 1.26 76 73 0.0 .839 
 High (due to blinding) 0.73 0.36 1.48 119 117 0.0 
 Total 0.78 0.53 1.14 195 190 0.0 — 
Time to reach full feed, d         
 Human −2.87 −6.02 0.28 123 122 34.3 b 
 Bovine −9.60 −19.46 0.26 21 19 — 
 BW <1000 g −4.55 −14.33 5.23 31 30 72.5 .794 
 BW ≥1000 g −3.17 −6.65 0.31 113 111 29.8 
 Low risk of bias −4.19 −9.40 1.03 69 62 48.8 .854 
 High (due to allocation concealment) −3.47 −9.06 2.13 75 79 48.1 
 Low risk of bias −4.55 −14.33 5.23 31 30 72.5 .794 
 High (due to blinding) −3.17 −6.65 0.31 113 111 29.8 
 Total −3.55 −6.77 −0.33 144 141 38.3 — 

BW, birth weight (mean, as reported in RCTs); ES, effect estimate (weighted mean difference for time to reach full enteral feed, and RR for the remaining outcomes); —, not applicable.

a

We did not perform any subgroup analysis for duration of hospital stay and incidence of feeding intolerance because there were ≤3 studies containing reports of those outcomes.

b

Because of the small number of trials, we did not perform a statistical test of interaction between the 2 groups.

In the 4 RCTs in which researchers reported NEC-related mortality,13,14,16,26 no infant died as a result of developing NEC in colostrum or placebo or usual care group (moderate CoE; Table 2).

All-cause mortality was reported in 7 RCTs,13,14,16,26,27,29,30 and the results of meta-analysis revealed no effect for colostrum compared with placebo or usual care (RR 0.84, 95% CI: 0.39 to 1.82; I2 = 0%; moderate CoE; Fig 3, Table 2). There was no evidence of any subgroup effect (Table 3, Supplemental Table 6). In 2 studies, researchers reporting this outcome used bovine colostrum. The indirect comparison of human and bovine colostrum revealed no difference (Supplemental Table 7).

FIGURE 3

Forest plot revealing RR for mortality for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled RR was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis.

FIGURE 3

Forest plot revealing RR for mortality for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled RR was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis.

In the 8 studies in which researchers reported on culture-proven sepsis,13,14,16,26,30 the risk of developing sepsis for infants who received colostrum was 22% less than those who received a placebo or usual care (RR 0.78, 95% CI: 0.53 to 1.14; I2 = 0.0%; moderate CoE, Fig 4, Table 2). We found no evidence of subgroup effect for this outcome (Table 3, Supplemental Table 6). In 2 studies, researchers reporting this outcome used bovine colostrum. The indirect comparison of human and bovine colostrum revealed no difference (Supplemental Table 7).

FIGURE 4

Forest plot revealing RR for culture-proven sepsis for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled RR was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis.

FIGURE 4

Forest plot revealing RR for culture-proven sepsis for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled RR was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis.

Of the 2 studies in which feeding intolerance was reported, researchers for 1 study used human colostrum26 and the other researchers used bovine colostrum.30 None of these researchers reported a benefit for using colostrum, and the pooled estimate was not significant (RR 0.97, 95% CI: 0.37 to 2.56; I2 = 55.5%; low CoE; Supplemental Fig 8, Table 2).

Time to reach full enteral feeding was reported in 6 studies.13,14,26,28,30 On average, infants receiving colostrum reached full feed 3.5 days earlier (95% CI: −0.33 to −6.77; I2 = 38.3%; moderate CoE; Fig 5, Table 2). We found no evidence of subgroup effect for this outcome (Table 3, Supplemental Table 6). The results of indirect meta-analysis revealed larger effect for human colostrum, but the difference was not statistically significant (mean difference −7.1 days, 95% CI: −18.2 to 3.9; Table 3).

