Blood Pressure Part Two: Lower is Better19 minute read

Steven Kornweiss, MD cardiovascular Leave a Comment

Summary First

  • Guidelines tell us that relatively young and healthy people with mildly elevated blood pressure only need to try exercise, a DASH diet, and decreased salt intake in order to decrease blood pressure. If those don’t work, keep trying until you become “high-risk.”
  • People in this category can live for 30-40 years with mildly elevated blood pressure or stage one hypertension without treatment.
  • There is strong evidence that even mildly elevated blood pressure is associated with an increased risk for just about every chronic disease, but most of all for heart attack, stroke, and death.
  • There is also strong evidence that mildly elevated blood pressure plays a causal role in accelerating vascular diseases, and that treating this blood pressure aggressively is a good idea.
  • In this piece, I review the guidelines, define the problem with specific patient examples, and review the evidence that supports the above points.

The Problem

The 2017 American College of Cardiology Guidelines for the management of blood pressure is 122 pages long. It describes the management of blood pressure for individuals in many scenarios. However, there is one important scenario that remains unaddressed. It accounts for millions of people, myself included. The problem is that people with "normal" or merely “elevated" blood pressure, while at higher risk for cardiovascular disease, stroke, and death than their counterparts with lower blood pressure, are not managed with lifetime prevention and longevity in mind.

The discussion of this scenario accounts for the majority of this article, but before I get into details, I want to quickly review the less controversial aspects of blood pressure management. I think this review is necessary for the rest of this discussion to make sense.

First, if you haven’t already done it, I recommend reading part one of this series on blood pressure in which I discussed the following points and questions:

In case you haven’t read part one, or even if you have, it might be worth another look at this chart to familiarize yourself with the categories of blood pressure. In the literature, you’ll find different terminology for blood pressure ranges, but this chart from the 2017 ACC Guidelines is probably the simplest, most up-to-date, and most widely used, so I’ll use these categories.

There is one study that I refer to later that uses a different classification, but I’ll draw your attention to that when we get to it.

Now I’ll dive into the details, starting with some uncontroversial points.

The Uncontroversial

Guidelines clearly indicate that we should initiate blood pressure lowering medications when systolic blood pressure surpasses 140 mm Hg or diastolic blood pressures exceeds 80 mm Hg, whether or not other risk factors for cardiovascular disease or stroke are present. In other words, if your blood pressure (either top or bottom number) is greater than 140/80 mm Hg, the guidelines recommend lifestyle interventions and medications to lower blood pressure.

A lower threshold for pharmacologic treatment exists for individuals at "high-risk" of having heart attacks and strokes in the next ten-years. In these individuals, the threshold is drawn at 130 mm Hg systolic and 80 mm Hg diastolic.

In order to be considered "high-risk," you’ve got to have some combination of the following:

  • Age over 40
  • dyslipidemia
  • diabetes
  • history of smoking
  • treatment with medications (i.e. anti-hypertensives, statins, or Aspirin)

Sex and race also play a role, with men being considered at higher risk than women, and with certain races being considered higher risk than others.

Risk calculators like the ACC Atherosclerotic Cardiovascular Disease (ASCVD) Risk estimator are commonly used to appraise ten-year risk for ASCVD.

It’s important also to know that risk calculators only apply to individuals who are greater than 40 years old because:

  1. There is almost no randomized clinical trial data looking at cardiovascular risk in people under the age of 30. This is because so few people have heart attacks at that age, and it’s impractical to study people for two or three decades.
  2. The risk-calculators generate a 10-year risk prediction score. So, even if we could extrapolate from data obtained in older individuals and apply it to younger people, average 10-year risk for anyone in their 30s would be near 0 regardless of other factors.

People in their 30s and early 40s almost never have heart attacks no matter how high their blood pressure or how poor their health.

So, an individual in his or her thirties, forties, or fifties who has few risk factors for heart attack or stroke, but whose blood pressure is between 120-139 mm Hg systolic and 80-89 mm Hg diastolic, will not be prescribed medications to lower blood pressure.

By the guidelines, non-pharmacologic “lifestyle" interventions should be prescribed to these individuals, but if those don’t work, then what? Guidelines are silent on this question.

