EDITOR’S NOTE – This conversation with Ron Krauss took place in late March 2012. It focuses on the health questions involving saturated fat and red meat that have been in the news of late. Ron is a Senior Scientist and Director of Atherosclerosis Research at Children’s Hospital Oakland Research Institute, Adjunct Professor in the Department of Medicine at UCSF and in the Department of Nutritional Sciences at UC Berkeley, and Guest Senior Scientist in the Department of Genome Sciences of Lawrence Berkeley National Laboratory. What follows is a paraphrased transcript of the interview. I’ll post the actual audio sometime soon. – Shelley
Ron Krauss, there’s so much debate about saturated fat. Some studies say it’s perfectly fine to have it, and yet as a standard policy in most health clinics and with institutional recommendations, it’s, “keep the saturated fats low.” Meanwhile, your research has been a little frustrating. You don’t come out clearly saying one way or the other. Instead you say, it depends. It depends on what the saturated fat is eaten with.
RON KRAUSS
I don’t know if I have to apologize for the way things really are, but that’s the way they are. And it’s not my fault. (laughs) My job is to help people understand how saturated fats work in the body, and explain it in the simplest possible way. We’re trying to work towards giving people that understanding, but simply the idea that saturated fat itself has an effect on health, outside the context of what one’s eating, is already a naive concept. Our studies are showing that how saturated fats interact with other foods is really more important than even we had realized when we started this work. And what we’ve learned about saturated fat, over the years, has shifted what we think. My first studies were driven by the hypothesis that lowering fat its good for you. This was in the old days, and I came out of that background of expecting lower fat diets to be healthier, and we wanted to study lower fats diets.
When was that?
RON KRAUSS
We did our first studies in 1989, so it’s been over 20 years ago. I’m somewhat shocked to realize how long I’ve been doing this kind of work, but along the way we’ve learned things we didn’t expect to see. In fact for me the most important advances and the most interesting ones certainly are the ones that come out with opposite results from what you started with, where you expected one result and the results came out differently. And that happened when we first studied lower fat diets, thinking they would benefit the individuals who had higher heart disease risk, such as people who had the more dangerous, small particle LDL cholesterol, the pattern B profile, for example. What we found, to our surprise, initially, was that when we fed these low fat diets and reduced the fats by substituting carbohydrates, which was at that time and still remains the current paradigm, we really didn’t achieve what we had wanted to achieve. There was some improvement in the overall amount of cholesterol in the very small percentage of individuals who had very high amounts of small LDL particles in their blood already, but what was really astonishing to me at the time was that the majority of people we studied, the high percentage of people who had the normal metabolic profile, with more of the safer, Pattern A, larger particle LDL, a very large percentage of those people we studied actually shifted into the riskier, pattern B mode when we reduced their saturated fat intake. We did the equivalent of changing their hair color from red to blonde, or maybe blonde to red, by putting them on a lower fat, higher carbohydrate diet, so we elicited what we think now is an underlying genetic susceptibility which is very common in the population.
And we’re not just talking hair color here. We’re talking about something that actually changed their health, because the increase in the smaller particle patten B LDL lipoproteins were an indication of greater health risk. So it was bad news. It seemed that giving most people the supposedly healthier, lower fat, higher-carb diet made them more at risk for heart disease.
RON KRAUSS
Yeah, and we were certainly concerned about increasing heart disease risk, so we turned our attention ultimately away from feeding higher carbohydrate, lower fat diets, to doing the reverse, to lowering carbohydrate and raising fats, and that’s where we intersected with the world of people very interested in very low carbohydrate diets. We sort of worked our way into studies along those lines and ultimately published a study five years ago, in which we systematically compared different dietary approaches, keeping in one case, saturated fat constant and in another loading the diet up with saturated fat, and what we found is that in our study groups, the reduction of carbohydrate alone improved metabolic profile in the majority of individuals, independently of saturated fat intake.
Was that the 2006 American Journal of Clinical Nutrition Study? In that study, you had a range of carbohydrates that people ate from 54% of calories as carbohydrates to 26% of their calories as carbohydrates.
RON KRAUSS
We had three levels. The low was 26% carbohydrates, the high was 54%, and then we had 39% in between.
And you also varied the amount of saturated fat. In some of those people you fed them 8% of their calories as saturated fat and in some you fed them 15%.
