Korsakoff’s syndrome (KS), a residual syndrome after an incomplete recovery from Wernicke’s encephalopathy (WE), is traditionally defined as “an abnormal mental state in which memory and learning are affected out of all proportion to other cognitive functions in an otherwise alert and responsive patient, resulting from nutritional, i.e. thiamine, depletion.” This definition certainly contains the key elements of an operational definition, but its usability as such is prevented by the fact that a reliable operationalization of “disproportionate memory disorder” is still lacking. As long as this situation continues, we should concentrate on the formulation of a useful and comprehensive conceptual definition of KS, to ensure its correct identification, to demarcate it from other forms of alcohol-related brain damage (ARBD), and also as a first step towards the formulation of a reliable operational definition. However, the above cited definition of KS does not satisfy these goals. In this presentation, we argue why and how it should be extended to satisfy the requirements of a useful conceptual definition.

A systematic identification, analysis and integration of the pertinent evidence from neuropathological, neuroimaging, clinical, neuropsychological, neuroanatomical and animal studies with respect to clinico-anatomical correlation in WE and KS was performed using an exhaustive PubMed literature search.

Neuropathological and neuroimaging studies have shown that in WE and KS lesions are not confined to the anterior thalamic nuclei and the mammillary bodies. The mediodorsal thalamic nuclei are damaged as well, probably even in advance, and often also the periaqueductal grey, the raphe nuclei, and the cerebellar vermis. Neuroanatomical and animal studies have established that, consequently, not only memory circuits are affected, but also several circuits responsible for executive functions (through damage in the mediodorsal thalamic nuclei) and emotion (through damage in the periaqueductal grey, the raphe nuclei and/or the cerebellar vermis). Clinical and neuropsychological studies have demonstrated that this manifold neuropathology is correlated with executive dysfunction, flattened affect, apathy, fantastic confabulations and lack of illness insight in KS, in addition to the memory disorder. Finally, studies in non-alcoholic WE and KS, in alcoholic patients without any evidence of WE, and converging evidence from other pathologies in subcortical grey matter structures, have confirmed that these pathological lesions and the accompanying symptoms are characteristic for thiamine deficiency (i.e., for WE or KS) and not for ethanol neurotoxicity or other forms of ARBD.

In KS, a manifold pathology leads to a multitude of symptoms. A conceptual definition of KS will therefore only enable a useful identification of KS and a demarcation from other forms of ARBD when it includes these symptoms, even though not all of them will always be present in every KS patient. Our findings are helpful for establishing clear diagnostic criteria for diagnosing KS, which are currently lacking as well.