The truth about Fat: its not what you think
YOU CAN’T TALK ABOUT CHOLESTEROL WITHOUT ALSO TALKING ABOUT FAT, which is convenient, because it’s exactly what we’re going to discuss in this post.
When you’re done reading it, you may have an entirely different perspective on fat and a much more accurate notion of what the terms “good fat” and “bad fat” mean. And no, we’re not just going to tell you the stuff you’ve heard a million times, such as “fat from fish is good” (completely true) and “saturated fat is bad” (very far from always true).
But let’s not get ahead of ourselves.
According to conventional wisdom, fat and cholesterol are the twin demons of heart disease, linked together in our minds as firmly as Hell and Damnation or Bonnie and Clyde. We’ve been admonished to lower our cholesterol and stop eating saturated fat. These two mandates are the basis of the diet–heart hypothesis, which has guided national health policy on healthy eating for decades and basically holds that fat and cholesterol in the diet are a direct and significant cause of heart disease.
Okay, so fat and cholesterol (whether they show up in your diet or in your bloodstream) are pretty much kissing cousins.
We’ve discussed cholesterol in the previous post, so let’s clear up some misconceptions about fat—what it is, what it does, what it doesn’t do—and why all this matters in the first place. Once we’ve done that, we’ll be able to look at the relationship among heart disease, fat in the diet, and cholesterol in the blood with completely new eyes.
Let’s get to work!
WHAT EXACTLY IS FAT, ANYWAY?
Fat is the collective shorthand name given to any big collection of smaller units called fatty acids. You can think of “fat” and “fatty acids” as analogous to paper money and a bunch of coins. The dollar bill is the “fat” and the coins are the “fatty acids.” Just as a dollar can comprise different combinations of coins—one hundred pennies, four quarters, ten dimes, twenty nickels, and so forth—a “fat” comprises different combinations of fatty acids.
There are more fatty acids in a big fat blob of butter than there are in a spoonful of butter, just as there are more coins in $5 than there are in $1, but whether you’re dealing with a spoonful of butter, a tub of lard, or a tablespoon of fish oil, all fat on earth is composed of fatty acids. The only difference between the fat in olive oil and the fat in lard is that if you looked at them under a microscope, you’d see that each is made up of a different mix of fatty acids (i.e., nickels, dimes, quarters, etc.).
There are three families of fatty acids: saturated fatty acids, monounsaturated fatty acids, and polyun-saturated fatty acids. (There’s actually a fourth class of fatty acids called trans fats, a kind of “Franken-fat,” but we’ll address that later.) In this section we’ll concentrate primarily on saturated fat, but keep a place on your dance card for two members of the polyunsaturated family called omega-3 fatty acids and omega-6 fatty acids. They’re of special importance, and we’ll be talking about them in depth later on.
Now a word of complete candor from your authors. We wrote this category and website for our families. We wanted the average intelligent person who didn’t have a background in science to be able to follow the basic arguments and have a clear sense of the takeaway messages. We wanted the discussions within the website to be simple enough that they could be easily grasped by nonmedical people. And, frankly, fat is complicated.
So this is the part of the website where we could easily slip into a short course on the biochemistry of fats. It’s interesting to write about, it fills a lot of pages—and it’s deadly dull for readers. Don’t worry, we’re not going to write about the chemical structure of fat, and here’s why. What makes one fatty acid “saturated” and another “unsaturated” has to do with fairly intricate details of fat architecture and composition that, frankly, most folks couldn’t care less about. (If you’re really dying to know, it has to do with the number of chemical double bonds that exist in the fatty acid’s molecular chain. Mono-unsaturated fats have one double bond. Polyunsaturated fats have more than one. There. Now you know.)
And as much as we enjoy talking about this stuff and would be happy to chat about it if you met us at a cocktail party, the truth is it causes many people’s eyes to glaze over pretty quickly. So if you’re interested in reading about double bonds, saturation, chain length, and other cool biochemical stuff, please, by all means, be our guest! That information is widely available. It’s not controversial, it’s not debated, and it’s not really germane to our story. So, mercifully, we’ve decided to forgo it here and instead give you the big picture—what you really need to know about saturated, polyunsaturated, and monounsaturated fats.
SATURATED FAT 101: EVERYTHING WE LEARNED WAS WRONG!
Saturated fats are primarily found in animal foods (meat, cheese, butter, eggs) and, less often, in certain plant foods, such as coconut, coconut oil, and palm oil. They tend to be solid at room temperature (think butter) and soften when warm.
WHAT YOU NEED TO KNOW
• Saturated fat has been wrongfully demonized.
• Saturated fat raises “good” (HDL) cholesterol.
• Saturated fat tends to change the pattern of your “bad” (LDL) cholesterol to the more favorable pattern A (big, fluffy particles).
• Several recent studies have shown that saturated fat is not associated with a greater risk of heart disease. One study from Harvard concluded that “greater saturated fat intake is associated with less progression of coronary atherosclerosis, whereas carbohydrate intake is associated with a greater progression.”7
• In the Nurses’ Health Study, refined carbohydrates were independently shown to be associated with an increased risk for coronary heart disease.
• Omega-6 fats—e.g., vegetable oils—are pro-inflammatory.
• The balance between omega-6 and omega-3 is far more important than saturated fat intake is.
• Low-fat diets work because they reduce omega-6 fats, not because they reduce saturated fat.
Chart by Michelle Mosher.
