Low-Carb Better for Heart Disease Risk

Low-Carb Beats Low-Fat Yet Again

muffin and protein on balance scale
Photo © Karen Struthers

Well, it's September of 2014 and we have yet another diet comparison study which has been reported all across the news media, this time in the Annals of Internal Medicine (Effects of Low-Carbohydrate and Low-Fat Diets: A Randomized Trial). This study attempts to do a few things differently, but it's mostly similar to studies in the past, and with (surprise!) similar results.

Why the constant attention on limiting carbohydrates vs fats?

It's mostly historical. At one time, the conventional wisdom was that limiting starchy and sugary foods was the way to lose weight, but about 35 years ago a theory emerged that too much fat in the diet caused heart disease. Ever since, researchers have been trying to prove that this is true, but the fruits of their labors have mostly suggested otherwise. Meanwhile, there is mounting evidence that eating too much carbohydrate could be causing health problems of various types.

You might think that with all the talk of "healthy fats" that people would have given up the idea of a low-fat diet. But even though recent health recommendations have veered away from recommending them, a July 2014 Gallup Poll showed that almost 60% of Americans avoid dietary fat (almost 30% avoid carbs). So low-fat eating is not dead by any means, and definitely an issue that people wonder about.

This study, led by Dr. Lydia Bazzano of Tulane University with funding from the National Institutes of Health, attempted to correct two common problem with similar diet studies: 1) Usually, African-Americans are underrepresented, whereas in this study about half of the participants were black, and 2) in this study, calories were not restricted for any of the participants (whereas usually at least the low-fat group, and sometimes the low-carbohydrate group, are following a diet that is also low in calories).

The low-fat group followed a diet recommended by the National Cholesterol Education Program, because the researchers were specifically looking at how the diets affected the risk of heart disease. The low-carb group was instructed to restrict carbs to under 40 grams of digestible (net) carbs, and increase intake of sources of monounsaturated and polyunsaturated fats.

The study ran for 12 months.

Participants: Most of the participants had obesity in the mild-moderate range. They did not have heart disease or diabetes, and most of them did not appear from the data to have prediabetes or metabolic syndrome. This could be significant, since people with difficulties processing glucose (the so-called diabetes spectrum) are likely to benefit most from carbohydrate reduction.

Compliance: Compliance was "OK". The low-carb group, on average, did not reach the goal of 40 grams of daily net carb, but were around 90-100 grams for the first half of the study, edging up to 127 grams by the end of the year (34% of calories). This was still a significant reduction, as they were eating about 240 grams before the study started. The low-fat group did better, staying under 30% of calories from fat, but this wasn't as big a change for them, as they started out at 35% of calories from fat. Notably, this group also reduced carbs significantly, by about 20%. This could be from following the guidelines of the National Cholesterol Education Program (TLC Diet), which advises a low-fat diet that is also lower in added sugars.

It's important to keep this in mind, as some of their results could have been from the carbohydrate reduction. For example, both groups spontaneously reduced calories, the low-carb group more than the low-fat group.

Results: I have to be honest and say that the results were a little hard to interpret in depth, as the paper mostly did not give straightforward numbers, but instead presented statistical manipulations of those numbers. Also, on most of the measures (e.g. blood pressure) the participants started out in the normal range, so they were less likely to experience a change than people who start out with abnormal numbers. (Interestingly, about a third of the participants were on blood pressure medication, but the researchers did not report whether anyone lowered their dose over the course of the study.)

That said, the main significant results described will be familiar to those of us who pay attention to the science regarding low-carb diets:

Weight: As usual for a study of this ilk, the average weight loss was modest: 12 lbs for the low-carb group and 4 lbs for the low-fat group. One disappointing thing about the paper is that they only reported averages, and not ranges. Presenting ranges often gives us a different picture - in one famous study (led by Christopher Gardner) there was a 66 lb variation in the amount of weight lost, even though the averages were similar to this study. So a lot of information is lost when only the averages are reported.

Cardiovascular Risk Factors: The low-carb group had significantly better improvements in HDL cholesterol ("good" cholesterol), triglycerides, and C-reactive protein (an inflammation marker). There was no change in either group in most of the other measures (which, as I mentioned, started out normal anyway), but notably the diet specifically formulated to improve cardiovascular risk (the low-fat TLC diet) did not reduce risk, whereas the low-carb diet did reduce risk, as measured by the 10-year Framingham CHD Risk Score. These improvements are absolutely typical when following a low-carb diet.

Final Thoughts

Although I know the researchers intended to find out something new, they did not. (We might assume that result that the African-American participants responded similarly to everyone else, but the paper didn't address this.) Yes, it's good to replicate results in different groups, but can't we please move on? There is so much we need to know about individual variation in response to diet! Seriously! For example, we KNOW that people with metabolic syndrome tend to respond especially well to low-carb diets. So why not take some people on the diabetes spectrum and start looking at different lower-carb approaches? Paleo vs Atkins anyone?  Medium-carb vs lower-carb vs ketogenic?  People who are diabetic vs people who are mildly insulin resistant?  Jeff Volek at the University of Connecticut is doing some great work in this area, but he needs company (and funding).

There was an interesting but sort of frustrating discussion in the New York Times as a follow-up to this study that illustrates my point. Featured in the piece were the opinions of Dr. David Katz, a prominent doctor who continues to argue for "not focusing on a particular nutrient" (meaning carbohydrate or fat) but on whole unprocessed foods instead. While I'm all for eating whole unprocessed food, it doesn't do anyone any good to pretend that we all do best on the same diet, or ignore that some of us do a lot better without a bunch of glucose (carbs) in our diets. (Dr. Katz is the developer of a food rating system used by some grocery stores called NuVal that reduces every food to a number.  This is helpful if you're seeking a one-size-fits-all solution, but my point here is that different sizes call for different solutions.) In any case, I came close to standing up and cheering when the voice of reason, in the form of Christopher Gardner (I am liking that guy more and more!) came in at the end of the article saying, "trying to find “the best diet” [for everyone] is misguided. 'The better question now should be ‘what is the best diet for different individuals, and how can we match them to those diets?'" Exactly!! Let's do it!

Continue Reading