FIGURE 5

Forest plot revealing the weighted mean difference in mean time to reach full enteral feed for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled mean difference was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis. WMD, weighted mean difference.

FIGURE 5

Forest plot revealing the weighted mean difference in mean time to reach full enteral feed for colostrum versus placebo groups. Horizontal bars denote 95% CIs. Studies are represented as squares centered on the point estimate of the result of each study. The area of the square represents the weight given to the study in the meta-analysis. The pooled mean difference was calculated by DerSimonian-Laird random-effects model. The diamond represents the overall estimated effect and its 95% CI in total (center line of diamond, dashed line). The solid vertical line is the line of no effect. Weights are from random effects analysis. WMD, weighted mean difference.

In 3 studies, researchers reported on duration of hospital stay.13,14,27 The results of meta-analysis didn’t reveal any significant difference between the duration of hospitalization between the infants who received colostrum and those who received a placebo or usual care (mean difference 1.3 days, 95% CI: −13.7 to 16.3; I2 = 41.1%; low CoE; Supplemental Fig 9 and Table 2). Juhl et al30 reported 4 cases of intraventricular hemorrhage (3 grade I and 1 grade II) in the placebo or usual care group (21.1%), whereas no infants in the bovine colostrum group were reported to develop intraventricular hemorrhage. None of the 5 studies in which researchers assessed the occurrence of adverse events contained reports of any serious adverse events associated with the intervention.13,14,26,27,30 

To our knowledge, this is the first systematic review and meta-analysis of human and bovine colostrum administration in preterm infants for NEC, mortality, and related health outcomes. In our review, we synthesized the evidence from 8 RCTs, including 394 infants, to describe the effect of bovine and human colostrum administration in preterm infants. On the basis of low to moderate CoE, we found that colostrum has no effect on mortality or morbidities in preterm infants. Nonetheless, colostrum administration resulted in less time to get full enteral feeding (moderate CoE). We explored and found no evidence of subgroup effect for any of the outcomes, and our univariate metaregression was not significant for any of the covariates (birth weight, gestational age, Apgar score at 1 and 5 minutes, proportion of infants delivered in a cesarean delivery procedure, and duration of treatment). These findings reveal that there is no effect of colostrum on NEC-related outcomes.

Although we did not find differences in culture-proven sepsis, there was a trend toward a positive effect. This effect, if present, may be related to the immune effects of colostrum. The lack of effect may be related to the lack of power given the relative low number of subjects studied. Additional clinical trials will increase the number of patients and may change the results for this outcome.

Colostrum contains numerous protective immune and trophic factors that seem to play an important role in the first days of extrauterine life.9,11 Mother’s milk provides the ideal form of administration of colostrum. However, considering that the content of bovine colostrum has been described as similar in many components to the human colostrum,31 when the latter is not available, bovine colostrum might be considered a good alternative.

Both types of colostrum have been used in different ways. Human colostrum has been administered by using an oropharyngeal method, whereas the bovine colostrum administration has been administered enterally. The rationale behind the oropharyngeal administration is that, because of the gastric tube feeding, preterm infants are not being exposed to the effect of protective bio-factors on the oropharyngeal associated lymphoid tissue to obtain an effect on their immune system.13 Thus, the administration of small amounts of colostrum in the oral mucosa aims to provide that exposure and to produce a positive impact on the immune system and therefore on the incidence of NEC-related health outcomes. Although this rationale has not been specifically described for bovine colostrum, it may be possible that this type of colostrum may also have an effect on the lymphoid tissue if administered via oropharyngeal.

In contrast, the bovine colostrum has been administered via an enteral route.17,29 For example, Juhl et al30 administered colostrum as a reconstituted colostrum powder to reach the required energy density, whereas Balachandran et al29 used small amounts of a different product powder that was mixed with expressed human milk and given 4 times per day. In this case, the aim was to produce an effect on the maturation of the gut as has been described on infant piglets.32,33 

Although they have been administered through different routes, in our review, we considered both types of colostrum. We hypothesized that the beneficial effects from colostrum contents could be similar; however, acknowledging the potential differences among both interventions, we conducted a between-study subgroup analysis and an indirect comparison to determine potential differences on the basis of colostrum type. In the direct comparisons of bovine and human colostrum, we did not find any difference.

The only outcome in which we found differences was the time to achieve full-enteral feeding and the CoE was judged as moderate. Colostrum administration reduced the time to achieve the enteral feeding by ∼3.5 days. Although the definition of full-enteral feeding provided by authors varied, ranging from 100 to 150 mL/kg per day, these results may be clinically relevant. The exact cause of this effect is not clear, but 1 potential explanation could be the presence of trophic factors on colostrum that may enhance intestinal maturation.9,33,34 

Achieving full-enteral feeding in a shorter period of time is with no doubt an important effect because it will relate to an earlier removal of central lines and, perhaps, to less associated infections. We identified fewer cases of culture-proven sepsis in colostrum groups but without statistically significant differences, which, as we described above, could be caused by a lack of power. In the subgroup analysis, we did not find differences by type of colostrum. However, the results, based on this limited number of trials, trended toward a treatment effect but were nonsignificant. We await published data from ongoing trials to further assess the potential of colostrum for time to achieve full-enteral feeding.

The certainty across the body of evidence was judged to be moderate or low. The reasons for rating down the CoE were due to heterogeneity and imprecision. Potential reasons for heterogeneity were explored by using both metaregression and subgroup analyses, and all analyses were nonsignificant. Imprecision was the reason for rating down because of the lack of significant effects (CIs ranged from values suggesting a substantial benefit to values suggesting substantial harm) and the modest sample size.35 Additional RCTs with more participants and more events will likely have an impact on the precision of estimates, which in turn will improve our certainty in evidence.

To date, there are at least 4 ongoing RCTs in which researchers are comparing colostrum to placebo that are registered in clinical trials register platforms (WHO and clinicaltrials.gov). In 3 trials, researchers are currently comparing oropharyngeal administration of human colostrum with placebo,2,36,37 and in 1 trial, researchers are comparing bovine colostrum with infant formula.38 In total, these researchers will analyze >1300 patients. Certainly, incorporating the results of these trials will add precision to the estimates, which in turn will provide higher certainty in our estimates of effect.

The strengths of this review include explicit eligibility criteria; a comprehensive search developed with a research librarian, with no language or publication status restriction; duplicate assessment of eligibility and independent data abstraction, risk of bias, and CoE assessment by using the GRADE approach; summarizing evidence for both human and bovine colostrum; and consideration of possible subgroup effects. Currently, there is a protocol for a Cochrane review that is designed to summarize the evidence of the administration of oropharyngeal human colostrum on morbidity and mortality in preterm infants39; however, this review is not considering bovine colostrum. To our knowledge, this is the first review that synthesizes the evidence from bovine colostrum.

The limitations of our review have to do with the underlying evidence. The total number of patients was modest, whereas the number of NEC-related events was low, which along with the heterogeneity led to low CoE. The inclusion of future ongoing trials will likely lead to more precise estimates and more confidence in the results. We pooled the evidence for both types of colostrum, although they are different interventions administered through different routes. To explore the potential heterogeneity related to type and administration route of colostrum, we performed subgroup analysis and indirect comparisons to evaluate the differences between both interventions, and our results demonstrated nonsignificant differences.

Moderate to low CoE suggests that human and bovine colostrum have no effect on NEC incidence, mortality, length of stay, and culture-proven infections among preterm infants. Colostrum may reduce the time for achieving full-enteral feeding. Future researchers need to confirm whether the effect on this outcome is similar between both types of colostrum or is limited only to human colostrum. Data from at least 4 ongoing trials will be useful in providing more patients to improve the precision of estimates for each of our outcomes. Given the interest in this topic, readers should look for review updates.

