With >24 million people affected worldwide, dementia is one of the main public health challenges modern medicine has to face. The path leading to dementia is often long, with a wide spectrum of clinical presentations, and preceded by a long preclinical phase. Previous studies have demonstrated that clinical strokes and covert vascular lesions of the brain contribute to the risk for developing dementia. Although it is not yet known whether preventing such lesions reduces the risk for dementia, it is likely that starting preventive measures early in the course of the disease may be beneficial. Echocardiography is a widely available, relatively inexpensive, noninvasive imaging modality whereby morphologically or hemodynamically derived parameters may be integrated easily into a risk assessment model for dementia. The aim of this review is to analyze the information that has accumulated over the past two decades on the prognostic value of echocardiographic factors in cognitive impairment. The associations between cognitive impairment and echocardiographic parameters, including left ventricular systolic and diastolic indices, left atrial morphologic parameters, cardiac output, left ventricular mass, and aortic root diameter, have previously been reported. In the light of these studies, it appears that echocardiography may help further improve currently used risk assessment models by allowing detection of subclinical cardiac abnormalities associated with future cognitive impairment. However, many limitations, including methodologic heterogeneity and the observational designs of these studies, restrict the scope of these results. Further prospective studies are required before integrating echocardiography into a preventive strategy.
Since the early 1990s, many cross-sectional and longitudinal epidemiologic studies have revealed an association between vascular factors, including clinical, biologic, and echocardiographic parameters, and risk for cardiovascular events. The results of these studies have allowed the stratification of patients according to their individual risk, the aim being to develop treatments and preventive measures for use at an early stage of the disease.
With >24 million people affected worldwide and considering the increasing life expectancy of the population, dementia is becoming one of the main public health challenges modern medicine has to face. It is one of the primary causes of loss of autonomy and institutionalization of elderly patients in many countries. The path leading to dementia is often long, with a wide spectrum of initial clinical presentations. Moreover, an extended period exists during which subjects do not exhibit clinically noticeable symptoms despite the presence of neurodegenerative lesions. Identifying early determinants of this process may lead to the initiation of specific treatments to slow the progress of the disease. More particularly, clinical strokes and covert vascular lesions of the brain contribute to the risk for developing all types of dementia. Although it is not yet known whether preventing such lesions reduces the risk for dementia, it is very likely that starting preventive measures early in the course of the disease may be beneficial. Prevention has had a significant role in the decline of many cardiovascular diseases, and the application of such measures to cognitive impairment could have major medical and economic impacts.
An increasing volume of information has accumulated over the past two decades on the prognostic value of echocardiographic factors in cognitive impairment. In this article, we seek to provide an up-to-date review on the link between the two.
Clinical Determinants of Cognitive Impairment
The main clinical factors associated with cognitive impairment are advanced age, hypertension, diabetes mellitus, high plasma cholesterol levels, obesity, tobacco smoking, and the presence of atherosclerotic vascular disease. In addition to these risk factors, hyperhomocysteinemia, diets comprising high levels of saturated fats, excessive alcohol consumption, and physical inactivity increase the risk for developing Alzheimer’s disease. These risk factors are also associated with an increase in the incidence of cerebrovascular diseases. Although the exact mechanisms are still unknown, heart failure and the occurrence of acute coronary syndromes are also associated with risk for cognitive impairment, probably because of the increased risk for vascular disease in this category of patients. The occurrence of cardioembolic strokes and reduced cardiac output (CO) leading to decreased cerebral perfusion may also explain the increased risk for cognitive impairment in patients with heart failure. Atrial fibrillation (AF) is also independently associated with the onset of cognitive impairment and Alzheimer’s disease, although the exact mechanism explaining this link remains unclear. With an annual incidence of 4.5%, AF is the primary cause of ischemic stroke, causing up to 15% of reported events. The development of brain imaging techniques including magnetic resonance imaging (MRI) has revealed the occurrence of silent brain infarctions in patients with AF, causing damage to the brain parenchyma, leading to further cognitive impairment. Moreover, patients with mild cognitive impairment face an increased risk for conversion to dementia if they have associated diagnoses of AF. Another important determinant of cognitive impairment is the presence of cerebral white matter lesions detected on brain MRI. These lesions are associated with reduced performance of many cognitive functions, such as executive functions, processing speed, and immediate and delayed memory.
Cognitive Impairment Risk Evaluation Models
It is now accepted that a preclinical phase exists in dementia, hence the early identification of patients at risk for developing the disease is likely to be a key step. In 2006, Kivipelto et al . proposed a cognitive impairment risk evaluation model, the Cardiovascular Risk Factors, Aging and Dementia risk score, to predict the 20-year risk for dementia in middle-aged people. This model stratifies patients by age, sex, education, hypertension, body weight, physical activity, hyperlipidemia, and apolipoprotein E genotype. Since then, several studies have assessed the usefulness of different risk scores to predict the onset of cognitive impairment and dementia. It is worthy of note that most of the risk factors associated with dementia used in these algorithms, such as diabetes mellitus, hypertension, hypercholesterolemia, and obesity, are modifiable and are shared with cardiovascular diseases, illustrating the close link between the brain and the heart. This link was further strengthened when the Framingham general cardiovascular disease risk score and the Framingham stroke risk score were compared with the Cardiovascular Risk Factors, Aging and Dementia risk score. Following the examples of risk scores used in AF, acute coronary syndromes, or cardiac surgical risk evaluation algorithms, creating a multimodal risk evaluation model on the basis of various parameters including ultrasound-based measures is an appealing idea that may help further enhance our capacity to predict cognitive impairment. However, the usefulness of integrating echocardiographic parameters in a cognitive impairment risk evaluation model has never been assessed.
Cognitive Impairment Risk Evaluation Models
It is now accepted that a preclinical phase exists in dementia, hence the early identification of patients at risk for developing the disease is likely to be a key step. In 2006, Kivipelto et al . proposed a cognitive impairment risk evaluation model, the Cardiovascular Risk Factors, Aging and Dementia risk score, to predict the 20-year risk for dementia in middle-aged people. This model stratifies patients by age, sex, education, hypertension, body weight, physical activity, hyperlipidemia, and apolipoprotein E genotype. Since then, several studies have assessed the usefulness of different risk scores to predict the onset of cognitive impairment and dementia. It is worthy of note that most of the risk factors associated with dementia used in these algorithms, such as diabetes mellitus, hypertension, hypercholesterolemia, and obesity, are modifiable and are shared with cardiovascular diseases, illustrating the close link between the brain and the heart. This link was further strengthened when the Framingham general cardiovascular disease risk score and the Framingham stroke risk score were compared with the Cardiovascular Risk Factors, Aging and Dementia risk score. Following the examples of risk scores used in AF, acute coronary syndromes, or cardiac surgical risk evaluation algorithms, creating a multimodal risk evaluation model on the basis of various parameters including ultrasound-based measures is an appealing idea that may help further enhance our capacity to predict cognitive impairment. However, the usefulness of integrating echocardiographic parameters in a cognitive impairment risk evaluation model has never been assessed.
Conventional Echocardiographic Parameters of the Left Ventricle and Cognitive Impairment
The presence of a heart disease may alter left ventricular (LV) function, leading to a reduction in cerebrovascular blood flow causing subclinical brain injury and cognitive impairment. It was described previously that patients with severe chronic heart failure may have reduced cerebral blood flow, by approximately 30%, despite the activation of physiologic neurohormonal counterregulatory mechanisms such as the renin-angiotensin system and the sympathetic nervous system.
LV Systolic Function
LV systolic dysfunction, measured by the LV ejection fraction (LVEF), appears to be particularly correlated with cognitive impairment ( Table 1 , Figure 1 ). Jerskey et al . found that reduced sustained attention and vigilance is correlated with an LVEF ≤ 55%. In the Framingham Heart Study cohort, comprising participants free of stroke or dementia, a nonlinear association between LVEF and measures of cognitive impairment was suggested. In that study, LVEF was not linearly associated with white matter lesions on brain MRI or with any neuropsychological variables, although patients in the lowest and highest LVEF quintiles had abnormal cognitive changes. Interestingly, patients belonging to the highest LVEF quintile also had a significant association with reduced cognitive functions in verbal and visuospatial memory, executive functioning, and visuoperceptual abilities. This association persisted despite adjustment for many confounding factors and after the inclusion of multiple covariates. The exact mechanism explaining this phenomenon is unknown. Furthermore, an increased risk for future cognitive impairment and stroke is associated with white matter lesions and silent brain infarcts. It is worth mentioning that in several studies, an association between white matter lesions on brain MRI and LV systolic dysfunction has been demonstrated.
