The sodium debate: heat or light?

Lowering sodium intake has been a constant in dietary guidelines for decades and is widely recommended as part of healthy eating advice. The conventional view is that eating too much sodium increases blood pressure and thereby increases the risk for cardiovascular disease. Yet the benefits of sodium reduction are being hotly debated with experts around the globe trading arguments in journals, on-line debates and the mainstream media. What’s driving this debate? 

In November 2011, O’Donnell and colleagues published a study on the relationship between urinary sodium and the risk for cardiovascular disease. Urinary sodium is a good way to estimate how much sodium people are eating as, generally speaking, ‘sodium in equals sodium out’. As higher sodium intake is known to be linked with higher blood pressure, the expectation might have been that the higher the sodium excretion, the higher the risk for cardiovascular disease. However, the relationship turned out to be J-shaped. In other words, high intakes of sodium were indeed associated with higher risk for cardiovascular disease, but so were low intakes of sodium. How can that be? Does eating too little sodium cause harm?

Just another essential nutrient?

In an on-line debate, Professor Michael Alderman from the Yeshiva University in the USA argued that the links between dietary sodium and health were similar to those of other essential nutrients – harm at low levels of intake (deficiency), optimal for health over a wide range of normal intakes, and harm at high intakes (toxicity). He argued that the notion that lowering sodium intake lowered risk for cardiovascular disease was simplistic and unproven. While not disputing that sodium intake was linked to blood pressure, he noted that sodium intake had many other effects on metabolism, neural systems, hormonal systems, insulin resistance and triglycerides. Some of these effects may have adverse implications for the risk for cardiovascular disease and could offset the beneficial effects on blood pressure. Alderman argued that intervention trials assessing the effects of lower sodium intakes on cardiovascular events were required.

Image: source

An alternative view

In the same on-line debate Professor Lawrence Appel of DASH Diet fame argued that an intervention trial of sodium and cardiovascular events would never be conducted. It would require too many people, be too long and too expensive. However, he noted that the Trials of Hypertension Prevention (Phase 3) had shown a significant 25-30% fall in cardiovascular events in those who had participated in earlier studies to lower blood pressure by sodium reduction. And in the study by O’Donnell and colleagues higher intakes of sodium were associated with all measures of cardiovascular disease assessed – myocardial infarction, stroke, hospitalisation with congestive heart failure and cardiovascular death.

But what about the apparent increase in cardiovascular risk at low urinary sodium? Appel suggested that some of the apparently low intakes of sodium could actually be due to ‘under collection’ and had given rise to a spurious relationship. Others have suggested that perhaps these subjects were eating low sodium diets on purpose because they had cardiovascular disease and had been advised to adopt a low salt diet. If so, is it any surprise that they these sick people had more cardiovascular events? In other words, the low sodium intake may have been associated with their ill health but was not the cause of it. Maybe those with low sodium intakes just had low calorie intakes. If so, why? Again, were they sick?

Trends in sodium intake: more debate

There are no good data on trends in sodium intake in Australia though a recent US study provided some interesting insights and generated further passionate debate. Harvard researchers Berstein and Willett, reviewed a variety of US studies carried out in the past half century in which 24-hour urinary sodium excretion had been measured. They found there has been little change in sodium consumption over this period, findings that mirrored those of a similar study conducted in the United Kingdom. Surprisingly, the mean sodium excretion in both countries was almost exactly the same and also similar to those observed in the international INTERSALT study, despite the wide differences in food cultures.

Image: source

As the period of investigation covered a time in which there was a significant shift from food preparation in the home to the consumption of more processed and fast food, the results came as a surprise. Isn’t processed and fast food supposed to be full of salt? The other take-home message from this study was that the rise in the prevalence of hypertension in the US is not being driven by changes in sodium intake. The likely cause is the rising prevalence of obesity.

An accompanying editorial took a very different line. Faced with the narrow range of urinary sodium levels and the lack of change over time, the authors suggested that human sodium intake may be a parameter that even the most well intentioned public policy cannot modify in most people. It was suggested that such policies are in fact ‘doomed to failure’. The authors cite sodium’s fundamental role in the regulation of extracellular fluid volume and the likelihood of failsafe mechanisms to ensure sufficient sodium availability. Indeed, when sodium intake falls below 120mmol/day a hormonal system switches on, reabsorbing more sodium from the urine. They argued that any recommended intakes below this level assume that the basic biology of humans can be ignored. The authors suggest that the range of sodium intakes in the US, between 120-210mmol/L should be considered ‘normal’. Current dietary recommendations are 100mmol/day.

Dietary sources of sodium in Australia

Researchers from Deakin University recently published a paper outlining the major sources of sodium in the diets of Australian children, based on the Australian 2007 Children’s Nutrition and Physical Activity Survey. Table 1 shows the data for 14-16 year olds.

Making sense of the sodium debate

I enjoy debates like this. It’s the stuff of science – established principles and policies being challenged and discussed as new data are published. There is seldom absolute certainty in nutrition as the ideal studies are often impractical to conduct. We are left to form a view based on the various lines of evidence that are available. But it has to be said that the evidence linking salt intake with blood pressure and that linking blood pressure with cardiovascular disease remains strong. Even if a reduction in population sodium intake to 100mmol/L may be out of reach at present, any fall in the current mean intake would be expected to bring benefit. It is not impossible that low sodium intakes have adverse health effects but these will have to be demonstrated, not assumed from epidemiology.