They also have a few other characteristics worth mentioning. Saturated fats are very stable. They’re tough—when exposed to high heat they don’t “mutate” or “damage” as easily as their more delicate cousins, the unsaturated fats do. That’s one reason why lard (with its high concentration of saturated fatty acids) is actually a better choice for frying than the cheap, processed vegetable oils that gradually replaced it as restaurants tried to be more health conscious.
The problem with vegetable oils is that they’re nowhere near as resistant to damage as saturated fats are. When you heat and reheat them for frying, as virtually every restaurant in America does, it causes the formation of all sorts of noxious compounds, including carcinogens. Compared to saturated fat, the unsaturated fatty acids in vegetable oils are much more easily damaged by high heat and more susceptible to oxidation and the production of free radicals. Those vegetable oils transform into all sorts of mutant molecules under the stress of high heat and reheating, but when high heat is applied to saturated fat, it behaves like the strong, silent uncle at the family gathering; everyone else is going nuts, but he’s calm and serene! (We’ll talk about some of the other problems with the overuse of vegetable oils in our diet later on.)
Now let us ask you a question, and please answer honestly: Did you just shudder in horror when we implied a few sentences ago that using lard for cooking might actually be a good idea? You probably thought to yourself, “Now they’ve gone too far. Did they really say lard is better to fry with than canola oil? That’s nuts!”
We’d be surprised if you didn’t recoil in horror. Most people would do just that—and it’s because most people have totally bought into the idea that saturated fat is the worst thing on the planet.
The idea that lard—with its high content of saturated fat—could ever be a better choice than those high-omega-6 vegetable oils that are continually pushed on us is in direct opposition to fat theology, the deeply held belief that saturated fat and cholesterol are the root of all heart disease evil. That notion has been the prevailing dogma about saturated fat, cholesterol, and heart disease for decades. By now you’re more than familiar with this notion, known as the diet–heart hypothesis—it’s the mantra that has guided public policy on diet and heart disease for virtually every major governmental and mainstream health organization, such as the American Heart Association.
There’s only one problem.
It isn’t true.
Despite its horrible reputation, saturated fat is far from a dietary demon. More and more health professionals, researchers, scientists, doctors, and nutritionists are beginning to reexamine the case against saturated fat, and they’re finding that it’s based on very little solid evidence (and a lot of guilt by association).
Saturated Fat and Heart Disease: Where’s the Evidence?
Look, there is no shortage of studies pointing to an association between increased saturated fat intake and cardiovascular risk, but there are a few things to know about those studies.
When I was in fifth grade back in Queens, New York, there was a kid named A.J. who was always, and I mean always, getting in trouble. But it was for the most minor stuff: coming in a couple of minutes late from recess, whispering in class, or, worst case scenario, throwing a spitball. There could be five other kids doing the same thing, but A.J. would always be the one to get caught. Singled out, reprimanded, parents called in to school, the whole humiliating deal.
But there were a couple of other kids in the class who were real pieces of work. One kid, Gilbert, compulsively lit firecrackers, scaring everyone to death, and then disappeared before he could be caught at the scene of the crime. Another kid named Howie took delight in breaking people’s windows with rocks. A third one, Corky, was a bully. And yet none of them ever managed to get caught. Rarely did any of these kids even get a stern talking–to. The role of the “bad kid” in the class was played by A.J., who would have to serve detention, sit in the corner, and be yelled at in front of the class, all for fairly meaningless infractions, while the kids who were doing all the really bad stuff got off scot-free.
Now it’s not that old A.J. didn’t do anything wrong. But unlike the other kids, he never beat anyone up, he never did anything mean, he never destroyed anyone’s property—and yet whenever there was trouble, he was always the scapegoat.
I think saturated fat is like that kid A.J. It’s not that it’s perfect. It’s just that it’s far less important than the stuff we ignore—such as high intakes of omega-6 fatty acids, low intakes of omega-3s, and obscene intakes of sugar and processed carbs.
Is saturated fat so wonderful that we should all resolve to melt a ton of butter and add it to our smoothies right this minute? No, of course not. Saturated fat has some negatives. It is mildly inflammatory. It may contribute to insulin resistance.
If the dietary dictocrats are going to warn us against inflammatory food components, why choose saturated fat, a relatively minor factor in inflammation compared to the omega-6 to omega-3 ratio? If they’re going to warn us about saturated fat because of its purported connection to insulin resistance, why do they continue to promote ridiculously high carbohydrate intakes, which are demonstrably worse?
Saturated fat is a lot like A.J. Not perfect, but it doesn’t deserve to get beat up. And the irony is that while everyone’s pushing him around and blaming him for everything bad that happens, the real culprits are getting away.
A WORD ABOUT META-ANALYSES AND WHY THEY’RE IMPORTANT
A little backstory about meta-analyses and why people do them. Say you want to learn about the sex habits of college students. There are probably a couple dozen relevant studies you could look at, but as with any other area of research, there’s no guarantee that all the studies will reach the same conclusions. In fact, it’s almost certain that they won’t. One study might find, for example, that college kids are having more sex, while another study might find that they’re actually having less. (A critical look at these two studies might uncover the fact that researchers in the two studies used slightly different definitions of the term “sex” when they surveyed the students, something that might account for the difference in results.)
Sometimes researchers overlook an obvious variable that could skew the results. Although researchers always try to control for these variables (such as age, sex, and smoking) and generally “match” subjects by the most important criteria, they don’t—they can’t—always control for every variable that might make a difference (and this is particularly true in diet research). The point is, if you look at anything worth studying you’re going to find a whole bunch of research on it, and among those research studies you’re almost guaranteed to encounter conflicting findings and areas of disagreement about how to interpret those findings.