     
  • CI

    confidence interval

  •  
  • CoE

    certainty of evidence

  •  
  • GRADE

    Grading of Recommendations, Assessment, Development, and Evaluation

  •  
  • NEC

    necrotizing enterocolitis

  •  
  • RCT

    randomized controlled trial

  •  
  • RR

    relative risk

  •  
  • WHO

    World Health Organization

Dr Sadeghirad conceptualized and designed the study, coordinated and supervised the systematic review, designed the search strategy, selected the articles, extracted the data, assessed the quality of the evidence, performed the data analysis, interpreted the data, and drafted the initial manuscript; Dr Morgan coordinated and supervised the systematic review, selected the articles, extracted the data, assessed the certainty on the evidence, and interpreted the data analysis; Ms Zaraatkar and Ms Zea selected the articles, extracted the data, assessed the quality of the evidence, and interpreted the data analysis; Ms Couban designed the search strategy and interpreted the data analysis; Dr Johnston assessed the certainty of evidence and interpreted the data analysis; Dr Florez conceptualized and designed the study, coordinated and supervised the systematic review, selected the articles, extracted the data, assessed the certainty on the evidence, interpreted the data analysis, and drafted the initial manuscript; and all authors reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

This protocol has been registered with the PROSPERO International prospective register of systematic reviews (registration CRD42018085566).

FUNDING: No external funding.

