Wednesday, September 23, 2009

Should Saturated Fat Be Avoided in Low-Carb Diets?

Butter fat is about 12% myristic acid, which increases both LDL and HDL.
Butter fat is about 12% myristic acid, which increases both LDL and HDL. (Photo by No. Meds)

Browsing through older issues of Nutrition & Metabolism, I came across an interesting article on saturated fats. It's a commentary to another article, titled "The case for low carbohydrate diets in diabetes management" (link). In the original paper, the authors argue that low-carb diets are better than traditional low-fat, high-carbohydrate diets for weight loss, blood pressure, metabolic syndrome and dyslipidemia. They also note that low-carb diets reduce triglycerides and increase HDL.

However, the authors are skeptical of the popular Atkins diet, because of its potentially high saturated fat content. Their recommendation is to limit saturated fat consumption and to use monounsaturated and polyunsaturated fats instead. So, while they succesfully dispel common myths about low-carb diets, they still propagate the myths about saturated fat being detrimental to health.

The authors behind the commentary, Volek and Forsythe, go over several myths about saturated fat and cholesterol (link). While they commend the authors of the first paper on their observations on low-carb diets in general, they note that restricting saturated fat is unnecessary.

The reason is that much of the bad reputation of saturated fat comes from overgeneralizations and misinterpretations of the data. To make a statement like "saturated fat is bad for you" doesn't really tell us much. Even if we narrow it down to saturated fat being bad for cholesterol, it still just leaves us with more questions.

Is all saturated fat bad or just some? Does the amount of saturated fat matter? Do combinations of different saturated fats behave differently than single saturated fats? And importantly, what does "bad for cholesterol" mean exactly – does it increase total cholesterol, increase LDL, decrease HDL, or something else entirely?

While the commentary doesn't answer every possible question on saturated fat and cholesterol, it does shed light on some of these questions. Since I was so inspired by this article, I've also referenced quite a few related studies not mentioned in the commentary.

First, there are different types of saturated fat. One way of categorizing fats is according to their carbon chain length, and this factor plays a part in how fats affect cholesterol levels. For example, stearic acid, which has an 18-carbon-chain (C18) and is found especially in meat and cocoa butter, does not raise cholesterol levels (link). Palmitic acid (C16) and myristic acid (C14) seems to raise cholesterol the most (link, link). Depending on the study, the cholesterol-raising effect of lauric acid (C12) is either slightly lower (link) or higher than that of palmitic acid (link). Clearly, saying that all saturated fats behave the same way is an oversimplification.

Second, the dietary context in which saturated fat is consumed plays a crucial part. Nobody eats only saturated fat. The lipid profile will depend not only on the amount of saturated fat, but also on the ratio of protein, fat and carbohydrates. Replacing carbohydrates with any fat will lower fasting triglycerides, and replacing carbs with saturated fat also raises both HDL and LDL (link). When total fat intake is relatively low, increased saturated fat intake is associated with less progression of atherosclerosis (link). Interestingly, the same study also showed that when replacing carbohydrates or protein, polyunsaturated fats were not associated with progression of atherosclerosis, but when they replaced other fats, they were.

In addition, it might makes a difference which other fatty acids are consumed with saturated fats. For example, when palmitic acid is consumed with linoleic acid (an unsaturated omega-6 fatty acid), it ceases to increase cholesterol levels (link). When dietary olive oil is replaced with palm oil, which is high in palmitic acid, no change in cholesterol is seen (link), possibly because other fatty acids are also present in the average diet.

The form in which the saturated fatty acid comes is also important. In a synthetic form, myristic acid raises total cholesterol less than palmitic acid (link). However, when compared to palm oil, myristic acid raises total cholesterol more (link). All this goes to show that it's surprisingly difficult to draw conclusions about how saturated fat from natural sources affects cholesterol levels based on studies using single fatty acids.

