Will the Positive Findings From the SSaSS Trial on Salt Substitution Silence the Salt Skeptics?

John M. Mandrola, MD; Bruce Neal, MB ChB, PhD


September 02, 2021

This transcript has been edited for clarity.

John M. Mandrola, MD: Hey everyone, this is John Mandrola from the | Medscape Cardiology, and I'm delighted to be with Professor Bruce Neal, who is professor of medicine at University of New South Wales in Australia. He is the primary author of the Salt Substitute and Stroke Study (SSaSS), which was presented as a hotline at the European Society of Cardiology meeting and published in The New England Journal of Medicine. This is a really important trial, so welcome, Professor Neal.

Bruce Neal, MB ChB, PhD: Thanks, John.

Mandrola: Can you give us the key findings of this important study?

Neal: We randomized more than 20,000 people to a salt substitute, which was a reduced sodium/potassium alternative to regular salt, and we followed them for about 5 years. We accrued about 3000 strokes, 4000 deaths, and 5000 major adverse cardiovascular events during the study. And we saw a clear 14% reduction in the risk for stroke, with a P value of .006, and we saw clear protection also against the two secondary outcomes: a 13% reduction in total major adverse cardiovascular events and also a 12% reduction in premature death. And both of those results had a P value < .001. So, pretty clear evidence of efficacy.

The other key observation was that we didn't see any increased risk in hyperkalemia, which had been a concern with a salt substitute. The risk ratio was 1.04. So it's a pretty compelling result.

Mandrola: And this was an extremely unique trial environment, wasn't it?

Neal: Yes. We did the study in rural China because to test the hypothesis of whether reducing sodium and increasing potassium would actually change outcomes, you of course need to be able to reduce sodium and add potassium to the diet. In rural China, about 50% of the dietary sodium that people eat is actually consumed as salt that you get at the table for seasoning or when cooking or when preserving food. In more developed countries, a lot more of the dietary sodium comes from processed and packaged food, and it's much harder to replace.

Mandrola: There was also a unique cluster randomized trial design. Most cardiovascular trials use individual randomization. Tell us about the cluster randomization approach.

Neal: We didn't randomize 21,000 people individually to salt or salt substitute; instead we had 600 villages. And in each of those villages we recruited 35 people. And then we randomized the villages — so, that group of 35 — to either salt or the salt substitute.

Now, you lose a bit of statistical power when you use this sort of design, but you make up for it in other regards. The reason we did the cluster randomization was because it was much easier for us to deliver the salt substitute intervention to a group of people in a village, and not have the situation of two people in the same village, with one using salt and the other using the substitute, and them asking questions and the whole thing getting really complicated. That was why we did it, and if you've got a large number of clusters, you don't miss out that much.

Mandrola: You had a significant robust reduction in hard outcomes from a salt substitute. I noticed that the patients were pretty high risk; this was clearly a secondary prevention or high-risk population. Tell us about the patients.

Neal: The patients were adults who either had a history of prior stroke, or they were 60 years or older with poorly controlled blood pressure. We chose that group not because we believe that's the only group who might benefit but because that's the group that's going to have the large number of events that we need to test — what was frankly a relatively modest relative risk reduction of 13%-14%.

What we did have was great power to look at the effects in subgroups. So while the study was done in a mostly high-risk population and we saw benefits in people who had a history of stroke, we also saw benefits in those who didn't have any history of stroke. We saw benefits in hypertensive people as well as those without hypertension.

Mandrola: I noticed from the New England paper and your presentation that the mean blood pressure reduction was kind of modest at only 3 mm Hg. How do you explain such a robust reduction in hard outcomes with that modest degree of blood pressure reduction?

Neal: People forget just how important a risk factor blood pressure is. While 3 mm Hg systolic doesn't sound like much when you're sitting in the clinic, actually 3 mm Hg systolic blood pressure lowering is pretty powerful. It's not that different to what you actually observe in some of the large-scale trials of blood pressure lowering with angiotensin inhibitors and angiotensin receptor blockers that probably have most of their effects through blood pressure lowering.

In this study, we were studying stroke, and stroke is uniquely responsive to blood pressure lowering. In China you also get a slightly greater proportion of hemorrhagic strokes. And hemorrhagic strokes are even more sensitive to blood pressure lowering, so I'm pretty sure that this is a blood pressure lowering effect that we see in SSaSS and not something else.

Mandrola: Some colleagues asked me to ask you about the operational challenges of working with cluster randomization in villages in rural China and whether you are confident in the results.