RON KRAUSS
That’s right. When we lowered carbohydrates in their diets, we raised the fats. So, on the lowest carbohydrate diet, the 26% carbohydrate diet, we jacked up the fat content with either monounsaturated or saturated fat. That is, either an olive oil type of monounsaturated fat versus, in this 2006 study, a saturated fat, mostly dairy fat.
And you didn’t see much difference when your study subjects switched from eating more monounsaturated fats to eating more saturated fats. But when their carbohydrate intake went up or down, you saw significant changes in their blood cholesterol.
RON KRAUSS
Yes. The improvement was clearly due to the carbohydrates.
Meaning lowering the carbohydrates generally lowered the risk factors that showed up in their blood.
RON KRAUSS
That’s right. The improvement was clearly due to reducing the carbohydrates.
It might seem silly to emphasize that, but we really need to say it more than once — that the improvement was due to increasing fats and reducing carbs, because that’s backwards from how most people think it works.
RON KRAUSS
Yeah. I like to think the world has come around but obviously that’s not true. We looked carefully at the saturated fat effects. With more saturated fat in the diet, we did see a signal for an increase in the overall amount of cholesterol in their blood. But when we looked more carefully, that slightly increased amount of total cholesterol was not being carried by more of the dangerous, small particle LDLs. It seemed to be carried more by larger particles. Actually, in the people eating more fat, and fewer carbs, the total particle concentration, which most people in our field think is a stronger signal of risk that total cholesterol, the total number of particles did not go up.
Let’s go into a little more detail here, because this idea about “total cholesterol” versus “total particles” is worth understanding. You’re saying that in people who ate more fat and less carbohydrate, and in fact, who ate more saturated fat, they had a little more total cholesterol fat in their blood, but it was mainly being transported in big, fluffy LDL particles, which are the kind that scientists who study heart disease consider pretty safe. In this study, the people eating more saturated fat, did not have an increase in the kinds of LDL particles you consider more dangerous, which as the small, BB-sized, pattern B LDL particles.
RON KRAUSS
Yes. When people ate more fat and less carbohydrate, the number of small particle LDLs remained low, and switching from monounsaturated to saturated fat didn’t increase their number at all. In fact, when people switched from mono- to saturated fat in this study, the large particle LDLs might have gone up a little bit, and the small particles went down. So by anybody’s current criteria about whats’s important for heart disease risk, saturated fat caused no increase in risk. That was clear to people who understand the role to the lipoprotein particles, as opposed to the overall cholesterol level, which I’m sure for some people is a subtle distinction.
And it all starts with how the cholesterol fats are carried through the blood. They’re carried in particles, which each include one protein called . . . .
RON KRAUSS
It’s Apolipoprotein B (ah-PILL-oh-PROE-teen Bee) Also known as Apo B
Apo B. Just one protein per particle. And that protein is like a mesh bag that is holding a whole bunch of fat inside of it – a lot of cholesterol fat. So one Apo-B protein per bag. Some Apo-B proteins enclose small amounts of fat — those are small-particle LDLs. Some enclose large amounts of fat–those are large particle LDLs. So . . . if you poured out all those bags at once, and measured the total amount of cholesterol that was in all of those bags, that would be the “total cholesterol” number you can see on a typical heart disease risk test. Just the cholesterol poured out of all those bags. But you seem to be saying that it’s equally important to simply count the number of bags — that is, the number of Apo-B particles that had been carrying the cholesterol. And also looking at the size of each bag. How about it? When it comes to heart disease risk, is the total amount of cholesterol a key piece of data? Or is it more important to count how many LDL particles, that is, how many “bags” are holding the cholesterol?
RON KRAUSS
Both numbers can be helpful. But most people certainly in the field of cholesterol and heart disease understand that the number of particles matter more than just how much cholesterol they carry overall. In terms of health, the first order of business is to make sure the total number of particles in a person’s blood is maintained in the healthy range, because that’s what dictates heart disease risk. Beyond that, when the total amount of cholesterol is unusually high, you’ll almost always find that the total number of LDL particles is high. And looking even deeper, when the total number of LDL particles is higher than normal, I and many of my colleagues would argue that the biggest concern is warranted when the number of smaller particles is high, not the larger ones.
