More evidence that probiotics may have a role in treating fatty liver disease

Abstract

Data continue to emerge linking the gut-liver axis and nonalcoholic fatty liver disease (NAFLD). More than 500 bacterial species exist within the alimentary tract, the composition of which has been implicated in obesity and its downstream complications. Early animal studies showed that transplantation of gut microbial species from conventionally raised mice into germ-free mice led to increased fat storage and insulin resistance (1). In nonalcoholic steatohepatitis, small intestinal bacterial overgrowth and increased TNF-a production have also been observed. Elevated cytokine concentrations are thought to increase intestinal permeability via disruption of intracellular tight junctions, resulting in progressive inflammation and fibrosis within the liver (2, 3). Interestingly, the host’s interaction with the gut microbiome may be a 2-way street because recent studies have shown that the gut inflammatory response can also influence the components of the microbiota in mice (4). Modulating the gut microbiome via probiotics and prebiotics has been shown to be beneficial in NAFLD in both animal and human studies (5, 6). Probiotics, which are live commensal organisms, and prebiotics, which are nondigestible oligosaccharides and polysaccharides that can stimulate the growth of intestinal bacteria, can be used together in what is known as a synbiotic. A recent meta-analysis reported favorable results with the use of probiotics, specifically declines in concentrations of alanine aminotransferase, aspartate aminotransferase, and TNF-a and measures of insulin resistance (6). Of 4 randomized controlled trials conducted, 2 trials included the use of synbiotics by cotreatment with fructooligosaccharides (FOS), a prebiotic (7, 8). FOS can promote potentially beneficial growth of bifidobacteria, and a trial using FOS alone was reported to cause improvement in aminotransferases (9). Of note is that one of the previous trials in patients with nonalcoholic steatohepatitis included liver biopsies that showed improved liver histology after 6 mo of a synbiotic containing Bifidobacterium longum and FOS (7). In this issue of the Journal, Eslamparast et al (10) add to this evidence with the results of a study that used a synbiotic formulation in NAFLD in a double-blind, randomized, placebocontrolled trial. Fifty-two nondiabetic patients with newly diagnosed NAFLD by ultrasound and an alanine aminotransferase concentration .60 U/L were randomly assigned to receive a synbiotic formulation including Lactobacillus casei, Lactobacillus rhamnosus, Streptococcus thermophiles, Bifidobacterium breve, Lactobacillus acidophilus, B. longum, and Lactobacillus bulgaricus with FOS or placebo and were followed for 28 wk. Important findings of this study were improvements in aminotransferases as well as a decline in baseline mild fibrosis as estimated by FibroScan (echosens). Whereas this study did not include liver histology as an endpoint, the authors provided evidence that systemic inflammation decreased with a decline in circulating TNF-a concentrations and diminished nuclear transcription factor jB activation in circulating mononuclear leukocytes. These observations are consistent with previous findings with both probiotics and prebiotics (6). Whether these measures represent inflammation at the level of the gut, the liver, adipose tissue, or throughout the body remains to be determined. Without specific interventions to manipulate components of the inflammation (eg, anti–TNF-a therapies or other specific pharmacologic antiinflammatory approaches), which components of inflammation play a causal role and which are consequences of liver injury is yet to be sorted out. It is tempting to attribute causality to the inflammatory axis as the authors of this study did, but it may be premature to do so without confirmatory data. The use of FibroScan to measure liver stiffness in this trial yielded provocative data. There appeared to be a reduction in liver stiffness, but it is worth noting that both the preand posttreatment liver stiffness measures were near normal. Liver stiffness measures are also influenced by the degree of steatosis and other factors such as the fasting state of the subject (11). Without corresponding liver biopsies, it is difficult to know the meaning of improved liver stiffness values when the values are near the normal range. Although the number of subjects treated was relatively small, this study supports the safety of synbiotic use and corroborated the findings of previous trials with improvement in not only aminotransferases but also measures of fibrosis, cytokine concentrations, and insulin sensitivity. Future studies with long-term follow-up after stopping treatment to assess if the benefits of