Table 1: % sodium intake by food category (14-16 year old children)

Food category                                        % of total sodium intake

Cereal-based products and dishes                   23.6
• Biscuits, cakes, pastries, etc                             (10.9)
• Pizza, hamburger, savoury noodles, etc            (11.3)
Breads and cereals                                             20.2
• Breads, muffins, etc                                           (16.1)
• Breakfast cereals/bars                                         (4.7)
Meat, poultry, fish                                                16.7
• Sausages, processed meat                                 (9.4)
Milk products                                                         8.9
• Cheese                                                                 (4.0)
Savoury sauces/condiments                                8.4
Vegetable products                                               4.3
Snack foods                                                            2.9
Soup                                                                        2.8
Beverages                                                              2.2
Fats, oils                                                                  1.1
Miscellaneous                                                        5.1

8 thoughts on “The sodium debate: heat or light?

  1. I think it depends on the person, their lifestyle and activity levels. For the average person who exercises minimally or not at all, they’re most likely either getting enough or too much sodium in their diet (when otherwise unrestricted). Sodium is everywhere because any food you eat has some amounts of sodium, eg. vegetables and grains, on top of the high intake of commercial breads, cereals, processed foods, take-away foods etc. people eat. I think for the average person low sodium or sodium deficiency is not a general concern (unless they’re ill and cannot keep foods or fluids down) because we get well enough of it.

    On the other hand, people who are very active like athletes, or young people who train and work out a lot, who sweat a lot, would apply to the danger of low-sodium theory. If their bodies use and excrete a large amount of sodium, without replenishing it from the diet, then of course there is going to be an imbalance. Sodium deficiency can be very acute and serious, hence the use of electrolyte fortified sports drinks among people who do exercise a lot. This group probably don’t have to worry about eating too much sodium as much as the rest of us are. If you use more, you’ll need more of it, no need to obsess. Rather, they may want to make sure they’re getting enough of it.

    Overall a great article. However I don’t think it should be targeted at the general population, but more at those who actually may be at increased risk of low-sodium, such as athletes.

  2. Thanks for another well researched post Bill. I understand being a reductionist is an integral part of nutrition science, however, don’t you think that all of this data, out of the context of potassium, calcium, magnesium intakes is just a little stretchy. Similar to the saturated fat replaced with polyunsaturated fat argument, do you think we should be looking at sodium replaced with potassium as the way forward. This one nutrient hypothesis is just fraught with confounders that are very difficult to control for.

  3. There seems to be far too much medical advice given on the basis of the mean. Blood pressure is a perfect example. I think pretty much every researcher knows that some people are salt sensitive and others aren’t. Based on the mean, everyone gets reduced salt advice. In a perfect world, the salt sensitive would get the greater attention.

    Similar things apply to RCT results. Huge standard deviations aren’t even commented on, and plots of individual data points are rare. Then the “proof” is labelled “significant” if there is less than 5% chance of the mean having occurred by random chance.

    Hopefully, medicine and nutrition will improve in treating the individual rather than the mean. Otherwise, gluten will remain good food for celiacs (statisically)!

    • Are you saying that salt restriction should be managed at the individual level? Should we have no public health strategy? Regards, Bill

    • There is definitely a link between a fall in blood pressure, amongst other parameters in blood such as cholesterol, triglycerides, inflammatory factors etc. when there is a reduction in excess calories eaten, whether from carbohydrate or fat. Even if calories are not relevant, a redistribution of carbohydrate, fat, protein ratio in the diet can also significantly improve problems such as high blood pressure and poor blood lipids, etc.

      So it’s not just the salt that matters, no. I do think salt plays a large role in cardiovascular disease and blood pressure, but that doesn’t mean that ‘low-salt fixes everything’. For example, people who cut sodium probably also cut foods high in sodium that are also unhealthy and high in calories and fat, such as processed meats, potato chips, fries etc. Simply saying that they ‘cut their sodium improved X’ doesn’t really convince me that salt itself did it all, if their diet changed substantially and had als become healthier overall, eg. no more excess calories or fat, not just lower in salt.

      • Hello MC.
        You are right that manipulating salt is not the be-all and end-all of blood pressure. The original DASH study (Appel LJ, 1997) showed that low fat dairy products and fruits/vegetables were important. The DASH-Sodium study (Sacks FM, 2001) showed that sodium had an independent effect. Replacing carbohydrate with protein also lowers blood pressure. High intakes of alcohol and of course obesity have adverse effects. It’s complicated but it also suggests that there is considerable potential to lower individual and population blood pressures by dietary means. Regards, Bill.

    • Hi Rob
      I hadn’t seen Taubes’ article on salt. I think it is legitimate for him to ask whether the apparent up-tick in risk associated with low salt intakes is real or an epidemiological aberration, but Taubes just assumes it’s real.
      However, by far the majority of people have intakes of salt well above this query zone and are in the area where salt reduction will lower blood pressure. Everything we know suggests that this will bring benefit. Taubes says the final proof is not in i.e. that there is no RCT evidence saying salt reduction lowers cardiovascular events. Fair enough, but as mentioned in my blog this evidence will never be gathered. So we are left asking whether lower blood pressure is better for you than higher blood pressure. I say yes. What do you think? Regards, Bill.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>