Multicenter Study of Microbial Profiles for Management of Hepatic Encephalopathy in Veterans with Cirrhosis

Cognitive impairment is epidemic in Veterans with cirrhosis, which poses a major psychosocial, financial, and clinical burden. While it is assumed that hepatic encephalopathy (HE, both minimal, MHE or overt forms) is the major cause of this impairment, recent work by our group has shown the frequent other causes/contributors in up to half of Veterans with cognitive complaints referred to our VAs specialized testing clinic had a non-HE related cause. The differential diagnosis for non-HE cognitive complaints includes mild cognitive impairment (MCI) and dementia, mood disorders, post-traumatic stress disorder (PTSD), obstructive sleep apnea (OSA), and Parkinson’s. A major issue is the lack of confirmatory test(s) for HE and the often-flawed current mode of clinical diagnosis. As a result, most clinicians treat every Veteran with cirrhosis with cognitive complaints for HE which results in therapies that are either poorly tolerated (lactulose), or expensive (rifaximin), while ignoring underlying causes of cognitive complaints. Specialized cognitive testing clinics, detailed inquiry of symptoms, co-morbid conditions, and interpretation are ideal but not practical without training, expertise, and significant financial investment. Therefore, we need better strategies to reliably diagnose/exclude HE, which will minimize unnecessary poorly tolerated and expensive therapy, avoid misdiagnosis in roughly half of patients, and guide clinicians taking care of Veterans with cirrhosis. Microbiome composition is altered in patients with HE (both MHE and overt) and predicts clinically relevant outcomes. Our recent research showed that the gut microbial analysis accurately differentiates between microbial profiles due to addiction disorders, PTSD, MCI/dementia, and rules out MHE-related cognitive impairment. Specificity of a clinical HE diagnosis was 50% versus >90% for microbiome profiling when both were compared to the detailed neuropsychological evaluation. As a result, reliable microbiome profile analyses could enhance clinical decision making regarding HE in Veterans with cirrhosis. Our preliminary data, shows that Veterans with cirrhosis were comfortable with stool collection using standardized kits across the three proposed study sites. Using machine learning methods in >300-patient training cohort, we found that 40% of those treated for HE did not have a microbial profile predicted for HE patients, which led to the re-evaluation of these patients’ need for HE treatment. These profiles remained stable over time. However, to operationalize this diagnostic test, it needs to be extended further to help clinicians examine or re-examine the HE diagnosis and need for poorly tolerated and/or expensive therapy. Our hypothesis is “Rapid gut microbial profiling in Veterans with cirrhosis can guide clinicians towards better management of cognitive complaints rather than needless hepatic encephalopathy treatment”. We will test the following aims in a multi-center prospective study across 3 large VA centers (Richmond, West Haven, and North Texas) in 300 Veterans with cirrhosis: Aim 1: Determine the stool microbiome composition using rapid sequencing and compare the results of cognitive testing performed at the time of stool collection in a multi-center study of Veterans with cirrhosis. We will enroll 300 Veterans outpatients with cirrhosis and analyze stool by rapid sequencing using nanopore. All patients will undergo detailed patient-reported outcomes, cognitive testing, neuropsychological assessment, and interview to determine presence of and causes of cognitive impairment. The primary outcome is the specificity of the microbiome-related MHE diagnosis compared to the neuropsychological gold standard. Aim 2: Determine the impact of the microbiome profile test results on clinical practice and patient-reported outcomes in Veterans with cirrhosis. The primary outcome is the proportion of patients with a discordant microbiome MHE profile from their neuro Project Number: 1I01CX002863-01A1 | Fiscal Year: 2026 | NIH Institute/Center: Veterans Affairs (VA) | Principal Investigator: Jasmohan Bajaj | Institution: VA VETERANS ADMINISTRATION HOSPITAL, RICHMOND, VA | Activity Code: I01 | Study Section: Special Emphasis Panel[ZRD1 GAST-L (01)] View on NIH RePORTER: https://reporter.nih.gov/project-details/11183157