Even something that now seems as clearly connected as the link between smoking and cancer started out as a hypothesis and had to be tested in all sorts of populations under all sorts of conditions. Studies can and do reach different conclusions depending on the statistical measures used, the populations studied, and even the definition of terms. (Is a “smoker” defined as anyone who has even one cigarette a week? Or is a “smoker” defined as someone who smokes at least half a pack a day?)
Which brings us, finally, back to meta-analysis.
Sometimes researchers gather up a whole bunch of these individual studies whose results are clustered all over the place like pins on CNN’s election maps. Then they’ll ask, “What do these studies, taken together as a whole, really tell us about what’s going on?” They’ll gather up all the studies on, say, smoking and cancer, college students and sex, or saturated fat and heart disease. They’ll examine them scrupulously, tossing out any studies whose methods, designs, or data don’t meet the highest standards of research excellence. (Meta-analyses typically exclude small pilot studies, unblinded studies, studies with too few participants, or studies that do not collect data on something the researchers consider important.)
Once the “best-of-the-best” studies are selected for inclusion (and lesser studies are eliminated), the researchers go to work and apply every statistical manipulation you can imagine to tease out the real relationships from the mass of accumulated data. They look at the findings of the individual studies and compare them. They pool the subjects from all the studies. They look for trends, directions, statistical significance, and hidden relationships. And though meta-analyses themselves are not infallible, they’re a great way to look at the big picture to gauge what’s really going on.
Number one, the associations are far weaker than one might suspect, given how entrenched the belief is that saturated fat clogs your arteries. In many of these studies, the major “risk” examined was cholesterol, so we wind up with a circular argument in which higher saturated fat intake increases the risk for heart disease, but only if you accept the use of cholesterol levels as a stand-in for heart disease. Studies that measure the effect of saturated fat on heart disease and mortality directly—rather than indirectly by measuring its effect on cholesterol—are few and far between. But there are some important ones, which we’ll discuss in a moment.
Number two, as scientists have looked more carefully at the association between saturated fat in the diet and levels of cholesterol in the blood, they are beginning to see that even here the relationship is murky. Saturated fat, as we’ve pointed out, does in fact raise overall cholesterol levels, but its effect is still more positive than negative, because it causes HDL levels to go up more than LDL levels. Even more important, saturated fat has a positive effect on the particle sizes of both LDL and HDL, making more of the big, fluffy, benevolent particles and much less of the small, dense, inflammatory particles (such as LDL pattern B and HDL-3). (It’s called shifting the distribution of LDL particles.) And, as we’ve been saying, the particle size of cholesterol molecules is far more important than their sheer numbers. Later, when we examine the twin principles of fat theology, you’ll learn exactly why this is so and exactly why particle size is what we should be looking at.
One of the basic tenants of fat theology is that saturated fat increases the risk of heart disease. In the scientific literature, this issue is as far from being settled as you might think from listening to CNN. Recently, Patty Siri-Tarino, Ph.D., and Ronald Krauss, M.D., of the Children’s Hospital Oakland Research Institute together with Frank B. Hu, M.D., Ph.D., of Harvard, decided to do a meta-analysis—a study of studies. In this case, they looked at all previously published studies whose purpose was to investigate the relationship of saturated fat to coronary heart disease (CHD), stroke, or cardiovascular disease (CVD). Note that this is one of those hard-to-find studies we mentioned earlier: a study of the direct effect of saturated fat on health. The researchers weren’t just interested in the effect saturated fat had on cholesterol—they wanted to know the effect saturated fat had on heart disease. (Remember, they are not the same thing!)
Twenty-one studies qualified for inclusion in their meta-analysis, meaning these studies met the criteria for being well designed and reliable. All in all, the twenty-one studies included 347,747 subjects who were followed for between five and twenty-three years. Over this period of time, 11,006 of the subjects developed coronary heart disease (CHD) or stroke.
Ready for the findings?
How much saturated fat people ate predicted absolutely nothing about their risk for cardiovascular disease. In the researchers’ own words, “Intake of saturated fat was not associated with an increased risk of coronary heart disease (CHD) or stroke, nor was it associated with an increased risk of cardiovascular disease (CVD).” Those folks consuming the highest amount of saturated fat were statistically identical to those consuming the least amount when it came to the probability of CHD, stroke, or CVD. Even when the researchers factored in age, sex, and study quality, it didn’t change the results. Saturated fat did bupkis—it didn’t increase or decrease risk in any meaningful way. Period.
“There is no significant evidence for concluding that dietary saturated fat is associated with an increased risk of CHD or CVD,” the researchers concluded.
Now—and this is a very important point—it’s not that there’s no evidence that saturated fat doesn’t raise cholesterol. There is, and we’ll examine that more in a moment. But the above meta-analysis didn’t just look at cholesterol levels; it looked at what we really care about—heart disease and dying. So never mind whether saturated fat raises my cholesterol level. What I really want to know is, what does eating saturated fat do to my chances of getting a heart attack? The meta-analysis looked at exactly that real-life endpoint we truly care about, and on that all-important metric, it found that saturated fat in the diet has virtually no effect.
That meta-analysis is hardly the only study that has found saturated fat innocent of any direct involvement in cardiovascular disease. In the fall of 2011, a new study came out in the Netherlands Journal of Medicine titled “Saturated Fat, Carbohydrates, and Cardiovascular Disease.” Like the above-discussed meta-analysis, its purpose was to examine the current scientific data on the effects of saturated fat, looking at all the controversies as well as the potential mechanisms for the role of saturated fat in cardiovascular disease.
