187 - Early Gut Microbiome and Retinopathy in Preterm Infants

Publication Number: 187.5066

Faizah Bhatti, University of Oklahoma Health Sciences Center, Oklahoma City, OK, United States; Eleana Cabello, University of Oklahoma Health Sciences, Oklahoma City, OK, United States; Lise DeShea, University of Oklahoma College of Medicine, Oklahoma City, OK, United States; Hande Acar Kirit, University of Oklahoma Health Sciences, Institutional Research Core Facility, Oklahoma City, OK, United States; Phillip I.. Tarr, Washington University in St. Louis School of Medicine, St. Louis, MO, United States; Barbara Warner, Washington University in St. Louis School of Medicine, St Louis, MO, United States

Background: Proliferative vasculopathy is the hallmark of retinopathy of prematurity (ROP), the leading cause of acquired childhood blindness. Oxygen-dependant vascular growth factors, systemic inflammation and redox imbalance are all contributing risk factors. Systemic inflammation is known to increase the risk of progression of ROP from mild to severe stages. The preterm neonatal gastrointestinal (GI) microbiome influences not only gut health and necrotizing enterocolitis, but also sepsis and growth. Data from small case series show enrichment of GI Enterobacteriaceae at 28 weeks’ PMA in infants with treatment warranted ROP.

Objective: We hypothesize that infants with higher stages of ROP will have a distinct less diverse early GI microbiome profile compared to infants with no ROP and that this may be an independent risk factor for severity of disease. This will be tested in a large cohort of infants from three neonatal units (NICU) in the US.

Design/Methods: Fecal microbiota and clinical data were analyzed for 82 preterm patients (38 with and 44 no ROP) as reported in Gibson et al., 2016. To explore potential associations between early GI microbiome and ROP status, shotgun metagenome sequencing data from first patient fecal sample were collected, before antibiotic treatments, and were analyzed using MetaPhlAn v4.0.6. Microbiome diversity indices of species richness and Shannon were calculated based on relative species abundances. Patients were grouped according to the highest ROP stage diagnosed and their diversity metrics were compared with Wilcoxon Rank Sum Test .

Results: Compared with the non-ROP group, babies with ROP were born significantly earlier and smaller, both p<.001 (Table 1). The species richness was significantly higher in non-ROP subject’s vs Stage 3 (p = 0.033) (Figure 1). However, Shannon diversity indices were significantly higher in subjects with no ROP vs Stage 2 ROP (p = 0.026) but not for Stage 3. Data are also shown for impact of delivery mode, birth weight, gestational age on diversity measures (Figure 2).

Conclusion(s): Infants with increasing severity of ROP appear to have lower trends in diversity of early fecal microbiota specimens prior to initiation of antibiotics during their NICU stay. In addition, infants with severe ROP born via vaginal delivery have less diversity of their gut microbiota compared to non-ROP subjects. Further analysis of this cohort exploring changes in microbiome over time and after exposure to inflammatory stimuli is being conducted. This is critical to establish modifiable risks for vascular dysgenesis in preterm infants.

Table 1. Maternal and Infant Baseline Characteristics
Demographic data show no significant differences in maternal race, use of antenatal steroids, or mode of infant delivery . As expected, infants with ROP had lower BW and GA.Number of infants with different stages of severity of ROP were also relatively distributed.

Figure 1. Relative Enrichment of GI Microbiome in All Infants with and Without ROP
Graphs depict microbial diversity for infants with increasing severity/stages of ROP. Tests for Richness and Shannon Diversity both trend towards a decrease in diversity with increasing severity of ROP. Comparison of enrichment with severity of ROP is shown in the table below. Stage 3 ROP had the least diversity in early microbiome compared to subjects with no ROP by Wilcoxson Rank Sum Test, with trends towards decrease in Shannon diversity with increasing severity of ROP; p<0.05 is significant.

Figure 2. Diversity of Early GI Microbiome in Infants and ROP: Stratified by Other Perinatal Risk Factors
Infants with all stages of ROP are further stratified by gestational age, birth weight, and mode of delivery with diversity mapped for both alpha enrichment and Shannon diversity in each grouping. This analysis was performed in order to account for the other early perinatal factors which may drive increasing severity of ROP (i.e. lower GA, BW and C-section delivery). Overall, the box plots appear to show a trend to a decrease in total microbial diversity. However, there are a few outliers, including infants born via vaginal delivery with Stage 3 ROP, the reason which is not immediately clear.