The first step when using animals to study a cognitive dysfunction is to characterize the disorder. The one we are presently concerned with is “alcoholic dementia.” Not everyone who drinks chronically develops dementia, and the major risk factors for disease development and the mechanisms by which this occurs have remained unclear. There are data indicating that abuse of alcohol alone does increase the risk of dementia, whereas the potential deleterious or beneficial effects of light to moderate drinking remain unclear (1–3). Alcohol abuse may contribute to cognitive impairment in several different ways. Acute use of alcohol (acute intoxication) impairs attention, memory, executive functions, and visuospatial skills (4–9), while chronic abuse causes neurocognitive deficits in memory, learning, visuospatial functions, psychomotor speed processing, executive functions, and decision-making and may lead to persistent amnesic disorder and alcoholic dementia (5, 10, 11). Among the several kinds of dementia, we cite substance-induced persisting dementia, which may be due to drug abuse, medication, or toxin exposure. A number of toxins have been reported to cause dementia, but the principal one in most societies is alcohol (ethanol). Alcoholic dementia is a disorder characterized by multiple cognitive deficits that include memory impairment associated with one or more cognitive disturbances listed in Table 1.

Cognitive impairment may range between severe dementia and subtle deficits, which may be reversible (at least in part) or permanent. Although chronic alcoholism is associated with cognitive impairments, there are still conflicting opinions about the pathogenesis of alcohol-related memory impairments (12). Much of this debate has revolved around the relative contributions of nutritional deficiency and alcohol to the neuropathological lesions (13, 14). One of the most important questions concerning alcoholic dementia is whether long-term alcohol use may have direct neurotoxic effects on the brain, leading to a characteristic dementia syndrome. Clinical and neuropsychological data point toward the existence of an alcoholic dementia without secondary complications. Improvements in neuroimaging technology, such as magnetic resonance imaging, magnetic resonance spectroscopy, and positron emission tomography have contributed significantly to unravel structural abnormalities in uncomplicated alcoholics (with no hepatic or thiamine deficiency problems) who are cognitively impaired (2, 15, 16). The existence of specific neurotoxic effects of alcohol on the central nervous system (CNS) has been approached in a number of studies (2, 17, 18).

To reduce subjective bias and standardize DSM-IV criteria (see Table 1) for alcohol-induced persisting dementia, Oslin et al. (19) proposed some criteria to improve the validity and reliability of the diagnosis of “Alcohol-related dementia” (ARD) (13). These authors reported that individuals with ARD are less cognitively impaired than patients with Alzheimer’s disease and that both their cognitive and functional status frequently stabilizes. They consider ARD to differ from alcohol-induced dementia (DSM-IV) in that the presence of language impairment (aphasia) is less likely in ARD. This absence of aphasia in ARD is of clinical importance for the potential differentiation of the latter from Alzheimer’s disease, frontotemporal dementia, and vascular dementia (20).

Brain neuropathological lesions observed in demented alcoholic patients are more extensive in patients who have additional vitamin B1 (thiamine) deficiency, i.e., Wernicke–Korsakoff Syndrome (WKS). In alcoholics, several factors may contribute to thiamine deficiency and to dysfunctions of the thiamine-dependent or -metabolizing enzymes: (i) nutritional thiamine deficiency due to poor eating habits often seen in alcoholics; (ii) impairment of thiamine absorption from the intestine (21); (iii) changes in thiamine-metabolizing enzyme activity in the brain, e.g., a significant decrease in thiamine pyrophosphokinase activity caused by excessive alcohol consumption (22); (iv) changes in activity of thiamine-dependent enzymes that participate in the metabolic pathway (pyruvate dehydrogenase, α-ketoglutarate dehydrogenase, and transketolase) (23).

