Scolaris Content Display Scolaris Content Display

Oral lactoferrin for the prevention of sepsis and necrotizing enterocolitis in preterm infants

This is not the most recent version

Collapse all Expand all

Abstract

Background

Lactoferrin, a normal component of human colostrum and milk, can enhance host defense and may be effective in the prevention of sepsis and necrotizing enterocolitis (NEC) in preterm neonates.

Objectives

Primary objective
To assess the safety and effectiveness of oral lactoferrin in the prevention of sepsis and NEC in preterm neonates.

Secondary objectives

1. To determine the effects of oral lactoferrin used to prevent neonatal sepsis and/or NEC on duration of positive‐pressure ventilation, development of chronic lung disease (CLD) or periventricular leukomalacia (PVL), length of hospital stay to discharge among survivors, and adverse neurological outcomes at two years of age or later.

2. To determine the adverse effects of oral lactoferrin in the prophylaxis of neonatal sepsis and/or NEC.

When data were available, we analyzed the following subgroups.

1. Gestational age < 32 weeks and 32 to 36 weeks.

2. Birth weight < 1000 g (extremely low birth weight (ELBW) infants) and birth weight < 1500 g (very low birth weight (VLBW) infants).

3. Type of feeding: breast milk versus formula milk.

Search methods

We used the search strategy of the Cochrane Neonatal Review Group (CNRG) and updated our search in July 2014. We searched the databases Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, PREMEDLINE, EMBASE, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL), as well as trials registries and conference proceedings.

Selection criteria

Randomized controlled trials (RCTs) evaluating oral lactoferrin at any dose or duration to prevent sepsis or NEC in preterm neonates.

Data collection and analysis

Review authors used standard methods of the CNRG.

Main results

Four RCTs are included in this review. Oral lactoferrin supplementation decreased late‐onset sepsis (typical risk ratio (RR) 0.49, 95% confidence interval (CI) 0.32 to 0.73; typical risk difference (RD) ‐0.09, 95% CI ‐0.14 to ‐0.04; number needed to treat for an additional beneficial outcome (NNTB) 11, 95% CI 7 to 25; four trials, 678 participants, moderate‐quality evidence), NEC stage II or greater (typical RR 0.30, 95% CI 0.12 to 0.76; typical RD ‐0.05, 95% CI ‐0.08 to ‐0.01; NNTB 20, 95% CI 12.5 to 100; two studies, 505 participants, low‐quality evidence), and "all‐cause mortality" (typical RR 0.30, 95% CI 0.12 to 0.75; typical RD ‐0.05, 95% CI ‐0.08 to ‐0.01; NNTB 20, 95% CI 12.5 to 100; two studies, 505 participants, low‐quality evidence).

Oral lactoferrin supplementation with a probiotic decreased late‐onset sepsis (RR 0.27, 95% CI 0.12 to 0.60; RD ‐0.13, 95% CI ‐0.19 to ‐0.06; NNTB 8, 95% CI 5 to 17; one study, 321 participants, low‐quality evidence) and NEC stage II or greater (RR 0.04, 95% CI 0.00 to 0.62; RD ‐0.05, 95% CI ‐0.08 to ‐0.03; NNTB 20, 95% CI 12.5 to 33.3; one study, 496 participants, low‐quality evidence), but not "all‐cause mortality."

Oral lactoferrin with or without probiotics decreased fungal sepsis but not chronic lung disease or length of hospital stay (from one study, low‐quality evidence). No adverse effects were reported. Long‐term neurological outcomes or periventricular leukomalacia was not evaluated.

Authors' conclusions

Evidence of moderate to low quality suggests that oral lactoferrin prophylaxis with or without probiotics decreases late‐onset sepsis and NEC stage II or greater in preterm infants without adverse effects. Completion of ongoing trials will provide evidence from more than 6000 preterm neonates and may enhance the quality of the evidence. Clarifications regarding optimum dosing regimens, type of lactoferrin (human or bovine), and long‐term outcomes are still needed.

PICOs

Population
Intervention
Comparison
Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

Plain language summary

available in

Oral lactoferrin for prevention of sepsis and necrotizing enterocolitis in preterm infants

Premature babies are at risk for blood infection (sepsis) and/or gastrointestinal injury (necrotizing enterocolitis, or NEC). A number of babies with sepsis or NEC die or develop long‐term brain and lung injury despite treatment with antibiotics. Lactoferrin, which is present in human milk, has been shown to be effective against infection when tested in animals and in the laboratory. Lactoferrin also enhances the ability of babies to fight infection. We found four studies that enrolled 1103 preterm babies. Evidence of moderate to low quality suggests that oral lactoferrin with or without a probiotic decreases sepsis and NEC in preterm infants with no adverse effects. When given alone, lactoferrin decreases deaths among preterm infants. We also found large studies that are ongoing, and their results when available may increase the strength of our analysis. Clarification regarding dosing, duration, type of lactoferrin (human or bovine), and development of preterm babies is still needed.

Authors' conclusions

Implications for practice

We found moderate‐ to low‐quality evidence to suggest that oral lactoferrin prophylaxis decreases late‐onset sepsis, NEC ≥ stage II, and "all‐cause mortality" in preterm infants without adverse effects. Low‐quality evidence indicates that lactoferrin in combination with probiotics decreases late‐onset sepsis and NEC ≥ stage II in preterm infants without adverse effects. Although oral lactoferrin holds great promise in the prevention of neonatal sepsis and NEC, questions regarding optimal dosage and type (bovine or human recombinant lactoferrin), or whether it should be regulated as a food additive or as a medication, remain.

Implications for research

Completion of all ongoing trials and registered trials will result in data from more than 6000 preterm neonates and will enhance the quality and applicability of evidence for oral lactoferrin prophylaxis in preterm infants. Effects of exclusive maternal milk feeding and addition of probiotics to lactoferrin supplementation should also be clarified. Clinical randomized trials evaluating lactoferrin prophylaxis should assess not only short‐term beneficial effects, but also long‐term neurodevelopmental and pulmonary outcomes.

Summary of findings

Open in table viewer
Summary of findings for the main comparison. Oral lactoferrin compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Oral lactoferrin compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Patient or population: preterm infants
Settings: neonatal intensive care units
Intervention: oral lactoferrin alone
Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

Number of participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

Placebo

Lactoferrin alone

Any late‐onset sepsis ‐ all infants

Study population

RR 0.49
(0.32 to 0.73)

678
(4 studies)

⊕⊕⊕⊝
Moderate1,2

181 per 1000

89 per 1000
(58 to 132)

Moderate

202 per 1000

99 per 1000
(65 to 147)

Fungal sepsis

Study population

RR 0.08
(0.01 to 0.64)

321
(1 study)

⊕⊕⊝⊝
Low3

77 per 1000

6 per 1000
(1 to 50)

Moderate

77 per 1000

6 per 1000
(1 to 49)

All‐cause mortality

Study population

RR 0.3
(0.12 to 0.75)

552
(2 studies)

⊕⊕⊝⊝
Low1,4

67 per 1000

20 per 1000
(8 to 50)

Moderate

55 per 1000

17 per 1000
(7 to 41)

NEC ≥ stage II

Study population

RR 0.3
(0.12 to 0.76)

552
(2 studies)

⊕⊕⊝⊝
Low4

67 per 1000

20 per 1000
(8 to 51)

Moderate

127 per 1000

38 per 1000
(15 to 97)

Chronic lung disease

Study population

RR 0.71
(0.34 to 1.48)

441
(2 studies)

⊕⊕⊝⊝
Low4

70 per 1000

50 per 1000
(24 to 104)

Moderate

101 per 1000

72 per 1000
(34 to 149)

Threshold retinopathy of prematurity

Study population

RR 0.35
(0.14 to 0.85)

321
(1 study)

⊕⊕⊝⊝
Low3

113 per 1000

40 per 1000
(16 to 96)

Moderate

113 per 1000

40 per 1000
(16 to 96)

Length of stay among survivors

Mean length of stay among survivors in the intervention groups was 1.8 higher
(2.23 lower to 5.83 higher)

505
(1 study)

⊕⊕⊝⊝
Low3

*The basis for the assumed risk (eg, median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1Methods of randomization and allocation concealment are not available for 3 studies. 2 studies are in abstract form and the risk of bias cannot be completely assessed.
2Blinding of the healthcare provider and blinding of outcome assessment unclear.
3Only 1 study.
4Only 2 studies.

