Probiotics for Preterm Infants: A Strain-Specific Systematic Review and Network Meta-analysis

Area of Science:

• Neonatal Medicine
• Microbiome Research
• Clinical Epidemiology

Background:

• Randomized controlled trials (RCTs) on probiotics for preterm infants show inconsistent results.
• Existing meta-analyses group strains, hindering identification of the most effective ones.
• Clinicians need evidence-based guidance on optimal probiotic strain selection.

Purpose of the Study:

• To identify specific probiotic strains with the greatest efficacy in reducing mortality and morbidity in preterm infants.
• To address the limitations of traditional meta-analyses by employing a network meta-analysis (NMA) approach.
• To provide clinicians with more precise information for selecting effective probiotic treatments.

Main Methods:

• Conducted a systematic literature search for placebo-controlled or head-to-head RCTs of probiotics in preterm infants.
• Extracted data on mortality, necrotizing enterocolitis, late-onset sepsis, and time to full enteral feeding.
• Applied Bayesian hierarchical random-effects models to construct a network meta-analysis.

Main Results:

• Included 51 RCTs with 11,231 preterm infants; most strains were studied infrequently.
• Only 3 of 25 probiotic combinations significantly reduced mortality.
• Seven treatments reduced necrotizing enterocolitis, 2 reduced late-onset sepsis, and 3 reduced time to full enteral feeding; no single strain excelled across all outcomes.

Conclusions:

• Network meta-analysis indicates efficacy in a minority of studied probiotic strains/combinations for preterm infants.
• Inconsistent results may stem from insufficient trial numbers/sizes or genuine lack of effect for certain species.
• Larger, powered RCTs focusing on the most promising strains are required to define optimal probiotic treatment strategies.