Regional cerebral glucose utilization in chronic organic mental disorders associated with alcoholism

Temporal Cognitive and Brain Changes in Korsakoff Syndrome

OBJECTIVE

To investigate cognitive and brain changes in patients with Korsakoff syndrome (KS) over months and up to 10 years after the diagnosis.

METHODS

Two groups of 8 patients with KS underwent neuropsychological, motor, and neuroimaging investigations, including structural MRI and ¹⁸F-fluorodeoxyglucose-PET. The KS^C group, recruited at Caen University Hospital, was examined early after the KS diagnosis (KS^C-T1) and 1 year later (KS^C-T2). The KS^R group, recruited at nursing home at Roubaix, was evaluated 10 years after the diagnosis. Longitudinal comparisons in KS^C explored short-term changes, while cross-sectional comparisons between KS^C-T1 and KS^R informed about long-term changes.

RESULTS

No cognitive, motor, or brain deterioration occurred over time in patients with KS. There was no clear improvement either, with only modest recovery in the frontocerebellar circuit. Compared to the norms, KS^C-T1 had severe episodic memory impairments, ataxia, and some executive dysfunctions. They also presented widespread atrophy and hypometabolism as well as cerebellar hypermetabolism compared to 44 healthy matched controls. Episodic memory remained significantly impaired in KS^C-T2 and KS^R. Contrary to KS^C at T1 and T2, KS^R had preserved inhibition abilities. Atrophy was similar but less extended in KS^C-T2 and even more limited in KS^R. At all times, the thalamus, hypothalamus, and fornix remained severely atrophied. Hypometabolism was still widespread in KS^C-T2 and KS^R, notably affecting the diencephalon. Cerebellar metabolism decreased over time and normalized in KS^R, whereas motor dysfunction persisted.

CONCLUSION

In KS, structural and metabolic alterations of the Papez circuit persisted over time, in accordance with the irreversible nature of amnesia. There was neither significant recovery as observed in patients with alcohol use disorder nor progressive decline as in neurodegenerative diseases.

Neural correlates of visuoperceptive changes in severe alcohol use disorder: A critical review of neuroimaging and electrophysiological findings

Visuoperceptive deficits are frequently reported in severe alcohol use disorder (SAUD) and are considered as pervasive and persistent in time. While this topic of investigation has previously driven researchers' interest, far fewer studies have focused on visuoperception in SAUD since the '90s, leaving open central questions regarding the origin and implications of these deficits. To renew research in the field and provide a solid background to work upon, this paper reviews the neural correlates of visuoperception in SAUD, based on data from neuroimaging and electrophysiological studies. Results reveal structural and functional changes within the visual system but also in the connections between occipital and frontal areas. We highlight the lack of integration of these findings in the dominant models of vision which stress the dynamic nature of the visual system and consider the presence of both bottom-up and top-down cerebral mechanisms. Visuoperceptive changes are also discussed in the framework of long-lasting debates regarding the influence of demographic and alcohol-related factors, together stressing the presence of inter-individual differences. Capitalizing on this review, we provide guidelines to inform future research, and ultimately improve clinical care.

The mammillary bodies and memory: more than a hippocampal relay

Although the mammillary bodies were one of the first neural structures to be implicated in memory, it has long been assumed that their main function was to act primarily as a hippocampal relay, passing information on to the anterior thalamic nuclei and from there to the cingulate cortex. This view not only afforded the mammillary bodies no independent role in memory, it also neglected the potential significance of other, nonhippocampal, inputs to the mammillary bodies. Recent advances have transformed the picture, revealing that projections from the tegmental nuclei of Gudden, and not the hippocampal formation, are critical for sustaining mammillary body function. By uncovering a role for the mammillary bodies that is independent of its subicular inputs, this work signals the need to consider a wider network of structures that form the neural bases of episodic memory.

Alcohol decreases baseline brain glucose metabolism more in heavy drinkers than controls but has no effect on stimulation-induced metabolic increases

During alcohol intoxication, the human brain increases metabolism of acetate and decreases metabolism of glucose as energy substrate. Here we hypothesized that chronic heavy drinking facilitates this energy substrate shift both for baseline and stimulation conditions. To test this hypothesis, we compared the effects of alcohol intoxication (0.75 g/kg alcohol vs placebo) on brain glucose metabolism during video stimulation (VS) versus when given with no stimulation (NS), in 25 heavy drinkers (HDs) and 23 healthy controls, each of whom underwent four PET-(18)FDG scans. We showed that resting whole-brain glucose metabolism (placebo-NS) was lower in HD than controls (13%, p = 0.04); that alcohol (compared with placebo) decreased metabolism more in HD (20 ± 13%) than controls (9 ± 11%, p = 0.005) and in proportion to daily alcohol consumption (r = 0.36, p = 0.01) but found that alcohol did not reduce the metabolic increases in visual cortex from VS in either group. Instead, VS reduced alcohol-induced decreases in whole-brain glucose metabolism (10 ± 12%) compared with NS in both groups (15 ± 13%, p = 0.04), consistent with stimulation-related glucose metabolism enhancement. These findings corroborate our hypothesis that heavy alcohol consumption facilitates use of alternative energy substrates (i.e., acetate) for resting activity during intoxication, which might persist through early sobriety, but indicate that glucose is still favored as energy substrate during brain stimulation. Our findings are consistent with reduced reliance on glucose as the main energy substrate for resting brain metabolism during intoxication (presumably shifting to acetate or other ketones) and a priming of this shift in HDs, which might make them vulnerable to energy deficits during withdrawal.

Reduced fronto-cerebellar functional connectivity in chronic alcoholic patients

BACKGROUND

Alcohol dependence is associated with neurocognitive deficits related to neuropathological changes in structure, metabolism, and function of the brain. Impairments of motor functioning in alcoholics have been attributed to well-characterized neuropathological brain abnormalities in cerebellum.

METHODS

Using functional magnetic resonance imaging (fMRI), we studied in vivo the functional connectivity between cerebellar and cortical brain regions. Participants were 10 uncomplicated chronic alcoholic patients studied after 5 to 7 days of abstinence when signs of withdrawal had abated and 10 matched healthy controls. We focused on regions of prefrontal, frontal, temporal, and parietal cortex that exhibited an fMRI response associated with nondominant hand finger tapping in the patients but not in the controls. We predicted that fronto-cerebellar functional connectivity would be diminished in alcoholics compared with controls.

RESULTS

Functional connectivity in a circuit involving premotor areas (Brodmann Area 6) and Lobule VI of the superior cerebellum was reduced in the patients compared with the controls. Functional connectivity was also reduced in a circuit involving prefrontal cortex (Brodmann Area 9) and Lobule VIII of the inferior cerebellum. Reductions in connectivity were specific to fronto-cerebellar circuits and were not found in other regions examined.

CONCLUSIONS

Our findings show a pattern in recently abstinent alcoholic patients of specific deficits in functional connectivity and recruitment of additional brain regions for the performance of a simple finger-tapping task. A small sample, differences in smoking, and a brief abstinence period preclude definitive conclusions, but this pattern of diminished fronto-cerebellar functional connectivity is highly compatible with the characteristic neuropathological lesions documented in alcoholics and may reflect brain dysfunction associated with alcoholism.

Altered fronto-cerebellar connectivity in alcohol-naïve youth with a family history of alcoholism

Fronto-cerebellar connections are thought to be involved in higher-order cognitive functioning. It is suspected that damage to this network may contribute to cognitive deficits in chronic alcoholics. However, it remains to be elucidated if fronto-cerebellar circuitry is altered in high-risk individuals even prior to alcohol use onset. The current study used functional connectivity MRI (fcMRI) to examine fronto-cerebellar circuitry in 13 alcohol-naïve, at-risk youth with a family history of alcoholism (FH+) and 14 age-matched controls. In addition, we examined how white matter microstructure, as evidenced by fractional anisotropy (FA), related to fcMRI. FH+youth showed significantly reduced functional connectivity between bilateral anterior prefrontal cortices and contralateral cerebellar seed regions compared to controls. We found that this reduction in connectivity significantly correlated with reduced FA in the anterior limb of the internal capsule and the superior longitudinal fasciculus. Taken together, our findings reflect associated aberrant functional and structural connectivity in substance-naïve FH+adolescents, perhaps suggesting an identifiable neurophenotypic precursor to substance use. Given the role of frontal and cerebellar brain regions in subserving executive functioning, the presence of premorbid abnormalities in fronto-cerebellar circuitry may heighten the risk for developing an alcohol use disorder in FH+youth through atypical control processing.

Recruitment of additional brain regions to accomplish simple motor tasks in chronic alcohol-dependent patients

BACKGROUND

Chronic alcohol-dependent patients (ALC) exhibit neurocognitive impairments attributed to alcohol-induced fronto-cerebellar damage. Deficits are typically found in complex task performance, whereas simple tasks may not be significantly compromised, perhaps because of little understood compensatory changes.

METHODS

We compared finger tapping with either hand at externally paced (EP) or maximal self-paced (SP) rates and concomitant brain activation in ten pairs of right-hand dominant, age-, and gender-matched, severe, uncomplicated ALC and normal controls (NC) using functional magnetic resonance imaging (fMRI).

RESULTS

Mean tapping rates were not significantly different in ALC and NC for either task, but SP tapping variances were greater in ALC for both hands. SP tapping was more rapid with dominant hand (DH) than non-dominant hand (NDH) for both groups. EP and SP tapping with the non-dominant hand demonstrated significantly more activation in ALC than NC in the pre and postcentral gyri, inferior frontal gyrus, inferior parietal lobule, and the middle temporal gyrus. Areas activated only by ALC (not at all by NC) during NDH tapping included the inferior frontal gyrus, middle temporal gyrus, and postcentral gyrus. There were no significant group activation differences with DH tapping. No brain regions activated more in NC than ALC. SP tapping in contrast to EP activated fronto-cerebellar networks in NC, including postcentral gyrus, anterior cingulate, and the anterior lobe and vermis of the cerebellum, but only parietal precuneus in ALC.

CONCLUSIONS

These findings with NDH finger tapping support previous reports of neurocognitive inefficiencies in ALC. Inferior frontal activation with EP in ALC, but not in NC, suggests engagement of regions needed for planning, organization, and impulse regulation; greater contralateral parietal lobe activation with SP in ALC may reflect right hemispheric impairments in visuospatial performance. Contrasting brain activation during SP and EP suggests that ALC may not have enlisted a fronto-cerebellar network as did NC but rather employed a higher order planning mode by recruiting parietal lobe functions to attain normal mean finger tapping rates. Elucidation of the compensatory neural mechanisms that allow near normal performance by ALC on simple tasks can inform functional rehabilitation of patients in recovery.

Development and resolution of brain lesions caused by pyrithiamine- and dietary-induced thiamine deficiency and alcohol exposure in the alcohol-preferring rat: a longitudinal magnetic resonance imaging and spectroscopy study

Wernicke's encephalopathy (WE) is characterized by lesions in thalamus, hypothalamus (including mammillary nuclei), and inferior colliculi, results in serious disabilities, has an etiology of thiamine deficiency, is treatable with thiamine, and occurs most commonly with alcoholism. Despite decades of study, whether alcohol exposure exacerbates the neuropathology or retards its resolution remains controversial. To examine patterns of brain damage and recovery resulting from thiamine deprivation with and without alcohol exposure, we conducted in vivo magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) at 3 T in alcohol-preferring (P) rats, which had voluntarily consumed large amounts of alcohol before thiamine manipulation. A total of 18 adult male P rats (nine alcohol-exposed) received a thiamine-deficient diet for 2 weeks: 10 (five alcohol-exposed) received intraperitoneal (i.p.) pyrithiamine (PT) and eight (four alcohol-exposed) received i.p. thiamine supplementation. Neurological signs developed by day 14. Rats were scanned before thiamine depletion and 18 and 35 days after thiamine repletion. Two-dimensional J-resolved MRS single-voxel spectra with water reference were collected in a voxel subtending the thalamus; metabolite quantification was corrected for voxel tissue content. MRI identified significant enlargement of dorsal ventricles and increase in signal intensities in thalamus, inferior colliculi, and mammillary nuclei of PT compared with thiamine-treated (TT) groups from MRI 1-2, followed by significant normalization from MRI 2-3 in thalamus and colliculi, but not mammillary nuclei and lateral ventricles. Voxel-by-voxel analysis revealed additional hyperintense signal clusters in the dorsal and ventral hippocampus and enlargement of the fourth ventricle. MRS showed a significant decline and then partial recovery in thalamic N-acetylaspartate, a marker of neuronal integrity, in PT compared with TT rats, with no change detected in creatine, choline, or glutamate. PT rats with prior alcohol exposure exhibited attenuated recovery in the thalamus and arrested growth of the corpus callosum; further, two of the five alcohol-exposed PT rats died prematurely. Parenchymal and ventricular changes with thiamine manipulation concur with human radiological signs of WE. The enduring macrostructural and neurochemical abnormalities involving critical nodes of Papez circuit carry liabilities for development of amnesia and incomplete recovery from other cognitive and motor functions subserved by the affected neural systems.