This next part is where the controversy begins.

The Controversial

If you’re one of these people with a merely “elevated” blood pressure, or if you have stage one hypertension, but you’re low-risk for heart attacks and strokes in the next ten years, there is no treatment for you beyond “lifestyle" or "non-pharmacologic" interventions. If these don’t work, guidelines suggest monitoring the situation, but for what? I suppose we’re monitoring and waiting for blood pressure to climb higher, or for some other factor to make a person fall into the high-risk category. Until that time, no treatment will be initiated.

If you’re otherwise healthy, you might not be "high-risk" until your 65th birthday if you’re a man, or your 71st if you’re a woman.

Look at this sample risk-calculation I created to illustrate this point:

Based on the ACC ASCVD (atherosclerotic cardiovascular disease) risk estimator, this 64 year-old white man, never-smoker, without diabetes or dyslipidemia, but who has a blood pressure of 139/89, has a 10-year heart-attack-risk of 9.4%. This falls below the ACC threshold whereby anti-hypertensive medication is recommended for primary prevention of atherosclerotic cardiovascular disease. In other words, you’ve got to wait until you have a one-in-ten chance of having a heart attack in the next decade until a doctor who follows guidelines will address risk contributed by “elevated” blood pressure or stage one hypertension.

This 64 year old man might have had elevated blood pressures for 30 or 40 years before someone treats him for it. It won’t be until his 65th birthday that the doctor will be permitted by guidelines to do something about it other than continuously recommend non-pharmacologic interventions. Of course, doctors don’t have to follow guidelines, but most tend to. It’s the easiest way to administer rapid, acceptable, standard care that you won’t be questioned for if anything goes wrong. Most doctors also don’t have the luxury of dissecting the literature and reviewing risks and benefits of treatment with their patients for twenty or thirty minutes. There just isn’t time in a busy schedule. Guidelines are written by subject matter experts, and so bedside clinicians do well to follow them.

If the damage caused by elevated blood pressure is cumulative, which we have reason to believe it is, then a guideline based approach loses out on maybe 30 or 40 years during which an individual could have enjoyed blood pressures of 115/70 instead of 135/89.

This is the controversial situation that remains to be addressed. It’s the issue that every physician considers when they’re seeing a 40 year old with a blood pressure of 135/85.

I don’t know anyone who thinks it’s better to have a blood pressure of 135/85 than it would be to have 115/70. But, the guidelines don’t support treating this blood pressure in "low-risk" patients. Insurance companies won’t pay for it. And patients don’t expect it or ask for it because it’s out of the norm.

Next, I’ll discuss a few other scenarios in which a patient is unlikely to receive pharmacologic blood pressure lowering treatments.

A Few Scenarios

To demonstrate how bad things can become before you’ll be offered medications, I went to the risk calculator and created the following scenarios:

Scenario 1

57-year-old white female with a blood pressure of 139/89 (the top of the stage 1 hypertension range), with a total cholesterol of 250, HDL-C of 35, LDL-C of 99, a history of diabetes, never-smoker, who is already on a statin and Aspirin. Her 10-year ASCVD risk is calculated at 9.6%, and thus she would not be offered pharmacologic therapy for her hypertension if her physician follows the guidelines.

Scenario 2

A 42-year-old “African-American” man with a blood pressure of 139/89, Total cholesterol of 299, HDL-C of 35, LDL-C of 189, diabetes, never-smoker, who is on a statin and Aspirin. His 10-year ASCVD risk is calculated at 9.4%, so he also will not be offered anti-hypertensive medications if his doctor follows the guidelines.

Scenario 3

A 36-year-old man with stage one hypertension and every single risk factor for cardiovascular disease except for diabetes and chronic kidney disease will not be treated because 10-year ASCVD risk cannot be calculated for individuals under the age of 40. Therefore, nobody under the age of 40 is likely to be treated with anti-hypertensives even if they fall at the top end of the stage one hypertension range. There are some exceptions, but for generally healthy non-diabetic patients, this is true almost across the board.