Neal: I'm really confident in the results. I've had the good fortune to be working on the ground in China for almost 25 years with colleagues there. We have really strong teams and really good relationships. One of the things that really helped us though was that the Chinese have a universal health coverage insurance system in the rural areas. What that means is if you've had a serious event and end up in hospital, you get most of your healthcare costs for that event reimbursed. More than 99% of people in the study were signed up to that system, so it gives you this phenomenal capacity to track people and know whether they're dead or alive, and also to identify whether they've been to hospital or not.

We were able to get phenomenal follow-up. We recorded complete follow-up of vital status in this study. I've never been able to do that before, and I've done a series of large-scale trials. So I’m pretty confident about the quality of the data.

Mandrola: Let me just ask you to speculate. I mean, so much has been said about salt and its role in hypertension and health. How do you think these findings inform the sodium debate?

Neal: The debate is around whether there is a risk associated with reducing dietary sodium, and frankly, a lot of that is based on some fairly weak observational epidemiology. I've never been a great believer in that as a concern, but it's a very real concern out there. SSaSS is not going to answer every skeptic's questions because we didn't lower dietary sodium to really low levels. But we clearly showed that reducing dietary sodium, and adding dietary potassium, is a very safe, effective thing to do.

I would love this to be the final word on the debate, but it's not going to be. I think that there is a message here for many people, because almost everyone eats salt, and almost everyone eats more salt than they should. More likely than not, the SSaSS trail is relevant to all those people. This is something that I would do if I did eat salt, which I don't, but I will be recommending it to all my friends.

Mandrola: You did two things: You reduced sodium and you added potassium. Will there be substudies that will be able to tell us which of the effects predominates, or is it both?

Neal: We wrote the paper very carefully to try to stay out of that debate. We just said that we used a salt substitute and it was really effective and you should do that, because we think that's the most pragmatic message to take from the paper.

That said, the population that we studied eats quite a lot of salt but they don't eat that much dietary potassium. There is no doubt that this was an effect that was driven by both reducing dietary sodium and supplemental dietary potassium. If you started off eating a different ratio of sodium/potassium, you might get a different proportion of the benefits from taking out the sodium vs adding the potassium. It's something we haven't worked out for this study yet but it is certainly on the list of things to do.

Mandrola: In the United States and elsewhere, so much of our sodium comes from processed food and we can't really control that. Do you think that this study informs policy decisions in terms of our food supply? A few years ago we took trans fats out of the food supply and that seemed to reduce events quite significantly. Could the same thing be said for salt substitutes in processed foods?

Neal: We think there is definitely a message for the processed food industry. They've probably not felt particularly strongly to act on dietary sodium in processed foods because the evidence was a bit lacking and there's been a lot of debate there. We hope that this will provide a new impetus to the food industry, and to people who regulate the food industry, to say, you really need to put less sodium in and you need to try and stop all that potassium that was in the fresh fruits and vegetables at the beginning from being leached out during processing.

How might you regulate that? Well, there are a few countries that have tried to set standards around maximum acceptable levels of sodium. I don't know that anyone has done anything around potassium levels. We can, of course, label foods for overall healthiness or specific constituents like sodium — that's a possibility.

I think that the bigger piece, in terms of rollout and implementation, that we'd like to see is a focus on low- to middle-income countries where more than 50% of dietary sodium comes from discretionary salt that could be directly replaced with a salt substitute. And the key there is probably going to be making the salt substitute the same price as regular salt, as well as making it widely available on supermarket shelves.

Mandrola: My final question, which also came from colleagues, is about the taste difference with the salt substitute. If there is a taste difference, could there have been an issue with blinding in the trial? If people knew they were on the salt substitute, might they do other things that help reduce their risk?

Neal: First of all, this was an open study. We only provided the salt substitute; there was no blinding. The people in the control group continued using their regular salt. To control for that lack of blinding, we measured hard outcomes. It's difficult to make a mistake about whether you had a stroke; it's very hard to make a mistake about whether you died or not. I don't think the results have been adversely influenced by the fact that it's an open and not a blind trial.

In terms of taste, we chose a 75/25 mix of sodium chloride/potassium chloride because it does taste very similar to regular salt if you cook with it. If you have a lot of it, you can taste a slight bitterness. If you use slightly more sophisticated products available on the shelves of some Western supermarkets, with amino acids or some flavor enhancers, you can pretty much get the flavor identical. Taste shouldn't be an issue, and 92% of people were still adherent in the intervention group at the end of 5 years. I don't know another trial, certainly of drug therapies which have no taste, that achieved that.

Mandrola: Well, Professor Neal, it's really an honor to talk to you. Congratulations on this really important study. We review a lot of studies and this is a really special one. So, thanks for being with us.

Follow | Medscape Cardiology on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.
Post as: