Vascular dementia (VD) or multi-infarct dementia is one of the most common causes of cognitive decline in adult humans, secondary only to AD. This is an increasingly diagnosed condition, where diffuse microvascular ischemic disease of the brain leads to progressive motor impairment and dementia. The wide range of clinical phenotypes has inspired numerous classification schemes. However, despite the efforts of many researchers, there are no universal clinical diagnostic criteria for VD. In addition, histopathologic data make this picture even more complex, showing that in many cases vascular pathology coexists with the pathology of AD. Although “pure” vascular pathology is relatively uncommon, it is widely accepted that VD can be split into cortical versus subcortical dementias: the former has a prevalent involvement of the cortical gray matter, whereas the latter is mostly characterized by lacunar infarcts and deep white matter changes.
The sensitivity for detecting vascular brain injury has widened significantly with the advent of modern neuroimaging. In patients with VD, quantitative MRI studies have consistently identified general brain atrophy and loss of cortical gray matter, enlargement of the ventricular CSF spaces, and even atrophy of the hippocampus and amygdala. Since all of these findings can also be seen in AD patients, the MRI comparison between these two types of dementia is usually inconclusive, especially at the early stages of each disease. Compared with normal subjects, an increased load of white matter lesions on T2-weighted images has been identified consistently in VD. The term leukoaraiosis has been suggested to describe the white matter changes associated with cerebrovascular disease, but the etiology and clinical relevance of these changes remain to be determined in most cases. The severity and frequency of leukoaraiosis increase with advancing age, risk factors for stroke, previous strokes (particularly of the lacunar type), and dementia. On the other hand, however, patients who meet neuroimaging criteria for leukoaraiosis may also be free of clinical signs of dementia, suggesting that the imaging changes are pathophysiologically nonspecific. Although the lack of pathological specificity of conventional MRI does not allow for the differentiation of the heterogeneous pathological processes associated with VD, neuroimaging still plays an important role in the diagnostic process of this disease. One of the most used diagnostic criteria, the so-called NINDS-AIREN criteria, includes brain imaging to support clinical findings, on the basis of lesion location and pattern typically associated with VD. Neuroimaging confirmation of cerebrovascular disease in VD provides information about the topography and severity of vascular lesions. It may also assist with the differential diagnosis of dementia associated with normal pressure hydrocephalus, chronic subdural hematoma, arteriovenous malformation, or tumoral diseases. In this context, MRI is preferred over CT scan because of its superior soft-tissue contrast. New MR modalities such as diffusion tensor imaging (DTI) have been reported to detect abnormalities not shown with conventional acquisition sequences and even attempt differential diagnosis, for example between AD and VD, on the basis of the brain regional differences in diffusion abnormalities.

1H-MRSpectroscopy

As mentioned above, the differentiation between AD and VD is challenging. In particular, since the patient may have mixed pathology, one may want to identify how much, if any, of the two pathologies are contributing to AD or VD, so that appropriate therapies can be planned. In this difficult context, the metabolic information provided by 1H-MRS can have some value. Levels of NAA and NAA/Cr are similarly diffusely reduced in the gray matter regions of patients with VD and AD. However, cortical levels of mI/Cr are normal in VD patients and elevated in AD patients. This can help in the differential diagnosis and, perhaps, to even identify concomitant AD in a demented patient with cerebrovascular disease. In addition, white matter levels of NAA/Cr are generally lower in patients with VD than in those with AD, reflecting the WM ischemic damage in VD compared to the cortical degenerative pathology in AD. This might be particularly useful for early diagnosis and in the context of specific familial conditions such as cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL).
The technical difficulties and limitations in performing an MRS examination described in AD are also relevant in this context. The use of short TE (~ 20–35 ms) is recommended. Finally, given the importance of assessing metabolic changes in cortical gray matter structures without contamination from other brain tissues, the use of MRI segmentation to allow tissue specific spectroscopic measures is strongly suggested.