COVERT HEPATIC ENCEPHALOPATHY: DIAGNOSIS, REVERSIBILITY AFTER LIVER TRANSPLANT AND ROLE OF GUT MICROBIOTA

Background. Hepatic encephalopathy (HE) produces a wide spectrum of nonspecific neurological and psychiatric manifestations. Mild HE that does not produce disorienting in time/space or asterixis is called covert HE (CHE). It occurs in 20%–80% of patients with cirrhosis. While significant progress has been made in understanding the importance of CHE, to date there is no consensus guidelines regarding the screening procedures for CHE and all the recognized techniques, although more or less sensitive and objective, require some kinds of equipment. Furthermore, the diagnosis of CHE can be confounded by other factors (as chronic alcohol misuse or HCV infection), which can cause cognitive alterations. It is also important to identify CHE in patients awaiting liver transplantation (LT) for the proper interpretation of cognitive disorders that may occure after transplantation. Infact, the influence of LT on mental performance is debated; as is the role of pretransplant HE. Notable, recent data on LT suggest that the cognitive dysfunction may not be totally reversible. Another important issue regards the treatment of CHE, in particular the effect of treatments on gut microorganism and ammonia production by microbial activity. HE treatment with prebiotics, antibiotics and probiotics, generally evidences a reduction of photogenic/ ammoniagenic bacteria and an increase in neurocognitive tests and mental status in patients. Nevertheless from a strictly microbiological point of view, little is known about the dynamics, interaction and metabolite production of the main bacterial groups in liver diseases. Aims. The present study aims to: 1) test a simple verbal psychometric test, called Animal Naming Test (ANT1), in the detection of CHE; 2) to evaluate the role of alcohol misuse, HCV infections, diabetes, aging and level of education as potential confounding factors in the diagnosis of CHE; 3) to evaluate the time course of the neuropsychological and electroencephalogram (EEG) features of patients with cirrhosis before and after LT with respect to prior HE and 4) to investigate how the HE treatments (by the use of lactulose, rifaximin and a probiotc mixture) affects gut microbial composition, determining changes in ammonia production. Materials and methods. For Aim 1: 208 healthy subjects and 327 consecutive patients with cirrhosis underwent the ANT1. Patients with cirrhosis were assessed by the Psychometric Hepatic Encephalopathy Score (PHES), a subgroup of 146 underwent also a quantified EEG and 95 the Critical Flicker Frequency. 202 patients were followed up for the occurrence of overt HE (OHE) and death. For Aim 2: a comprehensive neuropsychological profile and EEG spectral parameters were obtained in six age-matched groups of 30 subjects each: (i) HCV-related hepatitis without cirrhosis, (ii) chronic alcohol abusers, (iii) patients with HCV-related cirrhosis, (iv) alcohol-related cirrhosis, (v) cirrhosis not related to alcohol or HCV and (vi) healthy subjects. Cirrhotic patients were matched for MELD score. For Aim 3: the study population included 65 patients with cirrhosis on the transplant waiting list; 23 had a history of OHE. Each patient underwent an extensive psychometric assessment (10 tests, including paper and pencil tests and a computerized test) and an EEG before and 9 to 12 months after LT. EEGs were analyzed spectrally, and the mean dominant frequencies were obtained. For Aim 4: independent batch culture fermentations with controlled pH (6.8) were inoculated with fecal samples from six patients with cirrhosis (age 66±3.3 years; Child-Pugh A n=5 and B n=1); average MELD score 9±2.8). Seven different treatments with lactulose, rifaximin and VSL#3 or their combination were performed. Microbial populations were enumerated using flow cytometry Fluorescent in Situ Hybridization, while an ammonia concentration was determined at 0, 4, 10 and 24 hours. Results. Aim 1: in controls, the ANT1 was found relevantly affected only by extremely low levels of education (80 yrs; p15, 1=S-ANT1 10-15, 2=S-ANT1<10 sensitivity 83%, specificity 84%, respectively vs. PHES) that was correlated both to PHES (p<0.0001) and EEG (p=0.007). In the follow up, the S-ANT1 resulted to have prognostic value on the risk of OHE and death, so that a prognostic index could be computed. Aim 2: the factor ‘cirrhosis’ was associated with low Phonemic Verbal Fluency (PVF) and Difference between Trail Making Test B and A (TMT B-A) (p<0.001). Chronic alcohol misuse was associated with low PVF, TMT (B-A), Memory with Interference Task at 10 (ITM 10) and 30s (ITM 30) (all p<0.05). An interaction was found between the factors ‘cirrhosis’, ‘alcohol misuse’ and tests (p<0.01). HCV hepatitis reduced ITM 10 (p<0.05), but no interaction was found between ‘cirrhosis’, ‘HCV infection’ and tests (p= 0.14). The EEG parameters were mainly influenced by ‘cirrhosis’ (p<0.05), and EEG alterations were more pronounced in patients with alcoholic cirrhosis (p=0.04). Aim 3: Patients with a history of OHE before LT had worse cognitive performances (p<0.001) and EEG performances in comparison with their counterparts with a negative history. They also showed greater cognitive improvement after LT (p<0.01); however, their global cognitive performance remained slightly impaired (p <0.01). After LT, EEGs normalized for 98% of the patients (p