This is a follow up to my now-notorious diary concluding that you are probably drinking the exact amount of water that you need. I hope some of you find this useful.
High blood pressure (hypertension) is a global epidemic, and is responsible for substantial excess mortality everywhere it has been studied [1]. According to some estimates, hypertension accounts for around 50% of stroke and coronary heart disease [2] and similar numbers are obtained by other investigators. The burden of hypertension is greater in poorer countries and in the US it takes a heavier toll on people of color and people who lack adequate access to health care. Arguably there are very few health conditions where it is easier to point to social determinants in the outcomes, including morbidity and mortality.
The regulation of blood pressure is physiologically complex, and includes several independently acting negative-feedback loops. Your home air conditioning system is an example of a negative-feedback loop. It has sensors to measure the temperature in various parts of the house, there has to be a set-point (think about the temperature you set on your thermostat) and there have to be effector systems that lower the temperature if it gets to high or increases it if it gets too low relative to your chosen set-point (your AC and furnace). Your body has pressure sensors (baroreceptors) in various parts of your vascular system. Actually some of them detect high pressures in the arterial system and some that detect the lower pressures found in venous system, which tends to relate primarily to overall volume of blood. The effector systems include the autonomic nervous system (especially the sympathetic) which tends to work on fast time scales (like when you stand up quickly). There are also multiple hormonal systems that function as crucial effectors, especially something called the renin-angiotensin-aldosterone system (RAAS).
Eons ago when I was a student it seemed like the majority of the research on blood pressure control focused on neuronal and smooth muscle mechanisms, including the role of the baroreceptors, the sympathetic nervous system (including its CNS components), and studies of vascular smooth muscle. It made a lot of sense since a lot of the drugs that were initially used for hypertensive therapy impacted primarily on neuronal systems (and tended to be quite poorly tolerated). Research was also focused on the smooth muscle cells that make up the walls of the blood vessel, the idea being that they were too reactive and clamped down too hard, and that endothelial cells (the inner lining of blood vessels) were not functioning to properly inhibit the muscle. Of course it was known that arterial stiffening due to formation of atherosclerotic plaques played a role. Basically, the dominant paradigm was that if the vessels had too much resistance to blood flow for whatever reason, then you would have hypertension. Those studies weren’t wrong, but they tended to focus on the tail of the elephant, so to speak.
A fairly simple concept that was grasped from the beginning is that over long time periods, the volume of blood has to play a role overall blood pressure. The simple analogy is what happens to a balloon when you pump more air into it. And that brought attention to the kidneys, which are the major regulators of blood volume. In fact, drugs that are still widely used today, the various diuretic agents, can lower blood pressure by causing the kidney to excrete more fluid (and sodium). Along with this, it has long been known that people with chronic kidney disease often have elevated blood pressure as a result. Maybe this is the trunk of the elephant.
In the early 1970s, a physiologist named Arthur Guyton and his coworkers proposed a highly influential model based on a phenomenon known as pressure natriuresis [3]. This translates in English to “when blood pressure goes up the kidney excretes more sodium until the blood pressure comes back down”. It is a feedback loop that occurs at many levels, some of which are entirely contained within the kidney itself, but which we now know is also the primary role of the RAAS. There is a way to summarize this (as a first approximation) that makes a lot of things easy to understand: The kidneys control blood volume (and therefore pressure) by regulating the amount of sodium excreted. The kidneys control blood osmolarity by regulating the amount of water secreted. The kidneys can regulate these things independently. The blood pressure set-point is determined by what the kidneys do. It is not entirely intuitive but if you keep those ideas in mind, you tend to get the right answer. It is a direct consequence of pressure natriuresis. Diuretic agents and inhibitors of the RAAS lower blood pressure primarily by increasing the amount of sodium that you excrete. These drugs are quite effective in a lot of people. Nearly all of this pertains to the output end.
What about the input end? If people consume more sodium, will this have an effect similar to the kidney retaining more sodium? A consensus on this was a long time in coming, but at this point the answer seems unequivocal. On average, a restriction in salt intake will result in a decrease in blood pressure, and this occurs in people with normal or elevated blood pressure. This is the effect in populations. Now, the extent to which this occurs in any one person may vary because it remains the case that there are many different mechanisms that could lead to elevations in blood pressure (we are dealing with an entire elephant). For example, in some people changes in the contractility of the blood vessels themselves may be more important than in others (there is reason to think that this is especially important in African-Americans, for example). So restrictions in salt intake may not produce a clinically significant drop in blood pressure in everyone, but almost always there will be at least some drop. On this basis, since the 1980s, there have been strong recommendations to reduce salt intake. To see this effect most clearly, you have to follow people for a long time — months at least, and that is what makes these kinds of studies hard to do.
One issue that remains controversial, though, is whether or not lower salt diets actually end up reducing all-cause mortality. You would think it would, right? High blood pressure increases mortality, salt intake increases blood, pressure, so dropping salt intake should reduce mortality.
There still seems to be some disagreement on this. In fact, some studies have concluded that there are increases in mortality associated with either very high or very low sodium intake, with the sweet spot being at low but not too low. There are a host of methodological issues that entail how people actually measure salt intake and compliance, confounding factors, etc. To be honest, a lot of it gets into epidemiological details that are a bit beyond my ability to evaluate knowledgeably. However, a landmark paper a few years back in the Journal of the American College of Cardiology [4] and an accompanying commentary [5] summarized decades of research on this. Without question, high-sodium diets increase mortality.
The problem is that American diets have a lot of salt in them. We often crave salt (consider that in most locales throughout human history it was a difficult thing to obtain). People eat too much sugar, but they damned sure eat too much salt. As someone here commented earlier, in a grocery store you can find low-sugar, low-carb, and gluten-free foods of all sorts, even though a lot of people buying them are not diabetic and don’t have celiac disease. But it is a lot harder to find low-sodium foods.
For most people salt intake really is something to be mindful of, and definitely it is more important than obsessing about how much water you drink.
Edit added: And to be more clear, by salt I really mean sodium.
1. Bundy et al. (2017) Systolic Blood Pressure Reduction and Risk of Cardiovascular Disease and Mortality: A Systematic Review and Network Meta-analysis. JAMA Cardiology 2(7):775-781.
2. Arima et al (2011) Mortality patterns in hypertension. Journal of Hypertension 29 Suppl 1:S3-7.
3. Guyton et al. (1972). Circulation: overall regulation. Annual Review of Physiology 34: 13–46.
4. Cook et al. (2016) Sodium Intake and All-Cause Mortality Over 20 Years in the Trials of Hypertension Prevention Journal of the American College of Cardiology 68: 1609-1617
5. Mente et al. (2016) How Robust Is the Evidence for Recommending Very Low Salt Intake in Entire Populations? Journal of the American College of Cardiology 68: 1618-1621