So, if you have a lot of cholesterol fat in your blood, and it’s mainly being carried in a “normal” amount of LDL particles which are large, fluffy ones, that’s less of a concern than when the same amount of cholesterol is being carried by an unusually high number of little tiny particles.
RON KRAUSS
It’s definitely associated with lower heart disease risk if the cholesterol is carried in larger particles, and that’s because there are fewer of them. Some people who study heart disease report that total cholesterol levels are not very different between patients with and without heart attacks. If you look deeper, you may find that those patients who have heart attacks, generally have more LDL particles in their blood, compared to people who don’t have heart attacks, even when their cholesterol levels are similar.
So you’re saying that if a person just counted the number of Apo-B protein particles in their blood, that would be a better indicator of heart disease risk than measuring the total cholesterol in their blood. If you have a whole bunch of those, you probably have a lot of small particles.
RON KRAUSS
Apo-B is a a sort of first approximation of the particle concentration. It’s not measuring it exactly, but it’s a whole lot more accurate at predicting risk for heart disease, than just measuring the total amount of cholesterol itself. This is being acknowledged more widely in the first of heart disease assessment.
Has anybody told the American Heart Association this?
RON KRAUSS
The AHA, the American Heart Association, and I’ve served on their advisory board, the AHA, does two things It promotes scientific research and it makes statements from time to time.
But their policy statements don’t reflect the importance of counting LDL particles yet.
RON KRAUSS
No There’s still a lot of debate about whether we should be advancing beyond the old LDL “total” cholesterol risk assessment, to understanding it based on particle concentrations, but I think we’re moving in the right direction. Getting back to how all this applies to what kinds of fat to eat, and how much fat to eat, in our 2006 Study in the American Journal of Clinical Nutrition, we included one branch of the study that involved eating a really high saturated fat diet, with lots of dairy fat, which is the main dietary source of saturated fat for most people. Beef also contains some saturated fat, but in dairy, it’s much higher. That higher amount of saturated fat seemed to cause no adverse increases in LDL particle numbers, compared to feeding people higher monounsaturated fats, such as olive oil. And in our 2006 study, the blood work was better when feeding people higher fat diets, than when feeding them lower fat and more carbohydrates. But keep in mind, this was in the setting of lower carb and a mixed protein diet, proteins from various sources from white meat and dark meat and chicken and fish and beef.
Tofu
RON KRAUSS
Tofu. It was just a mixed diet all together. In that setting, with carbohydrate intake kept moderately low, saturated fat did not raise Apo-B. It didn’t raise the number of LDL particles. It didn’t increase inflammatory markers either. It didn’t raise any of the really meaningful basis of heart disease risk.
So that was an interesting study which showed that eating more saturated fat does not increase heart disease risk. But then, there’s that newer study you’ve done that involves saturated fat and red meat. And it’s a fascinating study because of some clues it gives about how health may be affected by both saturated fat and red meat. Right now there’s a great deal of concern that eating red meat may be dangerous for people’s health. But the question is why. In your recent study, you hint at a reason why.
RON KRAUSS
We published a paper this past fall in the Journal of Nutrition, in which we reported the results of the study that we carried out as a followup to the one we just discussed. Now, in the interest of full disclosure, I have to say that the first study was funded the National Dairy Council, and we used fairy fat and dairy products liberally in that study, since they’re high in saturated fats. The second, more recent study was funded by the National Cattleman’s Beef Association because they felt, and frankly we felt at the time, based on the evidence we had, that feeding a high saturated fat and low carbohydrate intake would have the same benefit on a high beef diet as as on a mixed protein diet, and bottom line is that when we did the study, we found out that was not the case.
So using what you learned from your 2006 study of a mixed-protein diet and high saturated fats, in this new study, you kept carbohydrates somewhat low, and fats somewhat higher, just as you did in 2006. Really, the main difference was that this time, you didn’t feed a variety of protein sources. Your test subjects just ate lots and lots of beef. And this time, you found that “healthy” blood work depended not only on what kind of protein people ate, but what kind of fat the people WITH the protein. So if you get out your Sherlock Holmes hat and pipe, what were the clues and what did they mean?