Open in table viewer
Summary of findings 2. Oral lactoferrin + probiotics compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Oral lactoferrin + probiotics compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Patient or population: preterm infants
Settings: neonatal intensive care units
Intervention: lactoferrin + probiotics
Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

Number of participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

Placebo

Lactoferrin + probiotics

Any late‐onset sepsis ‐ all infants

Study population

RR 0.27
(0.12 to 0.6)

319
(1 study)

⊕⊕⊝⊝
Low1,2

173 per 1000

47 per 1000
(21 to 104)

Moderate

173 per 1000

47 per 1000
(21 to 104)

Fungal sepsis

Study population

RR 0.26
(0.07 to 0.88)

319
(1 study)

⊕⊕⊕⊝
Moderate1

77 per 1000

20 per 1000
(5 to 68)

Moderate

77 per 1000

20 per 1000
(5 to 68)

All‐cause mortality

Study population

RR 0.54
(0.25 to 1.18)

496
(1 study)

⊕⊕⊝⊝
Low1,2

70 per 1000

38 per 1000
(17 to 82)

Moderate

70 per 1000

38 per 1000
(18 to 83)

NEC ≥ stage II

Study population

RR 0.04
(0 to 0.62)

496
(1 study)

⊕⊕⊝⊝
Low1,2

54 per 1000

2 per 1000
(0 to 34)

Moderate

54 per 1000

2 per 1000
(0 to 33)

Chronic lung disease

Study population

RR 0.74
(0.21 to 2.58)

319
(1 study)

⊕⊕⊝⊝
Low1,2

36 per 1000

26 per 1000
(7 to 92)

Moderate

36 per 1000

27 per 1000
(8 to 93)

Threshold retinopathy of prematurity

Study population

RR 0.76
(0.39 to 1.49)

319
(1 study)

⊕⊕⊝⊝
Low1,2

113 per 1000

86 per 1000
(44 to 169)

Moderate

113 per 1000

86 per 1000
(44 to 168)

Length of stay among survivors

Mean length of stay among survivors in the intervention groups was 2 higher
(1.88 lower to 5.88 higher)

496
(1 study)

⊕⊕⊝⊝
Low1,2

*The basis for the assumed risk (eg, median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1Blinding of the healthcare provider and blinding of outcome assessment unclear.
2Data from a single study.

Background

Description of the condition

Neonatal sepsis is the most common cause of neonatal death worldwide (Lawn 2006). The incidence of neonatal sepsis in the developed world is reported to be between one and four cases per 1000 live births (Stoll 2004b). In the developing world, the rate of neonatal sepsis is significantly higher (6.5 to 38 per 1000 live hospital births) (Zaidi 2005). Sepsis is a particular problem in very low birth weight (VLBW) infants (birth weight < 1500 g); early‐onset sepsis (sepsis in infants < 72 hours of life) occurs in about 1.5% and late‐onset sepsis in about 21% of VLBW infants (Stoll 2002; Stoll 2005). Most infections are caused by Staphylococcus and Candida species. Mortality and morbidity (including patent ductus arteriosus, prolonged ventilation, prolonged need for intravascular access, bronchopulmonary dysplasia, necrotizing enterocolitis, and increased length of hospital stay) are significantly increased among infected infants. In a large cohort study of infants born weighing less than 1000 g, infected infants had a significantly higher incidence of adverse neurodevelopmental outcomes at follow‐up when compared with uninfected infants (Stoll 2004a).

Necrotizing enterocolitis (NEC) occurs in 1% to 5% of admissions to the neonatal intensive care unit (NICU) (Lin 2006). The most consistent risk factors are prematurity and low birth weight. Gastrointestinal immaturity, enteral feeding (especially formula feeding), presence of bacteria, and inflammation in the gastrointestinal (GI) tract may all contribute to the development of NEC (Lin 2006). Host‐pathogen interactions trigger inflammation in the gut, which may contribute to the pathogenesis of NEC and septic shock (Blackwell 1997; Neish 2004). NEC significantly increases mortality (attributable mortality of 15% to 30%) and morbidity (including surgery in 20% to 40% of infants and delayed neurodevelopment) (Bell 1978; Lin 2006; Stoll 2004a).

Mortality and morbidity due to sepsis and NEC remain high despite the use of potent antimicrobial agents (Stoll 2002; Stoll 2005). Increased use of antimicrobials has led to the emergence of antibiotic‐resistant strains of bacteria (Levy 1998). Adverse pulmonary and neurodevelopmental outcomes after sepsis or NEC may be due to inflammatory injury (Adams‐Chapman 2006; Speer 1999). Agents that modulate inflammation and enhance host defenses may improve the outcomes of infants with neonatal sepsis or NEC.

Description of the intervention

The glycoprotein lactoferrin is a component of the innate immune response. It is found in significant concentrations in human colostrum and in lower concentrations in human milk, tears, saliva, seminal fluid, and secondary granules of neutrophils. Lactoferrin has broad‐spectrum antimicrobial activity against bacteria, fungi, viruses, and protozoa, which may result from its ability to sequester iron, or may occur as a direct lytic effect on microbial cell membranes (Valenti 2005). Proteolysis of lactoferrin under acidic conditions (as would occur in the stomach or in the phagolysosomes of neutrophils) yields peptides called lactoferricins, which have enhanced antimicrobial activity (Gifford 2005).

How the intervention might work

Lactoferrin inhibits the growth of Staphylococcus epidermidis and Candida albicans in vitro (Valenti 2005). It reduces the minimum inhibitory concentrations of vancomycin against S. epidermidis and antifungal agents such as azoles and amphotericin against Candida (Kuipers 1999; Leitch 1999). Lactoferrin and lactoferrin‐derived peptides are highly effective against antibiotic‐resistant Klebseilla and Staphylococcus aureus in vitro (Nibbering 2001).

Lactoferrin prophylaxis is effective in animal models of systemic and intestinal infection. In mice infected with Escherichia coli, pretreatment with lactoferrin improved survival from 4% to 70% (Zagulski 1989). In neonatal rats, lactoferrin reduced the severity of blood and liver infection after enteral infection with E. coli (Edde 2001). Parenteral prophylaxis with lactoferrin enhanced survival in a neonatal rat model of polymicrobial infection with C. albicans and S. epidermidis (Venkatesh 2007). In a germ‐free, colostrum‐deprived piglet model challenged with E. coli lipopolysaccharide, oral pretreatment with lactoferrin reduced mortality from 74% to 17% after challenge with E. coli lipopolysaccharide (Artym 2004). In animal colitis, lactoferrin reduced intestinal injury and inflammation (Togawa 2002). The systemic effects of oral lactoferrin generally are thought to be indirect and probably are initiated by contact with intestinal epithelial cells and gut‐associated lymphoid tissue (GALT). Lactoferrin modulates cytokine and/or chemokine production by GALT cells, which then enter the systemic circulation and influence circulating leukocytes (Bellamy 1992; Tomita 2002). Lactoferrin and other similar products in milk (prebiotics) create an environment for growth of beneficial bacteria in the gut, reducing colonization with pathogenic bacteria. The fact that intestinal receptors for lactoferrin have been demonstrated and lactoferrin has the ability to modulate intestinal cell differentiation and proliferation (Buccigrossi 2007) makes lactoferrin a promising agent in the prevention or treatment of NEC.

In adult humans, oral recombinant human lactoferrin has been found to be safe and well tolerated. Oral lactoferrin has shown promise as an antitumor agent (Hayes 2006) and has been shown to reduce viremia in chronic hepatitis C infection (Iwasa 2002; Tanaka 1999). In patients with acute myeloid leukemia and neutropenia, lactoferrin reduced the incidence, duration, and severity of bacteremia due to enteric pathogens (Trumpler 1989). To date, no significant adverse effects have been reported in animal or human studies.

Why it is important to do this review

The potential beneficial effects of lactoferrin make it a promising agent for prevention of neonatal sepsis and NEC. This review evaluated the role of oral lactoferrin in the prevention of neonatal sepsis or NEC in preterm neonates.

Objectives

Primary objective
To assess the safety and effectiveness of oral lactoferrin in the prevention of sepsis and NEC in preterm neonates.

Secondary objectives

  1. To determine the effects of oral lactoferrin used to prevent neonatal sepsis and/or NEC on duration of positive‐pressure ventilation, development of chronic lung disease (CLD) or periventricular leukomalacia (PVL), length of hospital stay to discharge among survivors, and adverse neurological outcomes at two years of age or later.

  2. To determine the adverse effects of oral lactoferrin in the prophylaxis of neonatal sepsis and/or NEC.

When data were available, we analyzed the following subgroups.

  1. Gestational age < 32 weeks and 32 to 36 weeks.

  2. Birth weight < 1000 g (extremely low birth weight (ELBW) infants) and birth weight < 1500 g (very low birth weight (VLBW) infants).

  3. Type of feeding: breast milk versus formula milk.

Methods

Criteria for considering studies for this review

Types of studies

Randomized or quasi‐randomized controlled trials that have been completed (published or unpublished).

Types of participants

Preterm (< 37 completed weeks of gestation) neonates (< 28 days).

Types of interventions

Oral lactoferrin at any dosage or duration used to prevent neonatal sepsis or NEC compared with placebo or no intervention. Separate analyses were performed for oral lactoferrin given with or without additional probiotics.

Types of outcome measures

Primary outcomes

1. Confirmed or suspected sepsis during hospital stay.
Confirmed sepsis is defined as clinical signs and symptoms consistent with infection and microbiologically proven with a positive blood culture, cerebrospinal fluid (CSF) culture, urine culture (obtained by a suprapubic tap), or culture from a normally sterile site (eg, pleural fluid, peritoneal fluid, autopsy specimens) for bacteria or fungi.

Suspected sepsis is defined as clinical signs and symptoms consistent with sepsis without isolation of a causative organism.

2. NEC Bell's stage II or III during hospital stay.
Necrotizing enterocolitis (NEC) (definitive NEC and perforated NEC, Bell's stage II or III) (Bell 1978) during hospital stay.

3. "All‐cause mortality" during hospital stay.