The third and final point is that the effect of saturated fat on total cholesterol levels is not very interesting, since total cholesterol is a poor predictor of disease. Rather, it's the effect on triglycerides, the LDL/HDL ratio, and the size of lipoprotein particles that we're mostly interested in. And, unfortunately for those who are scared of saturated fat, reducing saturated fat intake decreases HDL, especially large HDL (link). On the other hand, increasing saturated fat intake decreases triglycerides, increases HDL size (link) and increases LDL size during low-carb diets (link). High HDL combined with large LDL and HDL particle size is common among people who are long-lived (link).

As for how individual saturated fatty acids affect these factors, different studies report different findings. Most reviews on the subject agree that all saturated fatty acids except stearic acid increase LDL (link). Lauric acid has been shown to increase especially HDL (link), while stearic acid was shown to improve the ratio of total cholesterol to HDL more than palmitic or myristic acid (link). In fact, palmitic acid doesn't seem to change HDL one way or the other, whereas myristic acid increases it, albeit less than it increases LDL (link).

So, to wrap things up, the take home messages of the above are:
  1. The effect on cholesterol depends on the length of the carbon chain: in general, stearic acid is neutral, while palmitic acid, myristic acid and lauric seem to increase total cholesterol.

  2. Saturated fats don't exist in a vacuum: all the other stuff in the diet will also make a difference in how cholesterol levels are affected by saturated fat.

  3. Make sure you know what you're trying to achieve: if avoiding saturated fat, care should be taken to minimize reductions in HDL levels and lipoprotein particle size.
Of course, looking at individual fatty acids doesn't really tell us much about how saturated fats from food sources like meat, coconut, palm fruit, and butter affect cholesterol levels, but as I'm running out of time and space here, that'll have to be a subject for another post.

For more information on fat and cholesterol, see these posts:

SAs, MUFAs vs. PUFAs: Fat Storage Depends on Type of Fatty Acid in Rabbits
Blood Test Analysis: The Cholesterol and Saturated Fat Issue Revisited
Low-Carb vs. Low-Fat: Effects on Weight Loss and Cholesterol in Overweight Men
Anthocyanins from Berries Increase HDL and Lower LDL

4 comments:

  1. Minimizing saturated fat intake while increasing polyunsaturated fats, even on a low carb diet will decrease HDL. Simply looking at the total cholesterol number is an outdated concept. There is nothing inherently dangerous about saturated fats, except when they're combined with a high sugar diet.

    On the other hand, polyunsaturated fats will decrease cholesterol. So what? Cholesterol doesn't cause heart disease or cause you to go blind (oh wait, that's something else). Data from 164 countries show a mortality minimum at cholesterol levels between 200 and 220 (see Source: http://www.who.int/whosis/indicators/compendium/2008/1mst/en/).

    Saturated fats do improve the characteristics of LDL and HDL and will definitely increase HDL.

    ReplyDelete
  2. Ed Terry,

    Where do you get the information that minimizing SA's whilst increasing PUFA's on a low-carb diet = lower HDL?

    PUFA's are widely known to decrease LDL and increase HDL whether your source is flaxseed or fish oil.

    Also,
    Who said that its simply about looking at the cholesterol number?
    It just so happens that molecular chain length = size does matter, both for SA's and for PUFA's (EPA and DHA are long-chain.)

    I'm a big believer of biochemical individuality; in that how I respond will be different to how you or anyone else will respond, but these factors are not considered in these "SA's are bad" dogma propagator.

    There is far more to be considered.

    As Mark Pollan put it so well, "Foods are not merely the sum of their nutrient parts."

    ReplyDelete
  3. @Anonymous,

    Where do you get that all PUFAs increase HDL? Of course it depends on total fat intake, but compared to MUFAs and SAs, many PUFA-rich oils seem to reduce LDL and HDL. Flax seed oil and fish oil are the exceptions -- not all vegetable oils are high in n-3.

    - JLL

    ReplyDelete