The most characteristic symptoms of WKS are anterograde amnesia (i.e., the inability to learn and form new memories), retrograde amnesia (i.e., the loss of memories formed prior to the onset of WKS), and impairment of several cognitive processes (24). According to latter publication, these cognitive and memory deficits are accompanied by characteristic pathological damage to several brain structures. In these cases, alcohol may have a direct neurotoxic effect on cortical neurons, but much of the damage may be secondary­ to diencephalic pathology caused by thiamine deficiency, including loss of nerve cells in the thalamus, cell death and hemorrhages in the mammillary bodies, and enlargement of the third ventricle.

Results of a necropsy study by Harper (25) showed that 80% of 131 patients with WKS were not diagnosed as such during life. This is in accordance with some authors who have cast doubt on the existence of primary alcoholic dementia as a neuropathological entity and suggest that the majority of cases of alcoholic dementia are in fact WKS (26).

Wernicke’s encephalopathy is a neuropsychiatric condition that can lead to dementia and occurs as a result of thiamine deficiency. It has been described as an acute neurological disorder characterized clinically by oculomotor abnormalities, cerebellar dysfunction, and altered mental state, and characterized pathologically by hemorrhagic lesions in the wall of the third and fourth ventricles (27). Korsakoff’s syndrome is characterized by profound amnesia, disorientation, and frequent confabulation and is defined as a disproportionate impairment in memory, relative to other features of cognitive function, resulting from nutritional (thiamine) deficiency (28). Although the typical Korsakoff’s syndrome presents as a pure amnesic state without significant intellectual decline, some patients are reported to show remarkable and overall dementia. Scanning studies have shown that patients with Korsakoff’s syndrome have a more widespread cerebral and subcortical atrophy than alcoholic patients without amnesia (29). Because Korsakoff’s syndrome often follows or accompanies Wernicke’s encephalopathy, with the typical clinical pattern emerging when the acute global confusional state of the latter resolves, and both appear to share common pathological substrates, a number of researchers describe this continuum as WKS (26 28 30). The widely held view is that the cognitive deficits of varying types and severities found in alcoholics arise as a result of a coexisting thiamine deficiency syndrome (14). According to these authors, the fact that the degree of vulnerability of patients with Wernicke’s encephalopathy to Korsakoff’s syndrome appears to vary according to the history of alcoholism goes against the widely cited conclusion of Victor et al. (31) that Wernicke’s encephalopathy and Korsakoff’s syndrome are the acute and chronic phase of the same thiamine deficiency syndrome. Relating to the fact that alcohol consumption plays a relevant role in this syndrome, Martin et al. (32) consider that the most devastating neurological disorder caused by thiamine deficiency is WKS, which is seen most commonly in alcoholics. In short, clinical pathological issues regarding the so-called “alcoholic dementia” remain under debate. Although clinical observation favors the diagnosis of primary alcohol dementia, caused by direct alcohol neurotoxicity, further confirmation from the neuropathological and biochemical perspectives is required. Moriyama et al. (13) concluded that repeated episodes of subclinical WKS may partially account for the chronic state of primary alcoholic dementia, thus supporting the notion that primary alcoholic dementia exists in continuum with chronic and subclinical types of WKS. However, the role of thiamine deficiency remains undetermined in chronic “uncomplicated” alcoholism, but postmortem study indicates that Wernicke’s encephalopathy is underdiagnosed in life (16 32 33). He et al. (34) consider that these unsolved possibilities can be addressed with larger in vivo and postmortem animal studies that manipulate thiamine levels in the context of bouts of high alcohol exposure aimed at paralleling human alcoholism. Uncomplicated alcoholics likely sustain repeated bouts of less profound nutritional deficiency, which may contribute to their milder in vivo neuroradiological findings when compared with those marking WKS (14). McIntosh and Chick (35) state that improvements in imaging technology and the continuous progresses in brain neurochemistry knowledge have started to unravel the association between alcohol, thiamine, and memory. It has been proposed that most organic brain syndromes in alcoholic patients are variants of the WKS, and that there is no need to consider a separate category of “alcoholic dementia.”