Alcohol's Effects on the Brain: Neuroimaging Results in Humans and Animal Models

1Apart from direct effects on the brain, excessive alcohol consumption is associated with increased risk for trauma (i.e., traumatic brain injury), seizures, and stroke, each of which can have effects on brain structure independent of alcohol or each other. Furthermore, alcohol can alter the brain by affecting peripheral organs, including the digestive tract (e.g.,; ), liver (e.g., ), heart (e.g., ), pancreas (e.g., ), kidneys (e.g., ), and lungs (e.g., ). Mechanisms of these indirect effects of alcohol on the brain are likely mediated via soluble factors, such as ceramides (e.g., ).

2To evaluate alcohol’s central nervous system effects, researchers distinguish “uncomplicated alcoholism” (i.e., alcohol use disorder [AUD]) from the various clinically diagnosable consequences of chronic alcohol consumption, including Wernicke’s encephalopathy (WE), Korsakoff’s syndrome (KS), hepatic encephalopathy (HE), central pontine myelinolysis (CPM), alcoholic cerebellar degeneration (ACD), alcohol-related dementia (ARD), and Marchiafava-Bignami disease (MBD).1 The use of brain-imaging technology to evaluate clinically defined syndromes associated with chronic alcoholism, each with relatively unique radiological signatures (see table 1 and figure 1), provides guideposts for studying brain alterations associated with uncomplicated alcoholism.

3Approximately 7 percent of adults age 18 and older have an AUD. Prevalence estimates of alcoholism-related syndromes are difficult to ascertain. Incidence estimates often are based on postmortem findings. Postmortem evaluation indicates a prevalence of 2 percent of WE in the general population; however, as many as 12 to 18 percent of alcoholics can have postmortem evidence of WE. Based on observations that 80 to 85 percent of patients with WE can develop KS, the estimated prevalence of KS is 11 to 12 percent of the alcoholic population. Wernicke-Korsakoff syndrome (WKS) is used to refer to the presence of both WE and KS because of the close relationship between the two disorders.

4Estimates of HE are derived from estimates of alcoholic cirrhosis, which can range from 8 percent to 20 percent. Mild HE occurs in up to 80 percent of cirrhotic patients, and overt HE occurs in up to 45 percent of cirrhotic patients. One study estimated the incidence of CPM at 0.5 percent among the general population. However, prevalence is much higher (30 percent) among patients with liver transplants. For ACD prevalence, reports based on postmortem evaluation range from as low as 0.4 percent to as high as 42 percent of alcoholics. Rates of ARD can depend on the setting, with facilities specializing in early identification and treatment of memory disorders reporting rates of 3 percent and nursing homes reporting rates as high as 24 percent. Prevalence can also depend on the age of the population evaluated (i.e., higher prevalence of ARD is found in younger-onset [i.e., ages 45–64] dementia). MBD appears to be very rare, with only about 250 cases reported between 1966 and 2001.

5Human studies offer a full depiction of the consequences of chronic alcohol exposure but are limited by ethical considerations. That is, rigorous experimentation requires the ability to control for relevant variables such as the premorbid condition of the brain. The wide variation (or heterogeneity) of alcoholic populations examined with respect to genetic predisposition, age of onset, pattern of drinking, frequency of withdrawals, length of sobriety, nutritional, and hepatic status has hampered researcher attempts to isolate specific brain regions and mechanisms affected by alcohol, per se. This heterogeneity, and the complexity that it introduces, makes it difficult to thoroughly characterize the disorder. Animal models, in contrast to the indefinite natural course of alcohol use in humans, allow researchers to determine alcohol toxicity in a way that allows them to control for multiple genetic, environmental, and alcohol consumption factors. Animal models permit the study of underlying mechanisms, enabling researchers to better interpret findings from human studies.

6On the other hand, animal models also have limitations. Species differences in brain structure and function—among myriad other differences between humans and other animals—can give inadequate information when animal data are applied to human disease. For example, mice models fail to mimic human inflammatory disease with respect to genomic responses, and corticosteroids disturb development in animals but not in humans. Furthermore, researchers have hypothesized that the design, conduct, and analysis of a mainstay of animal experiments are questionable and rarely undergo meta-analytical review for consensus.

7This article reports key findings in humans, from macrostructural findings using magnetic resonance imaging (MRI), microstructural findings using diffusion tensor imaging (DTI), and metabolic findings from MR spectroscopy (MRS). Studies of alcohol-related central nervous system disorders are used as a framework for findings in uncomplicated alcoholism. The article also examines studies of abstinence and relapse and current imaging studies of animal models of alcoholism and co-occurring brain disorders. The evidence suggests that human studies are necessary to identify and classify the brain systems modified by concomitants of alcoholism versus alcoholism, per se, and that animal models of alcoholism and its co-occurring brain disorders are essential for a mechanistic understanding of vulnerable brain systems.

