A common complication of advanced liver disease is a condition called hepatic encephalopathy, which leads to confusion. The current treatment options cause side effects, are costly, and do not always work. An abnormal population of bacteria in the intestines may be causing this condition, and transplanting bacteria from the colon of a healthy person may treat it. In this research study, the investigators will first find two healthy stool donors whose stool donation improves the gut bacteria of patients with advanced liver disease and helps them think more clearly. Then, in a randomized controlled trial, the investigators will compare the ability of stool donation from these two best donors versus a placebo to improve the neurological function of patients with advanced liver disease. If the investigators find the expected results, there will be a new effective therapy for patients with advanced liver disease and the very troublesome complication of hepatic encephalopathy.

Decades of investigation demonstrate that hepatic encephalopathy (HE), a common complication of cirrhosis characterized by impaired cognition, develops as a consequence of intestinal microbial products reaching the brain. Recent investigation has found that cirrhotic patients, especially those who have developed HE, have intestinal dysbiosis compared to normal controls. Several plausible mechanisms explain how intestinal dysbiosis could lead to HE. There is limited prior literature on the efficacy of FMT in cirrhosis. The largest documented study of 10 cirrhotic patients receiving a single FMT enema found no significant change in microbiome diversity as assessed by 16S rRNA sequencing. The investigators hypothesize that aggressive manipulation of the microbial composition with fecal microbiota transplant (FMT) will improve neurological function in patients with a history of cirrhosis and HE. The investigators additionally hypothesize that five oral FMT capsule administrations from a previously efficacious stool donor will significantly change the intestinal microbiome composition of a cirrhotic patient. The study will consist of a 10-patient open-label pilot study to identify efficacious stool donors, defined as donors who precipitate the largest improvement in recipient neurological function and microbiome composition. The two most efficacious pilot study stool donors will be selected to donate stool for the randomized controlled trial (RCT). The 20-patient RCT will investigate the effect of FMT on neurological outcomes in patients with cirrhosis and a history of HE. Subjects will be randomized to receive 5 doses of oral FMT capsules or placebo capsules over 21 days. Cognitive testing and stool collections will occur at 4 time points, to assess for changes in neurological function and microbiome composition. The primary outcome is change in neurological function after FMT. The main secondary outcome is change in microbiome composition after FMT. This study could provide valuable information about the ability of FMT to improve intestinal dysbiosis in cirrhosis and treat HE.

Inclusion Criteria:

• Diagnosis of cirrhosis: Based on liver biopsy or clinical assessment of a hepatologist based on history, exam, laboratory and radiographic evidence
• History of at least one episode of overt HE, defined by West Haven Criteria Grades II to IV; episodes of HE that were precipitated by gastrointestinal hemorrhage requiring transfusion of at least 2 units of blood, by medication use, by renal failure requiring dialysis, or by injury to the central nervous system will not be counted as previous HE episodes
• Compliant with lactulose and rifaximin treatment (lactulose: at least one dose at least 5 days per week; rifaximin: at least one dose at least 5 days per week)

Exclusion Criteria:

• Current episode of overt HE as defined by West Haven Criteria Grades II to IV
• Expectation of liver transplantation within two months of the screening visit
• Current infection
• Variceal bleeding in the last 4 weeks
• Gut-absorbable or intravenous antibiotic therapy (including ciprofloxacin for SBP prophylaxis) in the last 3 months
• Alcohol or illicit drug intake within 3 months, by history and available serum testing; alcohol use will be characterized as >1 alcoholic drink / month
• PSC as etiology of liver disease, as prior literature has suggested these individuals have a unique microbiome
• History of Roux-en-Y Gastric bypass
• On immunosuppressive medications
• Positive C. difficile test
• Scoring above a threshold cut-off on the Psychometric Hepatic Encephalopathy Score (PHES)
• MELD > 17
• History of spontaneous bacterial peritonitis
• History of low ascites protein ( ≤ 1g/dL) in the last year
• Hemodialysis in the last 30 days
• Other significant laboratory abnormalities: serum creatinine > 2.0 mg/dL, hemoglobin < 8 g/dL, serum sodium < 125 mmol/L, serum calcium > 11.0 mg/dL, serum potassium < 2.5 mmol/L
• Placement of a portosystemic shunt or transjugular intrahepatic portosystemic shunt
• Unstable doses of opiates, benzodiazepines or other sedating medication
• Unable to provide consent; a. If MMSE is < 18 or the patient is deemed to not have capacity by an investigator, a legally authorized representative (surrogate) will be allowed to provide consent

Randomization will be performed by the lab producing the FMT and placebo capsules. Participants, providers, investigators and research assistants will be blinded.