Defending the JAMA Study on Salt: A Talk With Jan Staessen, MD, PhD

Linda Brookes, MSc


June 14, 2011

In This Article

The Interview

Medscape: Your paper attracted a great deal of critical comment in the news media, primarily from individuals in the United States.

Dr. Staessen: For sure.

Medscape: Did you receive similar feedback from other regions of the world?

Dr. Staessen: I do not know why there was so much criticism from the United States. In my own country, Belgium, there has been quite a lot of interest, and my colleagues in Eastern Europe tell me there have been some reactions there.

Medscape: Have people expressed the same sorts of concerns as people in the United States?

Dr. Staessen: No; they were actually interested in the paper. It was not a negative reaction. I would like to share with you what one leading scientist from a Scandinavian country recently wrote to me: "...Good for you that we are not in the Middle Age -- you had probably been brought before the inquisition and burnt as a heretic. Now you get away with a warning in The Lancet that the results should not even change thinking. This must be the first time that The Lancet warns against thinking. Don’t stop thinking”

Medscape: So what did you make of all the negative reactions?

Dr. Staessen: I think the study has been criticized in the way that is not very scientific. Even the editorial in The Lancet[14] referred to The New York Times. If The New York Times becomes the source of scientific information, then I think something is completely wrong.

Medscape: In the original Times article,[7] Peter Briss, MD, DrPH, medical director at the Centers for Disease Control and Prevention (CDC), was quoted as saying, “The study was small; that its subjects were relatively young, with an average age of 40 at the start; and that with few cardiovascular events, it was hard to draw conclusions.”

Dr. Staessen: What he said was incorrect. He probably did not look at the eTables. There was also criticism in the Lancet editorial attributed to Walter Willett, MD, DrPH, chair of the Department of Nutrition at Harvard School of Public Health.

Medscape: According to the Harvard School of Public Health Website, he said, “Take this study with a huge grain of salt, and then dispose of it properly.”[15]

Dr. Staessen: In time we will answer all these comments in a scientific paper. Meantime, however, I can answer the criticism of the way we measured sodium excretion, by timed 24-hour urine sample collection. That is exactly the way it should be done; it is the gold standard. In fact, this method was used in most studies listed at the Harvard Website.

Medscape: There were comments that the people with the lowest sodium in the study appeared to provide the smallest urine samples, suggesting that they did not collect all the urine over a 24-hour period.[12,15]

Dr. Staessen: No, that is not true. The samples were slightly smaller. This, in fact, validates the urine collection, because to excrete salt one needs volume. The potassium output did not follow the sodium. If the samples were really undercollected, then the potassium should have also proportionally been decreased to the same extent as sodium and this was not the case. In addition, to counter this comment, we repeated the calculations using the sodium:creatinine ratio and the findings were exactly confirmatory. (This is not published yet.) So even if the collections were incomplete, the duration of collection was the same across the tertiles. However, they were complete. We took all necessary precautions to be sure about that.

The study was also criticized by people who said that one urine collection is not sufficient to characterize the sodium intake of an individual. That might be true, which is why we subdivided our population in tertiles. Then sodium output is sufficient to characterize a group[16] and to be sure that all those in the lowest tertile are at the lower end of the distribution and that those in the middle and upper tertiles fit. Proponents of this method, such as Nancy Cook, ScD (Brigham and Women's Hospital, Boston, Massachusetts) in the Trials of Hypertension Prevention (TOHP), all use 24-hour urine collection in studies; they assess sodium intake in exactly the same way as we did. I do not understand these comments.

Another comment was that our findings would not be applicable to the US population because the US population is more obese. (The average BMI of subjects in this study was 25.) But we have repeated our calculations in obese and nonobese subgroups, and actually the findings were much stronger in the overweight and obese subgroups.

It was also said that our study was done in a mainly young population, but that is also incorrect, because the findings were consistent in those younger than 60 years and those 60 years and above. Actually, most of the cardiovascular events occurred in those older people. There were 66 deaths due to cardiovascular causes in the 60-and-older age group vs 18 in the younger-than-60 group.[2] The hazard ratios in the low tertiles were 1.41 for the younger group and 1.52 for the older group.

We also did a sensitivity analysis as a reaction to some of the comments. We had about 980 hypertensive participants in the study, and we compared the results in normotensives and hypertensives. The findings were stronger in the hypertensives than in the normotensives.

Medscape: How did you measure salt sensitivity? I understand that there is no standard method.