8Since the early 1980s, conventional structural MRI has allowed researchers to visualize the living human brain. Detailed images of the brain are possible in part because the different brain tissue types (i.e., gray matter, white matter, and cerebrospinal fluid [CSF]) contain different proportions of water. With MRI, the brain can be viewed from bottom to top (axial), from front to back (coronal), from left to right (sagittal), or at any oblique angle to these planes. This flexibility also enables greater accuracy in aligning images with internal landmarks, an essential consideration for ensuring consistency of data from replicate images from the same individual.

9WE occurs with chronic alcoholism and thiamine deficiency. If untreated, WE patients can develop KS, a severe neurological disorder characterized by anterograde amnesia. Malnutrition, vomiting, and diarrhea are common in chronic alcoholism and can contribute to thiamine deficiency. Further, the gastrointestinal tract’s ability to absorb necessary quantities of thiamine is diminished in alcoholics, and the liver, which houses a large part of the body’s supplies of thiamine, may not be able to store thiamine in the same capacity if it is in a diseased state. Classical clinical signs of WE included visual, gait, and mental disturbances, but more recent assessments describe mild, moderate, and severe signs and symptoms including anorexia, loss of memory, and emotional changes. An MRI image of acute WE (see figure 2) has symmetrical bright spots, or hyperintensities, clearly visible on T2-weighted images, and those created by fluid attenuation inversion recovery2 (FLAIR). The bright spots appear in the midbrain gray matter surrounding the cerebral aqueduct (i.e., periaqueductal gray matter), mammillary bodies, and tissue surrounding the third ventricle3. These findings agree with postmortem diagnosis of WE, often requiring evidence of lesions in the mammillary bodies and periventricular areas (e.g., ). In addition, observed MR hyperintense areas in WE include the thalamus, cerebellar vermis, dorsal medulla, tectal plates, olivary bodies, and dorsal pons. MRI analysis of KS patients compared with unaffected research participants (i.e., nonalcoholic control subjects) revealed substantial volume shrinkage of the mammillary bodies in KS and a lesser but significant volume deficit in uncomplicated alcoholics (;; but see; ). In contrast with early MR studies suggesting that KS affects the mammillary bodies while sparing the hippocampi, more recent work demonstrates hippocampal volume deficits in KS. Other regions affected by KS are the thalamus, orbitofrontal cortex, cerebellum, and pons.

10HE, occurring in acute or chronic liver disease, including acute liver failure and cirrhosis, is believed to arise, at least partially, from high levels of ammonia circulating in the blood. HE patients may appear confused and disoriented and have poor coordination. T1-weighted images of HE show bilateral, symmetrical, and high-intensity signals in basal ganglia structures, particularly the globus pallidus and substantia nigra (see figure 3). T2-weighted FLAIR images show hyperintense signals along the corticospinal tract and diffuse increases in white-matter signal intensities in the cerebral hemispheres. These in vivo MR features correspond with evidence of increased numbers of nonneuronal (i.e., glial) cells called astrocytes in basal ganglia and cerebral cortex of HE brains. Although discriminating features of WE and HE have been outlined, these diseases can be difficult to differentially diagnose and distinguish, because patients can appear to have similar symptoms and comparable MRI results, especially among alcoholics.

11A significant proportion of CPM cases include a history of alcoholism. People with CPM, which is associated with electrolyte disturbances and specifically with aggressive correction of low sodium levels in the blood (i.e., hyponatraemia), may have symptoms such as the inability to control facial movements, decreased voluntary muscle control (i.e., ataxia), and acute changes in consciousness. Classically, CPM was characterized by the presence of a symmetric triangular or “bat-wing” lesion in the pons, with hypointense T1-weighted and hyperintense T2-weighted images (see figure 4) reflecting damage to the protective covering of nerve cells (i.e., demyelination) as noted postmortem. The term osmotic myelinolysis (e.g.,; ) was coined to reflect the fact that other brain regions (e.g., basal ganglia, thalami, and cerebral gray–white matter junctions) are affected in CPM (e.g.,;;;;; ), despite suggestions that pathology in these other regions may not strictly represent demyelination. Because a postmortem study of 112 autopsy cases of CPM patients reported that 28 percent could also be diagnosed with WE, pontine dysfunction should be regarded as a cardinal clinical sign of CPM.

12ACD patients most frequently display ataxia, although other symptoms can include uncontrollable and repetitive eye movement (i.e., nystagmus) and speech problems resulting from impaired muscle control (i.e., dysarthria). Neuroimaging in ACD demonstrates damage disproportionately apparent in anterior superior portions of the cerebellar vermis, with postmortem pathology indicating loss of cerebellar Purkinje cells.

13Alcoholic dementia, or ARD, a currently preferred term, remains a controversial diagnosis because of confounding syndromes such as WE and HE. Nevertheless, certain clinically distinguishing features of ARD exist. It often occurs in socially isolated men at younger ages of onset (i.e., younger than age 65) than other types of dementia; deficits in visuospatial, executive, and memory functions; slower progression compared with other types of dementia; and partial reversibility. ARD is considered a frontal dementia. In support of such categorization, forensic evaluation of a sample of alcoholic brains noted a consistent pattern of synaptic loss in the superior laminae of the frontal cortex (i.e., Brodmann area 10), not related to liver disease.

14MBD, a disease marked by mildly impaired mental status (e.g., confusion) and sometimes by dysarthria or ataxia, is poorly understood but may be related to nutritional deficiencies in addition to chronic alcohol consumption. Traditionally characterized by demyelination and necrosis of the corpus callosum, a number of reports identify cortical lesions in so-called MBD. Such data, however, represent single case studies and may reflect inaccurate MBD diagnoses. As observed in the pons in CPM, lesions appear hyperintense on T2-weighted images and hypointense on T1-weighted images and often are located along the entire extent of the corpus callosum.

15Given the aforementioned findings in clinically differential and diagnosable alcohol-related syndromes, the following section examines whether similar brain disorders also appear in alcoholics who do not manifest the full spectrum of symptoms present in these conditions. That is, how do the brains of uncomplicated alcoholics compare? Quantitative MRI has shown that relatively mild yet significant structural deficits characteristic of alcoholic syndromes can occur in uncomplicated alcoholics.

16Relative to findings in WKS, research demonstrates mild volume deficits in the mammillary bodies, hippocampi, and thalami in uncomplicated alcoholics compared with healthy controls. As shown in figure 5, these structures show a graded effect of volume deficits. That is, volume deficits are greatest in brains of subjects with KS (figure 5C) compared with brains of subjects with uncomplicated alcoholism (figure 5B) and brains unaffected by alcohol (figure 5A). Results suggest that mammillary-body damage is not prerequisite for the development of amnesia in alcoholism. MR findings also show hippocampal volume deficits in alcoholics compared with healthy controls. Hippocampal volume deficits in alcoholism are influenced by age, even though age-related decline is difficult to detect in cross-sectional studies. Although deficits in hippocampal volume are not related to seizure incidence, temporal-lobe white matter may be sensitive to alcohol-withdrawal seizures. Hippocampal volume shrinkage in alcoholism is attributed to loss of white matter and decreased axonal diameter. Glial cell loss or reduced incorporation of newly formed neurons to the dentate gyrus, however, could also affect hippocampal volume in alcoholism.

17Other regions selectively affected in WE and KS include the orbitofrontal cortices (KS), periaqueductal gray matter, and tissue surrounding the third ventricle (WE). Reports suggest that propensity to relapse following sobriety is related to pronounced atrophy in bilateral orbitofrontal cortices (;;; also see ). The third ventricle (i.e., enlargement) is sensitive to resumption of chronic alcohol consumption. There currently are no studies regarding periaqueductal gray-matter volume in uncomplicated alcoholics.

18Key regions affected in HE include the globus pallidus and substantia nigra. Volume effects on these two structures have not been reported in uncomplicated alcoholics; however, in children with fetal alcohol syndrome, globus pallidus volume is reduced in size compared with unaffected children. In contrast, other basal ganglia nodes of reward circuitry have been described as affected in uncomplicated alcoholism: MRI studies have revealed smaller volumes of caudate, putamen, amygdala, and nucleus accumbens, especially in more recently sober alcoholics compared with healthy controls. Given the role of the amygdala in emotional regulation and behavioral control (for review, see ), however, researchers have speculated that premorbid amygdala volume deficits put individuals at heightened risk for developing AUD.

19CPM targets the pons and ACD affects the cerebellum. Total infratentorial volume (including pons, cerebellar hemispheres, vermis, fissures, cisterns, and fourth ventricle) is significantly smaller in uncomplicated alcoholics than control subjects. The volume of the pons and cerebellum (i.e., hemispheres) is smaller in uncomplicated alcoholics than in normal controls. Alcoholism-related volume deficits are also prevalent in gray and white matter of the cerebellar vermis, predominately in anterior superior but not posterior inferior regions (see figure 6).

20The frontal cortex is notably damaged in ARD. With respect to cortical regions in uncomplicated alcoholism, various methods have shown significant, widespread shrinkage of both cortical gray and white matter with corresponding increases in CSF-filled spaces. In particular, older (older than age 50) but not younger adult alcoholics show disproportionate deficits in both gray- and white-matter cortical volume, especially in the frontal lobes, when volumes are statistically adjusted for brain tissue decline associated with normal aging. This is the case even in comparisons made in groups selected on alcohol consumption, where older alcoholics have consumed equivalent amounts over their lifetime as younger alcoholics.

21Thinning of the corpus callosum occurs in uncomplicated alcoholics and is more prominent in the anterior than posterior regions. As with WE and KS, evidence for MBD-like pathology in uncomplicated alcoholism raises the possibility that brain damage occurs on a continuum. The following section examines how brain structures and function respond when drinking stops.

22Longitudinal MRI investigations show that the ventricles become smaller following weeks or months of drinking cessation. Reduction of lateral ventricles precedes reduction of third-ventricular volume and may be related to improvements in hematocrit, hemoglobin, and red blood cell counts. The following brain structures increase in volume in response to abstinence: the entire cerebral cortex; temporal, insular, and anterior cingulate cortices; amygdala (a finding that would argue against a premorbid volume deficit); thalamus; hippocampus; brainstem; and cerebellar cortex.

