Intended for healthcare professionals


Managing blood pressure in older adults

BMJ 2018; 362 doi: (Published 06 July 2018) Cite this as: BMJ 2018;362:k2912
  1. Stephen Makin, clinical lecturer in geriatric medicine,
  2. David J Stott, David Cargill professor of geriatric medicine
  1. Academic Section of Geriatric Medicine, Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK
  1. Correspondence to: D J Stott david.j.stott{at}

Age alone is no barrier to treatment

In a recent paper in The BMJ, Liv and colleagues reported the results of a large cohort study investigating the link between blood pressure and mortality in community dwelling Chinese people with a mean age of 92 years.1 Studies of very elderly people are challenging and rarely performed so these data are of particular interest. Both high and low systolic blood pressure were linked to an increase in mortality (a “U shaped curve” relation). Systolic blood pressure >154 mm Hg was associated with increased cardiovascular mortality, and <107 mm Hg with non-cardiovascular death.

The finding that low systolic blood pressure predicts increased risk of death is consistent with many other epidemiological studies in older people. Although low blood pressure may in itself cause harm, it is also likely to be a marker of ill health, with systolic blood pressure falling for up to two years before death.2 Any link between high systolic pressure and mortality has been much less consistent in later life. In very elderly people (≥85 years) an inverse linear relation has been reported, with lowest mortality at the highest pressures,34 but health status may play a major part in driving these associations.4 This is in contrast to epidemiological data for younger or middle aged people, in whom high pressure is clearly and consistently associated with increased risk of fatal and non-fatal cardiovascular events.5

How should these data influence decisions on use of antihypertensive drugs in very elderly people? Here we must be mindful of the fact that observational data have serious limitations. Even the most rigorously adjusted analyses cannot rule out residual confounding and reverse causality. Randomised controlled trials remain the gold standard for informing treatment decisions, and a substantial body of such evidence exists to guide treatment of hypertension in older adults.

Evidence says aim low

In a Cochrane review and meta-analysis of randomised trials of treatment of hypertension in people aged over 60, antihypertensive drugs reduced cardiovascular morbidity and both total and cardiovascular mortality.6 People aged over 80 have also been shown to benefit from drug treatment of high blood pressure, although there are fewer data in this age group. In the hypertension in the very elderly trial (HYVET), people with sustained systolic blood pressure >160 mm Hg were randomised to antihypertensive drugs or placebo. Treatment reduced systolic blood pressure from a baseline 173 mm Hg to 143 mm Hg (compared with 158 mm Hg on placebo), leading to a decrease in stroke, heart failure, and total mortality.7 However, for selected patients, we should probably be aiming lower.

The Systolic Blood Pressure Intervention Trial (SPRINT) found benefits from targeting a systolic blood pressure <120 mm Hg in participants with mean age of 68 years and systolic >130 mm Hg8; in a prespecified subgroup of people aged over 75 intensive antihypertensive treatment achieved a mean systolic of 123 mm Hg and led to a substantial reduction in cardiovascular events and total mortality compared with standard care with mean systolic pressure of 135 mm Hg.9 In common with most randomised trials, SPRINT participants were generally fitter with less multimorbidity and polypharmacy than patients in a clinical setting. There was a penalty associated with intensive treatment—an increase in adverse events including hypotension, syncope, acute kidney injury, and electrolyte disturbance, seen across all trial participants.8

Consider overall health

For selected older people, reduction of blood pressure aiming to prevent cardiovascular disease and prolong life will be a high priority—and there is potential for further gain from setting a low treatment target. The most recent 2017 American College of Cardiology/American Heart Association Task Force guidelines (including the evidence from SPRINT)10 recommend a target systolic blood pressure of <130 mm Hg for ambulatory community dwelling people older than 65.

However for some, such as people who are very frail or have complex comorbidity or limited life expectancy, antihypertensive drugs are likely to be irrelevant or harmful. The above guidelines recommend a more guarded approach for people with a high burden of comorbidity and limited life expectancy; clinical judgment, patient preference, and a team based approach to assess risks and benefits should be used in decisions regarding intensity of treatment and choice of antihypertensive drugs.

What should be done for older people with low systolic blood pressure? Evidence is limited, but it seems reasonable to avoid prescribing medicines likely to further reduce blood pressure. For those taking antihypertensive drugs, the risks and benefits associated with reducing or stopping treatment are also unclear.11 However, it is sensible to reduce or stop antihypertensives in patients with adverse effects such as syncope or acute kidney injury associated with low pressure.

Shared decision making is particularly important when considering preventive treatments for very elderly people. The priorities for older people with hypertension (and their carers) can vary greatly. For some, the option of taking antihypertensive drugs for longer term gain will be attractive. Others look for faster and more noticeable benefit from any medicine taken, with clear improvement of symptoms and enhanced quality of life. The clinician’s role is to support patients navigating this decision, in line with their individual preferences.


  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription