Antihypertensive treatment and cognitive function: another tempest in a P pot?

Abstract

Dementia is a major public health concern, due to its growing prevalence in the aging population, and hypertension is recognized as a risk factor for both of the most common forms of dementia, vascular dementia and Alzheimer disease. Vascular dementia results from ischemic, hypoperfusive, or hemorrhagic brain lesions and is characterized by damage of attentive and executive function and by noncognitive features such as depression, apathy and psychosis [1]. Alzheimer-type dementia is characterized by severe memory deficit, and its clinical expression is influenced by vascular damage in the brain, even if its pathogenesis involves primarily neurodegenerative processes [2]. Hypertension accelerates the cognitive decline through the vascular consequences of the blood pressure load, including accelerated atherosclerosis, arterial stiffness, microangiopathy, and chronic subclinical ischemia of the cerebral white matter (clinically measurable as white matter alterations on neuroimaging). These processes cause basal ganglia dysfunction and cortical deafferentation. The vascular alterations also disrupt cerebral blood flow autoregulation and make the brain more susceptible to hypoperfusion during occasional hypotension that may be caused by inappropriate antihypertensive therapy. The effect of antihypertensive therapy in preventing or slowing down cognitive dysfunction has been investigated in several studies, but the results have been inconsistent [3]. The reasons of the inconsistency include the use of diagnostic tests, which are not sufficiently sensitive and specific to detect a slowly progressing clinical condition, such as the cognitive decline, in a relatively short time span, the heterogeneous impact of antihypertensive drugs on the brain and its vasculature, and the failure to recognize the impact of hypertension-induced brain damage on different cognitive domains due to the use of global cognitive function tests (as suggested by Birns et al.[4] in this journal). Another possible explanation is the nonlinear relationship of hypertension with cognitive decline [5]. Present evidence shows that hypertension in midlife, especially if not well treated, negatively affects cognition and contributes to the development of dementia later in life [6]. However, low blood pressure in the elderly is also associated with high prevalence of dementia. Some studies suggest a J-shaped or U-shaped relationship of blood pressure and cognitive performance in the elderly [5]. Therefore, still much is needed to clarify the role of antihypertensive therapy as such, and individual antihypertensive drugs, in the course of brain aging. In this issue of the journal, Hanon et al.[7] describe the effect of eprosartan-based therapy on blood pressure and cognition in a large group of hypertensive patients (>25000) older than 50 years (mean 64). Cognitive function was assessed by the Mini Mental State examination (MMSE) at baseline and after 6-month treatment. They show that eprosartan therapy is associated with a slight (less than one) but significant (P < 0.0001) improvement in the MMSE score. The extent of improvement is related with the magnitude of blood pressure lowering and is larger in the elderly. The strength of this study is in the large number of patients, as studies addressing cognitive function during the course of antihypertensive therapy in a large population are few. However, we are left with more questions than answers. The main questions are as follows: Is such a small change in MMSE score clinically significant? Is the change in the MMSE score due to blood pressure reduction or eprosartan-based therapy? The average MMSE score changed from 27.1 to 27.9 after follow-up, a statistically significant but small increase that is unlikely to have any clinical meaning. The increase in MMSE score was greater than two points in 22% of the patients. These patients were those with the worst baseline cognitive performance, and the largest on treatment blood pressure reduction, and were taking the largest number of antihypertensive drugs. The large study population provides strong support to the findings, but it has to be noted that, as the small change in global cognitive function was detectable after only 6-month treatment, it could be accounted for by regression toward the mean or getting used to MMSE and not to antihypertensive treatment. The lack of a control group makes it impossible to resolve the doubt. Therefore, we should regard these results only as an evidence of protective effect of therapy on cognitive decline rather than true improvement. We regret that MMSE was the only cognitive test, which is reasonable in a population setting but precludes any conclusion on the effects of blood pressure lowering on different cognitive domains. As for the second question, whether the effects on MMSE are specifically due to eprosartan or the blood pressure lowering per se, the observation that the greatest cognitive benefit was correlated with the baseline performance and the extent of blood pressure lowering, which was larger in the elderly who, presumably, were treated with the greatest number of antihypertensive drugs, casts doubts on the drug specificity of the effect. Again, it has to be noted that, as the elderly are susceptible to the adverse cerebral effects of blood pressure lowering due to microangiopathy and impairment of cerebral blood flow autoregulation, these findings clearly demonstrate the safety of antihypertensive treatment on cognitive function. In conclusion, an early reversal of some aspects of cerebral vascular remodeling by angiotensin receptor blockade can be foreseen from the results of Hanon et al.[7], but further investigations and a more comprehensive approach is needed to study the impact of blood pressure and antihypertensive drugs on brain anatomy and function [8]. By now, we are stuck to a P, which is conventionally significant but of little clinical significance after a 6-month follow-up. The evidence that a long-lasting treatment makes it significant also on clinical grounds is still awaited.