23Sober alcoholics reveal several associations between brain-volume gain, as determined by MRI, and improvement in neuropsychological test performance: Reduced lateral-ventricle volume is related to improved memory performance, reduced third-ventricle volume is related to improved nonverbal short-term memory performance, and reduced fourth-ventricle volume is related to improvement in measures of ataxia.

24The brain’s capacity to return to “normal” following long-term sobriety is unknown. Short-term (6 weeks) abstinence seems sufficient to observe some brain-volume recovery but does not result in equivalent brain volumes between recovering chronic alcoholics and healthy controls. It is difficult to determine whether recovery is complete. Aging is a factor. That is, older alcoholics exhibit reduced capacity for recovery compared with younger alcoholics. Longer periods of abstinence may be required for follow-up investigations. Some brain damage, such as neuronal loss, may be irreversible, even with extended abstinence.

25Despite evidence for recovery of brain volume with abstinence, the mechanisms accounting for recovery remain unclear. One hypothesis, brain rehydration, was not supported by early human research studies. An alternative explanation suggests that new neurons are created (i.e., neurogenesis) (e.g., ): It is unlikely, however, that enough neurons could be made to replace the volume loss observed in chronic alcoholism. Nor is it clear that new neurons can migrate from neurogenic zones to distant areas of volume loss. On the other hand, adequate volume recovery may be explained by white-matter regeneration, because glial cells (i.e., oligodendroctyes) have the capacity to repair myelin and remyelinate neurons, and oligodendrocyte progenitor cells have the potential to migrate long distances. Indeed, alcoholics who relapse have decreased white matter, whereas continued abstinence is associated with increased white matter, notably in the corpus callosum and subcortical white matter.

26Turning from studies with humans to animals, the following section examines imaging studies in models of alcoholism and related disorders.

27There are two experimental approaches to model WE in rodents. The slower approach uses a thiamine-deficient diet (i.e., feeding with a thiamine-deficient chow), which can take 3–4 weeks to produce symptoms. Behavioral symptoms can be achieved in ~2 weeks using a combination of a thiamine-deficient chow and intraperitoneal (i.p.) administration of a thiamine pyrophosphokinase inhibitor such as pyrithiamine. Both models result in symptoms that mimic those observed in humans with WE. Structural MRI findings in thiamine-deficient animals show similar patterns of brain changes, including hyperintense signals observed on T2-weighted images in thalamus, collicular bodies, hypothalamus, hippocampus, mammillary bodies, corpus callosum, and superior cerebellar peduncles. Thiamine deficiency may cause degeneration through neuroinflammatory mechanisms. In rats, inflammatory genes were highly expressed in vulnerable brain regions. MRI in animal models permits further probing of the effects of thiamine deficiency on the brain and can be used to determine susceptible brain regions as a function of time of insult as well as relationships between neuroinflammatory markers and brain insult. Such studies have also been used to confirm a mechanism of toxicity suspected based on research in humans (e.g.,;; ): that glucose loading in a thiamine-deficient state can precipitate WE, likely involving a breakdown of the blood–brain barrier. In animals, postmortem followup can be used to confirm and extend in vivo findings. For example, electron microscopy showed a higher percentage of small fibers and myelin thinning in the corpus callosa of thiamine-deficient animals relative to controls.

28Research with animals demonstrates that thiamine deficiency impairs several biochemical pathways requiring the thiamine derivative thiamine pyrophosphate (e.g., transketolase, pyruvate dehydrogenase, and α-ketoacid dehydrogenase), thereby interfering with carbohydrate metabolism (for energy production), lipid metabolism (for production and maintenance of myelin), and amino acid metabolism (for production of glucose-derived neurotransmitters; for example, glutamate and γ-aminobutyric acid [GABA]). Consequently, the function of essential thiamine-requiring enzymes in the brain (e.g., transketolase, pyruvate dehydrogenase, and α-ketoacid dehydrogenase) is compromised, leading to oxidative stress, cellular energy impairment, and eventually neuronal loss.

29Evidence also shows that thiamine deficiency alters norepinephrine, dopamine, serotonin, and histamine synthesis and catabolism pathways. Thiamine deficiency may target focal brain areas such as the thalamus because, relative to other brain structures, it has lower levels of monocarboxylic acid transporters and acetyl-CoA-synthetase. This makes these areas less capable of generating energy from acetate, which is a potential source of cellular energy in place of glucose in alcoholism.

30Current rodent models to study HE include models of acute and chronic liver failure. According to the International Society for Hepatic Encephalopathy, however, “At this time, there are no satisfactory animal models of Type C HE resulting from end-stage alcoholic liver disease or viral hepatitis, the most common etiologies encountered in patients” (, p. 783). In addition, no MR-imaging studies to date have used rodent models of HE. Imaging studies in cats, dogs, and monkeys typically recapitulate the human condition, showing nonspecific sulcal widening and hyperintensities in lentiform nuclei (i.e., putamen and globus pallidus of the basal ganglia). Animal models of HE have been used to evaluate potential mechanisms of pathology, such as the contribution of excess ammonia in the blood (i.e., hyperammonemia) or lactate. Animal models of HE have also been used to explore treatment strategies for HE (e.g., hypothermia).

