Clinical manifestations

Presentation and course

Newer diagnostic criteria for vascular cognitive impairment. Vascular cognitive impairment now refers to a continuum of cognitive disorders ranging from mild vascular cognitive impairment to vascular dementia (49; 95; 20; 103; 47). Patients with vascular cognitive impairment can be identified on inpatient stroke services (post-stroke dementia) or in outpatient memory clinics among those who have never had strokes or transient ischemic attacks (20). Besides memory loss, those with vascular cognitive impairment may have other cognitive symptoms, such as problems with executive function, attention and concentration, visual-spatial abilities, as well as slowing of motor processing speed (122; 47; 60). One of the newer papers outlining diagnostic criteria for mild and major vascular cognitive disorder was written in an effort to harmonize with DSM-V criteria for mild and major cognitive disorder (95; 103). The difference between mild and major vascular cognitive impairment was based on the ability (or not) to perform instrumental activities of daily living. In this construct, major vascular cognitive impairment equated with what was formerly known as vascular dementia. The earlier criteria for vascular dementia were thought to be too narrow because they did not include patients with mild cognitive impairment, cognitive deficits other than memory loss, or those who had features of vascular dementia mixed with those of Alzheimer disease (95). A more recent set of diagnostic criteria for vascular cognitive impairment separates mild vascular cognitive impairment from vascular dementia based on the number of cognitive domains that are involved (one domain for mild vascular cognitive impairment and two or more domains for vascular dementia) (20). As with the older criteria, the newer ones also require signs of cerebrovascular disease on neuroimaging studies (large vessel disease, small vessel disease, hemorrhage, or hypoperfusion).

Types of vascular lesions. One way to think about the different subtypes of vascular cognitive impairment is to consider the different cerebrovascular lesions involved (49; 95; 11; 88; 119): (a) large vessel disease can lead to a single large infarct or multiple cortical infarcts; (b) small vessel disease can produce multiple cerebral microinfarcts, small subcortical infarcts, or extensive subcortical white matter changes; (c) strategic infarcts can occur in key places (thalamus, hippocampus); (d) cerebral amyloid angiopathy, which is associated with multiple cerebral microbleeds and subcortical white matter hyperintensities, can result in low executive function scores and mild cognitive impairment (11; 88); (e) cerebral hypoperfusion can result in cortical laminar necrosis or hippocampal sclerosis; (f) cerebral hemorrhage can result from multiple types of insults; (g) microemboli from atrial fibrillation can result in cerebral microinfarcts; and (h) Alzheimer disease can occur in combination with any of these types of cerebrovascular disease to produce mixed dementia (29). Cerebral microbleeds in one memory clinic population were identified in association with two different pathologies: cerebral amyloid angiopathy and hypertensive arteriopathy (35). Dilated perivascular spaces have been associated with cognitive decline and incident dementia (79; 92).

Lesion location matters. In subcortical vascular mild cognitive impairment patients, gray matter atrophy in the hippocampus and insula correlates with changes in memory and attention (65). Patients with subcortical vascular dementia have more atrophy in these areas, as well as the anterior cingulate and inferior temporal gyrus. In a small cohort of first-ever stroke patients, Zhao and colleagues studied the location of white matter hyperintensities most likely to be associated with cognitive impairment at baseline and again at one year (132). They found clusters of white matter changes in the corpus callosum, corona radiata, and posterior thalamic radiations that were associated with cognitive decline in these stroke patients (infarct volume was only 2.1 ml on average whereas white matter hyperintensities volume was 8.6 ml). In a large cohort of nondemented older adults, location of white matter hyperintensities was shown to influence the type of cognitive changes documented on neuropsychological tests (60). For example, those with white matter hyperintensities clustered near the frontal horns demonstrated executive problems, whereas others who had white matter changes near the posterior horns exhibited more problems with memory. Participants with deep white matter changes in the corticospinal tracts were more likely to show slowing in speed of motor processing. Cortical microinfarcts are frequently observed in watershed areas, suggesting they are the result of hypoperfusion (24). In a study of patients with mild cognitive impairment or subjective cognitive decline, large posterior areas of white matter hyperintensity appeared to be more closely associated with beta-amyloid positivity or Alzheimer disease, whereas large frontal white matter hyperintensity volumes or widespread white matter burdens were associated with higher vascular dementia (60; 78; 52).

