Blood pressure in very elderly people

Blood pressure levels in very elderly people have an apparently paradoxical inverse relationship with future morbidity and mortality. It has been long recognized that, in very elderly subjects (e.g. over 85 years), those with the highest blood pressure have the lowest mortality, and those with lowest levels have the highest death rates [1]. By contrast, in healthy young and middle-aged subjects, blood pressure has a direct and positive linear association with non-fatal and fatal cardiovascular events [2]. A further different pattern occurs in cohorts that have a higher prevalence of underlying vascular disease. In such groups, a so-called J-shaped curve of risk has been observed, with increased rates of vascular events and cardiovascular death in those with low diastolic blood pressure [3]. These different patterns are likely to be due to a number of confounding factors, including greatly increased rates of underlying serious illness in very elderly subjects with low blood pressure [4]. In this issue of the journal, this subject is further explored by van Bemmel et al. [5]. Using a longitudinal cohort that has been carefully characterized and followed up, they found increased mortality in those with blood pressures below 140/70 mmHg at 85 years of age. This increased risk of death was particularly evident in those with a history of hypertension, and was present even after correction for underlying known ischaemic vascular disease. Further correction for markers of frailty, including cognitive function and the ability to perform basic and extended activities of daily living, reduced the strength of association of low blood pressure with mortality but did not remove it. This does not necessarily mean that low blood pressure is the direct cause of increased mortality. There remains the possibility of residual confounding. For example, covert or unrecognized ischaemic heart disease is very common in older people, and is likely to lower blood pressure and associate with increased risk of death. Other unrecognized serious illness may also be present, causing an increase in mortality in the short-medium term [4]. Nonetheless, the possibility remains that low blood pressure is intrinsically harmful for very elderly people. So why should elderly people tolerate low blood pressure differently from young or middle-aged subjects? The most plausible reason is the dramatic increase in underlying ischaemic vascular disease with ageing. Clinically significant coronary heart disease is almost the norm in elderly patients in the developed world [6]. Often, it is unrecognized because there may be no symptoms of angina pectoris, and no history of acute myocardial infarction. Such patients are highly dependent on perfusion pressure to maintain adequate coronary flow, and low diastolic blood pressure might be expected to have serious deleterious effects compared to the healthy heart [7]. There are similar plausible reasons for low blood pressure being harmful for the elderly brain. A degree of small vessel cerebral ischaemic damage in the deep subcortex is present in most subjects over the age of 75 years [8]. This is often visualized on head computed tomography as patchy low attenuation around the lateral ventricles, or high density changes on T2-weighted magnetic resonance imaging. These changes are associated with a narrowing of the penetrating arterioles that supply blood to this region, and an associated reduction in perfusion with low-grade ischaemic damage [9]. Blood supply through these narrowed arterioles is highly dependent on perfusion pressure, and therefore low blood pressure is likely to cause accelerated brain damage. Small vessel ischaemia is an important contributor to cognitive decline in old age, often in combination with Alzheimer's pathology [8]. Low blood pressure and postural hypotension are also risk factors for falls in older people [10], presumably due to an associated reduction in cerebrovascular reserve. Falls and resulting injuries are an important cause of death in the very old. Low blood pressure is therefore a potentially useful risk marker for mortality in very elderly subjects, whether or not it is directly causal. This high-risk subgroup is likely to benefit from careful clinical assessment, including the presence of underlying cardiac disease and other serious co-morbidity, and consideration given to reducing or stopping any anti-hypertensive drugs. However, none of these data provide a robust argument against the use of anti-hypertensive drug treatment for elderly people with a persistently raised blood pressure. There is convincing evidence of benefit from randomized controlled trials, with reductions of stroke risk, death from myocardial infarction and heart failure in older subjects, at least to 80 years of age [11]. Over the age of 80 years, there are more limited data, but there is still evidence for reduced morbidity, including stroke [12]. There is also some evidence for prevention of cognitive decline in older people from the use of a calcium channel-blocker for systolic hypertension [13], and by prevention of recurrent stroke by blood pressure lowering with a thiazide diuretic plus angiotensin-converting enzyme inhibitor [14]. Given all these likely benefits, we should not reduce our efforts to effectively treat elevated blood pressure in older patients, including the very elderly.