Background and aims

Minimal hepatic encephalopathy (MHE) is the mildest form of hepatic encephalopathy (HE) and is characterized by deficits in neurocognitive performance without any clinical symptoms of HE. In the current study, we aim to evaluate and compare the neurocognitive, biochemical, and brain magnetic resonance (MR) imaging changes between patients with cirrhotic MHE and extrahepatic portal vein obstruction (EHPVO) MHE.

Methods

Thirty-three cirrhotic and 14 EHPVO patients were diagnosed with MHE and were included in the analysis along with 24 normal healthy volunteers. All subjects underwent MR imaging including diffusion tensor imaging and proton MR spectroscopy (¹ H-MRS) followed by cognitive assessments, critical flicker frequency (CFF) measurements, quantification of blood ammonia, and serum proinflammatory cytokine levels.

Results

We observed abnormal neurocognitive functions and CFF measurements in both cirrhotic MHE and EHPVO MHE patients as compared with controls. Significantly increased blood ammonia, serum proinflammatory cytokines (IL-6, TNF-α) level, mean diffusivity in multiple brain sites, ¹ H-MRS derived glutamate/glutamine (Glx)/creatine (Cr), and significantly decreased ¹ H-MRS derived myo-inositol/Cr were observed in both cirrhotic MHE and EHPVO MHE compared with those of controls. Choline/Cr level was significantly decreased in cirrhotic MHE as compared with controls and EHPVO MHE.

Conclusions

Cirrhotic MHE showed more severe changes on mean diffusivity in multiple brain sites and inflammation as compared with EHPVO MHE. This study confirms that there are significant difference in neurocognitive, biochemical, and MR profile between cirrhotic MHE and EHPVO MHE, which may help to understand the pathophysiologies of these two types of MHE and may contribute to improve their clinical managements.

Background

Acute-on-chronic liver failure (ACLF) is a form of liver disease with high short-term mortality. ACLF offers considerable potential to affect the cortical areas by significant tissue injury due to loss of neurons and other supporting cells. We measured changes in cortical thickness and metabolites profile in ACLF patients following treatment, and compared it with those of age matched healthy volunteers.

Methods

For the cortical thickness analysis we performed whole brain high resolution T1-weighted magnetic resonance imaging (MRI) on 15 ACLF and 10 healthy volunteers at 3T clinical MR scanner. Proton MR Spectroscopy ((1)H MRS) was also performed to measure level of altered metabolites. Out of 15 ACLF patients 10 survived and underwent follow-up study after clinical recovery at 3 weeks. FreeSurfer program was used to quantify cortical thickness and LC- Model software was used to quantify absolute metabolites concentrations. Neuropsychological (NP) test was performed to assess the cognitive performance in follow-up ACLF patients compared to controls.

Results

Significantly reduced cortical thicknesses in multiple brain sites, and significantly decreased N-acetyl aspartate (NAA), myo-inositol (mI) and significantly increased glutamate/glutamine (glx) metabolites were observed in ACLF compared to those of controls at baseline study. Follow-up patients showed significant recovery in cortical thickness and Glx level, while NAA and mI were partially recovered compared to baseline study. When compared to controls, follow-up patients still showed reduced cortical thickness and altered metabolites level. Follow-up patients had abnormal neuropsychological (NP) scores compared to controls.

Conclusions

Neuronal loss as suggested by the reduced NAA, decreased cellular density due to increased cerebral hyperammonemia as supported by the increased glx level, and increased proinflammatory cytokines and free radicals may account for the reduced cortical thickness in ACLF patients. Presence of reduced cortical thickness, altered metabolites and abnormal NP test scores in post recovery subjects as compared to those of controls is associated with incomplete clinical recovery. The current imaging protocol can be easily implemented in clinical settings to evaluate and monitor brain tissue changes in patients with ACLF during the course of treatment.