I can tell you that even patients under forty with stage two hypertension are unlikely to be treated. At least in my clinical experience, many of these patients will be written-off as having white-coat hypertension. They might be told to monitor at home, and understandably, few of them will do this. Almost nobody under age 40 will be given blood pressure medications at any blood pressure unless they are markedly hypertensive, or they themselves, their family, and/or their physician is particularly attentive.

Scenario 4

An otherwise completely healthy 64-year-old man with no history of diabetes, who has never smoked, has never been on medications, and who has a normal lipid panel, will not be offered pharmacologic treatment. He will need to reach the age of 65 before his risk will be over 10%. This is true for both white and “African-American” men. The equivalent female counterpart will have to reach the age of 71 before her risk is calculated at over 10%. This is because women are generally at lower risk for ASCVD than men.

Again, I can tell you from clinical experience, that even then, most physicians may not treat these patients right away. Physicians are not used to seeing many elderly patients who are in good health. They’re used to treating chronically ill people. So, if you are a very healthy and functional 64 year old man or 70 year old woman, and you walk into your physician’s office every year with blood pressures of 135/85, he or she is likely to be ecstatic with your blood pressure simply because it’s not 180/110, which is what the patient before and after you came in with.

I want to acknowledge that the guideline writers have a big job, and the primary care physicians have an impossible job.

A lot rides on the guidelines, and the writers must be judicious. They have to use the best and most rigorous evidence. They can’t change things too much or too quickly. Physicians are expected to abide by guidelines, and insurance companies, the government, and most of all, patients end up paying for care recommended by the guidelines. Guideline writers are criticized enough as it is, so I’m certainly not trying to add to that criticism, nor am I bringing into question the expertise of the guideline writers. If the guidelines recommended even more aggressive treatment than they already do, the writers would likely receive endless floggings from the medical community. More aggressive guidelines would make the doctor’s job more challenging, would expand treatment, increase costs, and set off lots of debates. So, the current recommendations are understandable. The same goes for primary care physicians; if they go off-book and something bad happens, they could be in trouble. They’re in a tight spot.

But, the issue remains that there is a potential opportunity here that is not addressed, and marginal gains like this are what we’re looking for as people interested in improving our healthspan and lifespan.

Next, I’m going to begin to make the case for being more aggressive with the early treatment of “elevated” blood pressure and stage one hypertension in individuals who are otherwise considered to be “low-risk” for ASCVD by current ten-year risk calculators.

The Question

Let’s restate the question we’re trying to answer:

Should a person with "normal" blood pressure, "elevated" blood pressure, or "stage 1 hypertension," who is not considered at "high-risk" for cardiovascular disease, consider lowering their blood pressure in order to delay the onset of heart attacks and strokes.

I’ll reveal at the outset that there is not an obvious answer.

To try to answer it, we have to look at the totality of knowledge on blood pressure. We’ll then have to attempt to reason our way into the answer, and then carefully apply the answer we obtain on a case by case basis.

I’ll also tell you up front that I think the answer is yes. I believe there is strong evidence that in many cases, an individual who is interested in approaching blood pressure with longevity in mind should consider lowering blood pressure as much as possible and as early as possible, without causing themselves harm in another way.

I want to take you through the evidence and the reasoning that led me to this conclusion. I’ll start by sharing the four postulates that have to be true in order for my conclusion to be correct.

The 4 Postulates

What would we have to know about blood pressure in order for our hope to be true, that by lowering blood pressure into the normal or sub-normal range early in life, we can delay or prevent heart attacks and strokes later in life:

  1. Blood pressure is a direct contributor towards the development of cardiovascular and cerebrovascular disease.
  2. The negative effects of marginally elevated blood pressures start early in life (perhaps in someone’s third or fourth decade).
  3. By lowering blood pressures to "normal," "optimal," or "sub-normal," we will delay the onset of cardiovascular and cerebrovascular disease as compared to blood pressures in the normal or elevated range.
  4. By doing so, we don’t cause some other harm that outweighs the good we’ve done.

In this article, we’ll examine the ideas and evidence behind the first postulate.

1. Blood pressure is a direct contributor towards the development of cardiovascular and cerebrovascular disease.

There are multiple levels of evidence for this.