RON KRAUSS
To begin with, keep in mind, this was a very high beef diet. People were eating beef breakfast lunch and dinner. So this is really way outside of what we would ever consider to be a usual health practice. Maybe some people do it. But not many. We were really interested in the metabolic impact of this diet. To get as many clues as possible, we fed people in either the context of lean beef alone, or with extra saturated fat, mostly from diary products. Again, that’s because most of the saturated fat we get in our diet comes from dairy products. There’s some saturated fat in beef, but more in dairy fat. To make things as clear-cut as possible, in this study, we fed the same beef product to two groups. Lean beef, low in fat, without any added saturated fat. For one group, we added lots of dairy fat, to increase saturated fat. For the other group, we kept saturated fat low, but kept total fat basically the same by using an unsaturated fat–basically olive oil. So between the two groups, let’s say the difference was the equivalent of a cheeseburger versus a lean hamburger dressed with olive oil. That sort of describes, in a nutshell the kind of differences we were looking for. When we did blood work on the groups, the group who ate lots of beef with low saturated fat, meaning the olive oil, didn’t seem to have any adverse effects.
Meaning the blood work you did on that group didn’t reveal an increase in the LDL particles and other biomarkers that indicated heart disease risk.
RON KRAUSS
We didn’t see any adverse effect if we just fed a high beef diet in the absence of saturated fat.
So, in your study, a lean burger without cheese looked on the surface as the one to do.
RON KRAUSS
That’s true from our study. However, keep in mind, there may be other effects from eating red meat that still might mean it’s something to limit in the diet. After all, there are studies from the epidemiology world that are very convincing pointing to red meat itself as associated with many disease outcomes ranging form cancer to heart disease to diabetes. In our study, our measurements were strictly focussed on the metabolic risk factors for heart disease and diabetes. In that setting we couldn’t detect a real significant signal. That doesn’t mean there aren’t other adverse effects being revealed by other techniques. But in our case, we analyzed total cholesterol and the number of LDL particles and blood sugar measurements and inflammation measurements, and we didn’t see anything particularly dangerous occurring when people ate lots of red meat but kept the saturated fat very low.
What’s interesting is what the punchline is going be be. So keep going!
RON KRAUSS
The punchline is that we expected that because these diets have low carbohydrate, when we fed the high saturated fat level along with the red meat, we would see a pretty benign metabolic risk profile. Just as we did with low saturated fat and red meat.
But that’s not what happened!
RON KRAUSS
This is one of the surprises that keep life interesting for us for us as researchers and also for the world out there who happens to be looking over our shoulder. In this case, the surprise was that the combination of the high beef diet and the high saturated fat diet caused very serious increases in all of the cholesterol related risk factors we had been measuring, including total particle numbers, small LDL, total LDL cholesterol, inflammation, whatever we looked at, we saw an adverse effect.
Everything went wrong.
RON KRAUSS
This was in contrast with our earlier studies where the same amount of saturated fat and very similar carbohydrate intake but a diet not loaded up with red meat, had no adverse effect even if it had lots of saturated fat in it.
So if you put on your Sherlock Holmes hat and get our your magnifying glass, what is it you found was the likely smoking gun here?
RON KRAUSS
I wish I could give you a definite answer to the question. But fortunately, the fact that we had these two very different results with two different kinds of protein led us to propose to the National Institutes of Health, one of the nation’s leading funders of health research, which hopefully will stay that way, a new study. We are glad that it’s being funded by the National Institutes of Health, as a neutral ground if you will, between the world of sponsors from the food industry. NIH is allowing us now to investigate in a detective-like matter what is going on, and also test directly in a head to head manner, three different diets. All the diets will have the same amounts of saturated fat. But one is high in red meat, one is a non-meat, vegetarian diet where the protein comes from vegetable sources, and one is an intermediate diet, with meat from chicken, primarily. We’re doing this study to determine whether the source of protein influences the response to saturated fat in the way we suspect it will from these earlier results, and we’ll do some blood work that will allow us to investigate possible underlying mechanisms.
You think there are clues in the beef.
RON KRAUSS
Perhaps some other component of beef, not necessarily the protein, but something that comes along with it, such as iron would be one example, that may have an adverse effect in conjunction with saturated fat.
In a detective novel, the writer often throws in a red herring, and a red herring is designed to take someone off the track. But in your case, the color red, which is caused by the iron in meat may actually be the smoking gun.
