Reply

Abstract

We thank Rothi Methani et al for their interest in our article published in Gastroenterology.1Solé C. et al.Gastroenterology. 2020; 160: 206-218.e13Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar We agree with Methani et al that different storage conditions for fecal samples affect the quality of extracted nucleic acids and the stability of their microbial communities. We strictly fulfilled the recommendation of the International Human Microbiome Standards project that developed the standard operating procedures designed to optimize data quality and comparability in the human microbiome field (http://www.microbiome-standards.org/). Previous data showed that for optimal preservation to extract high-quality RNA for metatranscriptomic analyses, stool samples should be kept at room temperature and brought to the laboratory within 24 hours after collection, stored immediately at −20°C in a home freezer, and transported afterwards in a freezer, or homogenized with stabilization solution (RNAlater) and delivered between 24 hours and 7 days.2Cardona S. et al.BMC Microbiol. 2012; 12: 1Crossref PubMed Scopus (159) Google Scholar Regarding multiple factors that could affect the microbiome, it is well-known that antibiotics, diet, etiology of liver disease, sex, age, obesity, drugs, and others could have a role in shaping the gut microbiome.3Acharya C. et al.Gastroenterology. 2021; 160: 556-572Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar In fact, we collected a 24-hour recall questionnaire diet and a median of 1145 kcal/d was obtained. Unfortunately, other analyses were not performed due to the absence of a standardized analytical methods. We agree that a more holistic approach that would take into account all possible confounders would be ideal. However, the sample size is not big enough to dissect out all these particular characteristics in each group. Therefore, as we pointed out in the conclusions, future studies with higher number of patients including different characteristics should be done to confirm our results. Focusing on precipitant events, Methani et al. suggest that it would be interesting to know how many patients had a history of recent gastrointestinal bleed. Twenty-one patients had a recent gastrointestinal bleed. However, at the time of sample collection, only 8 samples presented macroscopically observable bleeding (melena) and any particular further analysis was performed. Methani et al remark the high percentage of patients with hepatic encephalopathy (HE) in the acute on chronic liver failure (ACLF) group. We agree that dysbiosis previously reported in patients with HE could play a role in decreasing richness of patients with ACLF. However, we disagree with the statement that HE is the primary determinant of loss of microbial richness in ACLF.4Moreau R. et al.Gastroenterology. 2013; 144 (1437.e1-9): 1426-1437Abstract Full Text Full Text PDF PubMed Scopus (1624) Google Scholar HE alone is not a sufficient criterion for ACLF and other organ failures should be added to fulfill the criteria of ACLF, meaning that other factors may also play a role in the significant loss of richness. We agree with Methani et al that CLIF Consortium Acute Decompensation and CLIF Consortium ACLF scores are better predictors of mortality of decompensated patients with cirrhosis.5Jalan R. et al.J Hepatol. 2014; 61: 1038-1047Abstract Full Text Full Text PDF PubMed Scopus (517) Google Scholar However, the Model for End-Stage Liver Disease is the criterion used most worldwide for prioritization for liver transplantation. As a consequence, we prefer to use this score to assess the accuracy of the models. We completely agree that the study was not designed to assess causality. This is the unsolved “chicken and egg” scenario. On the one hand, as liver cirrhosis progresses, it may cause dysbiosis and loss of bacterial richness, leaky gut, and bacterial translocation and complications of disease. On the other hand, there are factors such as biliary acids, alcohol, drugs, obesity, or others that may cause dysbiosis, which in turn may cause bacterial translocation, inflammation and progression of the disease. Until now, clinical studies were not able to demonstrate causality. Regardless, dysbiosis is a cause, a consequence, or both, manipulating microbiome (with probiotics, nonabsorbable antibiotics, or with fecal microbiota transplantation) to a more healthy microbiome seems to be promising.6Bajaj J.S. et al.J Hepatol. 2020; 72: 1003-1027Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar Restoring the gut microbiome with fecal microbiota transplantation in the field of patients with cirrhosis and recurrent HE has been shown to reduce hospitalizations and improve cognition and dysbiosis.7Bajaj J.S. et al.Hepatology. 2017; 66: 1727-1738Crossref PubMed Scopus (309) Google Scholar In addition, fecal microbiota transplantation in patients with alcohol use disorder is safe and is associated with a short-term reduction in alcohol craving and consumption with favorable microbial changes.8Bajaj J.S. et al.Hepatology. 2020 Aug 4; https://doi.org/10.1002/hep.31496Crossref Scopus (43) Google Scholar In summary, future studies including a multi-omics and functionality approaches and strategies for modulating gut microbiome need to be performed to answer the causality riddle and step forward the knowledge of the gut microbes.