Correlations of regional cerebral metabolism with memory performance and executive function in patients with herpes encephalitis or frontal lobe lesions

Cerebral [18F]fluorodeoxyglucose positron emission tomography (18FDG-PET) data from patients suffering amnesia following herpes encephalitis (n=7) or frontal lobe pathology (n=14) were compared with data from age-matched nonamnesic subjects (n=10). All subjects received structural MRI, resting 18FDG-PET scans, and neuropsychological evaluation. PET data were analyzed using complementary statistical parametric mapping and region-of-interest methods. Differential patterns of hypometabolism were found in patients relative to healthy controls. Factor analysis of the neuropsychological data revealed that memory performance was associated with retrosplenial and medial temporal metabolism, and executive function was associated with dorsolateral frontal metabolism. The association between memory performance and retrosplenial metabolism remained statistically significant after accounting for measures of cerebral atrophy using MRI. The significance of the retrosplenium as a major relay station between the thalamus and the medial temporal and frontal lobes--sensitive to changes in either--is discussed in the light of the findings.

Compromised pontocerebellar and cerebellothalamocortical systems: speculations on their contributions to cognitive and motor impairment in nonamnesic alcoholism

BACKGROUND

Corticopontocerebellar and cerebellothalamocortical circuits underlie a wide range of neuropsychological processes compromised by alcoholism. The analyses herein tested whether abnormalities of volumes of brain structures forming nodes of these separate feed-forward and feedback systems are selectively related to each other and whether any of these noncortical regions can account for cognitive and motor deficits occurring as sequelae of chronic alcoholism.

METHODS

Regional brain measures originated from our prior neuroimaging studies, showing in alcoholics significant volume deficits in the principal structures of interest: cerebellar hemispheres, vermis, pons, and thalamus as well as prefrontal, frontal, and parietal cortex. Neuropsychological functions targeted for analysis-problem solving, visuospatial ability, and static postural stability-showed 0.6 to 1.6 SD deficits in these alcoholic men.

RESULTS

In alcoholics, the patterns of correlations were consistent with dissociation of thalamic and pontine circuitry. Pontine and thalamic volumes were not correlated with each other. Pontine volumes correlated with white matter volumes of anterior superior vermis and gray and white matter volumes of the cerebellar hemispheres but not with cortical regional volumes. Thalamic volumes correlated with gray matter volumes of the cerebellar hemispheres, parietal cortex, and inferior posterior vermian lobule, which itself correlated with parietal, prefrontal, and frontal cortical volumes. Controls did not show these correlational patterns. Brain structure-function relationships in alcoholics examined with multiple regression identified anterior vermian but not prefrontal or parietal volume as a unique predictor of balance scores; vermian and thalamic but not prefrontal cortical volumes as predictors of card sorting scores; and cerebellar hemispheric white matter but not parietal cortical volume as a predictor of visuospatial ability.

CONCLUSIONS

Each major node of frontocerebellar circuitry shows volume deficits in alcoholics but can be independently compromised. Disruption of these circuits may underlie alcoholism-related neuropsychological deficits, either by abnormalities present in individual nodes or by disconnection via interruption of selective circuitry.

FDG-PET findings in the Wernicke-Korsakoff syndrome

This study reports FDG-PET findings in Wernicke-Korsakoff patients. Twelve patients suffering amnesia arising from the Korsakoff syndrome were compared with 10 control subjects without alcohol-related disability. Subjects received [18F]-fluorodeoxyglucose (FDG-PET) imaging as well as neuropsychological assessment and high-resolution MR imaging with volumetric analysis. Volumetric MRI analysis had revealed thalamic and mamillary body atrophy in the patient group as well as frontal lobe atrophy with relative sparing of medial temporal lobe structures. Differences in regional metabolism were identified using complementary region of interest (ROI) and statistical parametric mapping (SPM) approaches employing either absolute methods or a reference region approach to increase statistical power. In general, we found relative hypermetabolism in white matter and hypometabolism in subcortical grey matter in Korsakoff patients. When FDG uptake ratios were examined with occipital lobe metabolism as covariate reference region, Korsakoff patients showed widespread bilateral white matter hypermetabolism on both SPM and ROI analysis. When white matter metabolism was the reference covariate; Korsakoff patients showed relative hypometabolism in the diencephalic grey matter, consistent with their known underlying neuropathology, and medial temporal and retrosplenial hypometabolism, interpreted as secondary metabolic effects within the diencephalic-limbic memory circuits. There was also evidence of a variable degree of more general frontotemporal neocortical hypometabolism on some, but not all, analyses.