Cognitive assessment tools. Cognitive screening tests used in the assessment of Alzheimer disease, particularly the MMSE, are not ideal tools in vascular cognitive impairment. Instruments that include assessment of frontal, executive, and subcortical function are preferred. Modifications of some of the tests originally developed for Alzheimer disease, such as the Vascular Version of the Alzheimer’s Disease Assessment Scale-Cognitive Subscale (VADAS-Cog), may be helpful (128). Some of these may be able to separate Alzheimer disease and subcortical vascular cognitive impairment, but even where this has been done using Pittsburgh B compound negativity to eliminate Alzheimer pathology, some overlap persists between the cognitive changes of vascular cognitive impairment and Alzheimer disease (130). Some validation data now exist in favor of the Montreal Cognitive Assessment (MoCA) for separating cases of vascular cognitive impairment from controls (83; 125; 30; 47) and the Addenbrooke’s Cognitive Examination (Revised Edition) (84). The NINDS and Stroke Canadian Network developed a 5-minute (12 point) subset of the MoCA for identifying stroke patients who had developed vascular cognitive impairment (55). Like the MoCA itself, total scores on this screening test inversely correlated with age and positively correlated with education. The Oxford Cognitive Screen is a 15-minute test that has fewer language elements than the MoCA, making it particularly useful in acute stroke patients who may have language deficits (71). Jolly and colleagues used the Brief Memory and Executive Test (BMET) to compare the prevalence of vascular cognitive impairment among CADASIL patients (39.8%) compared to controls (10.2%) (51). BMET was not quite as sensitive in their series as MoCA, which yielded prevalence figures of 47.7% and 19.6%, after controlling for age and sex.

The description of dysarthria, mild hemiparesis, imbalance, pseudobulbar palsy, small stepping (magnetic) gait, emotional incontinence, some degree of dementia, and incontinence (the "lacunar state") dates back to the turn of the century and in its classical form describes only a small proportion of all cases of vascular cognitive impairment. However, milder degrees of this, particularly gait disturbance, are not uncommon as a result of small vessel cerebrovascular disease (subcortical cognitive impairment) and may be quite prominent even in the absence of readily apparent cognitive impairment (10).

Prognosis and complications

Prognosis for mortality. Vascular dementia and vascular cognitive impairment due to small vessel disease can shorten life expectancy, but females, those with higher education, and those who perform well on some neuropsychological tests do better over time because of improved resilience (124). About a third die from complications of the dementia itself, one third from cerebrovascular disease, 8% from other cardiovascular disease, and the rest from miscellaneous causes, including malignancy. Several studies have suggested that the overall effect of vascular dementia on mortality is similar to or mildly worse than that of Alzheimer disease. Severity greatly influences prognosis, more severe cases progressing much more rapidly than milder ones with a mortality that may more than double (90). Cerebral amyloid angiopathy–related inflammation (CAARI) can occur spontaneously in those who have cerebral amyloid angiopathy; these abnormalities can produce a rare autoimmune encephalopathy that has been shown to be associated with increased CSF levels of anti-amyloid autoantibodies (85). These CAARI changes are similar or identical to those seen in amyloid-related imaging abnormality (ARIA) reactions to anti-amyloid antibody therapies (59).

Prognosis for cognitive decline. The Canadian Study of Health and Aging showed that hypertension predicts cognitive decline in those with evidence of vascular cognitive impairment but not in those with amnestic cognitive impairment (77). Those with more severe white matter hyperintensities have a greater risk of further cognitive decline than those without white matter hyperintensities (102; 48; 101; 02). In the Baltimore Longitudinal Study of Aging, 665 cognitively normal participants were examined with diffusion tensor imaging scans and neuropsychological tests (126). At baseline, a higher vascular burden score was associated with higher mean diffusivity in several white matter structures. After controlling for age, sex, race, and scanner type, a lower vascular burden score at follow-up (3.6 years) was associated with slower cognitive decline. Type 2 diabetes and impaired renal function are independently associated with abnormal brain structure and poorer performance on cognitive tests two years after stroke (07). In patients with small vessel disease, the progression of cognitive decline is nonlinear with an accelerating course over time (118). A more rapid cognitive decline was observed in autopsy-proven patients with mixed Alzheimer disease and comorbid cerebrovascular disease compared to brain donors with Alzheimer disease alone (29). The mixed dementia cases in this study showed accelerating deterioration in cognitive performance over time in measures of processing speed, working memory, verbal fluency, and naming. The magnitude of cognitive decline in CADASIL patients varies according to the stage of disease that is being assessed (09). Among young stroke patients (aged 18-60 years), the risk for poststroke dementia was highest (HR=2.53) in those who experienced seizures within the first three years of stroke onset (63). Among pediatric stroke patients, cognitive decline is more severe in later years among those who experienced stroke in early childhood compared to those whose strokes occurred in the neonatal period or in late childhood (01). A large study of first-ever stroke patients showed that those months after stroke, those with nocturnal sleep time of greater than eight hours (OR=3.54) were at greater risk of cognitive impairment (131). By six months after stroke, those with sleep time of less than 7 hours (OR=6.50) were at greater risk. After four years had passed, nocturnal sleep disturbances were still associated with more cognitive impairments in these patients.