Here’s what the researchers wrote:
“The dietary intake of saturated fattty acids is associated with a modest increase in serum total cholesterol, but not associated with cardiovascular disease [italics ours].”
As we’ve been saying throughout this website, cholesterol is only used as a marker. (In other words, it’s a stand-in answer for what we really want to know—namely, what is the likelihood of developing heart disease?) But if you’re looking for a metric to predict who is and isn’t going to get heart disease, cholesterol—as we’ve seen on this website—is a lousy choice for a marker. If cholesterol really predicted heart disease (wrong belief number one), and if saturated fat really did terrible things to your cholesterol (wrong belief number two), then that might be reason to eliminate saturated fat from your diet.
But it turns out neither of those two things is true.
Let’s take those two notions one by one, because they are the bedrock beliefs of fat theology.
FAT THEOLOGY: TWO MAIN TENETS DEBUNKED
Researchers in Japan examined the first of those beliefs—that cholesterol is a good predictor of heart disease—with another meta-analysis. They searched for all studies that had examined the relationship of cholesterol to mortality, excluding any done before 1995 and any that had fewer than five thousand subjects. Nine studies met the criteria, but four had incomplete data and so were excluded. The researchers then performed a meta-analysis on the remaining five studies, which together involved more than 150,000 people followed for approximately five years.
The researchers placed everyone into one of four groups depending on their cholesterol levels: less than 160 mg/dL, 160 to 199 mg/dL, 200 to 239 mg/dL, and higher than 240 mg/dL. (These categories mirror the American Heart Association guidelines, which state that 200 mg/dL or lower is “desirable,” 200 to 239 mg/dL is “borderline high,” and higher than 240 mg/dL is bad news indeed.)
Which group do you think would have the worst possible outcomes?
According to everything we’ve heard from the cholesterol zealots, the answer is simple: Those whose cholesterol readings were the highest (240 mg/dL and over), and even those with cholesterol readings in the “borderline” category (200 to 239 mg/dL), should be expected to die at a higher rate than those with a cholesterol level of 160 to 199 mg/dL. And those in the under 160 mg/dL category should live longest of all!
That is precisely and exactly what did not happen.
In fact, the group with the lowest cholesterol levels died at the highest rate.
In scientific terms, the risk for dying from any cause whatsoever (called “all-cause mortality”) was highest in the group with low cholesterol. Compared with the reference group (160 to 199 mg/dL), the risk of dying from any cause whatsoever was significantly decreased in the group having “borderline high” cholesterol of 200 to 239 mg/dL and even further decreased in the group having “high” (greater than 240 mg/dL) cholesterol. In contrast, your risk of dying from any cause was the highest of all if your cholesterol was under 160 mg/dL!
Total cholesterol is so irrelevant as a metric that in 2007 the Japan Atherosclerosis Society stopped using it in any tables related to the diagnosis or treatment criteria in its guidelines.
So high cholesterol is associated with a reduced risk of death? Not exactly what you might expect but exactly what the study found.
Total cholesterol is so irrelevant as a metric that in 2007 the Japan Atherosclerosis Society stopped using it in any tables related to the diagnosis or treatment criteria in its guidelines. It’s not that the society abandoned the cholesterol theory, mind you. It just now relies entirely on LDL levels to determine who should be classified as having “high cholesterol,” reasoning that if total cholesterol is high simply because you’ve got a terrifically high HDL level, that shouldn’t be counted as a bad thing. Many American doctors—even the most conservative ones—would probably agree that the LDL number is the important one, even if they don’t fully embrace the notion that it is the type of LDL—not the LDL number—that matters the most.
But is the LDL level a better predicator of heart disease or mortality than the total cholesterol level?
Once again, let’s go to the videotape.
Researchers in Japan set out to answer this question in something called the Isehara Study. The Isehara Study was based on data collected from annual checkups of residents in Isehara, a smallish city (population: 100,000) located in the central Kanagawa Prefecture in Japan. A database of 8,340 men (average age sixty-four) and 13,591 women (average age sixty-one) was mined for cholesterol readings, and the 21,931 people were divided into seven groups ranked from lowest to highest LDL cholesterol levels (in mg/dL): <80, 80 to 99, 100 to 119, 120 to 139 (reference group), 140 to 159, 160 to 179, and >180.
In both men and women, overall mortality was significantly higher in the group with the lowest LDL cholesterol levels (under 80 mg/dL).
Although it’s true that in this study mortality from heart disease was greater in the group with the highest LDL levels (over 180 mg/dL, which is, admittedly, pretty darn high), this was only true in men. In women the opposite was so—fewer women died of heart disease in the group with the highest LDL levels. In any case, this increase in heart disease in the high LDL group of men was apparently more than offset by the increase in deaths from other causes.
Okay, hopefully this information will get you, and your doctor, to at least question the notion that cholesterol is an important marker or predictor of heart disease. But let’s say for the sake of argument that you, or your doctor, is not quite willing to throw out the cholesterol theory. Fine, no problem. After all, you, like most of us, have been indoctrinated with the idea that anything that raises your cholesterol is bad news, and that’s a hard thing to let go of, especially when you’ve been hearing it for your entire adult life.
But before you go back to demonizing saturated fat, let’s examine the second belief that constitutes the bedrock of fat theology, the idea that saturated fat does really bad things to your cholesterol.
When cholesterol was assessed in the old-fashioned way—”total,” “good,” and “bad”—this idea might have made sense, because a number of studies show that saturated fat does raise total cholesterol and LDL cholesterol. And if you bought into the theory that cholesterol is a big cause of heart disease, this would be a good enough reason to give up the butter. But saturated fat actually raises HDL (“good”) cholesterol more than it does LDL cholesterol, leaving the ratio between total cholesterol and HDL cholesterol—a ratio that’s accepted as a measure of heart disease risk by just about everyone—unchanged or even improved.
If you eat less saturated fat and your cholesterol goes down as a result, your doc may think that’s a good thing and stop looking any further. But that’s the point: You can’t just look at your LDL number and stop there. The reduction in LDLs that you may get from cutting out saturated fat, and the reduction in LDLs that makes everyone jump for joy and celebrate your newfound “health,” comes with a hefty price: a big decrease in precisely the LDL molecules that you want more of—the “good citizen” LDLs, those big, fluffy LDL particles that, when they’re predominant, make up a pattern A cholesterol profile.When the number of big, fluffy particles goes down, the proportion of your LDL population shifts in favor of the nasty, angry, atherogenic, BB gun pellet–type particles, giving them a kind of “majority rule.” Sure, your LDL number will go down and your doctor will be happy, but meanwhile, because of the shift in makeup of your LDL population, your risk for heart disease goes up.
Conversely, when saturated fat intake goes up—and carbohydrate intake goes down—the opposite happens. Now you’ll see a significant shift to more of those big, fluffy, harmless LDL particles and less of those small, dense, angry LDL particles. Your LDL population has just shifted, and the big, fluffy, harmless particles are now in the majority, leaving you in a significantly better place in terms of your heart disease risk. Sure, your overall LDL level may go up a bit, but what’s actually happened is that there are now many more “good citizens” among your LDL population and far fewer “bad” ones. In other words, you’re much better off.
The Carbohydrate Swap
For decades, most health professionals have told us that we’d be doing ourselves a huge favor if we just cut out saturated fat and replaced it with carbohydrates. And that’s exactly what most people did. After all, this idea fit nicely with the prevailing ethos: Saturated fat is bad, and “complex” carbohydrates are good. If we just swap ‘em, everyone will go home happy, and all will be right with the world.
So, as our old friend Dr. Phil might say, “How’s that working for you?”
The answer is, “Not so well.”
One important study shed light on the whole “carbs for saturated fat” swap but raised a lot of eyebrows because of its unexpected results. The study, titled “Dietary Fats, Carbohydrate, and the Progression of Coronary Atherosclerosis in Post-menopausal Women,” was conducted by the distinguished researcher Dariush Mozaffarian and his associates from Harvard Medical School.
As the study title suggests, Mozaffarian set out to investigate how various fats—saturated, polyunsaturated, and monounsaturated—influenced the progression of heart disease in postmenopausal women who ate a relatively low-fat diet. Noting that standard dietary advice has always been to eat less saturated fat, the researchers wondered exactly what terrific things would happen if you replaced terrible saturated fat with other food substances. According to the standard advice, replacing saturated fat with good stuff (e.g., carbs or “good fats” such as vegetable oils) should substantially reduce your risk for heart disease.
Except that it didn’t.
“Greater saturated fat intake is associated with less progression of coronary atherosclerosis, whereas carbohydrate intake is associated with a greater progression [italics ours],” the authors concluded. “Women with higher saturated fat intakes had less progression of coronary atherosclerosis.”
Greater saturated fat intake was also associated with higher HDL levels, higher HDL-2 cholesterol levels, lower triglycerides, and an improved total-cholesterol-to-HDL ratio. Saturated fat, at least in this study, was hardly the dietary demon it’s been made out to be.
And if this were not a knockout punch by itself, consider what was associated with a greater progression of coronary atherosclerosis.
Are you sitting down?
Especially the high-glycemic, processed variety of carbohydrates, which is exactly what we tend to eat when we replace saturated fat in the diet with so-called “complex” carbs such as breads, pasta, rice, and cereal.
“The findings also suggest,” wrote the researchers, “that carbohydrate intake may increase atherosclerotic progression, especially when refined carbohydrates replace saturated or monounsaturated fats.”
“Wait a minute,” you might well say. “When I take the saturated fat out of my diet and replace it with high-glycemic carbohydrates I’m actually increasing my risk for heart disease?”
Dr. Sinatra: The Case Against Canola Oil
Back in 1997, I wrote an article for Connecticut Medicine about oxidized LDL and free radicals. I was very gung ho about canola oil at the time—as were most of my colleagues—and I was emphatic in my recommendation of it.
But the paper was rejected.
A Yale professor of medicine who was on the peer review board—a biochemist, in fact—reviewed the paper and nixed it for publication. But he was kind enough to suggest some review articles on canola oil in the literature.
I read them.
My reaction: “What have I been smoking all these years?”
The success of canola oil and its reputation as the healthiest of oils is a triumph of marketing over science. It’s a terrible oil. It’s typically extracted and refined using very high heat and petroleum solvents (such as hexane). Then it undergoes a process of refining, degumming, bleaching, and—because it stinks—deodorization using even more chemicals. The only kind of canola oil that could possibly be okay is organic, cold-pressed, unrefined canola oil, and hardly anyone is using that.
Our friend Fred Pescatore, M.D., bestselling author of The Hamptons Diet and former medical director of the Atkins Center, is something of a cooking oil expert. Here’s what he had to say about canola oil: “I would never use this stuff!”