Disruption of frontocerebellar circuitry and function in alcoholism

This article represents a symposium of the 2002 joint meeting of RSA and ISBRA held in San Francisco. Presentations were Neuropathology of alcohol-related cerebellar damage in humans, by Antony J. Harding; Neuropathological evidence of cerebellar damage in an animal model of alcoholism, by Roberta Pentney and Cynthia Dlugos; Understanding cortical-cerebellar circuits through neuroimaging study of chronic alcoholics, by Peter R. Martin and Mitchell H. Parks; and Functional reorganization of the brain in alcoholism: neuroimaging evidence, by John E. Desmond, S.H. Annabel Chen, Michelle R. Pryor, Eve De Rosa, Adolf Pfefferbaum, and Edith V. Sullivan.

Advanced brain imaging procedures and human memory disorder

The impact of advanced brain imaging procedures in the field of human memory disorder is reviewed, with particular emphasis on current and potential applications that may impact upon the diagnosis and management of memory-disordered patients. While both advanced structural, resting physiological and functional physiological brain imaging procedures have been applied to conditions where memory disorder is a major feature, the specific implications of research findings for diagnosis and treatment in routine clinical practice remain tentative and promising, but not yet substantive enough to inform clinical decisions to a significant degree. In terms of diagnostic applications, several promising areas include dementia, epilepsy, and transient amnesic states. In the case of applications in treatment settings, advanced brain imaging procedures may help to monitor neural correlates of spontaneous recovery or progression of memory function, and may also help in the planning and monitoring of therapeutic intervention.

Brain imaging. Functional consequences of ethanol in the central nervous system

In recent years, sophisticated methods have been developed to view structure and function within the living brain. Functional imaging methods are used to visualize dynamic chemical processes that are linked to brain activity. Increased neural activity, for example, leads to greater glucose and oxygen consumption and greater regional rates of blood flow to meet elevated energy demands. Mapping these changes provides quantitative visual descriptions of localized changes in brain activity that result from behavioral or pharmacological manipulations. This chapter first describes several current methods and how they are used to study the effects of alcohol on brain function. In the second part, the effects of acute intoxication are discussed with emphasis on the complex nature of alcohol's effects in the central nervous system, which depend on dose, time since administration, and environmental context. In the final part, the functional consequences of long-term exposure to alcohol as well as diseases associated with chronic alcoholism are reviewed.

Visualizing the firestorms in the brain: an inside look at the clinical and physiological connections between drugs and violence using brain SPECT imaging

The connection between drugs and violence has been well documented. Understanding the intricacies of this connection is essential to finding effective interventions. Much has been written about the psychosocial causes of these problems, but there have been few studies exploring the biophysiological interface between drug effects, violent behavior and brain metabolism. Over the past eight years, The Amen Clinic has been extensively involved in the clinical use of brain SPECT imaging to evaluate complicated neuropsychiatric problems, especially related to the issues of both violence and substance abuse. From this work several clinical patterns, as well as brain SPECT imaging patterns, have been recognized that may help further our understanding of these problems. In this article, following a brief review of the literature on drugs, violence and the brain, five clinical examples are explored; the authors show how these support the clinical utility of incorporating SPECT imaging into psychiatric assessment of drug abuse and violence. Finally, a model is proposed to help explain the complex interaction between the brain, violence and drug abuse.

