Antibiotic "Dysbiosis" in Preterm Infants

Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Society for Pediatric Dermatology

Prolonged antibiotic use in preterm neonates has significant consequences on the developing intestinal microbiome, metabolome and host response, predisposing the neonate to various major morbidities, including necrotizing enterocolitis (NEC), late-onset sepsis, bronchopulmonary dysplasia (BPD), and mortality. The hypothesis is that early and prolonged antibiotic use in preterm neonates has significant consequences on the developing intestinal microbiome, metabolome and host response, predisposing the neonate to various major morbidities. It is possible that the effect of this widespread antibiotic use outweighs the potential benefits. This study will randomize preterm infants born at less than 33 weeks gestation to either pre-emptive antibiotics or no-pre-emptive antibiotics. The purpose of this research is to evaluate the risks and benefits of current practice to determine optimal levels of antibiotic use that protects the babies from infection with minimal effect on the microbiome and subsequent adverse outcomes related to overuse of antibiotics.

A majority of preterm very low birthweight (VLBW) infants are exposed to antibiotics. Surveys from large databases in the US show that the rate of culture proven bacteremia in these infants at birth is only between 1-2 percent. Antibiotic use, especially when repeated, induces a perturbation ("dysbiosis") in gut microbiota that may not recover to the basal state. Antibiotic use increases the risk of subsequent disease and adverse outcomes. The dependence of the developing immune system on the intestinal microbiota is supported by emerging evidence from studies in animals demonstrating decreased resistance to subsequent disease with early exposure to antibiotics. A retrospective review of 50,0261 neonates across 127 neonatal intensive care units (NICUs) from California showed a forty-fold variation in NICU antibiotic prescribing practice with similar burdens of proven infection and mortality. A large number of preterm infants are thus subjected to a potentially harmful course of antibiotics that provides no clear benefit. There remains a major gap in our understanding of antibiotic-related intestinal microbial dysbiosis and how this may result in disease. There will be two aims. In the first aim, a prospective, randomized pilot study, will test the effects of pre-emptive postnatal antibiotics on the microbiome, metabolome and inflammatory responses in the neonate during the NICU course. The second aim will assess the effects of pre-emptive postnatal antibiotics on adverse outcomes in the neonate while in the NICU. The hypothesis is that higher antibiotic use will not be associated with decreased early onset sepsis and in fact, will be associated with increased adverse outcomes including retinopathy of prematurity, necrotizing enterocolitis, spontaneous ileal perforation, late onset sepsis, chronic lung disease, bronchopulmonary dysplasia, intraventricular hemorrhage, periventricular leukomalacia, and mortality.

Enterocolitis, Necrotizing, Bacteremia, Bronchopulmonary Dysplasia, Intraventricular Hemorrhage, Periventricular Leukomalacia, Chronic Lung Disease, Ileal Perforation

These neonates have a clinical indication to receive antibiotics, such as maternal chorioamnionitis with fetal tachycardia. The standard of care antibiotics include Ampicillin and Gentamicin or Cefotaxime and as part of standard of care blood tests such as complete blood cell counts, blood cultures, and C-reactive proteins will be performed. Study interventions will include the collection of samples for the following: breast milk, gastric fluid and stool samples for analysis of the microbiome and metabolome.

These neonates show no signs of respiratory distress(RDS) or have no indications of maternal chorioamnionitis. Antibiotics is not indicated for this group as standard of care. Study interventions will include the collection of samples for the following: breast milk, gastric fluid and stool samples for analysis of the microbiome and metabolome, along with standard of care complete blood counts, blood cultures, and C-reactive proteins.

This group will be randomized to receive standard of care antibiotics which include Ampicillin and Gentamicin or Cefotaxime. Standard of care blood tests such as complete blood cell counts, blood cultures, and C-reactive proteins will be performed. Study interventions will include the collection of samples for the following: breast milk, gastric fluid and stool samples for analysis of the microbiome and metabolome.

This group will be randomized not to receive standard of care antibiotics. Study interventions will include the collection of samples for the following: breast milk, gastric fluid and stool samples for analysis of the microbiome and metabolome, along with standard of care complete blood counts, blood cultures, and C-reactive proteins.

Rates of composite morbidities and mortality, including necrotizing enterocolitis(NEC), late onset sepsis (LOS), bronchopulmonary dysplasia (BPD) and death

Enrolled subjects' medical record will be reviewed to determine the association between antibiotic administration and the components of the composite outcome

Enrolled subjects' medical record will be reviewed to determine the association between antibiotic administration and the development of bacteremia after the first week of life.

Pacific BioSciences or Illumina sequencing will be done and the data analyzed using metagenomics Rapid Annotation using Subsystem Technology (MG-RAST).

Microbial diversity is assessed using Chao1, Shannon, and ordination methods implemented using a software program called phyloseq package in R.42 Chao1 estimates the species richness for each sample, while the Shannon Index scores richness and abundance, though is not sufficient in assessing overall microbiome differences. Detrended Correspondence Analysis (DCA), a multivariate statistical method, will be applied to detect overall microbiome differences. Adonis methods were used to attribute additional variables' contribution to microbial variance.

Metabolites as biomarkers of microbial-host metabolism will be identified by nuclear magnetic resonance and mass spectrometry. Levels of vitamins(microMol per gram), polyphenols(microMol per gram), cholesterol (microMol per gram), and short chain fatty acids(microMol per gram) will be measured.

Enrolled subjects' medical record will be reviewed to determine the association between antibiotic administration and diagnosis of BPD.

Enrolled subjects' medical record will be reviewed to determine the association between antibiotic administration and spontaneous ileal perforation

Enrolled subjects' medical record will be reviewed to determine the association between antibiotic administration and intraventricular hemorrhage

Enrolled subjects' medical record will be reviewed to determine the association between antibiotic administration and intraventricular hemorrhage

Inclusion Criteria: All infants less than 33 weeks gestation. Exclusion Criteria: Infants who are non-viable at birth.

Russell JT, Lauren Ruoss J, de la Cruz D, Li N, Bazacliu C, Patton L, McKinley KL, Garrett TJ, Polin RA, Triplett EW, Neu J. Antibiotics and the developing intestinal microbiome, metabolome and inflammatory environment in a randomized trial of preterm infants. Sci Rep. 2021 Jan 21;11(1):1943. doi: 10.1038/s41598-021-80982-6.