Introduction

Carl Wernicke first described the syndrome in 1881. He referred to the disorder as acute superior hemorrhagic polioencephalitis. Some of the original patients he described included two male alcoholics and a women with esophageal stenosis. He described a clinic triad of encephalopathy, ophthalmoplegia, and ataxia.

Unfortunately, most diagnoses are not made clinically but rather at autopsy. This suggests that the classic clinical triad is rare, or that clinicians do not properly recognize the symptoms. In some clinical studies, only one third of patients diagnosed with Wernicke’s encephalopathy (WE) presented with the classical triad. The majority of the patients presented encephalopathy (characterized by disorientation, indifference, and inattentiveness). Ocular motor abnormalities (nystagmus, lateral rectus palsy, and conjugate gaze palsies) occurred in 96%, resulting from oculomotor, abducens, and vestibular nuclei lesions. Gait ataxia presented in 87%, probably due to a combination of cerebellar and vestibular involvement as well as polyneuropathy. However, an autopsy-based study revealed that while 82% had mental status abnormalities only 23% had ataxia, 29% ocular motor abnormalities, and 11% polyneuropathy. The clinical triad was identified in only 17% of autopsy cases, and 19% showed none of the classic symptoms. This discrepancy between the clinical and autopsy-based studies is probably due to exclusion of atypical cases in the clinical series and the underestimation in the autopsy series of classic signs that were not properly elicited, recognized, or recorded (1).

At autopsy the characteristic lesions of WK occur primarily in nuclei and structures surrounding the third and fourth ventricle and cerebral aqueduct. The mammillary bodies are affected in the vast majority of cases, usually showing atrophy (identified by MRI in 80% of alcoholics with WK). The dorsal medial thalamus, locus coeruleus, ocular motor nuclei, vestibular nuclei, and vermis of the cerebellum are also commonly involved. The lesions are identified by endothelial prominence, microglial proliferation, and sometimes petrichial hemorrhages (small purplish spots). In chronic lesions demyelination, gliosis, and loss of neuropil occur leaving most neurons preserved (1).

Historically, the only way to know for sure if a case was WE clinically was to treat the patient with thiamin (vitamin B). If the patient dramatically improved over a relatively short hospital stay, they were confirmed to have WE. This explains why alcoholics and malnourished patients will most likely present with WE. Because many alcoholics go undernourished, alcohol being a non-nutritive caloric diet and diuretic, thiamine deficiency results. With prompt treatment of thiamine, ocular symptoms usually improve within hours to days and ataxia and confusion within days to weeks. However a majority of patients are left with horizontal nystagmus, ataxia, and a disorder known as Korsakoff amnesia. This disorder is characterized by reduced anterograde and retrograde memory, apathy, intact sensorium, and selective preservation of their intellectual abilities. Yet, not all alcoholics have WE, nor do all alcoholics with thiamine deficiency develop WE. Furthermore, not all Korsakoff disorders initially presented with WE. So the differential diagnosis of WE, a neurological disorder of thiamine deficiency, is made difficult in alcoholic patients due to the neurotoxic effect of ethanol itself and the many complications of alcoholism (1).

Thiamine Deficiency

Due to the lack of specificity of symptoms of the disease and the wide diversity of patients, a more accurate diagnostic tool is sought. A recent study using high performance liquid chromatography (HPLC), has documented thiamine deficiency in acute WE. The study also analyzed the relative concentrations of thiamine as un-phosphorylated compound (T) and thiamine monophosphate (TP) in plasma and CSF, also as thiamine diphosphate (TPP) in whole blood. TP proved to be the most sensitive marker for thiamin deficiency in chronic alcoholics while a decrease in TPP concentrations may develop slowly. Plasma T has actually shown increases in some alcoholics, thus making this measurement an unreliable marker. Because thiamine is present in several body compartments and may be simultaneously active in one and not the other, and the onset of WE appears to depend on several factors including different degrees of thiamine deficiency , larger screening studies need to be undertaken. In one of the four patients of the study, thiamine therapy lacked a response. This is not an uncommon occurrence, and some suggest a genetic predisposition to explain the severity of the disease in certain individuals and why some nutritionally deprived individuals acquire it and not others (2).