Effect of cognitive reserve. The concept of cognitive reserve is that higher education, or other healthy lifestyle activities, can attenuate the effects of vascular brain injury on cognition. In one large international study, the authors used data from 10,916 participants (mean age 58.8 years, 55.8% female) to show that university/college education, moderate or higher leisure physical activity, being in a marital partnership, and participating in social groups were all factors independently associated with higher cognitive scores, whereas vascular cognitive injury (nonlacunar brain infarcts, or high white matter hyperintensity burden) was associated with lower cognitive scores (21).

Prognosis for psychiatric symptoms. When vascular cognitive impairment participants in a Canadian 30-month longitudinal cohort study were compared to patients with Alzheimer disease and normal older controls, 30% of those with vascular cognitive impairment had developed affective symptoms compared to only 15% of patients with Alzheimer disease or 12% of normal controls (< 0.01) (90). Depression has been reported in up to 20% of those with vascular cognitive impairment (108). Another study showed that 15% of patients with vascular dementia developed symptoms of psychosis (18). In one cross-sectional study of 437 stroke survivors, 115 participants (26%) had depression six months after stroke onset, and 133 (30%) had anxiety (127). Regression analysis revealed higher depression scores for those with MoCA-measured cognitive impairment, whereas no associations were found between cognitive impairment and anxiety.

Prognosis for white matter hyperintensities changes. Several large studies of aging have examined the progression of white matter hyperintensities volume changes per year and have found that it can vary from 4% to 37%/year (02). A few newer studies have shown evidence for reduction of white matter hyperintensities volume over time, provided there is consistent control of blood pressure (02; 124) and control of lipids with statin therapy (13). In a large (n=1230) study of elderly participants with five years of follow-up, white matter structural integrity over time was maintained better in those who had higher educational/occupational levels at baseline (120). Greater baseline white matter hyperintensity volumes are more likely to be associated with shorter times to mild cognitive impairment onset among those who also have high CSF tau levels (105). A study demonstrated that white matter hyperintensity volume changes measured one year after stroke were closely associated with cognitive changes (17).

Biological basis

Etiology and pathogenesis

Types of vascular changes seen in vascular cognitive impairment. Vascular cognitive impairment reflects the consequences of cerebrovascular disease in all its forms and degrees of severity upon cognition. Subcortical white matter hyperintensities, small subcortical infarcts, cerebral microinfarcts, small- and medium-sized hemorrhages, cerebral amyloid angiopathy, vasculitides, enlarged perivascular spaces, and the combination of Alzheimer disease and cerebrovascular changes are all etiological factors (49; 95; 119; 47; 08; 16; 29; 92; 12). Incomplete infarction comprises zones of partial neuronal or axonal loss with demyelination, increased perivascular spaces, a reactive astrocytosis, gliosis, and sparse macrophages (124). Silent and microscopic infarcts also contribute to cognitive decline, as do microhemorrhages. Cortical microinfarcts are increasingly recognized as important in cognition (45; 122; 119; 25; 24) and these, when multiple, can almost double the risk of dementia. Ischemic hippocampal sclerosis may account for a significant proportion of cases of vascular cognitive impairment (49).