Secondary outcomes

  1. Neurological outcome at two years of age or later (neurodevelopmental outcome as assessed by a validated test).

  2. Chronic lung disease (CLD) in survivors (CLD defined as oxygen requirement at 36 weeks' post‐menstrual age (PMA)).

  3. Adverse outcomes directly attributable to oral lactoferrin: increased gastric residuals (gastric aspirate > 10% of oral feed), vomiting, and other GI disturbances during hospital stay.

  4. Periventricular leukomalacia (PVL) (defined as necrosis of brain white matter in a characteristic distribution, ie, in the white matter dorsal and lateral to the external angles of lateral ventricles involving particularly the centrum semi‐ovale and optic and acoustic radiations and diagnosed by magnetic resonance imaging (MRI), or as periventricular cystic lesions by cranial ultrasound (Volpe 1995) at discharge or at neurodevelopmental follow‐up).

  5. Duration of assisted ventilation through an endotracheal tube measured in days during hospital stay.

  6. Length of hospital stay measured in days for survivors to discharge.

  7. Post hoc analyses of bacterial infection, fungal infection, threshold retinopathy of prematurity, and urinary tract infection were included.

Search methods for identification of studies

We used the standard search methods of the Cochrane Neonatal Review Group and updated the search in July 2014.

Electronic searches

We identified relevant trials by searching the following.

  1. The Cochrane Central Register of Controlled Trials (CENTRAL).

  2. Electronic journal reference databases: MEDLINE (1966 to present) and PREMEDLINE, EMBASE (1980 to present), and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) (1982 to present).

  3. The following websites for ongoing trials: www.clinicaltrials.gov, www.controlled‐trials.com, Australian and New Zealand Clinical Trials Registry (http://www.anzctr.org.au), and the World Health Organization (WHO) International Clinical Trials Registry and Platform (http://www.who.int/ictrp/search/en/).

  4. Abstracts of conferences: proceedings of Pediatric Academic Societies (American Pediatric Society, Society for Pediatric Research, and European Society for Pediatric Research) from 1990 to present in the journal Pediatric Research and Abstracts Online.

We used this search strategy for MEDLINE and PREMEDLINE (strategy was adapted to suit EMBASE, CINAHL, and the Cochrane Central Register of Controlled Trials).

  1. explode 'sepsis' [all subheadings in MIME, MJME]

  2. sepsis or septicemia

  3. septic

  4. NEC

  5. 'necrotizing enterocolitis'

  6. # 1 or # 2 or # 3 or # 4 or # 5

  7. explode 'infant ‐ newborn' [all subheadings in MIME, MJME]

  8. Neonat*

  9. Newborn*

  10. # 7 or # 8 or # 9

  11. # 6 and # 10

  12. "lactoferrin' [all subheadings on MIME, MJME]

  13. Talactoferrin

  14. # 10 or # 11

  15. # 9 and # 12

We applied no language restrictions. We searched randomized and quasi‐randomized trials identified by reviewing the abstracts.

Searching other resources

We contacted study authors who published in this field to ask about possible unpublished articles.

We performed additional searches of the reference lists of identified clinical trials and of the review authors' personal files.

Data collection and analysis

We used the standard methods of the Cochrane Neonatal Review Group (CNRG) for conducting a systematic review (http://neonatal.cochrane.org/en/index.html).

Selection of studies

Two review authors assessed the titles and abstracts of studies identified by the search strategy to determine eligibility for inclusion in this review. If this could not be done reliably by title and abstract, we obtained the full‐text versions for assessment. We resolved differences by mutual discussion. We obtained full‐text versions of all eligible studies for quality assessment.

Data extraction and management

We designed forms for documenting trial inclusion/exclusion, for extracting data, and for requesting additional published information from authors of the original reports. We independently extracted data using specifically designed paper forms.

Assessment of risk of bias in included studies

We used the standardized review methods of the CNRG to assess the methodological quality of included studies. We independently applied the criteria developed by the CNRG and assessed the quality of included studies using the criteria listed below.

  1. A = Adequate (low risk of bias) allocation concealment.

  2. B = Uncertainty about whether the allocation was adequately concealed.

  3. C = Inadequate (high risk of bias) allocation concealment.

  4. D = No allocation concealment.

We reported blinding of interventions, blinding of outcome assessment, and completeness of follow‐up for all randomly assigned infants. We added this information to the Characteristics of included studies table.

In addition, we evaluated the following issues and entered them into the "Risk of bias" table.

  1. Sequence generation: Was the allocation sequence adequately generated? 

  2. Allocation concealment: Was allocation adequately concealed? 

  3. Blinding of participants, personnel, and outcome assessors: Was knowledge of the allocated intervention adequately prevented during the study? At study entry? At the time of outcome assessment? 

  4. Incomplete outcome data: Were incomplete outcome data adequately addressed? 

  5. Selective outcome reporting: Are reports of the study free of the suggestion of selective outcome reporting? 

  6. Other sources of bias: Was the study apparently free of other problems that could put it at high risk of bias?

Rating quality of evidence

We used the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) approach (Guyatt 2008) to interpret the quality of evidence and GRADEprofiler software to import data to RevMan 5.3 for the purpose of creating "Summary of findings" tables, which included ratings of quality of the evidence. Briefly, we used the following five GRADE considerations to interpret evidence: risk of bias, imprecision, inconsistency, indirectness, and publication bias. We did not estimate publication bias, as fewer than 10 studies were included.

Measures of treatment effect

We performed statistical analyses according to the recommendations of the CNRG. We analyzed all randomly assigned infants on an "intention‐to‐treat basis," irrespective of whether or not they received their allocated treatment. We analyzed treatment effects in the individual trials. We used the statistical package (RevMan 5) provided by The Cochrane Collaboration. We reported risk ratios (RRs) and risk differences (RDs) with 95% confidence intervals (CIs) for dichotomous outcomes and weighted mean differences for continuous outcomes. We calculated and reported the number needed to treat for an additional beneficial outcome (NNTB) or the number needed to treat for an additional harmful outcome (NNTH) for statistically significant reductions in RD.

Assessment of heterogeneity

We assessed heterogeneity of treatment effects between trials using the I2 statistic to check the appropriateness of pooling data and performing meta‐analyses. Meta‐analysis was deferred if heterogeneity was high (> 75%). We used the following cutoffs to report the degree of heterogeneity: < 25% no heterogeneity; 25% to 49% low heterogeneity; 50% to 74% moderate heterogeneity; and > 75% high heterogeneity. If we detected statistical heterogeneity, we explored possible causes (eg, differences in study quality, participants, intervention regimens, or outcome assessments) using post hoc subgroup analyses.

Data synthesis

We used a fixed‐effect model for meta‐analysis when appropriate, using Review Manager software (RevMan 5.3) supplied by The Cochrane Collaboration. For estimates of typical relative risk and risk difference, we used the Mantel‐Haenszel method.

Subgroup analysis and investigation of heterogeneity

Key subgroups were based on the following.

  1. Gestational age.

    1. Preterm infants (32 to 36 weeks' gestational age).

    2. Preterm infants (< 32 weeks' gestational age).

  2. Birth weight.

    1. VLBW infants (birth weight < 1500 g).

    2. ELBW infants (birth weight < 1000 g).

  3. Feedings.

    1. Breast milk feeding.

    2. Formula feeding.

Results

Description of studies

Our search strategy identified four randomized controlled trials (published in six reports) that were eligible for inclusion. Three published reports described one multicenter trial of oral lactoferrin prophylaxis in premature neonates (Manzoni 2009; Manzoni 2012; Manzoni 2014) and the other three included studies, which enrolled preterm neonates from the United States (Sherman 2013), Turkey (Akin 2014), and Peru (Ochoa 2011). Refer to the Characteristics of included studies table for details.

Included studies

Manzoni 2009

Manzoni and coworkers randomly assigned VLBW infants (birth weight < 1500 g) to oral bovine lactoferrin alone or in combination with a probiotic (Lactobacillus rhamnosus GG) or to placebo in 11 Italian NICUs. Late‐onset sepsis, defined as isolation of a pathogen in the blood, peritoneal fluid, or CSF after three days of life, was the primary outcome of interest. Secondary outcomes that were assessed were gram‐positive, gram‐negative, or fungal sepsis; mortality before hospital discharge; urinary tract infection; fungal colonization; progression from fungal colonization to invasive fungal infection; bronchopulmonary dysplasia; severe intraventricular hemorrhage (grade III or IV); threshold retinopathy of prematurity; NEC ≥ stage II; alteration of liver functions; and adverse effects.

Manzoni 2012

This is a report of the secondary analysis of data from the Manzoni 2009 study pertaining to fungal colonization and invasive fungal infections. Intervention and patient populations are similar to those in the Manzoni 2009 study. Prophylaxis with antifungal drugs was an exclusion criterion and was not permitted by the study protocol. Primary outcomes assessed were incidence rates of fungal colonization and invasive fungal infection. Secondary outcomes were intensity of fungal colonization, rate of progression to infection in colonized infants, frequencies of single fungal species in all groups, and mortality related to invasive fungal infections.

Manzoni 2014

This is a continuation of the randomized trial of Manzoni 2009 and was conducted to enhance power for assessing the effects of oral bovine lactoferrin in the prevention of NEC. Thirteen NICUs in Italy and New Zealand participated, and neonates were enrolled from October 1, 2007, through July 31, 2010. Intervention and patient populations and outcomes were similar to those in the Manzoni 2009 study.

