Association between central blood pressure, arterial stiffness, and mild cognitive impairment

Hypertension is the most important modifiable risk factor for the development and progression of cognitive decline. Cognitive impairment can range from simple memory loss to deficits in executive function, manifesting as difficulties with planning and organizing tasks, managing time, self-care, and decreased processing speed [18]. In the Canadian Study of Health and Aging (CSHA), executive function was the earliest-affected cognitive domain and preceded the development of dementia [19]. CBP may be more relevant to the study of cognitive impairment than PBP since the blood is delivered to the brain through the large central arteries. It has been shown that in hypertensive subjects, target organ damage is mediated not only by steady pressure, but also by pulsatile hemodynamics and blood flow [20]. A systematic review showed arterial stiffness is associated with cerebral small vessel disease and decreased cognitive function [21]. In this study, we analyzed the associations between central and peripheral blood pressure indices and both global and domain-specific cognitive function, adjusted for age and sex. Standard measures of brachial and central BP were not found to be associated with global cognitive decline or with individual domains of cognition. In this study, we did not further adjust for other co-morbidities such as hypertension or diabetes status because these conditions are causes of arterial stiffness and mediate the co-linear relationship between blood pressure, arterial stiffness and outcomes.

The clinical evaluation of arterial stiffness and its pulsatile hemodynamics is complex. A few studies have shown a relationship between CBP parameters such as PP and PWV and indices of cognitive function [22–26]. Among the CBP indices we evaluated in this study, AI and PPA were found to be significantly associated with deficits in certain cognitive domains. A lower PPA, reflecting accelerated arterial aging, was associated with abnormal language function, and a higher AI was associated with both abnormal language and executive function. Different CBP indices like cfPWV, PPA, and AI may be measuring different aspects of arterial mechanics, and there is some evidence that AI may be a better marker for cardiovascular outcomes. For instance, AI has previously been shown to distinguish the effects of different BP medications, when brachial blood pressure and cfPWV were not able to [27]. Although Mitchell et al. [28] did not find a relationship between AI and cognitive function, their study only evaluated adults aged 69 and older, whereas our study included adults over the age of 50. With progressive aging, central impedance increases faster than peripheral impedance and AI can consequently become dissociated from cfPWV and other markers of arterial stiffness [28]. In addition, with older adults marked stiffening of the aorta may reduce wave reflection at the interface between aorta and proximal large arteries, which result in reduced local and global wave reflections which may contribute to the dissociation between measures of aortic stiffness and AI [28].

Tsao et al. in their study showed similar results to our study in that cfPWV was not associated with cognitive decline [29]. This lack of association may be attributable to differences in cognitive tests (MoCA was used in this study, whereas MMSE was used in other studies). Small sample size may have reduced the effect size and our sensitivity to detect association between all blood pressure indices and cognitive outcomes. Finally, carotid-femoral pulse wave velocity (cfPWV) seems to mainly measure arterial stiffness, whereas pulse pressure amplification (PPA) and Augmentation Index (AI) mainly measures the hemodynamic wave reflections from the arterial wall. So these BP indices may be measuring different aspects of the arterial wall mechanics as well as arterial flow dynamics and hence the difference seen in this study.

A strength of this study is its use of the MoCA for cognitive assessment, which is a more sensitive test for assessing mild cognitive impairment than the Mini-Mental State Examination (MMSE), and which also incorporates an assessment of executive function. We also employed the standardized Consortium Criteria for the diagnosis of MCI. To our knowledge, this is the first study to show an association between high AI and abnormal language and executive function.

Limitations of this study include its cross-sectional nature, which shows only association and not causation, and the small sample size. We did not calculate the sample size. However, we observed significant relationships between two independent variables (augmentation index and pulse pressure amplification) and two aspects of cognition (clock-drawing and language). Although there was a relationship between these variables and the primary end-point of the study (presence or absence of mild cognitive impairment), it did not reach statistical significance. In addition, as expected augmentation index (mean?±?SD) was higher in those with MCI (MCI present 26.27?±?13.81vs MCI absent 23.40?±?12.28), and pulse pressure amplification was to be lower in those with MCI (MCI present 1.34?±?0.13 vs MCI absent 1.37?±?0.14) indicating that our data fit the experimental model.