Functional Neuroimaging of Cortical Dysfunction in Alcoholic Korsakoff's Syndrome

Many neuropsychological investigations of human memory have focused on the amnesic deficits of alcoholic Korsakoff's syndrome. Structural neuroimaging suggests that the syndrome results from midline diencephalic damage, but functional neuroimaging has the potential to reveal additional neuropathology that may be responsible for cognitive dysfunction. Accordingly, high-resolution positron emission tomography (PET) was used to measure regional cerebral metabolic rates for glucose utilization in five alcoholic Korsakoff patients and nine alcoholic control subjects. Results from a continuous recognition test administered during the radiotracer uptake period indicated that all subjects performed normally with respect to immediate memory, whereas Korsakoff patients demonstrated a marked memory impairment in delayed recognition. PET results from the Korsakoff group showed a widespread decline in glucose metabolism in frontal, parietal, and cingulate regions, suggesting that these functional abnormalities in the cerebral cortex contribute to the memory impairment. Hippocampal glucose metabolism did not differ between the groups. Thus, the evidence did not support the hypothesis that parallel brain dysfunctions are responsible for the similar amnesic symptomatology after hippocampal and diencephalic damage. We hypothesize that the amnesic dysfunction of Korsakoff's syndrome depends on a disruption of thalamocortical interactions that mediate a function critical for normal memory storage.

Effects of ethanol on GLUT1 protein and gene expression in rat astrocytes

Effects of ethanol on glucose transporter gene expression were examined in cultured rat astrocytes. Exposure to 50 or 100 mM ethanol for 18 hours significantly inhibited hexose uptake and reduced the number of glucose transporters, as indicated by binding studies with cytochalasin B. Indirect immunofluorescence and immunoperoxidase staining showed marked reduction of the GLUT1 glucose transporter by exposure to 100 mM ethanol for 5 or 18 hours, but no obvious change in response to 50 mM ethanol. Western blot analysis showed GLUT1 protein levels to be decreased by 52 +/- 12% (p < 0.05) after exposure to 100 mM ethanol for 18 hours. In situ hybridization histochemistry indicated an increase in steady-state GLUT1 mRNA in astrocytes exposed to 50 or 100 mM ethanol for 5 or 18 hours. Quantitation of GLUT1 mRNA levels by northern blot analysis showed that GLUT1 mRNA levels were increased by 59 and 112% in cells treated for 5 h with 50 and 100 mM ethanol, respectively. A similar effect was observed after treatment for 18 hours, but ethanol did not alter actin gene expression. Experiments using actinomycin D to block RNA synthesis suggest that this increase in steady-state mRNA level results from increased message stability. These results suggest that ethanol acts on GLUT1 gene expression at the post-transcriptional level.

Severe global amnesia presenting as Wernicke-Korsakoff syndrome but resulting from atypical lesions

A female alcoholic presented with Wernicke's encephalopathy subsequent to administration of diazepam and glucose (without thiamine) for treatment of withdrawal seizures. Nystagmus and cerebellar ataxia quickly resolved when administered thiamine, although severe global amnesia consistent with Korsakoff's syndrome persisted. Magnetic resonance imaging (MRI) revealed infarction of the right temporal lobe with hippocampal atrophy, but no lesions of thalamus or atrophy of mammillary bodies. Positron emission tomography (PET) confirmed decreased cerebral metabolic rates for glucose (CMRglu) in the right temporal lobe corresponding to MRI findings, but also significant metabolic asymmetry of dorsal thalamus, i.e. reduced CMRglu in left versus right. This patient is unique in that neuroradiological findings revealed intact mammillary bodies and suggest asymmetrical dysfunctions (structural right temporal and functional left diencephalic) to produce her profound amnesia.

Effects of ethanol on glucose transporter expression in cultured hippocampal neurons

Glucose transport was studied in primary hippocampal neuron cultures exposed to ethanol. Immunofluorescent staining with antibodies against neuron-specific enolase and glial fibrillary acidic protein identified approximately 95% of the cultured cells as neurons. Western blot analysis was conducted with polyclonal antisera to glucose transporter isoforms GLUT1 and GLUT3. As previously seen in astrocytes, GLUT1 protein was regulated by the culture medium glucose content. Exposure to 50 and 100 mM of ethanol for 5 hr induced dose-dependent reductions in GLUT1 and GLUT3 protein. In contrast, GLUT1 mRNA abundance was increased relative to controls under the same conditions. Glucose uptake, measured with the nonmetabolized analog, 2-deoxy-D-glucose, was reduced by 50 and 100 mM of ethanol in four experiments. These results indicate a direct effect of ethanol on neuronal glucose transporter expression, which may play a role in the neurotoxic effects of alcohol.