Transketolase Allele

Transketolase requires TPP as a co-factor and catalyzes several reactions in the pentose phosphate pathway. Variants of the enzyme with reduced affinity (high Km) for TPP have been described in patients with Wernicke-Korsakoff-Syndrome (WKS). To test the hypothesis that differences exist between transketolase gene coding regions between Wernicke-Korsakoff and non-Wernicke-Korsakoff individuals, human fibroblasts were cloned. The study found that the transketolase gene exists as a single copy gene and while some nucleotide variations do occur, no amino acid differences resulted between WK and non-WK cells. It was concluded that the Km differences were not due to the nucleotide differences. The differences of Km may be due to some post-translational modification differences that have yet to be elicited (3).

Mammillary Bodies

In almost all cases of WE the mammillary bodies are involved. A recent study has challenged the concept that these lesions are characteristic of thiamin deficiency. The study tested this hypothesis by examining the brains of severe hypoxia-ischemia cases as occurs in myocardial infarct and cerebral stroke. It was observed that nearly identical lesions resulted. The lesion was localized exclusively in the medial nucleus preserving the lateral nucleus in all cases. The striking similarity was confirmed by immunohistochemical preparations. Under both conditions glial cells and myelin are affected with relative preservation of the neurons. This is not surprising because both conditions lead to lack of ATP (either Oxygen or NADH are not sufficiently provided). Other mechanisms may include tissue acidosis caused by excess lactic acid. The surprise, is the selectivity of involved structures. For yet unknown reasons, the thiamine deficiency never involves the cortical structures, while the hypoxia-ischemia does. Only severe hypoxia-ischemia involves the mammillary bodies. The high rate of metabolic turnover of the mammillary bodies may explain the vulnerability to a thiamine deficiency. The only noted differences between the hypoxia-ischemia and thiamine deficiency appears to be small hemorrhages that occurred in two of the cases of WE (4).

Thalamic Vacuolation

In a recent study using frozen sections to minimize artifacts, the thalamus of WE cases showed lesions different from those found in mammillary, periventricular, and brain stem structures. It is proposed that a specific neural pathway is involved in the pathology observed. The thalamic regions have dense cholinergic innervations from the brain stem and basal forebrain. The cholinergic brain stem neurons that project to the thalamus also produce nitric oxide. The basal forebrain also contains nitric oxide neurons which project primarily to mediodorsal nucleus. These pathways are known to be diminished in WE patients which may result in the observed vacuolation. This suggests that more than one pathological process is occurring in WE (5).

Excitotoxic Damage

Cell death may also be due to N-methyl-D-aspartate (NMDA) receptor-mediated excitotoxic damage. Since the tissue damage to particular brain strictures in WE resembles that seen in hypoxia-ischemia, one proposal suggests the involvement of excitatory amino acids. Age-cell death in glutamate toxicity is attributed to a massive influx of calcium ions through NMDA receptors. Treatment of animals with an NMDA receptor antagonist MK-801 led to a reduction of the lesions in the mammillary bodies (6).

Memory Impairment

It is known that lesions in mammillary nuclei in primates result in a form of special memory impairment, but not amnesia. However, lesions of the mediodorsal nuclei of the thalamus alone can produce a severe memory impairment. A large number of patients with WE develop the permanent WKS. This is characterized be severe impairment of memory both anterograde and temporal graded retrograde, as well as other frontal lobe disturbances. Studies have shown that WKS patients have extremely defective recognition memory as well as free recall. These tests were performed on a patient diagnosed with acute WE and who later had mental disturbances and memory problems. No diagnosis was made for WKS because she retained recognition memory. This indicated a retrieval deficit and not the full blown amnesia seen in the WKS. This study illustrates the criteria for diagnosis and multi-component nature of memory deficits seen in these patients (7).

