IDDF2020-ABS-0025 The role of gut microbiota in clinical complications and treatment response in alcoholic hepatitis – a CIRCOS®, linear discriminant analysis effect size biomarker and CONET® co-occurrence network analysis

BackgroundIn severe alcoholic hepatitis(SAH), we aimed to characterize significant bacterial communities associated with clinical events(CE) and define bacterial relationships associated with specific CE and outcomes at baseline and on corticosteroids.Methods16-s rRNA sequencing on stool samples(n=3...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Gut 2020-11, Vol.69 (Suppl 2), p.A23-A24
Hauptverfasser: Philips, Cyriac, Ganesan, Karthik, Ranade, Shatakshi, Chopra, Varun, Patil, Kunal, Shende, Sonie, Phadke, Nikhil, Augustine, Philip
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:BackgroundIn severe alcoholic hepatitis(SAH), we aimed to characterize significant bacterial communities associated with clinical events(CE) and define bacterial relationships associated with specific CE and outcomes at baseline and on corticosteroids.Methods16-s rRNA sequencing on stool samples(n=38) collected at admission and at last follow-up within 90-days was (N=26, 12 on corticosteroids). Visual-characterization was performed on QIIME data using CIRCOS™, linear-discriminant-analysis-effect-size (LEfSe) method was used to identify significant bacterial communities and their functional metabolites. Conet/Cytoscape® utilized to identify significant co-occurrence with respect to clinical events.ResultsAll were males with mean age 47.3±9.1 years, median discriminant function(DF) 64, Child-Pugh(CTP)12 and model for end-stage liver disease(MELD)25.5. At admission, 27%, 42%, and 58% had acute kidney injury (AKI), hepatic encephalopathy (HE) and infections respectively; 38.5% died at 90-days follow-up. Specific bacteria were found to be associated with HE [Bifidobacteriaceae,Coriobacteriaceae], sepsis [Veillonellaceae,Prevotellaceae], CTP>10 [Bifidobacteriaceae, Synergistaceae], MELD>25 [Dehalobacteriaceae,Turicibacteraceae] and death [Enterobacteriaceae,Peptococcaceae]. Significantly higher relative abundance(RA) of Lachnobacterium, Catenibacterium associated with HE at-admission while orally-represented bacteria were associated with infections at admission. Propionibacterium, Fusobacteria were associated with DF≥65 while Eubacterium, Capnocytophaga were associated with CTP. Enhydrobacter and Pediococcus were preferentially abundant post-steroid-therapy. Aerococcus was associated with post-treatment death. Prevotella was associated with survival post steroid. Upregulation of phenylpropanoid-biosynthesis(innate-immunity) in those without follow-up infections and glycerophospholipid-metabolism(cellular-integrity) in those who died were significant. Co-occurrence between Christensenella, Prevotella and mutual-exclusion between Megamonas, Citrobacter was associated with HE at admission. Mutual-exclusion between Coprococcus eutactus, Catenibacterium and Megamonas was associated with infections at admission while Enterococcus cecorum, Acinetobacter schindleri, Mitsuokella were associated with AKI at admission (figure 1).Abstract IDDF2020-ABS-0025 Figure 1Circos representation lefse analysis and conet result representationConclusionsSpecific gut-microbiota, their i
ISSN:0017-5749
1468-3288
DOI:10.1136/gutjnl-2020-IDDF.34