Risk factors. The degrees of severity of damage associated with early vascular cognitive impairment can be very subtle and may begin before visible damage occurs on MRI. Risk factors, particularly hypertension, without clear vascular events are associated with both a relative increase in the volume of the lateral ventricles, suggesting central atrophy, and cognitive decline (22). Hypertension was the most prevalent risk factor for poststroke dementia in a large population-based cohort study (54). Diabetes correlates with lacunar stroke and decreased fractional anisotropy (66) may be particularly important in the elderly, but vascular risk factors overall may be especially important in younger patients (41). Atrial fibrillation increases the risk for both acute and chronic microinfarcts (OR = 4.8) (122; 119). Among dementia-free participants in the Framingham Offspring Study (mean age = 69 years), higher aortic stiffness predicted an increased risk of mild cognitive impairment (HR = 1.40) and all-cause dementia (HR = 1.45) (81). In the ARIC study, atrial cardiomyopathy was also shown to be a risk factor for incident dementia (50). It appeared to occur independently of the presence of either atrial fibrillation or stroke. There is also good evidence that other factors increase the risk for vascular dementia: age, low levels of education, obesity, midlife hypertension, hyperglycemia, diabetes, and stroke (31; 47). Studies have shown that systolic blood pressure variability is also a risk factor for vascular dementia (129; 69). Recent studies have suggested that an indirect link between air pollution and dementia may exist and that the homocysteine/methionine cycle may be implicated along with cardiovascular burden (34). Another report showed signs of Alzheimer pathology changes (increased tau binding on tau-PET) in those with cerebral amyloid angiopathy with amnestic presentations (99).

Genetic factors. There is increasing evidence for a significant role for genetic factors in the subcortical white matter changes of small vessel disease for which heritable factors may account for 70% of variability (115; 96) and for lacunar infarction (58; 112). Some of this may be mediated through angiotensin receptor and endothelial nitric oxide synthase gene polymorphisms or abnormalities of fibrinogen levels and fibrin clot structure. There may be racial differences. Although gene associations have been found (68), meta-analysis for polymorphisms in association with subcortical white matter changes has found no convincing associations. Apolipoprotein E4 homozygotes have more subcortical white matter hyperintensities and more cortical microbleeds than those who are not carriers of this allele (33). On the other hand, the APOE-E2 allele has protective effects against disease progression in those with subcortical vascular mild cognitive impairment (57). Similarly, those with hyperhomocysteinemia in association with MTHFR pathogenic variants are at increased risk of leukoaraiosis and lacunar strokes, but not other stroke types (94). Single-gene causes are increasingly recognized, including CADASIL, CARASIL, CARASAL, a collagen chain gene COL4A1 mutation, HERNS, a Dutch cerebroretinal vasculopathy, and German, Swedish, and Portuguese forms that mimic CADASIL. There may also be other factors acting directly on the endothelium including amyloid beta 1-40 and circulating endothelial progenitor cells. The Rotterdam Scan study has suggested a locus for vascular dementia on the X chromosome, near the locus for the androgen receptor gene, and studies looking at subjects enrolled in the UK Biobank found a locus in the VCAN gene (93) and PLEKHG1 (113). A useful review has been provided by Markus and Schmidt (72). Among CADASIL patients with NOTCH3 variants, earlier onset of stroke and encephalopathy has been demonstrated in those whose gene variants are in the lower epidermal growth factor domains (EGFRs 1–6) (15). The prevalence of vascular cognitive impairment among CADASIL patients ranges from 40% to 48%; it is closely associated with the number of lacunes measured on the MRI scan (51). Stage of disease is important when selecting cognitive tests for CADASIL patients because their cognitive abilities change quite a bit from middle age to older adulthood (09).

Leukoaraiosis. The white matter hyperintensities seen on MRI are the equivalent of what used to be referred to as leukoaraiosis, the white matter changes seen on CT (02). White matter hyperintensities include a wide range of structural changes encompassing increased water content, lacunes, and microbleeds (110). The subcortical U-fibers are spared. Axons, myelinated fibers, and oligodendrocytes are decreased in the affected areas, and spongiosis is seen in the same areas. Small punctate lesions seen on MRI correspond to dilated perivascular (Virchow-Robin) spaces, which, when dilated over 3mm, correlate with subcortical infarcts. Nonmodifiable risk factors for white matter hyperintensities include age, Black race, female sex, APOE-e4 allele, and cerebral amyloid angiopathy (88; 46), whereas the most prevalent modifiable risk factors are hypertension and low-density lipoprotein (02; 13). White matter hyperintensity volume was associated with poor executive function in a cohort study where adjustments were made for age, sex, education, and ethnicity (40). Postmenopausal women experience more white matter hyperintensity increase with age than men, and there is acceleration after menopause (67). These findings of accelerated white matter changes with age were more evident in women who had uncontrolled hypertension.