31Using a rat model of CPM to study white-matter degeneration, it was found that blood–brain barrier breakdown, detected with MRI, was associated with a higher risk of developing demyelination, as detected using postmortem histopathology. This study demonstrated that blood–brain barrier disruption exposes oligodendrocytes to substances normally excluded from the brain. This supports hypotheses from human postmortem studies suggesting that damage to the pons may be linked to reduced blood flow, as indicated by findings that basilar artery architecture is altered in CMP.

32Although there are no known studies using structural MRI in animal models of ACD, ARD, or MBD, the following section examines animal studies in uncomplicated alcoholism.

33An important initial report in the rodent MRI literature was the demonstration that brain growth continues beyond what would be considered adulthood in rats bred to prefer alcohol (i.e., alcohol-preferring rats, or P rats). Indeed, whole-brain volume in such rats continued to grow until approximately postnatal day 450, well past adulthood, which is typically considered as postnatal day 90. Baseline studies (in the absence of alcohol [i.e., EtOH] exposure) also suggest that brains of alcohol-preferring rats are different relative to their wild-type counterparts, including reduced gray-matter volume in thalamus, ventral tegmental area, and insular and cingulate cortices.

34One of the most consistent findings in alcohol-exposed rodents, ventricular enlargement, varies with timing and method of alcohol exposure. It is far more pronounced in rats achieving average blood alcohol levels (BALs) of 250 mg/dL in just 4 days of involuntary binge-type administration of EtOH than in rats achieving average BALs of 200 mg/dL over 24 weeks using vapor EtOH exposure or P rats gradually achieving average BALs of 125 mg/dL with voluntary EtOH consumption, where only modest ventricular enlargement was noted (cf.,; ). Even repeated binge exposures (i.e., 5 cycles of 4 days of intragastric binge EtOH exposure with 1 week abstinence in between), do not result in persistent effects on the brain detectable with MRI (Zahr et al. 2015). Although ventricular size increases with each binge EtOH exposure, there is rapid recovery during each week of abstinence (Zahr et al. 2015). Such studies suggest that EtOH alone, at least in the exposure protocols evaluated with MRI, does not result in the characteristics observed in human alcoholics. Conversely, rats exposed to vaporized EtOH during adolescence are reported to show persistent effects (i.e., ventricular enlargement and deficits in hippocampal volume) into adulthood. Mice exposed to EtOH during adolescence are similarly purported to exhibit long-lasting regional brain-volume deficits in the olfactory bulb and basal forebrain. These results suggest that the adolescent rodent brain may be more vulnerable to enduring toxic effects of EtOH than the adult rodent brain.

35In monkeys trained to voluntarily consume alcohol, those that drank at least 3 g/kg EtOH per day for 15 months showed significant brain-volume shrinkage in the cerebral cortices. Because these animals were well nourished, these results suggest a direct relationship between oral EtOH intake and measures of decreased brain gray-matter volume.

36A number of sources provide extensive descriptions of the principles of DTI. Briefly, DTI takes advantage of the fact that MR images of the brain are predominantly maps of water protons with contrast created by their immediate environment and their motility. In regions with few or no constraints imposed by physical boundaries, such as CSF in the ventricles, water movement is random and uniform in every direction and is therefore isotropic. In contrast to CSF, the path of a water molecule along a white-matter fiber is constrained by physical boundaries such as the axon sheath, causing greater movement along the long axis of the fiber than across it. This movement is called anisotropic; diffusion along the long axis of a fiber (axial or longitudinal diffusion) is greater than diffusion across the fiber (radial or transverse diffusion).

37DTI findings are described in terms of diffusion. The magnitude of diffusion, referred to as mean diffusivity (MD) or the apparent diffusion coefficient (ADC), is calculated mathematically. Increased MD corresponds to white-matter damage. Fractional anisotropy (FA), ranging between 0 and 1, reflects axonal integrity, with lower integrity reflected by FA values closer to 0. Thus, disruption of white-matter microstructure detectable with DTI can reflect compromised myelin, cytoskeletal structure, or axonal density.

38Several approaches have been used to quantify DTI metrics. One of the more desirable approaches is the use of quantitative fiber tracking, which is able to evaluate fibers along their entire length and can thus detect compromised white matter. This technique can be used to depict selective commissures (e.g., corpus callosum), projection fibers, and association fibers.

39In one study, DTI in alcoholics with (n = 7) and without (n = 20) WKS showed FA deficits in the fornix and cingulum bundle of the Papez or medial limbic circuit, measured using tract-based spatial statistics (TBSS). These FA effects were greater in alcoholics with WKS relative to those without it. The number of tracts in the fornix appears to be reduced only in WKS patients.

40Studies of people who have alcohol-related cirrhosis with HE have reported elevated MD in several white-matter bundles, including the corpus callosum, internal capsule, and frontal white matter, and effects on both FA and MD of occipital white matter. HE caused by alcoholism compared with other forms of HE (e.g., as a result of viral infection or primary biliary cirrhosis) appears to have different effects on DTI parameters, with more widespread changes in FA and MD in alcoholic relative to nonalcoholic cirrhosis. When researchers induced hyperammonemia in cirrhotic patients, an increase in ADC in brain white matter was observed, supporting excess ammonia in the blood as a mechanism driving cerebral edema.