Akin 2014

This prospective, single‐center, double‐blind, randomized controlled trial was performed at Ankara University, Turkey, between December 2009 and January 2011. Inborn neonates born at < 1500 g or at gestational age < 32 weeks were randomly assigned to bovine lactoferrin (200 mg/d) or placebo (2 mL of saline), once a day until discharge. Exclusion criteria were lack of parental consent, severe congenital malformations, severe hypoxic ischemic encephalopathy (HIE), and death before 72 hours of life. Primary outcomes assessed were nosocomial sepsis as defined by the criteria of the Centers for Disease Control and Prevention and NEC stage II. Secondary outcomes were safety (feeding tolerance, abdominal distention, emesis, and gastric residuals), length of hospital stay, and maturation of Treg levels.

Ochoa 2011

Ochoa and coworkers randomly assigned 190 premature infants < 2500 g in five neonatal intermediate and intensive care units in Lima, Peru, who were admitted to the NICU in the first 72 hours of life. Neonates were randomly assigned to oral bovine lactoferrin (200 mg/kg/d divided into three doses) or to oral maltodextrin (200 mg/kg/d in three divided doses) for four weeks; both were dissolved in human milk or formula or 5% glucose solution. The primary outcome assessed was the number of confirmed episodes of late‐onset sepsis in the first month of life; secondary outcomes assessed were incidence of gram‐positive and gram‐negative bacterial sepsis, fungal sepsis, pneumonia, diarrhea, and mortality in the first month of life.

Sherman 2013

This randomized clinical trial of human recombinant lactoferrin (talactoferrin (TLF)) conducted in the United States enrolled a total of 120 neonates (60 in each group). Preterm infants with birth weight of 750 to 1500 g were randomly assigned to enteral TLF or to placebo from one to 29 days of life at a dose of 150 mg/kg every 12 hours. TLF was provided by Agennix, Inc. Primary outcomes assessed were bacteremia, meningitis, pneumonia, urinary tract infection, and necrotizing enterocolitis; secondary outcomes were sepsis syndrome and NEC scares.

Excluded studies

King 2007

Healthy, formula‐fed infants greater than or equal to 34 weeks' gestation and four weeks of age or younger were enrolled from a pediatric clinic. Infants received formula supplemented with lactoferrin (850 mg/L) or commercial cow milk‐based formula (102 mg/L) for 12 months. Growth parameters and information on gastrointestinal, respiratory, and colic illnesses were collected for the infants' first year. Review authors excluded this study, as most enrolled infants were beyond the neonatal period and trial authors did not assess our prespecified neonatal outcomes.

Ochoa 2013

This community‐based, randomized, double‐blind, placebo‐controlled trial compared supplementation with bovine lactoferrin versus placebo. Investigators randomly assigned 577 weaned children at 12 to 18 months and followed them for six months with daily home visits. The aim was prevention of diarrhea, and outcomes assessed were number of diarrheal episodes, longitudinal prevalence of diarrhea, and severity of diarrhea and dehydration. Review authors excluded this study, as participants were not neonates.

Studies awaiting classification

Barrington 2013

This randomized controlled trial of oral bovine lactoferrin in Montreal, Canada, enrolled 80 neonates between January 2011 and April 2013. Investigators randomly assigned preterm infants in the NICU at CHU Sainte Justine, with a gestational age at birth between 23 0/7 and 30 6/7 weeks, who were less than 48 hours of age, to oral lactoferrin or placebo. The exclusion criterion was the presence of intestinal abnormalities that would prevent enteral feeding such as gastroschisis. The intervention group received 100 mg per day of bovine lactoferrin, divided into two doses per day, starting on the first day of enteral feeding (day of enrollment) or at the latest at 48 hours of age and until 36 weeks' PMA or discharge home. The control group received milk without lactoferrin. Primary outcomes assessed were death or at least one health care‐associated infection (HCAI) before discharge and tolerance of lactoferrin. Secondary outcomes assessed were infections per 1000 patient‐days, NEC stage II or III, surgical intervention for NEC, and infection‐associated mortality. This study has been completed, and we have requested details from the principal investigator.

Buonocore 2010

This randomized controlled trial of lactoferrin supplementation included preterm infants with birth weight ≤ 1500 g and/or gestational age ≤ 32 weeks. Neonates were excluded if fetal‐onset disorders were recognizable at birth, and if milk intolerance, family history of allergy, and use of infant formula supplemented with lactoferrin were reported. The intervention group (n = 650) received a daily dose of 100 mg of lactoferrin, and the control group (n = 650) received only standard therapy. Primary outcomes to be assessed were the antioxidant effects of lactoferrin and its ability to reduce free radical‐related diseases in the newborn; these were assessed through neurodevelopmental follow‐up. The secondary outcome was identification of a panel of markers for assessment of oxidative stress and for correlation with the lactoferrin antioxidant effect. This study planned to enroll 1300 neonates starting January 2011. We have requested details of the study from the principal investigator.

van Zoeren‐Grobben 2008

This was a prospective, double‐blind, randomized, placebo‐controlled study of preterm infants (n = 60) with gestational age 26 ± 0 to 35 ± 6 weeks. Neonates were excluded if born < 600 g, or if they had life‐threatening congenital malformations, non‐Dutch or English‐speaking parents, or a history of allergy among parents or siblings. Infants were randomly assigned to standard preterm formula, standard preterm formula with prebiotics (galacto‐oligosaccharides 28.5%, lactose 9.5%, galactose 0.5%, minerals 3.5%, fat 1.5%, and water 3%), or standard preterm formula with dairy lactoferrin 1 mg/100 mL (n = 20 in each group). The primary outcome assessed was composition of the gut flora at six weeks of full enteral feeds, incidence of infection, oxidative stress, and iron status. Secondary outcomes assessed were growth (weight, length, and head circumference), feeding intolerance, and psychomotor development at one year of age. This unpublished study was completed in 2009. We have requested details of the study from the principal investigator.

Ongoing studies

El‐Mokadem 2013

This double‐blind, randomized, controlled trial included neonates between 500 g and 2500 g and at ≤ 36 weeks' gestation, who were born in or were referred to the NICU of one of the participating hospitals in the first 48 hours of life. Preterm neonates were randomly assigned to one of three groups: low‐dose lactoferrin (100 mg/d), high‐dose lactoferrin (150 mg/kg/twice daily), or placebo (distilled water). The primary outcome assessed was blood culture positivity; secondary outcomes were complete blood count with differential leukocytic count and C reactive protein quantitative assay. This study was scheduled to start in June 2013 and will enroll 180 preterm neonates through January 2016.

McGuire 2012

This multicenter, randomized, placebo‐controlled trial examined prophylactic enteral lactoferrin supplementation to prevent late‐onset invasive infection in very preterm infants (ELFIN) in the United Kingdom. Infants are eligible to participate if gestational age at birth is less than 32 weeks, if they are less than 72 hours old, and if written informed parental consent is obtained. Infants are randomly assigned to receive either lactoferrin (150 mg/kg/d to a maximum of 300 mg) or placebo. Primary outcomes assessed are the incidence of microbiologically confirmed or clinically suspected late‐onset infection from trial entry until hospital discharge. Secondary outcomes are "all‐cause mortality" before hospital discharge, necrotizing enterocolitis (NEC) Bell’s stage II or III, severe retinopathy of prematurity (ROP) treated medically or surgically, bronchopulmonary dysplasia (BPD), a composite of invasive infection, major morbidity (NEC, ROP, or BPD), and mortality, number of days of administration of antibiotics per infant from 72 hours until death or discharge from hospital, number of days of administration of antifungal agents per infant, and length of hospital stay. This study is coordinated by the National Perinatal Epidemiology Unit Clinical Trials Unit, at the University of Oxford, UK; it was scheduled to start in September 2013 and will enroll 2200 preterm neonates.

Ochoa 2012

Phase III randomized controlled trial of oral lactoferrin for prevention of sepsis in infants (NEOLACTO study) in Peru. Neonates with birth weight between 500 g and 2000 g and born in or referred to the neonatal unit of one of the participating hospitals in the first 72 hours of life are eligible. Preterm neonates are randomly assigned to oral bovine lactoferrin (200 mg/kg/d divided in three doses) or oral maltodextrin (200 mg/kg/d in three divided doses) for eight weeks. The primary outcome assessed is a composite outcome of first episode of late‐onset sepsis or sepsis‐associated death. Secondary outcomes are neurodevelopment at 24 months of corrected age assessed by the Mullen Scale for Early Learning. This trial started enrolling in May 2012 and targets to enroll 414 neonates through January 2016.

Tarnow‐Mordi 2011

The Lactoferrin Infant Feeding Trial (LIFT) to prevent sepsis and death in preterm infants is a double‐blind, randomized, controlled trial in Australia and New Zealand. Eligibility for inclusion is based on the following: (1) doctor and parents are substantially uncertain whether bovine lactoferrin (BLF) is indicated or contraindicated, (2) < 1500 g birth weight, (3) < 7 days old, and (4) parent giving written informed consent. Neonates will be randomly assigned to BLF at 200 mg/kg/d dissolved in breast milk or formula until a corrected gestational age of 34 weeks or hospital discharge or placebo (breast milk or formula (without BLF)). The primary outcome that will be assessed will be mortality or major morbidity before hospital discharge. Morbidity is defined as the diagnosis of sepsis, brain injury, chronic lung disease, necrotizing enterocolitis, or severe retinopathy. Secondary outcomes that will be assessed include mortality related to sepsis (as assessed by positive blood culture). This study started enrolling January 2014 and plans to enroll 1100 infants.

