Data indicate that hepatic encephalopathy (HE) is an important morbid complication of cirrhosis, occurring in an estimated 75% of those with cirrhosis. With HE diagnosis often being subjective and usually based on clinical assessment of the patient and the majority of HE often being subtle and difficult to diagnose with current tools, imaging modalities may offer a potential technique for objective assessment, explains Mark Danta, B Med, MPH, MD, FRACP. “Metabolite changes in the brain are key to the pathophysiology of hepatic encephalopathy, which is incompletely understood,” he adds. “Magnetic resonance spectroscopy (MRS) allows measurement of specific metabolites by brain region.”

Analyzing MRS for HE

For a paper published in Neurology, Dr. Danta and colleagues conducted a meta-analysis of 31 observational studies with the goal of evaluating whether a derangement of cerebral metabolites as measured by MRS occurs with HE. “In particular, we analyzed the capability of MRS to differentiate between cirrhotic individuals with no HE (NHE) and minimal HE (MHE), given its relevance in clinical practice,” Dr. Danta notes. Included studies had adequate ascertainment of cirrhosis and hepatic encephalopathy assessment, excluded patients with liver transplant or transjugular intrahepatic portosystemic shunt, had an appropriate comparator group for the same imaging modality, included statistics that allowed confidence intervals to be calculated, had a minimum of 10 cases, and specified the HE grade of patients. The standard meta-analysis was performed based on the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines.

In total, the studies included nearly 1,500 patients with analysis of cirrhosis-related HE, classified into healthy controls and patients with NHE, MHE, or overt HE (OHE). Most studies reported findings on at least three of the four metabolites glutamate + glutamine (Glx), myo-inositol (mI), choline-containing compounds (Cho), and N-acetylaspartate (NAA), with findings mainly stratified into the parietal, occipital, and basal ganglia brain regions, which were each analyzed separately. Studies were published from 1992 to 2016 and the majority were from Europe and North America.

A Pattern Emerges

“The analysis revealed that glutamine levels in the parietal lobe by MRS was effective in distinguishing patients with MHE from those with NHE (+0.82), without substantial heterogeneity or publication bias,” says Dr. Danta. “Comparing NHE with MHE, imaging targeting the parietal lobe was found to be effective in distinguishing grades of HE. Similarly, myo-inositol (-0.77) and choline (-0.36) were effective in parietal cortex imaging comparing MHE and NHE patients. Imaging targeting the basal ganglia and occipital lobes had weaker correlations. Importantly, the NHE vs healthy control results indicate that the derangement of metabolites begins in chronic liver disease before the development of a clinically detectable encephalopathic condition.”

The results reveal that glutamine change in the parietal lobe has the strongest association with HE, a relationship that increases with increasing severity of HE, according to Dr. Danta (Table). “While choline and myoinositol decreased with increasing HE severity, the relationship was less clear,” he adds. “Taken together it suggests that there is a specific identifiable pattern for HE and that MRS may be of value in the assessment of HE. Recognition of this pattern may allow improved diagnosis, monitoring, and study of interventions in patients with HE.”

Focusing the Use of MRS 

The finding of glutamine as the most consistent metabolite, with the strongest correlation in the parietal cortex, in differentiating between NHE and MHE identifies a specific region to focus the use of MRS in the diagnosis of HE, says Dr. Danta. “Interestingly, these changes may precede the development of detectable HE,” he notes. “It also provides opportunities to study the longitudinal progression or reversal of imaging markers over time, following HE treatment or transplant. To better facilitate direct comparison between imaging modalities in the future, researchers should design more multimodal studies that take advantage of contemporary and evolving imaging techniques.”