  • Mechanistic
  • Clinical Experience
  • Epidemiologic / Observațional Data

Mechanistic Evidence

Blood pressure is the pressure of the blood against the wall of the arteries as it flows from the heart to the organs. Like any fluid that flows through a conduit for a long period of time, blood will eventually cause erosion to the inner surface of the arteries, and may leave behind deposits in damaged or eroded areas. This is true in the plumbing of a house, the hoses of a car’s power-train, and equally true within the blood vessels of the human body. The higher the pressure, the more likely it is that the vessel will suffer wear and tear.

Further complicating the situation within our blood vessels is that blood does not flow continuously, it flows in a pulsatile manner. To accommodate a wide range of pressures, our arteries are elastic. That is, our arteries expand to accommodate increased pressure during a pulsation, but then rebound to a smaller size in between pulses. As the pressure of the pulse increases, it becomes more likely that the elastic tissues that form the wall of a blood vessel may stretch beyond their capacity, and thus will suffer damage. Over-time, the combination of the erosion of the inner surface, along with the over-stretching of the elastic fibers of the wall of an artery, can contribute to catastrophic damage.

A 2015 paper by Kohn et al. in Frontiers in Genetics thoroughly reviewed these pathophysiologic aspects of atherosclerosis:

“Elastic fibers have an extremely low turnover rate in vivo, and this longevity allows for the accumulation of age-related changes caused by fragmentation, calcification, and MMP-degradation (Schlatmann and Becker, 1977). As elastin fibers decay, they lose functionality and shift load bearing onto stiffer collagen fibrils, which directly contributes to significant increases in arterial stiffness. Fatigue failure from pulsatile wall stress can cause elastin fragmentation throughout the lifetime (O’Rourke, 1976; Avolio et al., 1998; Greenwald, 2007). Calcium in the arterial wall also increases with age facilitating the direct binding of calcium ions to elastin fibers causing calcification (Urry, 1971; Urry and Ohnishi, 1974; Otto et al., 1999). Animal models that induce elevated elasto-calcinosis show increased medial elastin fragmentation and arterial stiffness (Elliott and McGrath, 1994; Gaillard et al., 2005).”

Clinical Experience

I’ve stood at the bedside of many patients in the emergency department whose blood pressures are markedly elevated. In many cases, if blood pressures are elevated for a long period of time, these people have heart disease, kidney disease, and strokes. Because high blood pressure is related to many other chronic disease of the endocrine, metabolic, and cardiovascular systems, it’s impossible to isolate elevated blood pressure as a singular causative factor of disease, and it’s likely that hypertension is both a cause and effect of these diseases. But it’s still well understood that elevated blood pressures can cause acute decompensated heart, kidney, brain, and vascular failure, and that lowering the blood pressure can alleviate the symptoms of the disease.

For instance, in an extreme case, called a “hypertensive emergency,” markedly elevated blood pressures in the range of 180 mm Hg systolic or greater can cause a patient to have stroke-like symptoms or heart-attack like symptoms, both of which can spontaneously resolve within minutes as we lower the blood pressure with medications. It may sound quite scary, but to an emergency physician, it’s actually one of the most gratifying experiences we can have. We’re used to seeing patients with end-stage chronic disease that is largely irreversible, so when we can alleviate chest pain or make stroke symptoms vanish by injecting a few CCs of an IV medication, it’s a great satisfaction and relief.

This example is extreme, and thus doesn’t directly inform the thesis of this article which deals with generally healthy people and much lower blood pressures, but I think it’s still a useful illustration for non-physicians to understand that there is direct clinical experience that informs a doctor’s knowledge of the effects of hypertension.

Epidemiologic Data

There is very strong epidemiological data supporting the idea that high blood pressure is associated not just with strokes and heart attacks, but with all-cause mortality. That means, if you take a group of people and put them in order by blood pressure, on the average, the people with the highest blood pressure will die the earliest, and people with the lowest blood pressure will live the longest. This isn’t true in every individual case, but it is true on a population level. It’s obvious that someone with an excellent blood pressure could still die in a car accident, or might even die from a heart attack due to causes other than hypertension, but on the average, lower blood pressure predicts a longer life, fewer heart attacks, and fewer strokes.

Some of the best data for this comes from actuarial studies.

In 1978, the Society of Actuaries published the “New Build and Blood Pressure Study.”

The data from this study demonstrates that normal blood pressure in a young and healthy adult is likely around 115/70. You can also see in the table below that, on average, BP increases slightly with age into the 7th decade of life.


You can see in this table that all-cause mortality decreases in lock-step with blood pressure to a floor of 88-97/48-67 mm Hg.


This next table shows you the relative mortality rates for various blood pressure windows. The lowest blood pressure windows have the lowest mortality, and the highest, the highest.


This table is fascinating. It’s showing us that those with blood pressure below 118/73 have half the likelihood of developing clinical coronary disease and cerebral hemorrhage, three-quarters the chance of developing hypertensive heart disease, half the rate of nephritis, half the rate of digestive diseases, and three-quarters the rate of all cause mortality when compared to those within a range of 128/78-137/87 mm Hg. In other words, having a sub-normal blood pressure puts you into a lower risk category than if your blood pressure is “normal” or “elevated.”

I’d like to point out one other interesting item on this table. While the lower rates of cardiovascular disease, cerebral hemorrhage, and nephritis (a form of kidney disease), are expected, the lower rate of digestive diseases in patients with lower blood pressure may come as a surprise.

But, the fact that digestive diseases also correlate directly with blood pressure is not that surprising. There are at least two likely explanations for this finding:

  1. Hypertension is just a marker of an unhealthy person, and unhealthy people are more likely to have all kinds of diseases. If this is true, it’s no surprise that lower blood pressure correlates with fewer digestive diseases and vice versa.
  2. Hypertension causes digestive diseases. Though we don’t think of the digestive tract as being affected by high blood pressure, this probably isn’t entirely true. Perhaps the bowels and digestive organs aren’t as sensitive to elevated blood pressures as are the kidney, heart, and brain, but like all organs, the digestive organs depend on proper blood flow in order to work properly. In the most severe cases of restricted blood flow to the intestines, the intestines can suffer from mesenteric ischemia, which is like a “heart-attack” of the gut. This can be catastrophic and deadly.

It’s likely that both of these inferences are true, and that in combination, they account for the fact that digestive diseases, like many other diseases, track with blood pressure.

Epidemiological data comes not just from actuarial studies, but also from an ample body of medical literature.

In 2002, a meta-analysis published in the Lancet which reviewed 61 studies found that the risk of vascular mortality was lowest with people who had the lowest blood pressure. This held true for blood pressures all the way down to at least 115/75. Take note that this is the same average blood pressure found in young and healthy people in the New Build and Blood Pressure Study mentioned above.

Here are a two key excerpts from the meta-analysis:


“At ages 40–69 years, each difference of 20 mm Hg usual SBP (or, approximately equivalently, 10 mm Hg usual DBP) is associated with more than a twofold difference in the stroke death rate, and with twofold differences in the death rates from IHD and from other vascular causes.”

“Within each decade of age at death, the proportional difference in the risk of vascular death associated with a given absolute difference in usual blood pressure is about the same down to at least 115 mm Hg usual systolic blood pressure (SBP) and 75 mm Hg usual diastolic blood pressure (DBP), below which there is little evidence.”


Throughout middle and old age, usual blood pressure is strongly and directly related to vascular (and overall) mortality, without any evidence of a threshold down to at least 115/75 mm Hg.”

These findings are compelling and integrate well with the actuarial data discussed above.

Finally, in a 2014 Lancet study, the lowest risk for cardiovascular disease was found in a group with blood pressures that ranged from 90-114/60-74. The higher the blood pressure, the higher the risk for most vascular diseases. I won’t review this study in detail, but its results are consistent with those discussed above and with the rest of the medical literature. In an exhaustive search, I’ve been unable to find any data or expert opinions suggesting that lower blood pressure even down to “sub-normal” blood pressure in the neighborhood of 115/70 or even lower is undesirable for any reason.

