Antihypertensives for all?

A new study has confirmed what we already know – lowering blood pressure reduces the risk of death, with the greatest benefit seen in those at higher risk to begin with. Does this mean that everyone should be taking antihypertensives?

It is no secret that reducing risk factors for life-threatening diseases can reduce mortality. This is at the heart of modern medical thinking and public health initiatives, and is behind the substantial fall in cardiovascular mortality in recent years.

Cardiovascular disease is still one of the major causes of death in Western countries, and its main risk factors are smoking, diabetes (often associated with obesity), high cholesterol and high blood pressure. A generation ago, each risk factor was managed in isolation, with people who were above a threshold value for cholesterol or blood pressure receiving treatment for that specific risk factor. Today, the focus is on assessing an individual’s overall risk, and applying management strategies that are most likely to reduce disease and mortality. It was seeing the failings of the previous approach and wanting to evaluate the overall risk that motivated us at Crystallise to start mortality modelling in the 1990s, and you can explore the impact of different risk factors and their modification in our free Sonata Vivo mortality model.

It is therefore no surprise that Dena Ettehad and colleagues(external link) have found that reducing blood pressure can reduce overall mortality and deaths from various cardiovascular diseases, and that the benefit is greater the more the blood pressure was reduced.

The researchers combined data from 613,815 participants in 123 clinical studies of blood pressure lowering in a high quality systematic review and meta-analysis. They found that every 10 mm Hg reduction in systolic blood pressure significantly reduced the risk of all-cause mortality by 13%, with a relative risk of 0·87 (95% confidence interval 0·84 to 0·91). This was driven by a significant reduction in major cardiovascular disease events (relative risk 0·80, 95% CI 0·77 to 0·83), coronary heart disease (0·83, 0·78 to 0·88), stroke (0·73, 0·68 to 0·77), and heart failure (0·72, 0·67 to 0·78).

Our Sonata Vivo mortality model shows this benefit graphically. The chart below shows that a 50 year old man with average values of cholesterol and body mass index, who is not diabetic and a non-smoker, would gain an extra 0.5 years in life expectancy if his systolic blood pressure falls from 130 mmHg to 120 mmHg.

Life expectancy increase from a 10mmHg reduction in systolic blood pressure in a normotensive 50 year old man

How much we would value an additional half a year of life expectancy varies depending on the context. For many of the new cancer treatments, an additional 6 months of life would be seen as a very good outcome, and might justify the outlay of £30,000 or more per patient. However, for a healthy 50 year old man, the benefits may be less clear, as shown in the curve of deaths for all-cause mortality, below. Here, the two lines for before blood pressure reduction (blue) and with a 10 mmHg lower blood pressure (pink) are very close together.

Although all-cause mortality is decreased slightly, the chart below shows that it is the reduction in cardiovascular events that is the main driver of this improved survival.

Curve of deaths for all-cause mortality from 10 mmHg reduction in systolic blood pressure in a normotensive 50 year old man

Curve of deaths for cardiovascular mortality from 10 mmHg reduction in systolic blood pressure in a normotensive 50 year old man

So, should we all be taking antihypertensives to reduce our blood pressure? The studies were conducted largely in people with high blood pressure to begin with, so the likely benefits in those of us with lower blood pressure have to be inferred. But we know from studies of statins that people with normal cholesterol levels have a similar relative reduction in mortality from further lowering of their cholesterol to those with high baseline levels, and there is no reason to think that this would be any different with blood pressure. As with cholesterol, though, although the relative benefit may be similar, the absolute reduction in deaths becomes smaller at lower baseline risk, and there is a point where the difference becomes not clinically meaningful.

There are some important differences between population-level reduction of cholesterol with statins and lowering of blood pressure with antihypertensives. People with impaired circulation, such as older people with narrowed arteries, may get critically low blood flow to vital organs if their blood pressure drops too far, which could cause heart attacks or ischaemic strokes. Even healthy adults can get dizzy when standing up suddenly if their blood pressure is low. Ettehad’s analysis found a J shaped curve, with the lowest risk overall at a systolic blood pressure of 115 mmHg. However, the point of greatest benefit varies by age, and older people may have optimal outcomes at a higher systolic blood pressure. No such lower limit, below which the risks of mortality start to increase, has been accepted for cholesterol.

If the overall decision is that an individual may benefit from reducing their blood pressure, how is this best to be achieved? Cholesterol lowering is effectively accomplished by use of statins, a single drug class, but lowering blood pressure can involve a combination of diuretics, calcium channel blockers, beta blockers, and ACE inhibitors. The study by Ettehad and colleagues found that different types of antihypertensives had different benefits in the type of disease they prevented, with heart failure being best prevented by diuretics, and strokes by calcium channel blockers. So, the decision about which antihypertensive to take is complex, and depends on an individual’s risk of developing each type of disease, and their tolerance of potential side effects of treatment.

If we just focus on cardiovascular disease prevention, as shown in the chart above, the benefits of enthusiastic blood pressure lowering seem obvious. But, as the all-cause mortality curve of deaths show, we all die of something eventually – the lifetime risk of death is still 100%. So, if we are preventing cardiovascular deaths, there must be an increase in deaths from other causes. The chart below shows how reducing blood pressure increases the risk of dying of cancer by 2%, from 32.6% with a blood pressure of 130 mmHg, to 34.6% with a blood pressure of 120 mmHg – the Sonata Vivo model separates out lung cancer from other cancers, but the pattern is the same for both.

Curve of deaths for cancer mortality other than lung cancer from 10 mmHg reduction in systolic blood pressure in a normotensive 50 year old man

So, you pays your money and you takes your choice about how you’re most likely to die. And, of course, the decision about implementing a blood pressure-lowering policy across the healthy population should take into account the economic burden of such a policy. Even generic antihypertensives become expensive at a national level if half the adult population is taking them. Although healthy 50 year old men usually pay for their prescriptions, which could offset much of the medication costs, the additional resources needed by already stretched primary care services to see patients and review their prescriptions every 6 months, and to manage their dizzy spells and other side effects, would be substantial. As individuals, we may decide that taking one or more tablet a day is worth it, but this approach may not be the best use of limited NHS resources.

And, in the meantime, we can all do our best to reduce our other risk factors, by not smoking, taking regular moderate exercise, drinking sensibly and keeping our body weight in normal limits. Let’s start after Christmas and the New Year…

Originally published April 2016

« Back