Risk of bias in included studies

Selection bias: In the multicenter trial by Manzoni 2009, randomization was stratified by center, and randomization sequences were generated by computer software. The pharmacy at each center prepared the interventions and diluted them in milk feeds based on random sequences. Allocation concealment is unclear, as it is difficult to predict whether the pharmacy was aware of future allocations. Akin 2014 did not report random sequence generation or allocation concealment. Sherman 2013 randomly assigned enrolled neonates centrally using a permuted block method. Data from Ochoa 2011 were derived from conference proceedings, and risk of selection bias could not be assessed.

Performance bias: In the trial by Manzoni 2009, interventions were diluted in feeds, and clinical and research staff were blinded to the intervention. In the situation in which the infant was not fed and interventions were administered by orogastric tube without milk, it is not clear whether blinding was adequate. Performance bias was not noted in Akin 2014 and Sherman 2013 and could not be assessed in Ochoa 2011.

Detection bias: None of the included studies explicitly reported blinding of outcome assessors.

Completeness of follow‐up: Outcomes were assessed at hospital discharge, and incomplete data were adequately accounted for in the included studies.

Effects of interventions

See: Summary of findings for the main comparison Oral lactoferrin compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants; Summary of findings 2 Oral lactoferrin + probiotics compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Four randomized controlled studies (published as six reports) were eligible for inclusion, of which three reports described one multicenter trial of oral lactoferrin prophylaxis in premature neonates performed in Italy and New Zealand (Manzoni 2009; Manzoni 2012; Manzoni 2014). The other three included studies were conducted in the United States (Sherman 2013), Turkey (Akin 2014), and Peru (Ochoa 2011).

Lactoferrin alone versus placebo (comparison 1)

All four included trials provided outcome data for this comparison (Akin 2014; Manzoni 2009; Manzoni 2012; Manzoni 2014; Ochoa 2011; Sherman 2013).

Late‐onset sepsis (outcome 1.1)

All infants (outcome 1.1.1)

Oral lactoferrin supplementation in preterm infants decreases late‐onset sepsis (typical RR 0.49, 95% CI 0.32 to 0.73; typical RD ‐0.09, 95% CI ‐0.14 to ‐0.04; NNTB 11, 95% CI 7 to 25; four studies, 678 participants) (Figure 1). No heterogeneity (I2 = 0%) was noted among the four trials for this outcome. Evidence was downgraded to moderate because risk of bias was unclear in the four included studies.


Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.1 Any late‐onset sepsis.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.1 Any late‐onset sepsis.

Subgroup analyses for the outcome of late‐onset sepsis

Birth weight < 1000 g (outcome 1.1.2)

The estimated risk ratio for the outcome of late‐onset sepsis in ELBW infants was RR 0.31 (95% CI 0.14 to 0.70), RD ‐0.25 (95% CI ‐0.40 to ‐0.10), and NNTB 4 (95% CI 2.5 to 25) (one study, 113 participants) (Figure 1). Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Birthweight 1000 to 1500 g (outcome 1.1.3)

The estimated risk ratio for the outcome of late‐onset sepsis in this subgroup was RR 0.46 (95% CI 0.12 to 1.74) and RD ‐0.03 (95% CI ‐0.09 to 0.020) (one study, 208 participants) (Figure 1). Quality of evidence was downgraded to low because of unclear risk of selection and performance bias, and because data were available from only one study.

Exclusively maternal milk fed infants (outcome 1.1.4)

The estimated risk ratio for the outcome of late‐onset sepsis in exclusively maternal milk fed infants was RR 0.13 (95% CI 0.02 to 0.98), RD ‐0.17 (95% CI ‐0.30 to ‐0.03), and NNTB 5.8 (95% CI 3.3 to 33) (one study, 79 participants) (Figure 1). This suggests a decrease in late‐onset sepsis in preterm infants exclusively on maternal milk and supplemented with lactoferrin. Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Formula fed infants (outcome 1.1.5)

The estimated risk ratio for the outcome of late‐onset sepsis in formula fed infants was RR 0.23 (95% CI 0.03 to 1.90) and RD ‐0.14 (95% CI ‐0.32 to 0.04) (one study, 46 participants) (Figure 1). Quality of evidence was downgraded to low because of unclear risk of bias, and data were available from only one study.

NEC ≥stage II (outcome 1.2)

Oral lactoferrin supplementation in preterm infants decreases NEC ≥ stage II (typical RR 0.30, 95% CI 0.12 to 0.76; typical RD ‐0.05, 95% CI ‐0.08 to ‐0.01; NNTB 20, 95% CI 12.5 to 100) (two studies, 505 participants) (Figure 2). Moderate heterogeneity (I2 = 74%) was observed between the two trials for this outcome. Quality of evidence was downgraded to low because of unclear risk of bias in the two included studies and because data were available from only two studies.


Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.2 NEC ≥ stage II.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.2 NEC ≥ stage II.

All‐cause mortality (outcome 1.3)

Oral lactoferrin supplementation in preterm infants decreases "all‐cause mortality" (typical RR 0.30, 95% CI 0.12 to 0.75; typical RD ‐0.05, 95% CI ‐0.08 to ‐0.01; NNTB 20, 95% CI 12.5 to 1000) (two studies, 505 participants) (Figure 3). No heterogeneity (I2 = 0%) was noted among the two trials for this outcome. Quality of evidence was downgraded to low because of unclear risk of bias in the two included studies, and because data were available from only two studies.


Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.3 All‐cause mortality.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.3 All‐cause mortality.

Bacterial sepsis (outcome 1.4)

The estimated risk ratio for the outcome of bacterial sepsis in preterm infants was RR 0.49 (95% CI 0.23 to 1.05) and RD ‐0.06 (95% CI ‐0.12 to 0.00) (one study, 321 participants) (Figure 1). Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Fungal sepsis (outcome 1.5)

The estimated risk ratio for the outcome of fungal sepsis in preterm infants was RR 0.06 (95% CI 0.00 to 0.98), RD ‐0.05 (95% CI ‐0.09 to ‐0.02), and NNTB 20 (95% CI 11 to 50) (one study, 321 participants). This suggests a decrease in fungal sepsis among preterm infants who were supplemented with lactoferrin. Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Chronic lung disease (outcome 1.6)

The estimated risk ratio for the outcome of chronic lung disease was typical RR 0.71 (95% CI 0.34 to 1.48) and typical RD ‐0.02 (95% CI ‐0.06 to 0.02) (two studies, 441 participants) (Figure 4). No heterogeneity (I2 = 0%) was observed among the two trials for this outcome. Quality of evidence was downgraded to low because of unclear risk of bias in the two included studies, and because data were available from only two studies.


Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.6 Chronic lung disease.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.6 Chronic lung disease.

Duration of mechanical ventilation (outcome 1.7)

The estimated mean difference for the outcome of duration of mechanical ventilation in preterm infants was MD ‐1.40 (95% CI ‐3.44 to 0.64) (one study, 321 participants). Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Length of hospital stay among survivors (outcome 1.8)

The estimated mean difference for the outcome of length of hospital stay among survivors in preterm infants was MD 1.80 (95% CI ‐2.23 to 5.83) (one study, 505 participants). Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Threshold retinopathy of prematurity (outcome 1.9)

The estimated risk ratio for the outcome of threshold retinopathy of prematurity in preterm infants was RR 0.35 (95% CI 0.14 to 0.85), RD ‐0.07 (95% CI ‐0.13 to ‐0.02), and NNTB 14 (95% CI 8 to 50) (one study, 321 participants). Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Urinary tract infection (outcome 1.10)

The estimated risk ratio for the outcome of urinary tract infection in preterm infants was RR 0.44 (95% CI 0.14 to 1.37) and RD ‐0.03 (95% CI ‐0.08 to 0.01) (one study, 321 participants). Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.

Other outcomes

No adverse effects were reported in any study included in this comparison.

The following outcomes were not assessed in the included studies: neurological outcome at two years of age or older and periventricular leukomalacia.

Oral lactoferrin + probiotics versus placebo (comparison 2)

Outcome data for analyses for this comparison are derived from one trial (Manzoni 2009), in which preterm infants were randomly assigned to oral bovine lactoferrin or oral bovine lactoferrin in combination with the probiotic Lactobacillus rhamnosus GG or placebo. Subgroup analyses using birth weight and types of milk subgroups for late‐onset sepsis were performed for the outcome of "late‐onset sepsis." Data for subgroup analyses for other outcomes were not available.

Late‐onset sepsis (outcome 2.1)

All infants (outcome 2.1.1)

Oral lactoferrin supplementation in combination with probiotics in preterm infants decreases late‐onset sepsis (RR 0.27, 95% CI 0.12 to 0.60; RD ‐0.13, 95% CI ‐0.19 to ‐0.06; NNTB 8, 95% CI 5 to 17) (one study, 321 participants) (Figure 5). Quality of evidence was downgraded to low because data were available from only one study.


Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.1 Any late‐onset sepsis.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.1 Any late‐onset sepsis.

Subgroup analyses for the outcome of late‐onset sepsis

Birth weight < 1000 g (outcome 2.1.2)

The estimated risk ratio for the outcome of late‐onset sepsis in ELBW infants was RR 0.30 (95% CI 0.13 to 0.69), RD ‐0.26 (95% CI ‐0.40 to ‐0.11), and NNTB 5 (95% CI 2 to 9) (one study, 114 participants) (Figure 5). This suggests a decrease in late‐onset sepsis among ELBW infants who were supplemented with lactoferrin in combination with probiotics. Quality of evidence was downgraded to low because data were available from only one study.

Birth weight 1000 to 1500 g (outcome 2.1.3)

The estimated risk ratio for the outcome of late‐onset sepsis in preterm infants born with birth weight from 1000 to 1500 g was RR 0.16 (95% CI 0.02 to 1.27) and RD ‐0.05 (95% CI ‐0.11 to 0.0) (one study, 205 participants) (Figure 5). Quality of evidence was downgraded to low because data were available from only one study.

Exclusively maternal milk fed infants (outcome 2.1.4)

The estimated risk ratio for the outcome of late‐onset sepsis in preterm infants fed exclusively on maternal milk was RR 0.33 (95% CI 0.07 to 1.48) and RD ‐0.13 (95% CI ‐0.28 to 0.02) (one study, 69 participants) (Figure 5). Quality of evidence was downgraded to low because data were available from only one study.

Exclusively formula fed infants (outcome 2.1.5)

The estimated risk ratio for the outcome of late‐onset sepsis in preterm infants fed formula milk was RR 0.09 (95% CI 0.01 to 1.67) and RD ‐0.18 (95% CI ‐0.35 to ‐0.01) (one study, 48 participants) (Figure 5). Quality of evidence was downgraded to low because data were available from only one study.

NEC ≥stage II (outcome 2.2)

Oral lactoferrin supplementation in combination with probiotics in preterm infants decreases NEC ≥ stage II in preterm infants (RR 0.04, 95% CI 0.00 to 0.62; RD ‐0.05, 95% CI ‐0.08 to ‐0.03; NNTB 20, 95% CI 12.5 to 33.3) (one study, 496 participants) (Figure 6). Quality of evidence was downgraded to low because data were available from only one study.


Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.5 NEC ≥ stage II.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.5 NEC ≥ stage II.

All‐cause mortality (outcome 2.3)

The estimated risk ratio for the outcome of "all‐cause mortality" in preterm infants was RR 0.54 (95% CI 0.25 to 1.18) and RD ‐0.03 (95% CI ‐0.07 to 0.01) (one study, 496 participants) (Figure 7). Quality of evidence was downgraded to low because data were available from only one study.


Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.4 All‐cause mortality.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.4 All‐cause mortality.

Bacterial sepsis (outcome 2.4)

The estimated risk ratio for the outcome of bacterial sepsis in preterm infants was RR 0.28 (95% CI 0.11 to 0.72), RD ‐0.09 (95% CI ‐0.14 to ‐0.03), and NNTB 11 (95% CI 7 to 33) (one study, 319 participants). Quality of evidence was downgraded to low because data were available from only one study.

Fungal sepsis (outcome 2.5)

The estimated risk ratio for the outcome of fungal sepsis in preterm infants was RR 0.26 (95% CI 0.07 to 0.88), RD ‐0.06 (95% CI ‐0.10 to ‐0.01), and NNTB 16.6 (95% CI 10 to 100) (one study, 319 participants) (Figure 8). This suggests a decrease in fungal sepsis among preterm infants who were supplemented with lactoferrin in combination with probiotics. Quality of evidence was downgraded to low because of unclear risk of bias, and because data were available from only one study.


Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.3 Fungal infection.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.3 Fungal infection.

Chronic lung disease (outcome 2.6)

The study definition of chronic lung disease was oxygen requirement greater than 30% for 28 days, positive‐pressure ventilation at 36 weeks, or both. We have requested data from the study authors on infants who required oxygen at 36 weeks' corrected age.

The estimated risk ratio for the outcome of chronic lung disease in preterm infants was RR 0.74 (95% CI 0.21 to 2.58) and RD ‐0.01 (95% CI ‐0.05 to 0.03) (one study, 319 participants). Quality of evidence was downgraded to low because data were available from only one study.

Duration of mechanical ventilation (outcome 2.7)

The estimated mean difference for the outcome of "duration of mechanical ventilation" in preterm infants was MD ‐1.10 (95% CI ‐3.04 to 0.84) (one study, 321 participants). Quality of evidence was downgraded to low because data were available from only one study.

Length of hospital stay among survivors (outcome 2.8)

The estimated mean difference for the outcome of "length of hospital stay among survivors" in preterm infants was MD 2.00 (95% CI ‐1.88 to 5.88) (one study, 496 participants). Quality of evidence was downgraded to low because data were available from only one study.

Threshold retinopathy of prematurity (outcome 2.9)

The estimated risk ratio for the outcome of threshold retinopathy of prematurity in preterm infants was RR 0.76 (95% CI 0.39 to 1.49) and RD ‐0.03 (95% CI ‐0.09 to 0.04) (one study, 319 participants). Quality of evidence was downgraded to low because data were available from only one study.

Urinary tract infection (outcome 2.10)

The estimated risk ratio for the outcome of urinary tract infection in preterm infants was RR 0.67 (95% CI 0.25 to 1.79) and RD ‐0.02 (95% CI ‐0.07 to 0.03) (one study, 319 participants). Quality of evidence was downgraded to low because data were available from only one study.

Other outcomes

No adverse effects due to oral lactoferrin in combination with probiotics were reported in the study included in this comparison (Manzoni 2009).

The following outcomes were not assessed in the study: neurological outcome at two years of age or older and periventricular leukomalacia.

Discussion

Summary of main results

We identified four randomized controlled trials that enrolled 1103 preterm infants and evaluated oral lactoferrin or lactoferrin in combination with a probiotic (Lactobacillus rhamnosus GG) or placebo. Compared with control, oral lactoferrin supplementation decreased late‐onset sepsis (typical risk ratio (RR) 0.49, 95% confidence interval (CI) 0.32 to 0.73; typical risk difference (RD) ‐0.09, 95% CI ‐0.14 to ‐0.04; number needed to treat for an additional beneficial outcome (NNTB) 11, 95% CI 7 to 25) (four trials, 678 participants, moderate‐quality evidence), decreased necrotizing enterocolitis (NEC) ≥ stage II (typical RR 0.30, 95% CI 0.12 to 0.76; typical RD ‐0.05, 95% CI ‐0.08 to ‐0.01; NNTB 20, 95% CI 12.5 to 100) (two studies, 505 participants, low‐quality evidence), and decreased "all‐cause mortality" (typical RR 0.30, 95% CI 0.12 to 0.75; typical RD ‐0.05, 95% CI ‐0.08 to ‐0.01; NNTB 20, 95% CI 12.5 to 100) (two studies, 505 participants, low‐quality evidence). Supplementation with lactoferrin also decreased fungal sepsis (RR 0.06, 95% CI 0.00 to 0.98; RD ‐0.05, 95% CI ‐0.09 to ‐0.02; NNTB 20, 95% CI 11 to 50) (one study, 321 participants, low‐quality evidence) and threshold retinopathy of prematurity (RR 0.35, 95% CI 0.14 to 0.85; RD ‐0.07, 95% CI ‐0.13 to ‐0.02; NNTB 14, 95% CI 8 to 50) (one study, 321 participants, low‐quality evidence). In subgroup analyses, extremely low birth weight (ELBW) infants and those fed exclusively maternal milk had a reduction in late‐onset sepsis after oral lactoferrin supplementation (one study, low‐quality evidence). No differences in chronic lung disease, urinary tract infection, duration of mechanical ventilation, or length of hospital stay were reported.

Compared with control, oral lactoferrin supplementation with a probiotic decreased late‐onset sepsis (RR 0.27, 95% CI 0.12 to 0.60; RD ‐0.13, 95% CI ‐0.19 to ‐0.06; NNTB 8, 95% CI 5 to 17) (one study, 321 participants, low‐quality evidence), NEC ≥ stage II (RR 0.04, 95% CI 0.00 to 0.62; RD ‐0.05, 95% CI ‐0.08 to ‐0.03; NNTB 20, 95% CI 12.5 to 33.3) (one study, 496 participants, low‐quality evidence) and fungal sepsis (RR 0.26, 95% CI 0.07 to 0.88; RD ‐0.06, 95% CI ‐0.10 to ‐0.01; NNTB 16.6, 95% CI 10 to 100) (one study, 319 participants, low‐quality evidence). No differences in "all‐cause mortality," chronic lung disease, urinary tract infection, duration of mechanical ventilation, or length of hospital stay were reported.

No adverse effects related to lactoferrin or to the probiotic were noted. Long‐term neurological outcomes or periventricular leukomalacia were not assessed in any of the included studies.

Overall completeness and applicability of evidence

The four randomized controlled trials were performed in neonatal intensive care units in Italy, New Zealand, United States, Peru, and Turkey. Trials are currently ongoing in Australia and New Zealand, Egypt, United Kingdom, Peru, and the Netherlands. Oral lactoferrin has been evaluated in both the developing and the developed world. Completion of all ongoing and registered trials will yield data from > 6000 preterm neonates from across the globe and may enhance the quality and applicability of evidence related to the use of oral lactoferrin in preterm neonates.