Axonal impairment. Functional imaging (diffusion tensor MRI) shows abnormalities of diffusivity in normal-appearing white matter in patients with subcortical white matter hyperintensities, and diffusivity correlates better with cognition and gait than with simple lesion load (19). These changes seem to be a pathological process, not just part of normal aging, with demonstration of evolution of areas of normal appearing white matter evolving into areas of subcortical white matter hyperintensities (117).

Cerebrovascular reactivity. Functional MRI is also capable of assessing cerebrovascular reactivity. One study used this method to compare patients with cerebral amyloid angiopathy, Alzheimer disease, and healthy controls (107). They found that cerebral amyloid angiopathy patients had lower cerebrovascular reactivity, compared to those with Alzheimer disease, or healthy controls. Lower vascular reactivity measurements in this study were associated with higher white matter hyperintensity volumes.

Epidemiology

The incidence of vascular dementia is about 3.8 per 1000 per annum. The incidence in women rises from between 0.3 and 1.36 in those 65 to 69 years old to 9.3 in those 85 years or older. For men, the corresponding figures are between 1.3 and 2.2, rising to 9.3 and 15.9 in those over the age of 90 years. Dementia from all causes has a prevalence of about 8% of the population over the age of 65. Between 9% and 39% (but typically 13% to 19%) of these cases are vascular. One paper estimated the prevalence of vascular dementia to be 16% (103). The proportion of cases due to vascular dementia falls with increasing age, but even so, the prevalence of all dementia rises so rapidly with age that the prevalence of vascular dementia also rises, from 0% to 2% in the 60 to 69 years age group to up to 16% for males aged 80 to 89 years, though more typical figures are between 3% and 6%. Males are more commonly affected than females in most studies.

Dementia after stroke is now increasingly recognized (64; 103). In a small MRI series of hospitalized patients with small strokes, 23% met criteria for vascular dementia; another 23% had what today would be called mild vascular cognitive impairment, and 54% had no cognitive impairment (43). In the Oxford Vascular Study, which excluded prestroke dementia patients, 20.7% of acute stroke patients developed signs of poststroke dementia (54). In the Framingham study, the 10-year risk of poststroke dementia was similar at 19.3% of cases, and dementia developed in 11% of controls (03). In a Chinese community-based cross-sectional study, the prevalence of poststroke mild cognitive impairment was 81%, whereas poststroke vascular dementia was 32% (87).

Even in those without prior dementia, poststroke dementia correlates with prestroke cognitive status suggesting preexisting pathology in at least some cases. Preexistent disease may also be vascular, cognitive decline after stroke being much more common in those who have small-vessel cerebrovascular disease. Post-stroke medial temporal lobe atrophy predicts subsequent dementia and death, this again favoring an important role for degenerative dementia in post-stroke dementia (26). Black participants may be at greater risk of post-stroke dementia than White participants, particularly because mixed pathology is more likely in Blacks than in Whites (06). Mixed pathology is also more common in the oldest old (53). Even though midlife hypertension is more common in men, it appeared in one study to only be associated with an increased risk of dementia in women (31). These findings were obtained from a large database collected from a diverse integrated U. S. healthcare delivery system (Kaiser Permanente).

Prevention

Vascular risk factors. Many of the commonly recognized vascular risk factors have been identified as risk factors for vascular cognitive impairment. These include hypertension, heart disease, atrial fibrillation, diabetes, insulin resistance, hyperlipidemia, smoking, carotid bruits, age, male sex, race, prior stroke, ECG changes, low hematocrit, history of myocardial infarction, homocysteine, carotid atherosclerosis, high dietary fat, low-density lipoprotein cholesterol, isolated systolic hypertension, chronic kidney disease, and proteinuria (07; 20; 47). For most of these, evidence that risk factor modification protects against vascular cognitive impairment is lacking, but it is reasonable to do so on first principles. Most of the benefit is likely to be on cognitive impairment mediated by large artery disease, as vascular risk factors have only a limited correlation with small vessel disease (123). Even so, recently identified falls in the incidence of dementia have some correlation with falls in vascular risk factors, although the falls in vascular risk factors (which exclude diabetes and obesity) do not fully explain the fall in incident dementia (97). Conversely, good cardiovascular health appears protective (80). There is now good evidence that treatment of atrial fibrillation and smoking cessation produces significant reduction in the risk of vascular dementia (47).

