Особенности микробиома кишечника при некротизирующим энтероколите у недоношенных новорожденных и доношенных новорожденных с врожденными пороками сердца по данным метагеномного секвенирования

Abstract

Necrotizing enterocolitis (NEC) is a serious disease that predominantly affects premature infants with a birth weight of less than 1500 g. Newborns with ductus-dependent congenital heart disease (CHD) are also at risk of developing NEC, which is associated with impaired mesenteric perfusion and hypoxemia. Objective. To identify features of the intestinal microbiome composition of preterm neonates and full-term neonates with ductusdependent CHD, depending on the development of NEC. Patients and methods. A prospective observation of neonates and a retrospective “case-control” analysis (1:1 ratio) of the intestinal microbiome composition depending on the development of NEC were performed. In neonates with NEC, specimens were collected at the onset of NEC (±2 days), in patients without NEC – on the day of life, corresponding to the onset of the disease in newborns with NEC (±3 days) matched by gestational age. The intestinal composition was analyzed in 31 preterm newborns with stage IB–III NEC (subgroup N1 with early-onset NEC (days 1–7 of life) (n = 19); subgroup N2 with late-onset NEC (days 13–34 of life) (n = 12)), as well as in 31 preterm neonates without NEC (subgroup C1 of patients with specimens collected on days 2–7 of life (n = 19), subgroup C2 of patients with specimens collected on days 11–36 of life (n = 12)). Also, an analysis of the microbiome composition was performed in 14 full-term newborns with ductus-dependent CHD who developed NEC (group H1), and 14 full-term newborns with ductus-dependent CHD without NEC (group H2). Fecal specimens were collected from diapers into sterile tubes and stored at -40°C. Analysis of the intestinal microbiome was performed on Illumina MiSeq by prokaryotic 16S rRNA sequencing of amplicons with primers F515/R806 to the V4 16S rRNA variable region. A Naive Bayes classifier trained on the NCBI 16S RefSeq database was used for ASV taxonomic classification. Results. NEC stages of preterm neonates: IB – 5 children, IIA – 19, IIB – 4, III – 3. Stages of NEC in newborns with ductusdependent congenital heart disease: IB – 3 children; IIA – 11. There were no statistically significant differences in the Shannon alpha diversity index between groups (N1 and C1, N2 and C2, H1 and H2). Neonates with antibiotic exposure (ABT) at specimen collection had different beta diversity (PERMANOVA F-value: 24.534; R-squared: 0.21997; p = 0.001) and had a higher alpha diversity (Shannon index: 2.0 (0.8–2.8) vs. 1.2 (0.8–1.5), p = 0.004), primarily due to representatives of opportunistic flora. Analysis of the intestinal microbiome did not reveal bacteria, which could be interpreted as a universal marker for the development of NEC, but some features were identified. Among neonates from the “early” groups (N1 and C1), the dominant genera were more often Staphylococcus, Serratia, and Corynebacterium. Neonates with early-onset NEC (N1) had a significantly lower relative abundance of Enterococcus, Lacticaseibacillus, Clostridium, Fusobacterium, Limosilactobacillus, Agrococcus, Alkalibacterium, Ancrocorticia, Blautia, Brachybacterium, Brevibacterium, Carnobacterium, Citricoccus, Flaviflexus, Garicola, Lactococcus, Microbacterium, Phocaeicola, Propionimicrobium, Ruoffia, Stenotrophomonas compared to subgroup C1. Preterm neonates with late-onset NEC (N2) had a lower relative abundance of Bifidobacterium (p = 0.027) and Streptococcus (p = 0.037) than the comparison subgroup (C2) had. Neonates with late-onset NEC (N2) had a significantly higher relative abundance of Enterococcus than in early-onset NEC (N1) (7.83 (0.09–46.15) % vs. 0 (0–0.18) % resp., p = 0.0004), no differences were found between the comparison subgroups (C2 and C1). Preterm neonates with a longer NEC duration (n = 16, nil per os 20 (16–27) days) compared to children with a shorter duration (nil per os 10 (9.5–11) days) had several pathobionts/pathogens in fecal specimens at the NEC onset: Alkalibacterium, Brevibacterium, Flaviflexus and Ruo (Facklamia). Full-term neonates with ductus-dependent CHD and NEC (H1) had a higher relative abundance of Staphylococcus (p = 0.029), Serratia ( p = 0.011; OR = 14 (95% CI 1.43–137.3)), Pelomonas ( Roseateles) (p = 0.017), Methylobacterium (p = 0.017), and Nitrincola (p = 0.017) than neonates with CHD without NEC (H2). Conclusion. Intestinal dysbiosis may act as a predisposing background for the development of NEC, but not its main cause. The dysbiosis by antibiotic exposure is not a sufficient factor for NEC development; the mother's health, fetal development status, and the early postnatal period play a significant role. Early correction of antibiotic-induced dysbiosis may be a reasonable approach to preventing NEC. Probiotics, prebiotics, postbiotics/metabolites, live biotherapeutic products, and phages may be potential microbiome-modulating agents in neonates. Key words: necrotizing enterocolitis, ductus-dependent congenital heart disease, metagenomics, 16S rRNA sequencing, intestinal microbiome