Can blood pressure-lowering therapy reduce the risk of cognitive decline in the elderly?

Cognitive decline and dementia are inextricably associated with ageing and are increasingly contributing to the global burden of disease, as our populations live longer, right across the world. The most common forms of dementia are Alzheimer's disease, vascular dementia and mixed forms, and unfortunately, there are no proven treatments at present, once these are established. Our hopes for prevention are only marginally stronger and depend on treatment or avoidance of risk factors such as hypertension, diabetes, high total cholesterol and smoking. Indeed, the links between high blood pressure and cognition provide some cause for optimism, but at this time, there is no clear evidence that lowering the blood pressure will reduce the risk of dementia or cognitive decline. The optimism is mainly based on observational studies demonstrating a link between raised blood pressure or hypertension and the risk of dementia or cognitive decline in later life [1–10]. One of the main sources of evidence has come from ‘The Honolulu-Asia Aging Study’, which first demonstrated that higher levels of SBP (but not diastolic) in Japanese-American men living in Hawaii, and aged in their late forties or fifties, were associated with reduced cognitive function in later life, as observed in their late seventies [1,2]. They also demonstrated that the risks of both Alzheimer's disease and vascular dementia were much greater in men without antihypertensive therapy than those receiving treatment [1,2]. Very similar associations have been reported in studies from Finland, France, the Netherlands and North America in predominantly white population [3–6]. A recent report from the ARIC study (Atherosclerosis Risk in Communities) has taken this further by demonstrating that the relationship between dementia and risk factors such as hypertension is much stronger in men and women in middle life (<55 years) than those in later life (≥70 years) [7]. Similar findings were reported for diabetes, smoking and hypercholesterolemia [7]. These findings help to explain the lack of convincing evidence from randomized clinical trials that blood pressure lowering will prevent or reduce the risk of cognitive decline and dementia [11–20]. The main difficulty is that as most clinical trials span a maximum of 5 years, it is very difficult for them to bridge the time from maximum influence of risk factors (midlife) to the time of maximum incidence of cognitive decline or dementia (later life). To be successful, trials should either span a much longer period, which poses both logistic and financial problems, or finely balanced in the middle, with little guarantee of success. The main positive results have come from the Systolic Hypertension in Europe (Syst-Eur) trial and the Perindopril Protection against Recurrent Stroke Study (PROGRESS) [11–14]: In Syst-Eur, treatment with a regimen based on nitrendipine as well as enalapril and hydrochlorothiazide reduced the incidence of dementia by 50%, with only 11 cases versus 21 cases in the placebo group [11–13]. These benefits were maintained during posttrial follow-up of a further 2 years, with 21 cases versus 43 in the placebo group [13]. However, these numbers are very small and clearly in need of confirmation. In PROGRESS, a regimen based on perindopril as well as indapamide clearly reduced the risk of dementia and cognitive decline, associated with recurrent stroke, but did not do so in the absence of recurrent stroke [14]. A number of other large trials have failed to find significant evidence that blood pressure lowering per se will reduce the risk of cognitive decline in dementia. These include the Systolic Hypertension in the Elderly Programme (SHEP) trial, the Study on Cognition and Prognosis in the Elderly (SCOPE) trial, the Hypertension in the Very Elderly Trial Cognitive function assessment (HYVET-COG) trial, the Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS) trial, the On-going Telemisartan Alone and in Combination with Ramipril Global Endpoint (ON-TARGET) trial, the Telemisartan Randomised Assessment Study in ACE Intolerant Subjects with Cardiovascular Disease (TRANSCEND) trial and the Secondary Prevention of Small Subcortical Strokes (SPS3) trial [15–20]. A number of authors have also performed systematic reviews and meta-analysis of various groups of these studies, and these have generally shown little effect of blood pressure-lowering treatment or nonsignificant reductions of the order of 5 or 10% in the relative risk of cognitive decline or dementia [17,19,21,22]. In this issue of the Journal of Hypertension, Ruth Peters et al. [23] present observational evidence from the Newcastle 85+ Study. This addresses an important issue, whether the choice of blood pressure-lowering drug class can influence the rate of cognitive decline in very old people, aged 85 years. The authors perform an analysis of 238 white individuals, constituting an established population-based cohort of people born in 1921, living