1
World Health Organization
. Preterm birth. 2017. Available at: www.who.int/mediacentre/factsheets/fs363/en/. Accessed April 20, 2018
2
Rodriguez
NA
,
Vento
M
,
Claud
EC
,
Wang
CE
,
Caplan
MS
.
Oropharyngeal administration of mother’s colostrum, health outcomes of premature infants: study protocol for a randomized controlled trial.
Trials
.
2015
;
16
:
453
[PubMed]
3
Westerbeek
EA
,
van den Berg
A
,
Lafeber
HN
,
Knol
J
,
Fetter
WP
,
van Elburg
RM
.
The intestinal bacterial colonisation in preterm infants: a review of the literature.
Clin Nutr
.
2006
;
25
(
3
):
361
368
[PubMed]
4
Thyoka
M
,
de Coppi
P
,
Eaton
S
, et al
.
Advanced necrotizing enterocolitis part 1: mortality.
Eur J Pediatr Surg
.
2012
;
22
(
1
):
8
12
[PubMed]
5
Neu
J
,
Walker
WA
.
Necrotizing enterocolitis.
N Engl J Med
.
2011
;
364
(
3
):
255
264
[PubMed]
6
Raval
MV
,
Hall
NJ
,
Pierro
A
,
Moss
RL
.
Evidence-based prevention and surgical treatment of necrotizing enterocolitis-a review of randomized controlled trials.
Semin Pediatr Surg
.
2013
;
22
(
2
):
117
121
[PubMed]
7
Lin
PW
,
Stoll
BJ
.
Necrotising enterocolitis.
Lancet
.
2006
;
368
(
9543
):
1271
1283
[PubMed]
8
Christensen
RD
,
Gordon
PV
,
Besner
GE
.
Can we cut the incidence of necrotizing enterocolitis in half–today?
Fetal Pediatr Pathol
.
2010
;
29
(
4
):
185
198
[PubMed]
9
Radillo
O
,
Norcio
A
,
Addobbati
R
,
Zauli
G
.
Presence of CTAK/CCL27, MCP-3/CCL7 and LIF in human colostrum and breast milk.
Cytokine
.
2013
;
61
(
1
):
26
28
[PubMed]
10
Hettinga
K
,
van Valenberg
H
,
de Vries
S
, et al
.
The host defense proteome of human and bovine milk.
PLoS One
.
2011
;
6
(
4
):
e19433
[PubMed]
11
Rodriguez
NA
,
Caplan
MS
.
Oropharyngeal administration of mother’s milk to prevent necrotizing enterocolitis in extremely low-birth-weight infants: theoretical perspectives.
J Perinat Neonatal Nurs
.
2015
;
29
(
1
):
81
90
[PubMed]
12
Zhang
Y
,
Ji
F
,
Hu
X
,
Cao
Y
,
Latour
JM
.
Oropharyngeal colostrum administration in very low birth weight infants: a randomized controlled trial.
Pediatr Crit Care Med
.
2017
;
18
(
9
):
869
875
[PubMed]
13
Rodriguez
NA
,
Groer
MW
,
Zeller
JM
, et al
.
A randomized controlled trial of the oropharyngeal administration of mother’s colostrum to extremely low birth weight infants in the first days of life.
Neonatal Intensive Care
.
2011
;
24
(
4
):
31
35
14
Lee
J
,
Kim
H-S
,
Jung
YH
, et al
.
Oropharyngeal colostrum administration in extremely premature infants: an RCT.
Pediatrics
.
2015
;
135
(
2
). Available at: www.pediatrics.org/cgi/content/full/135/2/e357
[PubMed]
15
Ronayne de Ferrer
PA
,
Baroni
A
,
Sambucetti
ME
,
López
NE
,
Ceriani Cernadas
JM
.
Lactoferrin levels in term and preterm milk.
J Am Coll Nutr
.
2000
;
19
(
3
):
370
373
[PubMed]
16
Sohn
K
,
Kalanetra
KM
,
Mills
DA
,
Underwood
MA
.
Buccal administration of human colostrum: impact on the oral microbiota of premature infants.
J Perinatol
.
2016
;
36
(
2
):
106
111
[PubMed]
17
Schams
D
.
Growth factors in milk.
Endocr Regul
.
1994
;
28
(
1
):
3
8
[PubMed]
18
Bell
MJ
,
Ternberg
JL
,
Feigin
RD
, et al
.
Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging.
Ann Surg
.
1978
;
187
(
1
):
1
7
[PubMed]
19
Walsh
MC
,
Kliegman
RM
.
Necrotizing enterocolitis: treatment based on staging criteria.
Pediatr Clin North Am
.
1986
;
33
(
1
):
179
201
[PubMed]
20
Higgins
JPT
,
Altman
DG
,
Gøtzsche
PC
, et al;
Cochrane Bias Methods Group
;
Cochrane Statistical Methods Group
.
The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials.
BMJ
.
2011
;
343
:
d5928
[PubMed]
21
Akl
EA
,
Sun
X
,
Busse
JW
, et al
.
Specific instructions for estimating unclearly reported blinding status in randomized trials were reliable and valid.
J Clin Epidemiol
.
2012
;
65
(
3
):
262
267
[PubMed]
22
Guyatt
GH
,
Oxman
AD
,
Vist
GE
, et al;
GRADE Working Group
