Z Gastroenterol 2015; 53 - FV22
DOI: 10.1055/s-0035-1559018

Non-alcoholic fatty liver disease (NAFLD) is associated with a distinct fecal microbiome signature – a case control study

C Schulz 1, M Lerch 2, L Lahti 3, J Kühn 4, K Schütte 1, F Weiss 2, H Völzke 5, S Baumeister 5, S Fuentes 6, G Fluhr 2, W de Vos 7, J Mayerle 2
  • 1Otto-von Guericke-Universität Magdeburg, Universitätsklinikum Magdeburg, Klinik für Gastroneterologie, Hepatologie und Infektiologie, Magdeburg, Deutschland
  • 2Ernst-Moritz-Arndt-Universität, Universitätsmedizin Greifswald, Klinik für Innere Medizin A, Greifswald, Deutschland
  • 3University of Helsinki, Department of Veterinary Biosciences, Helsinki, Finnland
  • 4Ernst-Moritz-Arndt-Universität, Universitätsmedizin Greifswald, Klinik für diagnostische Radiologie und Neuroradiologie, Greifswald, Deutschland
  • 5Ernst-Moritz-Arndt-Universität, Universitätsmedizin Greifswald, Institut für Community Medicine, Greifswald, Deutschland
  • 6Wageningen University, Laboratory of Microbiology, Wageningen, Niederlande
  • 7University of Helsinki, Haartman Institute, Department of Bacteriology and Immunology, Immunobiology Research Program, Helsinki, Finnland

Background: Nonalcoholic fatty liver disease is now being recognized as the most common liver disorder worldwide. The majority of NAFLD patients are characterized by mere liver steatosis but up to one third progresses to non-alcoholic steatohepatitis (NASH). The underlying pathophysiology is poorly understood but changes in the gut microbiome have been suggested to be involved. Design: Using a case control design we recruited 84 subjects with liver steatosis and 83 controls from the population-based Study of Health in Pommerania. Subjects with diabetes mellitus, BMI > 25 kg/m2, immoderate alcohol intake or gallstone disease were excluded. Liver fat content was quantitated by confounder corrected chemical shift encoded MRI sequence at 1.5T. Cases with steatosis were defined as subjects with a mean liver fat content of 24.9% and controls with 2.2%. NAFLD and NASH were distinguished by using the FIB-4 score (Cut-off 1.3). Phylogenetic profiling of fecal samples was performed using the Human intestinal tract Chip (HITChip). For phenotypic correlation of the gut microbiome signature up to 224 host variables, including diet, were available and 38 reached significance. Results: By study design the extent of steatosis on liver MRI differed significantly between cases and controls (p < 10 – 6). Hierarchical clustering showed a clustering tendency. Random Forrest analysis revealed 69%± 14% 95CI classification accuracy on 130 genus-level taxa. Diet did not affect the classification accuracy. Reduced Shannon diversity (p = 0.046) and richness (p = 0.007) in cases were detected. PCA cluster analysis identified 4 out of 130 taxa discriminating between cases and controls (Prevotella oralis and P. melaninogenica, Sutterella wadsworthia, Uncultured Clostridiales) all of those with bimodal distribution. NASH cases showed a significantly increased abundancy of Gram-positive taxa as well as several Bacteroides spp. that could be used as a classifier. Conclusion: In the absence of metabolic syndrome NAFLD is associated with a distinct gut microbiome signature, which is unaffected by diet. Decreased abundancy of taxa, previously defined as tipping elements, points to a pathophysiological relevance. Progression to NASH is correlated with additional distinct changes in the microbiome.


Das Mikrobiom im Therapiefokus

Freitag, 18.09.2015/17:00 – 18:30/Saal 2