Nonvascular risk factors. Putative nonvascular risk factors include high alcohol consumption, psychological stress in early life, low education, lack of physical activity, poor diet, blue collar occupation, occupational exposure to pesticides, herbicides, liquid plastic, or rubber, and premorbid personalities that make subjects more liable to stress or psychosomatic reactions. Most of these require confirmation and quantification. In a prospective European study of older community participants (mean age = 74 years), physical activity (at least 30 min/d for 3 days/wk) was shown to reduce the risk of vascular dementia over three years (HR = 0.42; CI = 0.22–0.80) (121). The risk reduction produced by physical activity occurred independently of age, education, white matter hyperintensity volume, medial temporal atrophy, previous stroke, or diabetes. There is evidence from a 2-year trial that a multidomain intervention (diet, exercise, cognitive training, and vascular risk monitoring) was effective in producing a small amount of cognitive benefit in older subjects with several vascular risk factors (76). Regular exercise does reduce the risk of vascular dementia, Alzheimer dementia, and all-cause dementia (20). In a population-based sample of 800 women aged 38 to 54 years followed for 44 years, cognitive activity in midlife (artistic, intellectual, or religious) was associated with reduced risk of total dementia (HR = 0.66; CI = 0.49–0.89) during follow-up. Physical activity in midlife was associated with reduced risk of mixed dementia (HR = 0.43) and dementia with cerebrovascular disease (HR = 0.47) (75).

Antihypertensive drugs. The PROGRESS study provided evidence for the efficacy of treating hypertension in the secondary prevention of dementia after stroke or transient ischemic attack (86). The reduction, over 3.9 years of follow-up, in risk of dementia from 7.1% to 6.3% was not significant, but there was a reduction in cognitive decline from 11% to 9.1%, this benefit being attributable to the prevention of recurrent stroke. The Syst-Eur study showed that a reduction of 7 mmHg in systolic and 3.2 mmHg in diastolic blood pressure over 3.9 years halved incident dementia, although the absolute figures are a little less impressive at three cases per 100 patient years (28). Despite this association with the prevention of further stroke, there was a surprising lack of association of benefit with the drug regime used. The main PROGRESS study results had shown that combination perindopril and indapamide were more effective than monotherapy in preventing stroke, probably because of a greater reduction in blood pressure (86). A subsequent report from the PROGRESS trial showed that stroke patients with low baseline MMSE scores were at higher risk of subsequent dementia, even in the absence of recurrent stroke (89). A longitudinal cohort study in France showed that lower dementia risk over a median of 8.4 years’ follow-up was associated with the use of calcium channel blockers (HR = 0.56) or loop diuretics (HR = 0.45) after adjusting for blood pressure variability (114). The Systolic Blood Pressure Intervention Trial (SPRINT) tested the hypothesis that intensive blood pressure control would reduce cardiovascular disease outcomes in people with hypertension (106). The SPRINT MIND sub-study tested whether aiming at a lower systolic blood pressure target of 120 mm would reduce the risk of probable dementia better than a target of 140 mm. After one year of follow-up, the intensive control of blood pressure group (120 mm systolic) had a reduced risk of mild cognitive impairment (HR=0.81) and a combined reduced risk of both mild cognitive impairment and probable dementia (HR=0.85), but there was no significant reduction in the risk of probable dementia. A post hoc analysis of the SPRINT MIND trial showed that the best class of antihypertensive drugs to slow progression of small vessel disease was the ACE inhibitor class (lisinopril, etc) (32). The next best class of antihypertensive agents was calcium channel blockers (nifedipine, etc). The beneficial effects of these drugs on the slowing of small vessel disease occurred independently of either blood pressure control or age.