41DTI showed elevated MD in the middle cerebellar peduncles with no effects on corticospinal tracts in a study participant with CPM relative to three healthy comparison participants.

42The largest DTI study of MBD to date included six study participants, five with a history of chronic alcoholism. All six showed hyperintense signals on diffusion images and low ADC of the corpus callosum. Researchers observed cortical lesions in frontoparietal regions in three of six study participants with the poorest outcomes. Remaining DTI studies of MBD were case studies (e.g., ) showing low ADC along the entire corpus callosum, with FA values diminishing progressively from front to back. No known DTI studies have been conducted in patients with ACD or ARD.

43DTI has revealed microstructural damage related to alcoholism in cerebral areas that appear intact in structural MRI analyses (e.g.,;; ). Corpus callosum findings in uncomplicated alcoholics are common and, as observed for MBD, show greater anterior than posterior effects (e.g.,;;;; ). Quantitative fiber tracking has demonstrated greater FA deficits in anterior than in posterior fibers of supratentorial and infratentorial white-matter bundles in alcoholics compared with healthy controls, as well as low FA in tracts of the corpus callosum, centrum semiovale, internal and external capsules, fornix, superior cingulate, and longitudinal fasciculi. Frontolimbic, fronto-parietal, fronto-occipital, fronto-cerebellar, cortico-striatal, and cortico-pontine fibers are also affected in alcoholics relative to healthy controls.

44Studies have also examined DTI- function relationships in alcoholism. FA in anterior cingulate and motor areas correlates with executive and psychomotor performance, FA in the splenium correlates with working memory, and FA in several regions (corpus callosum, parietal, occipital, and frontal white-matter) correlates with performance on the Iowa Gambling Task. A double dissociation was found showing that higher diffusivity in sensory- motor and parietal bundles was associated with poorer balance but not psychomotor speed, whereas higher diffusivity in prefrontal and temporal bundles was associated with slower psychomotor speed but not balance. DTI changes in multiple supratentorial and infratentorial fiber systems in alcoholics correlated with impairment in speeded performance and postural stability, frontal fiber integrity connecting left and right hemispheres predicted performance on a coordinated psychomotor task, and number of reconstructed fibers running between the pons and the midbrain was related to cognitive flexibility performance. Gray-matter diffusivity in the hippocampus, which is lower in alcoholics than in healthy controls, is related to episodic memory impairment.

45Similar to structural MRI findings demonstrating pronounced tissue- volume shrinkage of orbitofrontal cortices in abstinent alcoholics who were likely to resume drinking (e.g.,;; ), DTI identified alcoholic individuals more likely to resume drinking 6 months following initial evaluation based on lower FA and higher diffusivity in frontal white matter at baseline. Increases in FA and decreases in diffusivity have been interpreted as evidence for white-matter recovery with abstinence. Studies have shown recovery in corpus callosum at 1 year compared with 2 weeks of abstinence and in frontal white matter at 1 month compared with 1 week of abstinence, at least in nonsmoking, sober alcoholics. Other reports suggest that some white-matter impairments persist after 6 to 30 months of recovery in alcoholics relative to healthy controls. In a seminal longitudinal study of 47 alcoholic and 56 healthy controls study participants, reported that, despite abnormally low FA, age trajectories of the alcoholics who abstained were positive and progressing toward normality, whereas those of the relapsing alcoholics and control subjects were negative.

46DTI data have been collected in animal models of WE but not in other concomitants of alcoholism.

47In the study in which WE was induced by thiamine deficiency, animals were imaged at baseline, presymptomatic stage (day 10), symptomatic stage (days 12 and 14), and after recovery on days 31 and 87. A decrease in FA in the inferior colliculi was first noted on day 10 but showed recovery on day 87. On the other hand, the FA decrease in the thalamus first noted on day 12 persisted through day 87. This model was also used in a pharmacological DTI study in which animals were exposed to rasagiline, a selective monamine oxidase B inhibitor, as a potential protective agent against thiamine-deficiency–induced brain damage. In addition to reducing ventricular enlargement, rasagiline appeared to ameliorate the effects of thiamine deficiency on the FA decrease in the thalamus. Histopathology showed that treatment with rasagiline reduced the lesions in thalamus and colliculi observed in the thiamine-deficient brain. Rasagiline has not been evaluated in human patients with WE.

48Adolescent animals exposed to intermittent EtOH and evaluated postmortem showed no effects on FA but reduced axial diffusivity (hippocampus, cortex, and cerebellum), reduced radial diffusivity (hippocampus and cortex), and reduced MD (cerebellum and corpus callosum) in several brain regions. Adult rats exposed to a single dose of EtOH showed a slight and transient reduction, relative to unaffected rats, in ADC in brainstem, frontal lobe, hippocampus, thalamus, and cerebellum. These findings were interpreted as reflecting the development of cytotoxic brain edema, as histological analysis showed cell swelling and narrowed extracellular spacing.