If you’d like to read more about the dark side of canola oil, check out the definitive paper by lipid biochemist Mary Enig and Weston A. Price Foundation president Sally Fallon. Widely available online, it’s called, tellingly, “The Great Con-Ola.”
As for my 1997 paper, I revised it, removing the recommendation to use canola oil. The paper was accepted and published.
Dr. Jonny: Good Carbs, Bad Carbs
Whenever I give a talk about healthy eating and I mention that a diet very high in carbohydrates is problematic for most people, I’m very careful to add the caveat: “I’m not talking about fruits and vegetables!” So here’s a quick cheat sheet on “good” versus “bad” carbs.
Good carbs include the following foods:
• Beans and legumes
Bad carbs, which cover almost all carbs that come in a box with a bar code*, include:
• White rice
• Snack foods
• Juice drinks
* There are exceptions in the categories of cereal and bread, but they are few and far between. Oatmeal is one example (but not the instant kind). Ezekiel 4:9 bread is another. But by and large if you stay away from most of the foods on the above list—or keep them to an absolute minimum—you’ll be much better off healthwise.
By the way, Mozaffarian and his research team didn’t just look at cholesterol. They looked at actual clinical events, such as heart attacks and deaths, from any type of cardiovascular disease. They also looked at lesser known metrics that only your doctor will appreciate (such as coronary revascularization and unstable angina).
Bottom line: Greater saturated fat intake didn’t increase the risk for any of them.
Vegetable Oils: Myths and Myth-Conceptions
The researchers also tested what happens when you replace saturated fat with polyunsaturated fat (such as vegetable oils), the conventional dietary advice given by just about every major health organization. Maybe high-sugar carbs aren’t so good for us after all, but what about the much-touted vegetable oils, which contain the “healthy fat” our doctors keep telling us about? Swapping saturated fat for a nice helping of healthy vegetable fat has got to be just the ticket to heart health, right?
So the researchers looked at the effect of replacing saturated fat with polyunsaturated fat. Just for fun, they also took a look at what happens when you swap carbs for polyunsaturated fat.
When carbs were replaced with polyunsaturated fat there was no change in atherosclerotic progression—in terms of heart disease risk, it was a wash. But when saturated fat was replaced with polyunsaturated fat, there was a big change—but not in the expected direction. Replacing saturated fat with polyunsaturated fat actually led to an increase in the progression of coronary atherosclerosis (This seemingly crazy finding will make a lot more sense when we discuss those special classes of polyunsaturated fat mentioned earlier in the post, omega-3s and omega-6s. Stay tuned.)
If you’re confused by these findings, you’re hardly alone. The American Journal of Clinical Nutrition devoted an entire editorial to the findings titled “Saturated Fat Prevents Coronary Artery Disease? An American Paradox.” But it’s only a paradox if we refuse to question the bedrock belief of fat theology that saturated fat consumption increases the risk for heart disease. The research is showing that it does not.
We worry deeply about the wholesale, unqualified recommendation to reduce saturated fat at all costs, because it invariably means that people will replace it with processed carbohydrates. That switcheroo is just about guaranteed to both reduce HDLs and increase triglycerides, and if you’re trying to prevent heart disease, those are very bad outcomes indeed. In the Nurses’ Health Study, for example, refined carb-ohydrates and their high glycemic load were independently shown to be associated with an increased risk for coronary heart disease.
We worry deeply about the wholesale, unqualified recommendation to reduce saturated fat at all costs, because it invariably means that people will replace it with processed carbohydrates.
GLYCEMIC INDEX AND GLYCEMIC LOAD
Glycemic index is a measure of how quickly a given amount of food raises your blood sugar (and keeps it elevated). Glycemic load is a related (and more accurate) measure of the same thing. High-glycemic foods—such as most white breads, white rice, and cereals—are simply those that send your blood sugar on a roller–coaster ride. Low-glycemic foods include most fruits and vegetables as well as beans and legumes.
Now don’t misunderstand us. If you wanted to swap some saturated fat out of your diet and trade it for some low-sugar, high-fiber, nutrient-rich carbohydrates, such as Brussels sprouts or kale, no one would complain. Substituting saturated fat with low-glycemic carbs such as vegetables doesn’t increase the risk of heart attacks at all, but substitution of saturated fat with high-glycemic carbs does—by a fair amount, actually. A study in the American Journal of Clinical Nutrition found that replacing saturated fats with high-glycemic index carbs was associated with a 33 percent increase in heart attack risk. Because most people replace saturated fat with exactly these kinds of processed, high-glycemic (high-sugar) carbs (e.g., breads, cereals, and pasta), the conventional wisdom to cut out saturated fat and consume lots of carbs instead is starting to look like an increasingly boneheaded notion. Although it’s not perfect, saturated fat does a number of good things in the body. Its wholesale replacement by the worst kind of carbohydrates is turning out to be a cure worse than the disease.
A recent Dutch study added to the list of accumulating research showing that when you substitute high-glycemic carbohydrates for saturated fat you actually increase cardiovascular risk. But the Dutch researchers had an interesting take on this, one that appreciates that an accumulation of saturated fat in the body is not necessarily the best thing in the world.
They pointed out that eating a high amount of carbs causes your body to hold on to the saturated fatty acids that you’re also consuming—and those saturated fats get preserved, stored in your body rather than burned for energy. Meanwhile, all those extra carbs you’re eating get converted into more saturated fatty acids in the liver. Now you’ve got a serious excess of saturated fatty acids—you’re holding on to the ones you’re eating, and your liver is creating even more of them, fueled by the carbs you’re consuming. Because large amounts of saturated fat can lessen the anti-inflammatory actions of HDL cholesterol, this isn’t a good situation.