Mechanistic evidence, clinical experience, and epidemiologic data support the idea that low blood pressures are better than higher blood pressures at every level. Starting at around 115/70, or perhaps a bit lower, we see the lowest rates of all types of chronic diseases and mortality at every age range. This is a good starting point, but it’s not proof that minimal elevations in blood pressure cause damage over time. Instead, it could be that lower blood pressure is a marker of better genetics, or that higher blood pressure is caused by a separate factor that also predisposes to disease and death. We still have work to do in order to demonstrate that by lowering “normal” and “elevated” blood pressure early in life we can prevent vascular damage and thus prolong healthspan and lifespan.

Blood pressure causes vascular damage, and lowering elevated blood pressure decreases the likelihood of disease.

One of the best studies that ties together the concept that blood pressure is damaging, and that bringing it down is beneficial, is the SPRINT trial.

In this trial, 9,361 individuals with systolic blood pressure of at least 130 mm Hg were randomized to two treatment groups. One group was treated with a target systolic blood pressure of 120 mm Hg (intensive treatment), and the other was treated with a target of 140 mm Hg (standard treatment).

The trial was stopped after 3.26 years because of a significantly lower rate of death and cardiovascular disease in the group who underwent intensive treatment.

The intensive treatment group achieved a mean blood pressure of 121.4 mm Hg, whereas the standard treatment group achieved a mean blood pressure of 136.2 mm Hg. There were 210 deaths in the standard treatment group as opposed to only 155 deaths in the intensive group.

In addition to proving the benefits of “intensive” versus “standard” treatment, by reducing confounders, this trial also demonstrates that blood pressure alone is likely a causal factor in vascular disease and death, and that lowering blood pressure is an effective treatment. What’s more, it demonstrates that lowering blood pressure down to a mean of 121.4 mm Hg in a group of over 4,000 people was safe and effective, at least on a population basis.

We always have to keep in mind when we’re reading studies, that the studies and statistics are all about groups of people. Even if something looks like it works for a group on the average, we have to remember that in every group, there’s at least one person (likely more) for whom the treatment didn’t work, or even was harmful, and some for whom it might have worked far better than average.

Regardless, this study, by keeping other variables constant, and treating only blood pressure, it shows us that blood pressure is likely causal in vascular disease and death and that it’s a good target for treatment.

The especially astute reader might ask a few more questions about this study. I won’t review every question, but here are a few.

  1. What were the baseline blood pressures?

    Baseline blood pressure for both groups was roughly 150 +/- a standard deviation of 15. Blood pressures were distributed as follows: a third of each group was below 132 mm Hg, a third between 132 and 145 mm Hg and a third over 145 mm Hg.

  2. Were there more adverse events in the intensive group?

    Yes, serious adverse events were higher in the intensive treatment group. An additional 27 patients had syncope, which refers to a transient loss of consciousness and is often accompanied by a fall. However, the standard treatment group actually had a higher number of "injurious falls." There were also an additional 37 cases of electrolyte abnormalities and 76 cases of acute kidney injury or acute renal failure in the intensive treatment group. There were 80 more cases of patients with sodium levels below 130, and 40 cases of patients with potassium levels below 3.0. Interestingly, the rates of orthostatic hypotension were actually higher in the standard group, which is unexpected.

    Though there were more adverse events, these were mostly minor, and the adverse events in the standard group were qualitatively worse than in the intensive group. What’s more, most of these adverse events can be mitigated if they’re anticipated and monitored closely on an individual basis.

Put simply, the SPRINT trial, observational data from the medical literature, epidemiological data from actuarial studies, and our mechanistic understanding and clinical experience with blood pressure suggests that lower is definitively better whether achieved “naturally” or pharmacologically. The SPRINT trial really drives home the idea that blood pressure is causal in vascular disease and death, not merely an effect or associated factor.

In the next part of this series, we’ll examine the idea that the negative effects of marginally elevated blood pressure start early in life, and that the damage of elevated blood pressure is cumulative.

Until next time,


Read “Blood Pressure Part One: What is ‘normal?’”

Featured Image Credit: Uwe Diegel

Leave a Reply