Lactoferrin's sepsis benefit in preterm infants shrank as trials grew

How does lactoferrin fight infection?

Lactoferrin sequesters the free iron that bacteria need to grow, and it has direct antimicrobial and immune-signaling activity in laboratory and animal work. Lactoferrin is an iron-binding glycoprotein found in milk, and its core biology is not in dispute.

That mechanistic story is why lactoferrin became a fixture of preterm-infant nutrition research.

Mechanism is not outcome. One trial of oropharyngeal colostrum confirmed that lactoferrin concentrations rise measurably in treated infants, exactly what the biology predicts, and found no significant difference in NEC, late-onset sepsis, or mortality compared with controls ^[6]. The protein did what the biochemistry says it should do, and clinical outcomes still did not move reliably.

What did the early lactoferrin trials show?

Early lactoferrin estimates were large. A 2018 meta-analysis of 9 randomized trials and 1,834 infants found lactoferrin associated with significant reductions in culture-proven late-onset sepsis (RR 0.47, 95% CI 0.33-0.67) and in NEC of stage II or worse (RR 0.40, 95% CI 0.18-0.86) ^[5].

Those two numbers built lactoferrin's reputation. They are still what most consumer-facing coverage cites.

How much does lactoferrin reduce sepsis in the full trial record?

The sepsis reduction survived in the larger evidence base at a smaller size and lower certainty. A 2021 meta-analysis of 10 randomized trials and 3,679 infants, which folded in the multicentre ELFIN trial of enteral lactoferrin, found lactoferrin with or without probiotics still associated with reduced sepsis (RR 0.56, 95% CI 0.36-0.86) ^[1]. It also found substantial heterogeneity across trials (I² = 58%) and graded the certainty of the evidence low ^[1].

The moderate-certainty subgroup that survived was lactoferrin without probiotics, and that pool covered only 8 studies and 1,209 subjects, so the largest and most heterogeneous trials sit outside the clean result ^[1].

Network meta-analyses have kept the sepsis signal alive, smaller again. A 2023 analysis found lactoferrin associated with lower sepsis odds (OR 0.74, 95% CrI 0.54-0.92) and not with lower mortality ^[9]. Cochrane's assessment of enteral lactoferrin for sepsis and NEC prevention has been revised three times as trials accumulated ^[4]⁠^[3]⁠^[2].

Across the sequence, the point estimate weakened from roughly RR 0.47 to the RR and OR range of 0.56 to 0.74 ^[1]⁠^[5]⁠^[9].

Does lactoferrin prevent NEC, death, or lung and eye complications?

No. Reduced NEC, reduced mortality, reduced bronchopulmonary dysplasia, and reduced severe retinopathy of prematurity all failed to reach significance once the bigger trials entered the pooled analyses ^[1]⁠^[8].

The 2021 meta-analysis found NEC no longer significantly reduced (RR 0.62, 95% CI 0.29-1.33), and all-cause mortality not reduced either (RR 0.74, 95% CI 0.36-1.53), along with null results for bronchopulmonary dysplasia and severe retinopathy of prematurity ^[1].

A separate network meta-analysis of 106 trials and 25,840 infants found that only multiple-strain probiotics, not lactoferrin, were associated with reduced all-cause mortality ^[8].

Why did lactoferrin's effect shrink?

Three factors line up: sample size, heterogeneity, and feeding context.

• Sample size. The trials behind the strongest early estimates totaled under 2,000 infants combined ^[5], while the pools that dilute the effect include contributions on the order of 3,700 infants ^[1]. A single large multicentre trial can shift a pooled estimate more than several small ones combined.
• Heterogeneity. The 2021 pooled sepsis estimate carries an I² of 58%, high enough to make "the trials are measuring the same thing" a shaky assumption. The smaller, more homogeneous subgroup (I² = 0%) happens to exclude the largest trials ^[1].
• Feeding context. The subgroup analysis comparing breast-milk and formula or donor-milk infants found different magnitudes of apparent benefit by feeding type ^[7]. Pooling trials with different baseline feeding practices can average away or exaggerate a real but conditional effect.

The same pattern of a benefit shrinking under better-controlled trials appears in IPP and VPP blood pressure research.

Does lactoferrin work better in formula-fed preterm infants?

Possibly. A secondary analysis comparing outcomes across the earlier RCT and the ELFIN trial suggests the benefit may concentrate in infants not exclusively fed mother's own milk ^[7].

Background feeding composition already contains lactoferrin, which could partly explain why pooled effects vary by trial population ^[7].

Does lactoferrin support immunity in healthy adults?

None of the trials cited below tested lactoferrin for immune support in healthy adults. Every trial in the lactoferrin sepsis literature enrolled hospitalized preterm infants.

Claims about immune support or iron absorption in general adult populations are separate from, and less evidenced than, the preterm-infant sepsis literature. The gap between a mechanism and a population that has actually been tested recurs across compounds, as KPV's preclinical biology and the human evidence gap shows.

What is still unknown about lactoferrin?

Three questions stay open even for the preterm population the trials studied:

• Whether the sepsis signal is real. Whether the surviving reduction is a feeding-context-dependent effect or largely a statistical artifact of pooling heterogeneous small trials is unresolved.
• Dose and duration. No trial establishes what dose and duration would maximize a real effect if one exists.
• Other populations. Healthy adults and iron status in pregnancy are untested in this literature, because every cited trial enrolled hospitalized preterm infants.