Statin drugs. Evidence that the statins protect against cognitive loss is relatively weak. Some studies have shown a benefit for all types of dementia, but others have not. The PROSPER study of 6000 individuals aged 70 to 82 years old was unable to demonstrate any benefit on stroke, subcortical white matter disease, cognition, or activities of daily living but did show a benefit on myocardial infarction and TIA (100; 109). In a 4-year observational study of 1000 postmenopausal women, statin users had a trivially (1%) higher score on a modified mini-mental state examination. The Cardiovascular Health study also showed no benefit. The MRC/BHF study of cholesterol lowering in 20,536 patients was also negative; but cognition was not formally evaluated prospectively in this study, and it is not clear how dementia was identified. Control of hyperlipidemia with statins is recommended for stroke prevention, even though the effects on vascular cognitive impairment are minimal (20). In a longitudinal study of 295 newly diagnosed mild cognitive impairment patients with low small vessel disease burden, those treated with statin drugs were less likely to have converted to dementia at two years (13).

Anticoagulants. There have been several studies showing that oral anticoagulant drugs reduce the risk of dementia in patients with nonvalvular atrial fibrillation (62). In this systematic review with meta-analysis, direct oral anticoagulants (DOACs) were associated with lower risk of dementia outcomes, compared with warfarin use (OR=0.56; CI=0.34-0.94). No significant differences were found between dementia types (vascular dementia vs. Alzheimer disease). There were also no differences between the effectiveness of two commonly used direct oral anticoagulants: apixaban reduced dementia risk (OR=0.58) by as much as rivaroxaban (OR=0.67).

Carotid endarterectomy. Because the effect of asymptomatic carotid revascularization on cognitive function was unclear, Ancetti and others performed a systematic review of 31 studies where cognitive performance was assessed before and after surgery (carotid endarterectomy) or carotid artery stenting (04). Their conclusion was that it was rare when either surgery or stenting resulted in cognitive improvement at 3 to 5 months. They also noted that the risk of cognitive decline was low in this group (less than 2%).

Differential diagnosis

The principal differential diagnoses are from Alzheimer disease and depression supervening on stroke, as depression commonly follows stroke. In 90% of cases in which multiple infarcts are responsible, there is also a history of stroke or of transient ischemic attacks. However, in cases in which subcortical ischemic change is the vascular mechanism, a history of stroke may be absent in up to 40%, and focal signs are also less common. Use of the ischemic scale score, which has 89% sensitivity and specificity (Table 1), can be helpful; scores of four or less suggesting Alzheimer disease and those of 7 or more supporting vascular dementia (39). However, distinguishing mixed dementia from either vascular dementia or Alzheimer disease remains difficult with poor specificity. Compounding the difficulty of identifying vascular dementia is its common coexistence with Alzheimer disease and the important effect of cerebrovascular disease to hasten the clinical manifestation of Alzheimer disease in what are, in effect, mixed cases even though there may be no specific history of stroke and the vascular disease may amount to only microinfarcts (111). Furthermore, many previous studies concerning the diagnosis of Alzheimer disease have dismissed small volumes of ischemic damage. In view of the interaction between cerebrovascular disease and Alzheimer disease, this now seems highly inappropriate, and about a quarter of cases of clinically diagnosed "pure" Alzheimer disease will have some vascular component.

Confusing conditions

Ischemic amnesia. Transient or persistent amnesia can sometimes be the main symptom of an ischemic stroke (74). The duration of symptoms in these patients was about five hours, and triggering psychological or physical factors were usually absent (unlike cases of transient global amnesia). On MRI, at least one ischemic lesion was localized to the posterior circulation in the majority of cases reported by Michel and colleagues (74). The ischemic lesions were most likely to be cardioembolic (53.8%) and multi-territorial (30.7%). During 12 months of follow-up, the patients with ischemic amnesia remained functionally independent, primarily because of the transient nature of their symptoms.

Diagnostic workup

Clinical evaluation. To diagnose vascular cognitive impairment, there must be the presence of a cognitive disorder (both subjective complaint and objective evidence of impairment), so patients will need to be tested with a cognitive screening instrument or will need neuropsychological testing. Memory loss is no longer required for the diagnosis of vascular cognitive impairment because many patients will have executive function problems as a more salient feature of their vascular cognitive impairment (95). The difference between mild or major vascular cognitive disorder depends on the patient’s functional independence, so the caregiver should be interviewed with an appropriate functional screening instrument to see whether they are independent with instrumental activities of daily living or not (42; 95).