Other theories have existed for memory impairment associated with WKS including lesions of the locus coeruleus. A recent study disproved the theory showing that neither alcohol neurotoxicity nor thiamine deficiency result in a reduction in the number of pigmented cells in the locus coeruleus (8).

The serotonergic system is also affected by alcohol by reducing serotonin levels. A recent study quantified this loss in chronic alcoholics compared to age-matched non-alcoholics. The alcoholics were further divided up into WE, WKS, and without neurological complications. The study tested whether the decrease in serotonin was greater in WKS. It was found that serotonin decreased equally in all of the chronic alcoholics. However, with the loss of cholinergic neurons in the basal forebrain known to occur in WKS, and the additional loss of serotonin, the combination may produce the amnesia. This would explain why thiamine deficiency alone does not produce WNS (9).

Conclusion

Wernicke’s encephalopathy is a rare disease which occurs primarily in chronic alcoholics. A positive correlation exists between thiamine deficiency and WE. However, it appears that other pathologies contribute to the disease. Since only a subset of chronic alcoholics actually come down with the disease, it appears that they have a predisposition toward it. This would also account for the observation that not all thiamine deficient individuals get WE. Questions for further research include why a thiamine deficiency is selectively targeting brain tissues, is a genetic post-translational event causing a reduced affinity for thiamine in certain individuals, and what is the main factor in the amnesia found in those individuals that go on to have the permanent condition of Korsakoff’s syndromes?

References:

1. Charness, M. E. Brain Lesions in Alcoholics. Alcohol Clin. Exp. Res., 1993 (Feb), 17(1): 2-11

2. Tallaksen, C., M. Bell, H. Bohmert. Thiamine and Thiamin Phosphate Ester Deficiency Assessed by High Performance Liquid Chromatography in Four Clinical Cases of Wernicke Encephalopathy. Alcohol Clin. Exp. Res., 1993 (Jun), 17(3): 712-716.

3. McCool, B. A., Plonk, S. O., Martin, P. R., Singleton, C. K. Cloning of Human Transketolase cDNA’s and Comparison of the Nucleotide Sequence of the Coding Region in Wernicke-Korsakoff and Non-Wernicke-Korsakoff Individuals. J. Biol. Chem., 1993 (Jan. 15), 268(2): 1397-1404

4. Vortmeyer, A. O., Hagel, C., Laas, R. Hypoxia-Ischemia and Thiamine Deficiency. Clin. Neuropathol., 1993 (July-Aug.), 12(4): 184-190.

5. Byrne, C., Halliday, C., Ellis, J., Harper, C. Thalamic Vacuolation in Acute Wernicke’s Encephalopathy. Metab. Brain Dis., 1993 (June), 8 (2):107-113.

6. Butterworth, R. F., Pathophysiology of Cerebellar Dysfunction in the Wernicke-Korsakoff Syndrome. Canad. J. Neuroscien., 1993 (May), 20 suppl. 3: sl23-126.

7. Parkin, A. J., Dunn, J. C., Lee, C., O’Hara, P. F., Nissbaum, L. Neuropsychological Sequelae of Wernicke’s Encephalopathy in a 20-Year-Old Woman: Selective Impairment of a Frontal Memory System. Brain Cognition, 1993 (Jan), 21(1): 1-19

8. Halliday, G., Ellis, J., Harper, C. The Locus Coeruleus and Memory: A Study of Chronic Alcoholics With and Without the Memory Impairment of Korsakoff’s Syndrome. Brain Research, 1992 (Dec), 598; 33-37.

9. Halliday, G., Ellis, J., Heard, R., Caine, D., Harper, C. Brainstem Serotonergic Neurons in Chronic Alcoholics With and Without the Memory Impairment of Korsakoff’s Psychosis. J. Neuropathol. Exp. Neurol., 1993 (Nov), 52(6):567-579.