A major concern in the initial trials was safety of oral lactoferrin in premature neonates, especially ELBW infants, who are at high risk of developing sepsis and NEC. In this review involving more than 1000 preterm neonates, no adverse effects due to oral lactoferrin have been reported. One trial evaluated human recombinant lactoferrin; all the others used bovine lactoferrin. Bovine lactoferrin has a 69% DNA sequence homology to human lactoferrin (Pierce 1991). Differences in glycosylation patterns of human recombinant and bovine lactoferrins may be responsible for differences in susceptibility to proteolysis and pathogen adhesion (Barboza 2012; Bellamy 1992). Whether human lactoferrin is as effective in vivo as bovine lactoferrin, or whether higher doses of human lactoferrin can be tolerated, needs to be confirmed in future trials.

The optimum timing of prophylaxis appears to be within the first three days of life based on the trial by Manzoni and coworkers (Manzoni 2009). The duration of prophylaxis with oral lactoferrin that has optimal benefits without adverse effects is still not clear in preterm neonates. Trials used oral lactoferrin for 28 to 45 days of life, and it is not clear that increased duration of prophylaxis is more effective in preventing late‐onset sepsis or NEC.

The current increase in interest in lactoferrin stems not only from improved understanding of its functions, but also from its increased availability in various forms and sources. Lactoferrin processed from bovine and human milk is available commercially as a food supplement (Swedish Dairies Association, Tatua Co‐operative Dairy Company in New Zealand, Lacto Bretagne Associes' in Belgium, Milei in Germany, Morinaga Industries in Japan, DoMO Food Ingredients, a subsidiary of Friesland Dairy Foods, in the Netherlands, etc). In the United States, human recombinant lactoferrin (talactoferrin from Agennix, Inc.) has an investigational new drug status for clinical research purposes. Lactoferrin expression in transgenic rice (Ventrus Biosciences) and transgenic maize (Meristem Therapeutics) is being researched. Bovine lactoferrin is less expensive than human lactoferrin and is affordable even in developing countries.

Quality of the evidence

We assessed the quality of evidence using GRADE (Grades of Recommendation, Assessment, Development, and Evaluation) methodology (Guyatt 2008); we downgraded the evidence to moderate to low on the basis of unclear risk of bias, and because data were available from only one or two studies, except for the outcome of late‐onset sepsis. We could not assess publication bias, as only four studies were included in the review. Manzoni 2009 and Sherman 2013 centrally randomly assigned the neonates and had low risk of selection bias. In the studies Akin 2014 and Ochoa 2011, generation of randomization sequences and allocation concealment could not be assessed and the risk of selection bias is unclear. Performance bias was noted to be at low risk or unclear risk for the included studies. In the trial by Manzoni 2009, interventions were diluted in feeds, and clinical and research staff were blinded to the intervention. In the situation in which the infant was not fed and interventions were administered by orogastric tube without milk, it is not clear whether blinding was adequate. Performance bias was not noted for Akin 2014 and Sherman 2013 but could not be assessed in Ochoa 2011. None of the studies explicitly report blinding of outcome assessors (detection bias). No attrition bias was noted in the included studies, as all outcome assessments were performed before hospital discharge, and incomplete data were adequately accounted for in the included studies.

Potential biases in the review process

We strove to decrease biases in the review process. Both review authors performed the literature search using an inclusive search strategy and combined their results. Our search strategy identified six reports from four randomized clinical trials on prespecified neonatal outcomes. Our post hoc analysis of evaluation of fungal sepsis, bacterial sepsis, threshold retinopathy of prematurity, or urinary tract infection did not change the conclusions of the review. We pursued the investigators of published randomized controlled trials and searched conference proceedings for data and missing information with limited success.

Agreements and disagreements with other studies or reviews

We did not identify any other review that synthesized data from trials of oral lactoferrin in preterm neonates by meta‐analysis. Ochoa 2014 reviewed the details of published and ongoing clinical trials on oral lactoferrin prophylaxis in preterm neonates. Lingappan 2013 reviewed and expanded on the biology, antimicrobial effects, and immunomodulatory effects of lactoferrin and commented on efficacy and safety related to its use in the newborn.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.1 Any late‐onset sepsis.
Figures and Tables -
Figure 1

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.1 Any late‐onset sepsis.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.2 NEC ≥ stage II.
Figures and Tables -
Figure 2

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.2 NEC ≥ stage II.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.3 All‐cause mortality.
Figures and Tables -
Figure 3

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.3 All‐cause mortality.

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.6 Chronic lung disease.
Figures and Tables -
Figure 4

Forest plot of comparison: 1 Lactoferrin alone versus placebo, outcome: 1.6 Chronic lung disease.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.1 Any late‐onset sepsis.
Figures and Tables -
Figure 5

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.1 Any late‐onset sepsis.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.5 NEC ≥ stage II.
Figures and Tables -
Figure 6

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.5 NEC ≥ stage II.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.4 All‐cause mortality.
Figures and Tables -
Figure 7

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.4 All‐cause mortality.

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.3 Fungal infection.
Figures and Tables -
Figure 8

Forest plot of comparison: 2 Lactoferrin + LGG versus placebo, outcome: 2.3 Fungal infection.

Comparison 1 Lactoferrin alone versus placebo, Outcome 1 Any late‐onset sepsis.
Figures and Tables -
Analysis 1.1

Comparison 1 Lactoferrin alone versus placebo, Outcome 1 Any late‐onset sepsis.

Comparison 1 Lactoferrin alone versus placebo, Outcome 2 NEC ≥ stage II.
Figures and Tables -
Analysis 1.2

Comparison 1 Lactoferrin alone versus placebo, Outcome 2 NEC ≥ stage II.

Comparison 1 Lactoferrin alone versus placebo, Outcome 3 All‐cause mortality.
Figures and Tables -
Analysis 1.3

Comparison 1 Lactoferrin alone versus placebo, Outcome 3 All‐cause mortality.

Comparison 1 Lactoferrin alone versus placebo, Outcome 4 Bacterial sepsis.
Figures and Tables -
Analysis 1.4

Comparison 1 Lactoferrin alone versus placebo, Outcome 4 Bacterial sepsis.

Comparison 1 Lactoferrin alone versus placebo, Outcome 5 Fungal infection.
Figures and Tables -
Analysis 1.5

Comparison 1 Lactoferrin alone versus placebo, Outcome 5 Fungal infection.

Comparison 1 Lactoferrin alone versus placebo, Outcome 6 Chronic lung disease.
Figures and Tables -
Analysis 1.6

Comparison 1 Lactoferrin alone versus placebo, Outcome 6 Chronic lung disease.

Comparison 1 Lactoferrin alone versus placebo, Outcome 7 Duration of mechanical ventilation.
Figures and Tables -
Analysis 1.7

Comparison 1 Lactoferrin alone versus placebo, Outcome 7 Duration of mechanical ventilation.

Comparison 1 Lactoferrin alone versus placebo, Outcome 8 Length of stay among survivors.
Figures and Tables -
Analysis 1.8

Comparison 1 Lactoferrin alone versus placebo, Outcome 8 Length of stay among survivors.

Comparison 1 Lactoferrin alone versus placebo, Outcome 9 Threshold retinopathy of prematurity.
Figures and Tables -
Analysis 1.9

Comparison 1 Lactoferrin alone versus placebo, Outcome 9 Threshold retinopathy of prematurity.

Comparison 1 Lactoferrin alone versus placebo, Outcome 10 Urinary tract Infection.
Figures and Tables -
Analysis 1.10

Comparison 1 Lactoferrin alone versus placebo, Outcome 10 Urinary tract Infection.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 1 Any late‐onset sepsis.
Figures and Tables -
Analysis 2.1

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 1 Any late‐onset sepsis.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 2 NEC ≥ stage II.
Figures and Tables -
Analysis 2.2

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 2 NEC ≥ stage II.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 3 All‐cause mortality.
Figures and Tables -
Analysis 2.3

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 3 All‐cause mortality.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 4 Bacterial sepsis.
Figures and Tables -
Analysis 2.4

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 4 Bacterial sepsis.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 5 Fungal Infection.
Figures and Tables -
Analysis 2.5

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 5 Fungal Infection.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 6 Chronic lung disease.
Figures and Tables -
Analysis 2.6

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 6 Chronic lung disease.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 7 Duration of mechanical ventilation.
Figures and Tables -
Analysis 2.7

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 7 Duration of mechanical ventilation.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 8 Length of stay among survivors.
Figures and Tables -
Analysis 2.8

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 8 Length of stay among survivors.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 9 Threshold retinopathy of prematurity.
Figures and Tables -
Analysis 2.9

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 9 Threshold retinopathy of prematurity.

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 10 Urinary tract infection.
Figures and Tables -
Analysis 2.10

Comparison 2 Lactoferrin + probiotics versus placebo, Outcome 10 Urinary tract infection.