in the Newcastle/Tyneside area of the UK, including some living in institutions. This cohort was limited to people with a diagnosis of hypertension who were taking blood pressure medications and who had completed a ‘standardised Mini-Mental State Exam’ (SMMSE) at both baseline and after 3 years of follow-up [23]. Following extensive multivariable adjustment, the authors report that the use of calcium channel blockers (CCBs) was associated with a slower rate of decline of the SMMSE over 3 years, than the use of the other four classes analysed angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), beta-blockers and thiazide-like diuretics [23]. The results were similar when all individuals were included and when only those with an SMMSE over 24 at baseline were included in the analysis. There was no formal control group, not taking any blood pressure-lowering medications, but a number of sensitivity analyses were reported, with comparisons against patients without hypertension and patients not receiving blood pressure-lowering treatment, and these are reported as being confirmatory [23]. The study has many positive features. It was a longitudinal study conducted in a stable population at community level, with meticulous follow-up and monitoring, and is rigorously analysed. However, it has some limitations. The first and most obvious is the very small sample size with only 238 individuals, and these split into five subgroups, in a field wherein large studies with many thousands of individuals have failed to find convincing evidence that blood pressure lowering per se will prevent or reduce the onset of cognitive decline or dementia. The second is that most individuals were taking more than one class of blood pressure-lowering drug, so that the comparisons are confounded by the use of combination therapy. Another is the inevitable confounding that is inherent in an observational study, and particularly the lack of a formal, matched control group not taking any blood pressure-lowering therapy. In assessing the weight that should be given to the present report, it is difficult to go past the conclusions of the systematic review by Ruth Peters herself published in this Journal a year ago, viz. ‘At present, there is no clear convincing evidence to suggest that CCB use increases or decreases the risk of cognitive decline or dementia in the very elderly. A robust clinical trial is now required to resolve this’ [24]. That remains the case today. Furthermore, it seems rather premature to be comparing the efficacy of various classes of blood pressure-lowering drugs before we have definitive evidence that the treatment of hypertension does indeed prevent, reduce or retard the onset of cognitive decline and dementia. That being said, there are now a few straws in the wind, each with its own limitations and each far from definitive. The results of the present study fit well with the positive effects obtained with CCB in the Syst-Eur trial [11–13], while the results from the PROGRESS trial support the use of ACEI along with diuretics in patients with recurrent stroke [14]. On the other hand, the results from the Honolulu-Asia Aging Study support the use of beta-blockers in elderly Japanese-Americans in Hawaii [25]. The problem of course is that like everyone else working in this field, I have an opinion and a wish. The opinion is one that many of us share that lowering blood pressure will eventually be shown to reduce the incidence of cognitive decline and dementia, and the wish is that robust clinical trials should be mounted to produce hard evidence that this is so. Unfortunately, there are considerable hurdles facing us all in providing this evidence. Partly because it is daunting to fund a study that reconciles the need to start treating those with raised blood pressure in middle age (or at least the young elderly, say between 60 and 70 years), and the fact that the incidence of significant cognitive decline and dementia is much greater in older adults, say those over 75 years. Another is that without a good new drug as a candidate, industry sponsorship is very unlikely and funding must rely on the public purse, and at present, the evidence to support the major outlay of funds that would be needed is likely to be insufficient. This suggests that we need to mount some smaller ‘proof-of-concept’ randomized trials as a first step, before we can hope to conduct large-scale, definitive trials. The sting in the tail of course is that convincing evidence that treating hypertension in middle age prevented dementia in old age would pose a major challenge for health economists and even bigger problems for government treasuries and health systems. We can only hope that they have to meet this challenge sooner, rather than later! ACKNOWLEDGEMENTS Conflicts of interest J.C. has received research grants, administered through the University of Sydney, as Principal Investigator for the PROGRESS and ADVANCE trials, and also received honoraria and travel support for speaking about these studies at scientific meetings.