However, the Dutch researchers correctly noted that cutting saturated fat out of the diet is not the most effective way to combat the accumulation of saturated fatty acids in the body. It’s far better, they suggested, to reduce dietary carbohydrates. This way, your body makes fewer saturated fatty acids, and its tendency to hold on to those you do eat is reduced. “Attention should be shifted from the harmful effects of dietary saturated fat per se to the prevention of the accumulation of saturated fatty acids (in the body),” the authors wrote. “This shift would emphasize the importance of reducing dietary carbs, especially carbs with a high glycemic index, rather than reducing dietary saturated fat.”
Carbohydrates have a nasty effect on cholesterol particle size, which, as you’ve seen, is of significantly greater importance than total cholesterol, LDL, or even HDL. Two researchers from the Department of Atherosclerosis Research, part of the Children’s Hospital Oakland Research Institute in California, decided to test the effect of dietary carbohydrates on the size and density of both LDL and HDL. They found that people who ate more carbohydrates—particularly simple sugars and starches with a high glycemic index—had significantly greater levels of those angry, dense, atherogenic particles of LDL (pattern B). They also had the greatest number of small, dense HDL particles.
Fat in the Diet: Our Perspective
We want to propose a different way of looking at fat intake. We think what we are about to suggest goes a long way toward explaining the contradictory findings, or apparently contradictory findings, on saturated fat, diet, fat reduction, and cardiovascular disease.
To do this, we have to briefly introduce the other two categories of fats besides saturated: monounsaturated fats and polyunsaturated fats. (Remember, all fatty acids fall into one of these three broad categories.)*
Monounsaturated fat is the fat that’s predominant in olive oil (as well as in nuts and nut oils, such as macadamia nut oil). Its health benefits have been well documented and are noncontroversial. Monounsaturated fat is the primary fat consumed in the highly touted Mediterranean diet, and it’s generally accepted that this kind of fat is perfectly healthy. For that reason, we won’t spend much time on it, because it is pretty irrelevant at this point to the case we’re about to make.
The real action is with polyunsaturated fats.
Remember, polyunsaturated fats, which are primarily found in vegetable oils, are the very ones we’ve been admonished to include more of in our diets. When lard was slammed back in the early part of the twentieth century, the health dictocrats started their cheerleading effort for vegetable fats. (The first major beneficiary of this all-out campaign to make vegetable fats synonymous with “healthy” fat was actually the trans fat–laden Crisco, the most popular vegetable shortening of its time.) Even now, most people believe that substituting vegetable oil for animal fats is universally a good thing.
But is it always?
Let’s, as they say, go to the videotape.
Polyunsaturated fats as a whole are divided into two subcategories: omega-3 fatty acids and omega-6 fatty acids. (For those who’ve always wondered what the heck an “omega” is anyway, you can think of the terms omega-6 and omega-3 as real estate terms; they’re simply descriptions of the location of certain chemical structures—called double bonds—within the fatty acid. An omega-3 has its first double bond at the third carbon atom in the chain, while omega-6 has its first double bond at the sixth carbon atom in the chain. Now, for our purposes, you can promptly forget all that and just concentrate on what these two types of fatty acids—omega-3s and omega-6s—actually do in the body.)
Omega-6s, as mentioned, are found primarily in vegetable oils and some plant foods. Omega-3s are found primarily in fish, such as salmon, and certain animal foods, such as grass-fed beef, as well as in some plant foods, such as flax and flaxseed oil. So far, so good.
Here’s where it gets tricky.
Both inflammatory and anti-inflammatory hormones, known as eicosanoids, are made in the body from polyunsaturated fats. (And to answer the inevitable question, yes, we actually need both. Inflammatory compounds are a necessary part of the immune system and play a big part in the healing process when you have a wound or other type of injury.)
Omega-6s are the precursors to the inflammatory compounds in our body—they’re the building blocks the body uses to make these inflammatory hormones (specifically series 2 prostaglandins). And omega-3s have the opposite function: The body uses omega-3s as building blocks for the anti-inflammatory compounds (known as series 1 prostaglandins and series 3 prostaglandins).
A ton of research has established that the ideal ratio of omega-6s to omega-3s in the human diet is somewhere between 1:1 and 4:1. This seems to be the best balance to keep inflammation in check and everything running smoothly. It’s the ratio found in the diets of both hunter-gatherers and healthy indigenous societies where heart disease is rare.
But the ratio of omega-6s to omega-3s in Western diets is anywhere from an astonishing 15:1 to an even more astonishing 20:1 in favor of omega-6s. If you think of the inflammatory and anti-inflammatory hormones as two armies that work together in the body to create balance in the body, that means we’re overfunding the inflammation army by 1,500 to 2,000 percent!
The Law of Unintended Consequences
Our extraordinarily high intake of vegetable oil has another unintended consequence, and one that may have a profound effect on cardiovascular health. To understand it, though, you have to take a short excursion into the world of omega-3 fatty acids. (Trust us, it’s a short and easy trip.)