Imaging. Imaging is required to confirm the presence of cerebrovascular disease and to exclude other structural causes of cognitive impairment (20). MRI will show the extent of subcortical white matter changes and small infarcts better than CT (23; 132; 02). MRI is also important to document the presence of enlarged perivascular spaces, cerebral microhemorrhages, cortical superficial siderosis and the results of hypoperfusion (laminar necrosis and hippocampal sclerosis) (95; 33; 79; 08; 16; 92; 12). Imaging with diffusion-weighted imaging MRI is needed in order to identify hyperacute (hyperintense) or chronic (isointense) microinfarcts (119). The spontaneous amyloid-related imaging abnormalities seen in patients with cerebral amyloid angiopathy are usually visible on MRI and can resemble the “ARIA-E” abnormalities seen in patients with Alzheimer disease who have been treated with anti-amyloid therapies. PET studies have shown that these abnormalities are associated with focal areas of microglial activation (85).

Laboratory. Other investigations should be directed by clinical suspicion. Not all laboratory investigations are routinely required. However, a complete blood count, erythrocyte sedimentation rate, glucose, and ECG should be done. Where appropriate, carotid duplex Doppler, chest x-ray, echocardiography, Holter monitoring, thrombophilia screen, lipid profile, lupus anticoagulant, thyroid function, vitamin B12, red-cell folic acid, and autoantibody screen are justifiable. Glycosylated hemoglobin may detect unsuspected diabetes. Cerebral angiography may be indicated if carotid surgery is considered or to demonstrate beading of the smaller cerebral vessels if a cerebral vasculitis is suspected, but it is not a routine investigation and can in any case miss an active vasculitis.

Management

Prevention. The treatment of vascular cognitive impairment is prevention by treating the underlying cerebrovascular disease and vascular risk factors, such as diabetes and hypertension (116; 47; 106; 104; 32). Lowering blood pressure to 120 mm systolic (vs. 140 mm) is associated with a lower risk of mild cognitive impairment, but not necessarily a lower risk of dementia (106). Statin therapy is associated with reduced conversion to dementia, when it is given to patients with mild cognitive impairment and low small vessel disease burden (13). Smoking cessation is recommended for those in the stage of mild vascular cognitive impairment because it will reduce the risk of vascular dementia (47). Regular physical activity should be advised for those who are in the stage of subjective cognitive decline because it is likely to prevent mild vascular cognitive impairment (47; 21). Multidomain intervention (diet, exercise, advice about vascular risks, and cognitive training) has been shown to have a modest benefit for reducing dementia risk in cognitively normal elderly people who have vascular risk factors (76). Treatment of atrial fibrillation has been demonstrated to reduce the incidence of vascular dementia by 29% (47).

Treatment. A few drug trials for vascular cognitive impairment have been completed. In the ONTARGET and TRANSCEND trials ramipril, telmisartan, and placebo were compared and cognition scored using the MMSE. None of the groups differed significantly, with 7% to 9% in each group developing cognitive impairment and 17% in each group developing lesser degrees of cognitive impairment over 56 months (05). The recent Canadian Consensus Conference on Diagnosis and Treatment of Dementia concluded that there was moderate evidence for use of cholinesterase inhibitors to improve cognition and activities of daily living in patients with vascular cognitive impairment (104).

Long-term low-dose aspirin in both the Women’s Health Study and Aspirin for Asymptomatic Atherosclerosis studies revealed no cognitive benefit. Clopidogrel or aspirin with dipyridamole given after stroke also failed to make any difference to cognition to a median of 2.4 years of follow-up, although the MMSE was the cognitive scale used. This finding was supported by the SPS3 trial, which found no benefit for cognition from either short-term dual antiplatelet treatment or greater antihypertensive therapy after lacunar stroke (82). The Canadian Consensus Conference recommended that aspirin use was reasonable for vascular cognitive impairment patients when there were signs of silent infarcts (104).

Extracranial-intracranial bypass after carotid occlusion offers no benefit over best medical therapy with respect to stroke or cognitive outcome (73). A randomized clinical trial confirmed that this particular bypass procedure offered no benefit of surgery over medical management for preventing stroke (70).

Special considerations

Anesthesia

In some circumstances, hypotension may exacerbate the condition and should be avoided.

Pediatric stroke

Investigators using the Swiss Neuropediatric Stroke Registry have shown that age at the time of pediatric stroke appears to influence long-term cognitive outcome (01). Those with early childhood stroke (onset between 29 days and 6 years) are less likely to become cognitively impaired than patients with neonatal or late childhood strokes.