| Study | Cardiac imaging parameters | Population | Age (y) ∗ | Cognitive parameters | Results |
|---|---|---|---|---|---|
| Russo et al . (2013) | LV GLS | 436 stroke-free patients without CVD | 69.3 ± 9.7 | White matter lesions and silent brain infarctions in brain MRI | • Lower GLS associated with silent brain infarctions and white matter lesions • LVEF not associated |
| Van den Hurk et al . (2011) | LV diastolic function: ratio of early mitral valve flow velocity and early diastolic lengthening velocity | 313 healthy people | 67 (59–87) | • Attention and executive functioning • Information-processing speed | LV diastolic dysfunction at baseline associated with lower scores on attention and executive functioning at follow-up |
| Jefferson et al . (2011) | LVEF | 1114 patients from the Framingham Heart Study Offspring Cohort free from stroke or dementia | 67 ± 9 | • Delayed memory, language, executive functioning, verbal reasoning, visuoperceptual abilities • White matter lesions | • LVEF not linearly associated with white matter lesions or neuropsychologic variables • Lowest and highest LVEF quintiles had abnormal cognitive changes |
| Jerskey et al . (2009) | LVEF | 67 outpatients with stable CVD without neurologic diseases | 68.5 ± 7.4 | Sustained attention and vigilance | Reduced LVEF associated with reduced sustained attention and vigilance |
| Suwa and Ito (2009) | LV diastolic function | 81 outpatients with stable CVD and LVEF ≥ 40% | Minimum, 65 | MMSE score ≤ 24 | LV diastolic dysfunction related to cognitive impairment |
| Moser et al . (1999) | LVEF | 93 patients with CVD enrolled in a cardiac rehabilitation program | 60.9 | • Response generation memory • Verbal abstraction • Verbal fluency | LVEF associated with cognitive impairment |
| Zuccalà et al . (1997) | LVEF | 57 patients with chronic heart failure | Mean, 76.7 | • MMSE score < 24 • Mental deterioration battery • CES-D score • Katz activities of daily living • Instrumental activities of daily living | LVEF associated with cognitive impairment |
∗ Data are expressed as mean ± SD or mean (range) unless otherwise stated.
Recently, Russo et al . published the first article studying the association between LV global longitudinal strain (GLS) and subclinical brain disease in a community-based cohort ( Table 1 ). The definition of an abnormal LV GLS value was based on a measure lower outside the 95th percentile of the LV GLS distribution in a subgroup of healthy participants and corresponded to LV GLS ≥ −14%. This article highlighted an association between lower LV GLS and the existence of subclinical cerebral white matter lesions on brain MRI in subjects without dementia symptoms and without overt cardiac disease independently of LVEF. Moreover, significantly lower LV GLS values were observed in participants with silent brain infarcts despite having similar LVEF values.
LV Diastolic Function
LV diastolic dysfunction appears particularly associated with lower cognitive function in patients with heart diseases ( Table 1 ). Van den Hurk et al . showed, in a cohort of healthy people, that LV diastolic dysfunction assessed by E/e′ ratio at baseline was associated with lower scores on attention and executive functioning at follow-up. Similarly, in a population of patients with cardiovascular diseases, risk factors, and LVEF ≥40%, Suwa and Ito found a significant association between diastolic function and cognitive impairment. More recently, Shimizu et al . identified an association between LV diastolic function and white matter lesions on brain MRI in a population of elderly patients with normal LV contraction and no signs or history of cardiac disease, stroke, or cognitive dysfunction.
Left Atrial Enlargement Parameters and Cognitive Impairment
Left atrial (LA) function can be considered as a three-component model. The first component occurs during LV systole and is called the reservoir phase, in which the left atrium receives blood from the pulmonary veins. The second component is called the passive conduit phase, occurring during LV early diastole and diastasis. The LA contraction phase is the third component, in which the left atrium acts as a contractile pump during late diastole LV filling, contributing to the increase in LV stroke volume. Silent brain infarcts caused by AF and reduced cerebral perfusion due to lowered CO are the two main possible mechanisms explaining cognitive impairment caused by LA enlargement. LA dilation is mainly observed in pressure- or volume-overload situations. The relationship between increased LV filling pressures, secondary to LV dysfunction or mitral valve disease, and increased LA size due to increased LA afterload has previously been described. The same applies to the association between LA enlargement and LA chronic volume overload observed in valvular regurgitations. LA enlargement usually involves several of the aforementioned parameters that are independently associated with cognitive impairment ( Figure 2 ). Thus, the relationship between increased LA diameter and the risk for developing AF has previously been described. LA enlargement is also a significant predictor of stroke and death and a strong predictor of cardiovascular outcome. Furthermore, LV pressures are increased in abnormal LV relaxation situations, causing an increase in LA pressure to maintain efficient LV filling. Therefore, the augmentation of atrial wall tension leads to LA dilation. Thus, increased LA volume was found to express the severity of LV diastolic dysfunction. Finally, LA enlargement has been found to be a predictor of congestive heart failure.
