Blood Microbiota As a Potential Noninvasive Diagnostic Biomarker for Liver Fibrosis in Severely Obese Patients: Choose Carefully

Zongxin Ling,Xia Liu,Yiwen Cheng,Li Shao,Haiyin Jiang,Lanjuan Li
DOI: https://doi.org/10.1002/hep.28987
2017-01-01
Abstract:We read with great interest the recent article by Lelouvier et al., who investigated the relationship between liver fibrosis in patients with severe obesity and blood microbiota. With well-characterized discovery and validation cohorts, the blood 16S ribosomal RNA concentration was shown to be increased in patients with liver fibrosis, whereas the bacterial diversity was decreased. Although specific bacterial taxa signatures in the discovery cohort were not confirmed in the validation cohort, the data nonetheless verified that blood dysbiosis was associated with liver fibrosis in obese patients, which could be used as a potential noninvasive biomarker. Obesity is one of the major predisposing factors for nonalcoholic fatty liver disease, and obesityassociated gut dysbiosis may contribute to increased intestinal permeability and bacterial translocation. The gut may be the major origin of blood microbiota; however, an oral source for blood microbiota cannot be ignored. Koren et al. suggested that the atherosclerotic plaque microbiota may at least in part be derived from the oral cavity. As an important risk factor for periodontal disease, obesity may also contribute to bacterial translocation from the oral cavity to the blood and participate in shaping the blood microbiota. Without excluding oral disease in the enrolled participants, a reliable correlation between blood microbiota and liver fibrosis cannot be established because specific bacterial signatures may also exist in patients with oral diseases but without liver fibrosis. Thus, the inclusion criteria should include the oral health status of the participants. Studying oral microbiota in addition to gut and blood microbiota may help clarify the role of key functional bacteria in liver fibrosis. In addition, bacterial fragments in blood cannot be equated with blood microbiota, which can be easily detected using sequencing techniques. However, the read numbers for each sample were not reported in the Lelouvier et al. study. Only one a-diversity index—the Shannon index—was used to define the decreased blood bacterial diversity, whereas other indices such as b-diversity and Good’s coverage were missing. Principal coordinates analysis can provide interindividual variations intuitively, whereas Good’s coverage estimator can evaluate the sequencing depth for blood microbiota. The operational taxonomic unit number and the overall composition in different taxonomic levels should also be provided to interpret the bacterial richness of blood microbiota and its taxonomic differences. The inclusion of all the aforementioned parameters will contribute to a comprehensive understanding of the overall structure and composition of blood microbiota. Moreover, blood microbiota in healthy participants should also be considered to exclude those commensal bacteria in the blood. In conclusion, blood microbiota analysis can be influenced by various factors, including the origin of the condition. Without considering these factors, it is impossible to demonstrate the relationship between blood microbiota and liver fibrosis, or arbitrarily use the specific blood taxonomic signatures as biomarkers.
What problem does this paper attempt to address?