LAUSR

The study from Bajaj and colleagues, evaluating gut microbiome–metabonome changes in patients with cirrhosis receiving fecal microbiota transplant (FMT),(1) merits further discussion of its conclusions. In this study, patients with cirrhosis and hepatic encephalopathy (HE) taking standard of care therapy were randomized to either continue with standard of care alone or to also receive 5 days of antibiotics prior to a single FMT enema. The investigators demonstrate partial recovery of gut microbiota, bile acid, and shortchain fatty acid profiles for patients in the FMT arm and conclude that FMT directly restores gut microbiome–metabonome interactions in antibiotic-treated patients with cirrhosis.(1) However, there was no arm in this study where patients received antibiotics but no FMT; therefore, it is impossible to distinguish whether any changes observed were directly caused by FMT as claimed or at least partly represent purely gut microbiota recovery after completion of antibiotics. There are certainly grounds for questioning the efficacy of the FMT. Specifically, despite selecting a stool donor with high relative abundance of Lachnospiraceae and Ruminococcaceae, FMT recipients actually demonstrate a significantly reduced relative abundance of these bacterial families in stool post-FMT (day 20) compared to baseline values (day 0) (see Bajaj et al., fig. 2B(1)). Furthermore, it is well established that human gut microbiota structure recovers to a profile comparable to that found pre-antibiotics within a short period following completion of an antibiotic course.(2) Additionally, while poorly explored in cirrhosis, short-chain fatty acid(3) and bile acid profiles(4) reconstitute over a time course of days to weeks after antibiotic cessation in healthy rodents without any specific intervention. If FMT directly treated HE through modulation of gut microbiota–metabonome interactions, then it would be expected that at least some of these variables would exceed baseline values post-FMT, rather than only match baseline levels at best, as shown. Adverse events associated with FMT are uncommon, but serious adverse events are described—including bacteraemia— in the population without cirrhosis;(5) such concerns are clearly amplified in patients with decompensated cirrhosis, given this group’s propensity to and outcomes from sepsis. Until the safety profile of FMT in patients with cirrhosis has been further evaluated—and a potential mechanistic link between FMT and improvement in HE defined—FMT will remain at too early a stage to consider more widely as treatment for HE.