Etiology of cirrhosis predicts deterioration of “brain reserve” after hepatic encephalopathy or transjugular intrahepatic portosystemic shunt

See Article on Page 1123 Neurological complications remain a common problem following liver transplantation affecting as many as 47% of transplant recipients. Cognitive abnormalities are more prevalent and more severe in patients who had suffered 1 or more episodes of overt hepatic encephalopathy (OHE) before transplant.1 Moreover, an alcoholic etiology for cirrhosis leads to a poorer posttransplant neurocognitive outcome.2 In a study published in this issue of Liver Transplantation, Ahluwalia et al.3 give evidence to support the notion that a worsening of “brain reserve” (BR) resulting from alcoholic liver disease (despite abstinence) has the potential to modulate the impact of hepatic encephalopathy (HE) and the course of the disease with a likely impact on transplant suitability. BR is a relatively new concept that is characterized in general terms by structural changes and is dependent on the etiology, severity, and progression of disease. Alterations of BR have been described in a range of neurological diseases including dementia, multiple sclerosis, and, more recently, HE. It should be noted that an entity related to BR and known as “cognitive reserve” exists, but unlike the structurally defined BR, it is a dynamic phenomenon based on socioeconomic and educational factors. Cognitive reserve has been shown to be a determinant of health‐related quality of life in patients with cirrhosis.4 Using a sophisticated multimodal magnetic resonance imaging (MRI) and spectroscopic series of techniques, Ahluwalia et al.3 assessed BR in abstinent alcoholic patients with cirrhosis compared to nonalcoholic patients with cirrhosis who were matched for age, Model for End‐Stage Liver Disease score, and venous ammonia. Results demonstrate that diminished BR was related to 3 factors identified by their magnetic resonance techniques, namely, structural changes in white matter (WM) and gray matter (GM) of multiple brain structures, the presence of alterations of water movement resulting in edema, and, using magnetic resonance spectroscopy, changes in brain metabolites reflecting the consequences of hyperammonemia on brain metabolism. Alcoholic patients with cirrhosis performed poorly on cognitive testing compared to nonalcoholics. Furthermore, magnetic resonance results suggested enhanced cortical damage in the alcoholics where the effect of HE status on cognitive tests and BR was also enhanced. A prospective study performed before and after transjugular intrahepatic portosystemic shunt (TIPS) placement revealed that the relative deterioration in the metabolic profile in alcoholic patients after TIPS was insignificant compared to nonalcoholics, a finding that was attributed to maximal induction of metabolites before TIPS.3 Consequently, it was suggested that the potential for BR changes in nonalcoholics may be greater with a potential for loss of function after TIPS. The notion of a deleterious effect of an alcoholic etiology upon cortical function in patients with cirrhosis is borne out by findings from a previous study using the technique of critical flicker frequency (CFF) for grading of HE severity in cirrhosis. Using this technique for the quantification of low‐grade HE, it was demonstrated that alcoholic etiology of cirrhosis resulted in significantly more disturbances of CFF compared to age‐matched nonalcoholics even though patients had been studied at comparable stages of minimal hepatic encephalopathy (MHE) and OHE that had been demonstrated using computerized psychometric testing together with an assessment of mental state by West Haven criteria.5 Flicker‐fusion frequency analysis reflects the working of the visual system as well as other components of the cerebral cortex and shows that an alcoholic etiology of cirrhosis has the potential to result in additional impairments of cortical function and BR due to effects of alcohol per se. Diffusion tensor imaging revealed alterations in WM suggestive of interstitial edema in the alcoholic group leading to the suggestion that microstructural damage in WM due to alcohol may combine with the effects of cirrhosis even after abstinence. Similar results were found in GM with lower GM density observed in reward‐related areas of the brain as well as in the anterior cingulate cortex, a brain structure associated with the control of attention and one of the earliest structures to manifest reduction in activity in patients with cirrhosis with MHE using the technique of the [18F]fludeoxyglucose–positron emission tomography scan.6 Chronic alcoholism is known to cause structural changes in the brain, the causes of which are diverse in nature involving direct effects of the ethanol molecule as well as effects mediated by thiamine deficiency leading to Wernicke's encephalopathy (WE). Ethanol causes thiamine deficiency by inhibition of gastric transport of the vitamin and by reduction of hepatic thiamine stores. Moreover, in 1 study, histopathological examination of brain tissue from patients with cirrhosis who died in hepatic coma revealed unsuspected thalamic or cerebellar neuropathology in 56% of patients, most of which was judged to be longstanding in nature.7 It is generally considered that the thalamic lesions and consequent impairment of thalamocortical inputs because of thiamine deficiency contributes to the encephalopathy that includes confusion/confabulation and memory impairment characteristic of WE. Similar symptoms have been reported as neurological sequelae following liver transplantation.8 Other neurological disorders that have been described in end‐stage liver disease include acquired hepatocerebral degeneration, which although rare, includes damage to posterior frontal, parietal, and occipital cortices.9 In another study, 22% of a group of patients with cirrhosis on the liver transplant wait list were found to have basal ganglia changes on MRI. Neurological examination revealed a disorder that closely resembled Parkinson disease10 subsequently becoming known as parkinsonism in cirrhosis. The destructive lesions in the disorder were attributed to manganese deposition rather than hyperammonemia,11 and in contrast to the findings in the Ahluwalia et al.3 study, the extent of neuronal damage in parkinsonism in cirrhosis was not dependent on the etiology of cirrhosis. In conclusion, it is evident that patients with alcoholic cirrhosis have diminished BR that results in impaired cognitive performance. BR in these patients continues to be modulated by HE despite abstinence. Other potential causes of diminished BR in alcoholics relate to the direct effects of ethanol on the brain and on inhibition of thiamine transport in the gastrointestinal tract together with reduced hepatic thiamine stores leading to thalamic neuronal cell loss characteristic of WE. In contrast to alcoholics, patients with nonalcoholic cirrhosis have an increased susceptibility to develop BR deterioration after HE or TIPS. It is hoped that further elucidation of the structural and functional basis of cognitive reserve and BR in patients with cirrhosis as a function of the etiology of liver disease will guide medical teams in the future effective management and support of caregivers and patients.