Date of Award

January 2015

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Guadalupe Garcia-Tsao

Subject Area(s)



Hepatic encephalopathy (HE) describes the constellation of neuropsychological and neurophysiological findings in patients with cirrhosis. HE consists of two subtypes: covert HE (mild) and overt HE (severe). Because a concrete understanding of covert HE is lacking, so is our framework for standardizing the assessment of such patients. In addition, studies of neurocognitive abnormalities in patients with cirrhosis have combined compensated and decompensated patients. These are two entirely separate disease entities with different survival times and predictors of death, therefore patients with compensated and decompensated cirrhosis should not be managed or investigated jointly. Thus, our objectives were 1) to evaluate the prevalence of neurocognitive abnormalities in patients with compensated cirrhosis (if present, characterize these abnormalities), 2) compare study subjects with and without neurocognitive abnormalities to determine differences in pathophysiological or other variables. We hypothesized that 1) a subgroup of patients with compensated cirrhosis will have neurocognitive abnormalities, 2) patients with neurocognitive impairment may have more severe liver dysfunction or portal hypertension compared to patients without impairment. Patients with compensated cirrhosis (no ascites, esophageal variceal hemorrhage (EVH), HE or jaundice) that met inclusion and exclusion criteria underwent neurocognitive testing from 10/2013 - 1/2015 at two tertiary care academic medical centers using ‘pencil and paper’ tests widely available in the United States. The neurocognitive test battery evaluated attention, learning/memory, visuospatial/visuomotor, language, psychomotor speed and executive function. Patients were evaluated for mild neurocognitive impairment (mNCI) based on test performance using two sets of criteria: mNCI-1 and mNCI-1.5. Demographic, clinical and biochemical variables were compared between patients with and without mNCI. Wilcoxon rank-sum test and Student’s t-test were used for continuous variables and Fisher’s exact test for categorical variables. 2-sided P-value ≤ 0.05 was considered statistically significant. 42 patients with compensated cirrhosis (median age 61 years, range 44 - 88) were included. Using mNCI-1 criteria, 10/41 patients (24.4%) had mNCI; domains most affected were learning and memory (21/42 [50.0%]) and psychomotor speed (14/41 [34.15%]). Study patients with mNCI using mNCI-1 criteria had statistically significant elevations in aspartate aminotransferase and Fibrosis-4. Using mNCI-1.5 criteria, 20/41 patients (48.8%) had mNCI; domains most affected were attention (17/42 [40.5%]), learning and memory (23/42 [54.8%]) and psychomotor speed (18/41 [43.9%]). Using both criteria, patients with mNCI had more severe liver dysfunction and portal hypertension than patients without mNCI. Increasing the sample size and following patients prospectively for development of overt HE or other decompensating events (ascites, EVH, jaundice) will inform which criteria for mNCI is optimal in predicting clinical course.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.