RON KRAUSS
That’s one possibility.
Why would the iron in red meat, when coupled with saturated fat, increase risk factors for inflammation, for small particle LDL, for higher blood sugars, and all the rest?
RON KRAUSS
This is just an idea. This is in our whole discussion the one area where we do’t have data . . . yet. But it’s an intriguing hypothesis. Because it’s known from genetic, metabolic and population studies that the iron content of the liver, which stores most of our iron for our needs for red blood cell production and many other metabolic processes . . . if that amount of iron within the liver is excessive, it can lead to impaired sugar metabolism and predispose even to diabetes in the extreme case. High levels of iron in the liver also can be associated with abnormal lipid profiles, with higher amounts of small particle LDL. So we already have evidence that high iron levels in the liver are a potential determinant of things that can influence risk for diabetes and heart disease. And since iron comes from what we eat, it could be related to the dietary response that we studied. Why beef should have a particularly high effect on hepatic iron is a possibility we’re interested in, along with the question of why beef would create this risk for us, in particular, when we consume saturated fat. So while we don’t have data yet, we do have an intriguing collection of clues that maybe would lead to definitive conclusions from Sherlock Holmes.
You’re still awaiting the results from your study, though you have some suspicions.
RON KRAUSS
It turns out that heme iron, the form of iron in red meat, is absorbed pretty efficiently into the body. That’s why red meat is considered a good, or maybe in this case, we should just say, an abundant, source of dietary iron. But even though it’s more bioavailable, heme iron still requires certain factors for absorption. In checking to see why saturated fat could potentially increase the amount of iron coming in from beef, what I discovered buried in the literature is that certain kinds of saturated fat, beef tallow being one of them, and the saturated fat called stearic acid being another, both promote the absorption of heme iron. And both saturated fat and stearic acid are found in dairy products and in the fat that comes with red meat. So it’s our hunch that the combination of eating both an abundant source of iron and the fat that helps the body absorb that iron, might be what converts this style of eating into a dangerous risk profile that raises small particle LDL, blood sugars, inflammatory markers, and so on.
And that risk might also be a consequence of modern day living. Looking back in time, people used to be in wars and accidents more often, and during those times, they had more parasites in their blood that would be going after the iron in the body. They might have been bleeding out more of their iron stores, and more vulnerable for iron deficiencies. So we may have evolved to conserve iron, and way back then, it may be that eating foods that promoted iron uptake was actually protective. But these days we don’t have the parasites to sip on our iron stores, and we don’t have the blood bleeding out of us as much. So Ron Krauss, to make up for our cleaner, more peaceable times, if someone likes cheeseburgers and they like them nice and juicy and covered with high saturated fat cheese, should they reduce their iron stores by giving blood more often?
RON KRAUSS
(LAUGHS) You’ve asked a question that has come up in other contexts as well. Does keeping people’s iron level low reduce risk of heart disease? For instance, there are people who claim that women who have a natural protection from heart disease when they’re menstruating lose that protection after they go through menopause, because without that monthly release of blood, their iron levels can build up.
So, does giving blood and reducing your iron stores that way reduce heart disease risk?
RON KRAUSS
We don’t know. The real way to check this would be to not focus on the iron in someone’s diet, but instead, do blood work to reveal how much of the potentially dangerous iron you have in your system. That would be an interesting study to do sometime. As for now, there’s absolutely no basis yet for therapeutically reducing iron stores, for instance, by giving blood, as a means of reducing heart disease risk. The best way to reduce your risk, right now, is to stay away from the cheeseburger diet.
But to get the answer through blood work might be challenging. It’s very hard to measure with any accuracy what the heme level of iron is in the liver. It’s a different measurement than the standard “annual checkup” measure of iron in the blood. Getting a good measurement of the liver’s iron . . . . It’s a very very hard bit of data to get.
RON KRAUSS
There are markers . . . and we haven’t measured all of them yet. Part of the funding we’re getting from the new NIH grant will allow us to measure some of these more accurate markers. So it may be possible for us to tease out more information. The standard iron tests in the blood are not sufficient. That’s where we need more information, and why we’re doing this new study.
33 comments for “Ron Krauss – Saturated Fat? Red Meat? It Depends . . .”