Brain proton magnetic resonance spectroscopy studies in recently abstinent alcoholics

Chronic alcohol-dependent patients have reduced brain volumes and concomitant neurobehavioral deficits that may recover during abstinence. In 10 chronic alcoholic patients, using localized proton magnetic resonance spectroscopy, we found reliable increases during the first 3-4 weeks of abstinence in the concentrations within the superior cerebellar vermis of choline (Cho)-containing compounds relative to the neuronal marker, N-acetylaspartate (NAA). Lesser changes were observed following 1 month of abstinence, and in one of the patients studied longitudinally over 3 months, a marked reduction in the Cho/NAA ratio was associated with relapse. After detoxification, the Cho/NAA ratio correlated with a composite clinical impression of brain functions. The lowest Cho/NAA was observed in a patient with persisting alcoholic dementia, in striking contrast to reduced relative concentrations of NAA reported in dementia of the Alzheimer's type. Possible molecular explanations for these brain metabolic changes are discussed.

The Korsakoff syndrome

BACKGROUND

Investigations of the Korsakoff syndrome by researchers from different disciplines have proliferated in recent years, making it apposite to review the various findings.

METHOD

This review is based on the author's knowledge of reports in the major clinical and neuropsychological journals, supplemented by Medline searches to update particular subtopics.

RESULTS

The Korsakoff syndrome is defined as a disproportionate impairment in memory, relative to other aspects of cognitive function, resulting from a nutritional (thiamine) depletion. The initial manifestations of the disorder are variable, and a persistent memory impairment can result from a non-alcoholic aetiology, although this seems to happen much less commonly than in the past - presumably because of generally higher standards of nutrition. Although there is agreement on the underlying neuropathology, the critical lesion sites for memory disorder have been debated. Recent evidence suggests that the circuit involving the mammillary bodies, the mammillo-thalamic tract and the anterior thalamus, rather than the medial dorsal nucleus of the thalamus, is particularly critical in the formation of new memories. The relationship of these deficits to thiamine depletion remains a topic of current investigation, as does the purported role of neurotransmitter depletions in the cholinergic, glutamate/GABA and catecholamine and serotonergic systems. Neuro-imaging studies have confirmed autopsy findings of more widespread structural and metabolic abnormalities, particularly involving the frontal lobes.

CONCLUSIONS

The relationship of these neuropathological, neurochemical, and metabolic abnormalities to cognitive functioning, with particular reference to specific aspects of memory processing, has been considered in some detail. Whereas structural and/or neurochemical abnormalities within the limbic/diencephalic circuits account for anterograde amnesia, some other factor, such as frontal lobe dysfunction, must underlie the severe retrograde memory loss which is characteristically found in this syndrome.

Decreased cingulate and precuneate glucose utilization in alcoholic Korsakoff's syndrome

Localized cerebral glucose utilization was determined for nine abstinent alcoholic men with Korsakoff's syndrome and 10 age-matched normal men who underwent positron emission tomography with [18F]2-fluoro-2-deoxyglucose (FDG). Patients with Korsakoff's syndrome showed relatively decreased glucose utilization in cingulate and precuneate areas. These decreases persisted even after correction for group differences in ventricular and sulcal cerebrospinal fluid measured on computed tomography. Electroencephalographic recordings at the time of FDG uptake showed no group differences, a finding that demonstrates that the metabolic differences could not be explained by differences in physiological arousal at the time of scanning. It is concluded that the decreased glucose utilization in the patients reflects a disruption of memory circuitry, the Papez circuit, caused by diencephalic lesions induced by thiamine deficiency.

Fully automated segmentation of cerebrospinal fluid in computed tomography

A method is presented for automated delineation and measurement of cerebrospinal fluid (CSF) regions in computed tomographic (CT) sections. Regions of skull and scalp are removed by using a linear discriminant analysis approach. Beam-hardening artifact is reduced by subtracting from each section the average radial intensity profile, characterized by a polynomial function. Remaining intensity gradients are suppressed by implementing CSF segmentation with a local thresholding technique based on maximum-entropy principles. CSF fractions from 12 regions of interest (ROIs) were measured in 10 patients with alcoholic Korsakoff syndrome and 9 normal volunteers. The same ROIs were also assessed by an interactive segmentation method, which enabled the operator to compensate for beam-hardening distortions by selecting suitable threshold values for each ROI. Both methods identified the same ROIs as displaying statistically significant differences between the two subject groups. However, interactive segmentation underestimated sulcal CSF by 20-70%, which was confirmed by applying both methods to CT scans of an anthropomorphic phantom. Hence, in contrast to interactive thresholding, unsupervised segmentation relies on firmly fixed criteria that reduce the influence of beam-hardening distortions and provide more objective results.