Summary of findings for the main comparison. Oral lactoferrin compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Oral lactoferrin compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Patient or population: preterm infants
Settings: neonatal intensive care units
Intervention: oral lactoferrin alone
Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

Number of participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

Placebo

Lactoferrin alone

Any late‐onset sepsis ‐ all infants

Study population

RR 0.49
(0.32 to 0.73)

678
(4 studies)

⊕⊕⊕⊝
Moderate1,2

181 per 1000

89 per 1000
(58 to 132)

Moderate

202 per 1000

99 per 1000
(65 to 147)

Fungal sepsis

Study population

RR 0.08
(0.01 to 0.64)

321
(1 study)

⊕⊕⊝⊝
Low3

77 per 1000

6 per 1000
(1 to 50)

Moderate

77 per 1000

6 per 1000
(1 to 49)

All‐cause mortality

Study population

RR 0.3
(0.12 to 0.75)

552
(2 studies)

⊕⊕⊝⊝
Low1,4

67 per 1000

20 per 1000
(8 to 50)

Moderate

55 per 1000

17 per 1000
(7 to 41)

NEC ≥ stage II

Study population

RR 0.3
(0.12 to 0.76)

552
(2 studies)

⊕⊕⊝⊝
Low4

67 per 1000

20 per 1000
(8 to 51)

Moderate

127 per 1000

38 per 1000
(15 to 97)

Chronic lung disease

Study population

RR 0.71
(0.34 to 1.48)

441
(2 studies)

⊕⊕⊝⊝
Low4

70 per 1000

50 per 1000
(24 to 104)

Moderate

101 per 1000

72 per 1000
(34 to 149)

Threshold retinopathy of prematurity

Study population

RR 0.35
(0.14 to 0.85)

321
(1 study)

⊕⊕⊝⊝
Low3

113 per 1000

40 per 1000
(16 to 96)

Moderate

113 per 1000

40 per 1000
(16 to 96)

Length of stay among survivors

Mean length of stay among survivors in the intervention groups was 1.8 higher
(2.23 lower to 5.83 higher)

505
(1 study)

⊕⊕⊝⊝
Low3

*The basis for the assumed risk (eg, median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1Methods of randomization and allocation concealment are not available for 3 studies. 2 studies are in abstract form and the risk of bias cannot be completely assessed.
2Blinding of the healthcare provider and blinding of outcome assessment unclear.
3Only 1 study.
4Only 2 studies.

Figures and Tables -
Summary of findings for the main comparison. Oral lactoferrin compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants
Summary of findings 2. Oral lactoferrin + probiotics compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Oral lactoferrin + probiotics compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants

Patient or population: preterm infants
Settings: neonatal intensive care units
Intervention: lactoferrin + probiotics
Comparison: placebo

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

Number of participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

Placebo

Lactoferrin + probiotics

Any late‐onset sepsis ‐ all infants

Study population

RR 0.27
(0.12 to 0.6)

319
(1 study)

⊕⊕⊝⊝
Low1,2

173 per 1000

47 per 1000
(21 to 104)

Moderate

173 per 1000

47 per 1000
(21 to 104)

Fungal sepsis

Study population

RR 0.26
(0.07 to 0.88)

319
(1 study)

⊕⊕⊕⊝
Moderate1

77 per 1000

20 per 1000
(5 to 68)

Moderate

77 per 1000

20 per 1000
(5 to 68)

All‐cause mortality

Study population

RR 0.54
(0.25 to 1.18)

496
(1 study)

⊕⊕⊝⊝
Low1,2

70 per 1000

38 per 1000
(17 to 82)

Moderate

70 per 1000

38 per 1000
(18 to 83)

NEC ≥ stage II

Study population

RR 0.04
(0 to 0.62)

496
(1 study)

⊕⊕⊝⊝
Low1,2

54 per 1000

2 per 1000
(0 to 34)

Moderate

54 per 1000

2 per 1000
(0 to 33)

Chronic lung disease

Study population

RR 0.74
(0.21 to 2.58)

319
(1 study)

⊕⊕⊝⊝
Low1,2

36 per 1000

26 per 1000
(7 to 92)

Moderate

36 per 1000

27 per 1000
(8 to 93)

Threshold retinopathy of prematurity

Study population

RR 0.76
(0.39 to 1.49)

319
(1 study)

⊕⊕⊝⊝
Low1,2

113 per 1000

86 per 1000
(44 to 169)

Moderate

113 per 1000

86 per 1000
(44 to 168)

Length of stay among survivors

Mean length of stay among survivors in the intervention groups was 2 higher
(1.88 lower to 5.88 higher)

496
(1 study)

⊕⊕⊝⊝
Low1,2

*The basis for the assumed risk (eg, median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence.
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

1Blinding of the healthcare provider and blinding of outcome assessment unclear.
2Data from a single study.

Figures and Tables -
Summary of findings 2. Oral lactoferrin + probiotics compared with placebo for prevention of sepsis and necrotizing enterocolitis in preterm infants
Comparison 1. Lactoferrin alone versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Any late‐onset sepsis Show forest plot

4

Risk Ratio (M‐H, Fixed, 95% CI)

Subtotals only

1.1 All infants

4

678

Risk Ratio (M‐H, Fixed, 95% CI)

0.49 [0.32, 0.73]

1.2 Birth weight < 1000 g

1

113

Risk Ratio (M‐H, Fixed, 95% CI)

0.31 [0.14, 0.70]

1.3 Birth weight 1000‐1500 g

1

208

Risk Ratio (M‐H, Fixed, 95% CI)

0.46 [0.12, 1.74]

1.4 Maternal milk fed infants

1

79

Risk Ratio (M‐H, Fixed, 95% CI)

0.13 [0.02, 0.98]

1.5 Formula fed infants

1

46

Risk Ratio (M‐H, Fixed, 95% CI)

0.23 [0.03, 1.90]

2 NEC ≥ stage II Show forest plot

2

552

Risk Ratio (M‐H, Fixed, 95% CI)

0.30 [0.12, 0.76]

3 All‐cause mortality Show forest plot

2

552

Risk Ratio (M‐H, Fixed, 95% CI)

0.30 [0.12, 0.75]

4 Bacterial sepsis Show forest plot

1

321

Risk Ratio (M‐H, Fixed, 95% CI)

0.49 [0.23, 1.05]

5 Fungal infection Show forest plot

1

321

Risk Ratio (M‐H, Fixed, 95% CI)

0.08 [0.01, 0.64]

6 Chronic lung disease Show forest plot

2

441

Risk Ratio (M‐H, Fixed, 95% CI)

0.71 [0.34, 1.48]

7 Duration of mechanical ventilation Show forest plot

1

321

Mean Difference (IV, Fixed, 95% CI)

‐1.40 [‐3.44, 0.64]

8 Length of stay among survivors Show forest plot

1

505

Mean Difference (IV, Fixed, 95% CI)

1.80 [‐2.23, 5.83]

9 Threshold retinopathy of prematurity Show forest plot

1

321

Risk Ratio (M‐H, Fixed, 95% CI)

0.35 [0.14, 0.85]

10 Urinary tract Infection Show forest plot

1

321

Risk Ratio (M‐H, Fixed, 95% CI)

0.44 [0.14, 1.37]

Figures and Tables -
Comparison 1. Lactoferrin alone versus placebo
Comparison 2. Lactoferrin + probiotics versus placebo

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Any late‐onset sepsis Show forest plot

1

Risk Ratio (M‐H, Fixed, 95% CI)

Subtotals only

1.1 All infants

1

319

Risk Ratio (M‐H, Fixed, 95% CI)

0.27 [0.12, 0.60]

1.2 Birth weight < 1000 g

1

114

Risk Ratio (M‐H, Fixed, 95% CI)

0.30 [0.13, 0.69]

1.3 Birth weight 1000‐1500 g

1

205

Risk Ratio (M‐H, Fixed, 95% CI)

0.16 [0.02, 1.27]

1.4 Maternal milk fed infants

1

69

Risk Ratio (M‐H, Fixed, 95% CI)

0.33 [0.07, 1.48]

1.5 Formula milk fed infants

1

48

Risk Ratio (M‐H, Fixed, 95% CI)

0.09 [0.01, 1.67]

2 NEC ≥ stage II Show forest plot

1

496

Risk Ratio (M‐H, Fixed, 95% CI)

0.04 [0.00, 0.62]

3 All‐cause mortality Show forest plot

1

496

Risk Ratio (M‐H, Fixed, 95% CI)

0.54 [0.25, 1.18]

4 Bacterial sepsis Show forest plot

1

319

Risk Ratio (M‐H, Fixed, 95% CI)

0.28 [0.11, 0.72]

5 Fungal Infection Show forest plot

1

319

Risk Ratio (M‐H, Fixed, 95% CI)

0.26 [0.07, 0.88]

6 Chronic lung disease Show forest plot

1

319

Risk Ratio (M‐H, Fixed, 95% CI)

0.74 [0.21, 2.58]

7 Duration of mechanical ventilation Show forest plot

1

319

Mean Difference (IV, Fixed, 95% CI)

‐1.10 [‐3.04, 0.84]

8 Length of stay among survivors Show forest plot

1

496

Mean Difference (IV, Fixed, 95% CI)

2.0 [‐1.88, 5.88]

9 Threshold retinopathy of prematurity Show forest plot

1

319

Risk Ratio (M‐H, Fixed, 95% CI)

0.76 [0.39, 1.49]

10 Urinary tract infection Show forest plot

1

319

Risk Ratio (M‐H, Fixed, 95% CI)

0.67 [0.25, 1.79]

Figures and Tables -
Comparison 2. Lactoferrin + probiotics versus placebo