Dr. Staessen: No one does it the same way. One is always thinking about sodium sensitivity, which illustrates how biased the nomenclature is. The distribution must be continuous and there are also people who are sodium losers; sodium is a regulated ion. Research on the exchange of sodium and circulating plasma volume from 30 years ago has shown that it is very tightly regulated to maintain exchangeable sodium and plasma volume constant.[9]

Medscape: You yourself said that you were not very surprised at the lack of an association between blood pressure and 24-hour sodium excretion, because you haven’t seen it in 25 years of your own research.[17] You did express surprise at finding an inverse relationship between sodium intake and cardiovascular death. When you first saw that result, which appeared to contradict other studies,[18,19] were you suspicious that there might have been an error somewhere?

Dr. Staessen: Yes; it was surprising, but the data were very strong. We did a lot of sensitivity analyses, we looked at the influence of different factors, and everything we did was confirmatory. We had to address many queries raised by the JAMA editor, but it is our intention to answer the criticism that was voiced after publication of our manuscript, but only in a scientific paper that will go through peer review again.

Medscape: It was also said that there is always one study that comes out against risk factors, like smoking, implying that this study on salt was the one like that.

Dr. Staessen: It is not the first. There are the National Health and Nutrition Examination Survey (NHANES) studies,[20,21,22] Alderman’s report,[23] and2 papers published recently in Diabetes Care in patients with type 1 or type 2 diabetes.[24,25] These also found increased mortality associated with low dietary sodium intake.

Medscape: Diabetic patients would presumably be hypertensive, so presumably they would need to reduce their sodium intake.

Dr. Staessen: They represent different general populations, but we excluded those with known cardiovascular disease at baseline.

Medscape: The comments seemed to be based on a fear that the results were contradicting dietary recommendations about salt intake, so people would interpret the study as showing that they need not follow recommendations to limit salt in their diet. You said in the paper that your results did not support “the current recommendations of a generalized and indiscriminate reduction of salt intake at the population level,” although they did not negate the blood pressure-lowering effects of a dietary salt reduction in hypertensive patients. So the general population might not need to worry about reducing salt, but others might?

Dr. Staessen: Exactly. We can extrapolate our findings to the general population, but we are not going to change the guidelines for hypertensive patients or for patients with heart failure or whatever condition, because this is not our task. The only thing we observed is that, to summarize, if you change sodium intake over time, you see small effects on systolic blood pressure, not diastolic blood pressure. This is of the same magnitude as in the normotensive individuals in the intervention trials, ie, about a 2-mm Hg increase in systolic blood pressure, but no significant change in diastolic blood pressure.[8] Sodium intake does not translate into increased incidence of hypertension because the change is confined to systolic blood pressure and is too small. The third finding, which was a surprise to us, was that low sodium intake is associated with higher cardiovascular mortality.

Medscape: Could that effect be related to increased insulin resistance caused by the lower sodium intake? Insulin resistance would increase coronary heart disease risk because of its association with increased triglycerides, decreased high-density lipoprotein cholesterol, and hypertension.

Dr. Staessen: Yes, low sodium increases insulin resistance, but it is also known to activate the renin-angiotensin system and the sympathetic nervous system, and some studies have shown that high plasma renin activity is a risk factor.

Medscape: So what do you think of current guidelines that recommend low sodium intake?

Dr. Staessen: I think people should keep an open mind. The American Heart Association is now proposing to lower sodium intake for the general population to less than 1500 mg per day by 2020.[26] Without considering whether this is even feasible, these experts made 2 extrapolations. The first one is the change in blood pressure seen in studies in hypertensive patients or randomized controlled trials in normotensive subjects, which was small, and they said that if this occurs in these trials it will also occur in the general population. The second extrapolation that they made is that the change in pressure, a few mm Hg, will lead to tremendous changes in morbidity and mortality and hospitalization. So they extrapolate at 2 different levels.

Now, from our paper you will see that there was no change on average in the sodium excretion of the total population or in the different subgroups between baseline and follow-up. In Belgium, the first patients were recruited in 1985, and over the past 25 years we have seen a decrease in cardiovascular morbidity and mortality in my country. So there is a contradiction here. The sodium intake or the sodium excretion is the same, but if you look at the mortality statistics in Belgium and also in many other countries in Europe and worldwide, including the United States, there is a decrease in cardiovascular mortality -- everywhere, actually.

Medscape: Is your study being extended so that you can have longer follow-up and more events?

Dr. Staessen: Yes; this is an ongoing study, so we will keep participants in follow-up in Belgium and the European countries where they were recruited.

Medscape: Presumably it is never going to be possible to answer these questions through randomized clinical trials instead of observational studies.

Dr. Staessen: That would be impossible and probably irrelevant. All foodstuffs carry a label describing the amount of salt they contain. The public should be properly educated about the pros and cons of a decrease in sodium intake, in particular if they are healthy, and left free to eat what they like.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.