49Whereas chronic exposure to vaporized EtOH did not result in detectable effects on FA or MD, binge EtOH exposure resulted in transient decreases in FA and transient increases in MD. Together, these results suggest that DTI can detect acute and subchronic effects on the brain, but that chronic exposure to EtOH can result in brain adaptations such that effects on FA and MD are no longer discernable.

50Although MRI primarily depicts the distribution of water protons, similar technology can also be used to obtain information about chemical constituents other than water, primarily due to a small frequency shift, or “chemical shift,” relative to the water signal. The acquisition of MR-detectable signals other than those of water and fat is referred to as MRS and is an in vivo application of traditional laboratory-based NMR spectroscopy.

51MRS reveals information about several biochemicals, or metabolites, in the brain. The largest signals arise from N-acetylaspartate (NAA), creatine and phosphocreatine (i.e., total creatine [tCr]), and choline-containing compounds (Cho). Signals from the combined resonances of glutamate (Glu) and glutamine (Gln) (i.e., Glx) are also sometimes reported, as are myo-inositiol (mI) and lactate (lac). Signals from Glu and GABA can also be detected under certain conditions.

52The predominant in vivo proton signal is NAA, with contributions from other N-acetyl compounds, especially N-acetyl aspartyl glutamate. NAA is found almost exclusively in neurons and, thus, is considered a measure of neuronal integrity. Postmortem (Cooper 1972;; ) and MRS studies have shown NAA levels to be higher in gray than in white matter in healthy study participants, as have in vivo studies.

53The tCr signal, generated by creatine and phosphocreatine, is influenced by the state of high-energy phosphate metabolism. In spectroscopy studies, it often is used as a reference for other peaks based on the incorrect assumption that its concentration is relatively constant (cf.,, ).

54The in vivo MRS-visible Cho peak is generated primarily by water-soluble choline-containing compounds (free choline, phosphocholine, and glycerophosphocholine) and is associated with cell-membrane synthesis and turnover. The Cho resonance also provides an index of cellular density in brain tumors and may be a marker of increases in glial density with age and disease. MRS-measured Cho concentration is higher in white than gray matter and increases with normal aging.

55Myo-inositiol is present in glial but not neuronal cell cultures and plays a role in maintaining cell volume. The concentration of mI is higher in gray than in white matter.

56Figure 7 shows a graph of MR spectra from the thalamus of a 55-year-old nonalcoholic woman. The major metabolites are color coded.

57As with the other imaging modalities, MRS reports of WE are primarily case studies. For example, a Japanese man who had consumed alcohol for 50 years and had eaten poorly for several days as a result of a cold presented with gait disturbances and incoherent speech. MRS before and after thiamine treatment found an initial low level of NAA/tCr in the thalamus, which appeared to increase with thiamine replacement. NAA/tCr levels in the cerebellum did not increase, although a lactate peak initially present in the cerebellum resolved. MRS conducted in two patients with non–alcohol-related thiamine deficiency (i.e., caused by gastric and pancreatic cancer) compared with five healthy study participants showed similar results: relatively low NAA/tCr levels in the thalamus resolved after treatment with thiamine.

58In a variety of brain regions, MRS findings in alcohol-related cirrhosis and HE are remarkably consistent and comparable with findings in nonalcoholic HE (e.g.,;; ), showing lower levels of Cho/tCr and mI/tCr and higher levels of Gln/tCr. Levels of mI and Cho are lowest and Glx highest in patients with HE.

59Mild swelling of astrocytes is proposed as the key event in the pathogenesis of HE (e.g., ). In cirrhosis, elevated blood level of ammonia is thought to result in elevated brain ammonia, which can be toxic. It often has been proposed that the brain’s response to elevated ammonia levels is to combine ammonia and glutamate to make glutamine using glutamine synthetase, found primarily in astrocytes. Thus, brain swelling in cirrhosis is thought to reflect an increase in astrocytic glutamine formation. The decrease in mI is thought to be a compensatory mechanism to counterbalance the osmotic effect of cerebral glutamine accumulation. Although some articles claim to measure in vivo glutamine (e.g.,;;;; ), it is unlikely that the MRS method used in these cases permitted the separate detection of glutamate and glutamine, which are strongly coupled and difficult to detect independently, even with very short echo times. A single study measured GABA levels in five alcoholics without HE and five study participants with both alcohol and non–alcohol-related HE. GABA levels were lower in the two patient groups relative to 10 comparison participants.

60In the only report of MRS conducted in a case of alcoholism-associated CPM, elevated Cho/tCr was found and interpreted as reflecting edema or demyelination in a 53-year-old man with gait disturbances and hearing loss. In 12 patients with chronic hyponatremia (nonalcohol etiology), MRS showed reduced Cho and mI relative to unaffected study participants, reflecting osmolyte disturbances.

61Research also has found compromised NAA/tCr levels in patients with cerebellar degeneration. Two MRS case studies of MBD showed reduced NAA/tCr and elevated Cho/tCr in corpus callosum splenium, findings consistent with demyelination (elevated Cho) and axonal injury (reduced NAA).

62Most MRS studies show lower levels of NAA in recently sober alcoholics relative to healthy controls in several brain regions, including frontal areas and cerebellum. Similarly, studies in AUD patients shortly following detoxification have found low levels of Cho, although Cho findings in AUD are less consistent (e.g.,; ). Because these findings are prominent in white matter, it is thought that the effects of alcoholism are greater in white than in gray matter.

63MRS studies suggest that NAA (e.g.,;; ), particularly in frontal and cerebellar regions and Cho levels (e.g.,;;;; ) show normalization (i.e., increase) with abstinence. Elevations in mI are not seen in long-term sober alcoholics. These findings suggest that low NAA levels initially observed in recently sober alcoholics reflect neurodegeneration without cell death, and increases with abstinence may reflect healing without cell generation. The disruption and recovery of Cho and mI levels suggest white-matter recovery with sobriety and the potential for remyelination.

64In rat models of WE induced using pyrithiamine, the dominant MRS pattern is a reduction in both NAA and Cho in several brain regions, including the thalamus. Researchers also frequently report elevations in lactate. Precipitation of WE with glucose (resulting in seizures) is associated with further decreases in NAA and Cho and, significantly, an elevation in lactate. Treatment with thiamine is associated with recovery in Cho levels.

65MRS has been used to evaluate models of HE achieved using various methods and most reports show similar findings. Hepatic devascularization, carbon tetrachloride treatment, bile-duct ligations (; Rackayova et al. 2015), and other means of promoting hyperammonemia (e.g., acute liver ischemia, urease, or methionine sulfoximine treatment) result in elevated levels of Gln and frequently lactate in rat brain (e.g., cortex). Additional effects reported include lower levels of NAA, mI, Cho, and Glu (;;;;; Rackayova et al. 2015; ). As in the human condition, a similar caveat holds: that is, it is not clear if Glu and Gln are clearly discriminated in many of these studies, and, often, reports more likely reflect Glx levels.

66Although in vivo MRS studies in both humans and animals have persisted in interpreting elevations in brain Gln as reflecting elevations in peripheral ammonia and brain edema, ex vivo carbon 13 nuclear MR studies have challenged the convention that glutamine accumulation is the major cause of brain edema in acute HE. Such studies instead indicate limited metabolic pathway reactions and capacity of astrocytes to detoxify ammonia by glutamine synthesis and emphasize distortions of energy and neurotransmitter metabolism.

67MRS can be used in animals to detect and quantify in vivo and real-time brain EtOH kinetics (e.g., rats:; monkeys: ). Unlike findings in long-term sober human alcoholics, nonabstinent chronic heavy drinkers and social and moderate drinkers show elevated levels of brain Cho. Elevated levels of Cho are also reported in the thalamus of rodents between weeks 16 and 40 of alcohol exposure. Neuroimaging research has been conducted with rodent models of binge, repeated binge (Zahr et al. 2015), and chronic alcohol exposure. In vivo MRS studies have consistently shown that a single 4-day binge exposure with BALs approaching 300 mg/dL is associated with reversible changes to the brain: levels of NAA are lower and those of Cho are higher following binge EtOH exposure. In the repeated-binge experiment, animals were exposed to 5 cycles of 4 days of intragastric EtOH treatment and 10 days of recovery. Changes in MRS metabolite levels again were transient: levels of NAA decreased, whereas those of Cho increased with each binge EtOH exposure cycle but then recovered during each abstinence period. Changes in response to EtOH were in expected directions based on the previous, single-binge EtOH exposure experiments but did not accrue with repeated-binge EtOH exposure (Zahr et al. 2015). In the chronic EtOH exposure study, NAA levels were lower in the EtOH-exposed relative to the comparison group but did not attain statistical significance, whereas levels of Cho appeared to demonstrate a dose-response curve (i.e., increasing levels with higher and longer EtOH exposure).

68Imaging investigations of alcohol-related brain disorders show unique neuropathology (as outlined in table 1), offering a framework for examining pathology in uncomplicated alcoholism. Because brains affected by AUD can show mild effects in the regions aggressively targeted by overt disease, animal models have been useful in distinguishing the etiology of pathology and differentiating brain regions specifically targeted by thiamine deficiency versus hyperammonemia, for example. Individuals with AUD may show more prominent effects in some regions compared with others, suggesting a propensity for one diagnosis over another (e.g., an alcoholic may be more vulnerable to thiamine deficiency than to liver damage). What remains unresolved, and what animal models can help determine, is why certain brain regions are differentially vulnerable to certain pathologies. For example, are the colliculi sensitive to thiamine deficiency because of their relatively high metabolic rate? Is the pons susceptible to CPM because of its proximity to the basilar artery? Does dopamine explain why basal ganglia are targets of liver disease?

69In vivo imaging studies in humans and animal models will continue to provide an evolving picture of the course of alcoholic brain disease through remissions and exacerbations as long-term studies follow human alcoholics as they age and as new initiatives evaluate adolescents before they are exposed to alcohol.