To the Editor: We read with great interest the article by Gabriel Gold, M.D., and coauthors (1). Since there is a remarkable lack of data on the neuropathology of vascular dementia, studies aimed at evaluating the pathological findings in patients with dementia and cognitive impairment are badly needed.

Dr. Gold and co-workers compared the postmortem diagnoses of dementia types with those clinically reached by the application of four well-known sets of criteria for vascular dementia (those from DSM-IV, ICD-10, the State of California Alzheimer’s Disease Diagnostic and Treatment Centers [ADDTC], and the National Institute for Neurological Disorders and Stroke—Association Internationale pour la Recherche et l’Enseignment en Neurosciences [NINDS-AIREN]). The authors reported that the sensitivity of these clinical criteria, when compared to the gold standard of postmortem examination, ranged from 0.20 (NINDS-AIREN criteria) to 0.70 (ADDTC criteria), while the specificity showed somewhat higher values, ranging from 0.78 to 0.93. We would like to contribute to the discussion with some comments.

First, as acknowledged by the authors, there are no established and validated pathological criteria for vascular dementia. In their study, the attribution of a patient to the vascular dementia group was based on the finding of multiple macroscopic and microscopic cortical infarcts involving three or more cortical areas exclusive of the visual cortex. Lesions confined to the subcortical structures were not taken into account; this was obviously arbitrary. As is largely accepted today, "vascular dementia" encompasses a wide spectrum of conditions, only some of them related to the occurrence of multiple cerebral infarcts. Unfortunately, almost all current criteria, while recognizing this aspect, outline clinical features (the presence of major cerebrovascular events) that are mostly related to the occurrence of large cerebral infarcts. As rightly stated by the authors, current criteria mainly identify cases of so-called "multi-infarct dementia." It is disappointing that these cases do not appear to be the most prevalent in the population, being probably numerically surmounted by cases characterized by the presence of subcortical lesions (multiple or strategic lacunar infarcts and diffuse white matter lesions) mainly linked with disease in the small vessels (2–4).

Second, cases of subcortical vascular dementia are probably the most difficult to clinically distinguish from degenerative cases, given the lack of a history of stroke in many instances and progressive cognitive decline. Only a thorough clinical and neuropsychological evaluation, perhaps with the help of neuroimaging, may clarify this distinction. That is seldom the case in retrospective and epidemiological studies. Accordingly, the sensitivity and specificity of current criteria could be even lower in examinations of patients with subcortical lesions.

A final consideration concerns the role of neuropathology in the field of dementia. Since dementia is a pure clinical definition, can the pathological examination be considered a true gold standard in this field? By no means; neuropathology has a great role in defining the lesions present in patients with cognitive decline. However, we feel that we are far from finding the precise structural correlate of dementia in the brain. In many instances, dementia derives from a functional rather than from an anatomical alteration. In other instances, the discovery of cerebral lesions is not reflected in clinically evident alterations. Accordingly, the diagnosis of dementia should remain essentially clinical. Notwithstanding these facts, there is a tremendous need to descriptively and systematically study from the pathological viewpoint the types, extent, location, and possible coexistence of cerebral lesions in cases of cognitive decline. However, agreement should be reached regarding what lesions are to be considered as vascular. Neuropathologists are called upon to merge their efforts with those of clinicians and neuroradiologists to better define the vascular burden on the brain.