The main studies that analyzed the link between LA parameters and cognitive impairment are detailed in Table 2 . Oh et al . showed that LA enlargement, defined as a LA diameter on two-dimensional echocardiography > 40 mm in men and > 38 mm in women, was correlated with impaired cognitive functions and white matter lesions. In addition, Karadag et al . found that in patients with cognitive impairment, LA volume index was increased by ≥34 mL/m 2 . Van den Hurk et al . showed that a higher LA volume index was associated with a lower information-processing speed.
| Study | Cardiac imaging parameters | Population | Age (y) ∗ | Cognitive parameters | Results |
|---|---|---|---|---|---|
| Karadag et al . (2013) | LA volume index ≥ 34 mL/m 2 | 108 stroke-free healthy people without CVD | 74.8 ± 6.9 | MMSE score ≤ 25 | Larger LA associated with cognitive impairment |
| Oh et al . (2012) | LA diameter: > 40 mm (men) and > 38 mm (women) | • 54 Alzheimer’s disease • 23 Parkinson’s disease • 16 vascular dementia | 74.1 (52–96) | • White matter lesions MMSE score | LA enlargement correlated with white matter lesions |
| Van den Hurk et al . (2011) | LA volume index | 313 healthy people | 67 (59–87) | • Attention and executive functioning • Information-processing speed | Higher LA volume index associated with lower information-processing speed |
Cardiac Output and Cognitive Impairment
The exact mechanism governing the link between cognitive impairment and reduced CO remains unknown. It has previously been reported that counterregulatory mechanisms contribute to the increase of cerebral blood flow during acute periods of reduced CO. However, chronic CO reduction may decrease cerebral blood flow, overwhelming autoregulatory phenomena and leading to brain injury, particularly in elderly individuals with hypertension, in whom a shift of the cerebral autoregulation curve is observed. Interestingly, cerebral blood flow appeared to increase significantly by 53.3%, and returned to normal values after cardiac transplantation in patients with refractory congestive heart failure.
The main studies that analyzed the link between CO parameters and cognitive impairment are shown in Table 3 . Cohen et al . found that CO was correlated with overall attention and executive functioning. In the Framingham Offspring Cohort, a subclinical decrease in MRI-assessed cardiac index (CO/body surface area) was associated with impaired cognitive function and positively related to total brain volume and lateral ventricular volume on brain MRI, both markers of brain aging. Reduced CO < 4 L/min was shown to be associated with poorer executive functioning, mainly sequencing and planning. The association between decreased CO and a lower Mini-Mental State Examination score was also recently reported. Furthermore, a decrease in CO is associated with the presence of white matter lesions adjacent to the subcortical nuclei in brain MRI.
| Study | Cardiac imaging parameters | Population | Age (y) ∗ | Cognitive parameters | Results |
|---|---|---|---|---|---|
| Oh et al . (2012) | CO | • 54 Alzheimer’s disease • 23 Parkinson’s disease • 16 vascular dementia | 74.1 (52–96) | • MMSE score • White matter lesions | Lower CO associated with MMSE score |
| Jefferson et al . (2010) | MRI-assessed cardiac index (CO/body surface area) | 1,504 patients from the Framingham Heart Study Offspring Cohort free from stroke and dementia | 61 ± 9 | • Verbal memory, visuospatial memory, verbal learning, executive functioning, information processing, language, object recognition • Brain MRI | Cardiac MRI-assessed cardiac index associated with neuropsychological and total brain volume and lateral ventricular volume in MRI |
| Cohen et al . (2009) | CO | 88 geriatric outpatients with documented stable CVD and no history of neurologic disease | 70 ± 7.7 | • MMSE score, language, visuospatial abilities, learning, memory, attention, executive function. • White matter lesions | CO correlated with attention and executive functioning |
| Jefferson et al . (2007) | CO | 36 patients with CVD and free of dementia | 71.5 ± 7.5 | White matter lesions | Inverse relationship between white matter lesions adjacent to subcortical nuclei and CO |
| Jefferson et al . (2007) | CO and LVEF | 72 geriatric outpatients with treated stable CVD and no history of neurologic disease | 69 ± 7.5 | Tasks sensitive to cognitive functions mediated by frontal-subcortical systems: executive functioning, other measures including learning and memory | • Reduced CO is associated with poorer executive functioning • No association with LVEF |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
