Date of Award

January 2018

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Guadalupe García-Tsao

Second Advisor

Erica Herzog


Patients with cirrhosis, particularly those with decompensated cirrhosis, are at increased risk of bacterial infection, and bacterial infections are, in turn, associated with the development of the complications of cirrhosis, which may lead to multiorgan failure and increased mortality. Bacterial translocation (i.e., the passage of microrganisms and/or their products to mesenteric lymph nodes or other extraintestinal organs or sites) and the resultant systemic inflammation are key events in the development of infection and “further” decompensation in patients with cirrhosis. Non-selective beta blockers, such as propranolol, may reduce bacterial translocation and inflammation, perhaps via effects on intestinal permeability and bacterial overgrowth.

The aim of this study is to compare markers of intestinal permeability, bacterial translocation and inflammation in patients with cirrhosis at different stages of disease (compensated vs. decompensated) and before and after treatment with propranolol. Adult patients (>18 years old) with cirrhosis undergo testing for the evaluation of: (a) intestinal permeability by measuring urinary excretion of orally-administered sucrose, lactulose, and mannitol (SLM test), (b) gut mucosal integrity via assessment of tight junction proteins (immunohistochemistry and RT-PCR) in gastric and duodenal biopsies, (c) markers of systemic inflammation (plasma IL-1β, IL-6, TNF-α, IL-10, CRP), and (d) indirect markers of bacterial translocation (serum lipopolysaccharide and lipopolysaccharide-binding protein).

The results in this thesis are based on baseline analysis of 23 patients with cirrhosis (9 compensated, 14 decompensated) and a subset of 10 patients (8 decompensated, 2 compensated) that had baseline and post-propranolol assessments. Enrollment in this study is ongoing, and a preliminary analysis of partial data (intestinal permeability and inflammatory markers) is presented. Urine sucrose recovery and sucrose/mannitol ratio (markers of gastroduodenal permeability) was higher in decompensated than in compensated subjects, while lactulose/mannitol ratio (a marker of intestinal permeability) was similar between prognostic stages. There were no significant differences in IL-1β, IL-6, TNF-α, or IL-10 between compensated versus decompensated patients, although there was a uniform tendency for higher levels in decompensated patients. Markers of gastroduodenal/intestinal permeability and systemic inflammation were unchanged after treatment with propranolol.

In conclusion, preliminary analysis of the data suggests an increase in gastroduodenal permeability in patients with decompensated cirrhosis that is unaffected by treatment with propranolol. Completion of patient enrollment and analysis of remaining parameters (markers of gut mucosal integrity, indirect markers of bacterial translocation) are eagerly awaited.


This thesis is restricted to Yale network users only. It will be made publicly available on 06/25/2100