You see, there are actually three omega-3 fatty acids—ALA (alpha-linolenic acid), EPA (eicosapentaenoic acid), and DHA (docosahexaenoic acid). The only one that is “essential” in the diet is ALA, which is found in green, leafy vegetables and in flaxseeds, chia seeds, perilla seeds, and walnuts. That doesn’t mean the other two aren’t important. In terms of their overall effects on human health, the other two are probably more important than ALA. The reason the other two—EPA and DHA—aren’t considered “essential” is that scientists use the word essential in a different way than regular people use it in ordinary conversation. In this context, essential simply means that it’s something the body can’t make, so you have to get it from your diet. Your body can make EPA and DHA, so technically they’re not classed as “essential.” Because the body can’t make ALA, however, it’s considered an “essential” omega-3.
But the fact that the body can make EPA and DHA from ALA doesn’t mean it does a particularly good job of it. It converts the ALA from the diet into EPA and DHA using enzymes and a complicated series of operations known as elongation and desaturation, the success of which is influenced by many different factors, including the amount of inflammatory omega-6’s in the diet. Even under the best of circumstances, only a small amount of ALA successfully gets converted into the very critical EPA and DHA.
Omega-6s and omega-3s compete for the same enzymes, and when omega-6 intake is very high, it wins the competition by default. A high intake of omega-6 reduces the conversion of ALA into EPA and DHA, which might be another reason why high omega-6 diets contribute to heart disease. So not only are those omega-6 fatty acids pro-inflammatory on their own, but they also reduce the body’s ability to produce two of the most anti-inflammatory substances on the planet: the omega-3s EPA and DHA. It’s a double whammy, and your heart is the loser.
No, the omega-6s that have been the darling of the high-carb, low-fat movement, the vegetable oils we’ve been told to use instead of animal fats—the very vegetable oils that “saturate” (no pun intended) our diet through their incorporation into virtually every baked, fried, and processed food available in the supermarket, the very vegetable oils that restaurants proudly boast of using because they’re so “healthy”—are actually turning out to be as bad as, or worse than, the original saturated fats (such as lard) that they replaced, just as margarine turned out to be far worse than butter.
The vegetable oils we’ve been told to use instead of animal fats are actually turning out to be as bad as, or worse than, the original saturated fats (such as lard) that they replaced, just as margarine turned out to be far worse than butter.
For example, the primary omega-6 fatty acid—linoleic acid—has been shown to increase the oxidation of LDL cholesterol, thus increasing the severity of coronary atherosclerosis. One research study showed that a diet enriched with linoleic acid increased the oxidation of the small, nasty LDL particles, precisely the cholesterol particles that are most dangerous and most involved in the formation of arterial plaque. Omega-6s even inhibit your body’s ability to fully incorporate the EPA you get from fish or fish oil supplements into the cell membranes, which is meaningful because EPA is the omega-3 that has the most profound effect on the heart.
Published values for omega-6 intake closely track observed coronary heart disease death rates for all sorts of populations worldwide. And in the famous MRFIT study, subjects with the lowest ratio of omega-6 to omega-3 (i.e., those with the lowest intakes of omega-6 relative to their omega-3 intakes) had the lowest death rate.
The Paradox of the Ultra-Low-Fat Diet
At this point you may well be wondering why low-fat, high-carb diets work at all when they do work. If saturated fat is not the bad guy we thought it was, and if carbohydrates aren’t always the good guys, why is it that some of these high-carb, super-low-fat programs seem to work sometimes?
Glad you asked, because we have a theory about that.
Although many people may believe that extremely low-fat diets work because they cut out saturated fat, we suspect the real benefit comes from reducing omega-6s. Omega-6 is the predominant fat we consume, and as we’ve seen, we consume way too much of it. When we follow a very low-fat diet we consume less of it, which automatically lowers the pro-inflammatory to anti-inflammatory ratio. The fact that saturated fat is lowered is actually incidental.
In addition, those famous low-fat, high-carb diets, such as those promoted by McDougall, Ornish, and Esselstyn, are remarkably low in sugar. The carb content may be high, but they’re not the carbs most people are gorging on. The carbs in these high-carb diets tend to be vegetables, fruits, and a smattering of starches, such as beans and brown rice. And although some of the starches may be high-glycemic (such as potatoes), they don’t contain a ton of fructose (as do most processed carbs and virtually all packaged goods). Fructose is the most metabolically dangerous of the sugars, and it is a very minor player in any of the low-fat, high-carb diets that are successful. We suspect that when very low-fat, high-carb diets work at all—and they frequently don’t—they work because of these three dietary factors: fewer inflammatory omega-6s, fewer high-glycemic carbs, and much less fructose or sugar. We believe that whatever benefits might sometimes accrue from extremely low-fat, high-carb diets could be easily achieved by simply reducing sugar and processed carbs, eliminating trans fats, increasing omega-3s, and decreasing omega-6s. Reducing saturated fat and dietary cholesterol intakes has virtually nothing to do with it.
Besides, what is the mechanism by which saturated fat could cause heart disease? In 2008, the distinguished biochemist Bill Lands attempted to answer this and other related questions about conventional dietary advice in a closely argued review (complete with 231 scientific references) that was published in the scientific journal Progress in Lipid Research.
Here’s what Lands had to say about saturated fat and heart disease:
“Advice to replace saturated fat with unsaturated fat stimulated my early experiments in lipid research. It made me ask by what mechanisms could saturated fats be ‘bad’ and unsaturated fats ‘good’ . . . Fifty years later, I still cannot cite a definite mechanism or mediator by which saturated fat is shown to kill people . . . The current advice to the public needs to identify logical causal mechanisms and mediators so we can focus logically on what food choices to avoid.”
When it comes to the theory that saturated fat kills people, Lands was essentially challenging his researcher colleagues to “prove it.”
And they haven’t.