.
GRADE: an emerging consensus on rating quality of evidence and strength of recommendations.
BMJ
.
2008
;
336
(
7650
):
924
926
[PubMed]
23
Altman
DG
,
Bland
JM
.
Interaction revisited: the difference between two estimates.
BMJ
.
2003
;
326
(
7382
):
219
[PubMed]
24
Sterne
JA
,
Sutton
AJ
,
Ioannidis
JP
, et al
.
Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials.
BMJ
.
2011
;
343
:
d4002
[PubMed]
25
Miladinovic
B
,
Hozo
I
,
Chaimani
A
,
Djulbegovic
B
.
Indirect treatment comparison.
Stata J
.
2014
;
14
(
1
):
76
86
26
Glass
KM
,
Greecher
CP
,
Doheny
KK
.
Oropharyngeal administration of colostrum increases salivary secretory IgA levels in very low-birth-weight infants.
Am J Perinatol
.
2017
;
34
(
14
):
1389
1395
[PubMed]
27
Romano-Keeler
J
,
Azcarate-Peril
MA
,
Weitkamp
JH
, et al
.
Oral colostrum priming shortens hospitalization without changing the immunomicrobial milieu.
J Perinatol
.
2017
;
37
(
1
):
36
41
[PubMed]
28
Zhang
Y
,
Ji
F
,
Hu
X
,
Cao
Y
,
Latour
JM
.
Oropharyngeal colostrum administration in very low birth weight infants: a randomized controlled trial.
Pediatr Crit Care Med
.
2017
;
18
(
9
):
869
875
29
Balachandran
B
,
Dutta
S
,
Singh
R
,
Prasad
R
,
Kumar
P
.
Bovine colostrum in prevention of necrotizing enterocolitis and sepsis in very low birth weight neonates: a randomized, double-blind, placebo-controlled pilot trial.
J Trop Pediatr
.
2017
;
63
(
1
):
10
17
[PubMed]
30
Juhl
SM
,
Ye
X
,
Zhou
P
, et al
.
Bovine colostrum for preterm infants in the first days of life: a randomized controlled pilot trial.
J Pediatr Gastroenterol Nutr
.
2018
;
66
(
3
):
471
478
[PubMed]
31
Davis
TA
,
Nguyen
HV
,
Garcia-Bravo
R
,
Fiorotto
ML
,
Jackson
EM
,
Reeds
PJ
.
Amino acid composition of the milk of some mammalian species changes with stage of lactation.
Br J Nutr
.
1994
;
72
(
6
):
845
853
[PubMed]
32
Jensen
ML
,
Sangild
PT
,
Lykke
M
, et al
.
Similar efficacy of human banked milk and bovine colostrum to decrease incidence of necrotizing enterocolitis in preterm piglets.
Am J Physiol Regul Integr Comp Physiol
.
2013
;
305
(
1
):
R4
R12
[PubMed]
33
Rasmussen
SO
,
Martin
L
,
Østergaard
MV
, et al
.
Bovine colostrum improves neonatal growth, digestive function, and gut immunity relative to donor human milk and infant formula in preterm pigs.
Am J Physiol Gastrointest Liver Physiol
.
2016
;
311
(
3
):
G480
G491
[PubMed]
34
Sisk
PM
,
Lovelady
CA
,
Gruber
KJ
,
Dillard
RG
,
O’Shea
TM
.
Human milk consumption and full enteral feeding among infants who weigh </= 1250 grams.
Pediatrics
.
2008
;
121
(
6
). Available at: www.pediatrics.org/cgi/content/full/121/6/e1528
[PubMed]
35
Guyatt
GH
,
Oxman
AD
,
Kunz
R
, et al
.
GRADE guidelines 6. Rating the quality of evidence–imprecision.
J Clin Epidemiol
.
2011
;
64
(
12
):
1283
1293
36
US National Library of Medicine
. Bovine colostrum for preterm newborns (PreColos-RCT). 2018. Available at: https://clinicaltrials.gov/ct2/show/NCT03085277?term=colostrum&age=0&rank=6. Accessed April 20, 2018
37
World Health Organization
. Oral immunotherapy in very low birth weight preterm infants - randomized double-blind, placebo-controlled clinical trial. 2018. Available at: http://apps.who.int/trialsearch/Trial2.aspx?TrialID=RBR-83dk7z. Accessed March 15, 2018
38
World Health Organization
. Efficacy of oropharyngeal administration of colostrum in reducing morbidity and mortality in very preterm infants: a randomized controlled trial - colostrum. 2018. Available at: http://apps.who.int/trialsearch/Trial2.aspx?TrialID=CTRI/2017/11/010396. Accessed April 20, 2018
39
Nasuf
AW
,
Ojha
S
,
Dorling
J
.
Oropharyngeal colostrum in preventing mortality and morbidity in preterm infants.
Cochrane Database Syst Rev
.
2015
;(
10
):
CD011921

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data