Recently many of us were provided an opportunity to watch a highly interesting video on the internet covering a talk given by Dr. Salim Yusuf at a recent Cardiology Update meeting. Dr. Yusuf is the Marion W. Burke Chair in Cardiovascular Disease at McMaster University Medical School in Hamilton, Ontario, Canada and current President of the World Heart Federation. In his talk, he presented data from the PURE study on the relationship between the consumption of different macronutrients and the risk of cardiovascular disease (CVD).
The video received enormous attention, mostly because the data seemed to seriously challenge current dietary guidelines and recommendations by respected professional societies such as the American Heart Association (AHA) and the World Health Organisation (WHO). Unfortunately, the video is not available on the internet anymore (although you may still find it in some “dark alleys”), due to a copyright claim by the Zürich Heart House.
It appears that a large part of the data presented by Dr. Yusuf is still to be published in peer-reviewed medical journals. Hence, widespread introduction of the findings may be a delicate issue. However, the presentation of unpublished data at meetings and conferences prior publication is indeed quite common. Considering that we live in the information age, it may be difficult (and possibly unethical), to block the spread of valuable knowledge from the rest of the medical community and the public.
PURE is an abbreviation for Prospective Urban and Rural Epidemiological Study. It is a large epidemiological study aimed at examining the relationship of societal influences on human lifestyle behaviors, cardiovascular risk factors, and incidence of chronic noncommunicable diseases (1).
The PURE study recruited 153.996 adults from 17 countries aged 35-70 years between 2003 and 2009 from urban and rural communities in low, middle, and high-income regions of the world (2). Later it was expanded to more countries and is still ongoing.
Several papers on the design of the study are available, but as of yet, results have not been widely introduced (3). However, last year I wrote a blog post addressing data from the PURE study presented at World Heart Federation’s World Congress of Cardiology & Cardiovascular Health in the summer of 2106 in Mexico City. The data suggested that, contrary to what has been publicly advocated, high-carbohydrate diets seem to be worse for blood lipids than high-fat diets (4).
Experts agree that lifestyle factors significantly affect the risk of developing CVD. When it comes to diet, the focus has often been on macronutrients and their effects on surrogate markers such as blood cholesterol. On order to keep fat intake at a minimum, the AHA, WHO and many other public health authorities have recommended that 55-60 percent of calories consumed should come from carbohydrates.
The rising prevalence of obesity and type 2 diabetes suggests that public health authorities have either failed to deliver the correct message to the food industry and the public or that their recommendations have simply not been implemented. It is for this reason that we so eagerly await the results from nutritional part of the PURE study. Obviously, it is of public interest that these data become widely available.
Before we go further, it is important to acknowledge that PURE is an epidemiological study and not a randomized trial, and as such, has a limited value when it comes to proving a causal relationship. Food Frequency Questionnaires (FFQs) were used to measure dietary habits. Although this method has several limitations, the benefits of using FFQ is that they are relatively simple and easy to administer. A pilot study suggested that the FFQ’s used in the PURE study could capture dietary intake adequately (5).
Let’s look at the data presented by Dr. Yusuf on the association between carbohydrate and fat consumption and cardiovascular disease. In the same lecture, he also presented data on salt intake, blood pressure and the role of fruits and vegetables. These data are not covered here.
Carbohydrates, Fats, and CVD in the PURE Study
Carbohydrate intake was divided into quintiles. Those in Q5 consumed most, and those in Q1 consumed least. As can be seen from the slide below, presented by Dr. Yusuf, using the lowest carbohydrate intake (Q1) as a reference, there is an increased risk of CVD with increasing carbohydrate consumption.
Dr. Yusuf pointed out that earlier dietary guidelines “said reduce fats and compensate for it by carbohydrates. So essentially we’ve increased carbohydrate intake in most Western countries, and this is likely damaging.”
Furthermore, Yusuf said: “We were in for a big surprise. We actually found that increasing fats was protective. Now, these are all fats. So this very first slide challenges the WHO and the AHA guidelines on diet.”
With regards to carbohydrates as a percentage of daily energy consumption, he also said: “once you get past about 40% or about 55% of carbohydrate intake as percent energy, there is a steep increase in the risk of CVD. The WHO guidelines say that up to about 75% of carbohydrates is ok. But that is wrong.”
Then he adds:
“Again, total fats, if anything, is protective. And the reasons for this are twofold. Too low fat is inadequate, too high fat is probably bad. And the original studies from Finland was at extremely high fat levels, not the usual fat levels that populations consume.
Does the Type of Fat Matter?
Current dietary guidelines recommend that we limit the intake of saturated fats and replace them by mono- and polyunsaturated fats. Low-fat dairy products are recommended for the purpose of avoiding saturated fats. Vegetable oils should be used instead of butter.
As Dr. Yusuf points out, these recommendations are not supported by data from the PURE study:
“Then if you look at the types of fats, saturated fats, you will see there is not really a clear pattern of anything. Within the normal range, saturated fats are not harmful. May even be slightly beneficial. But there is no harm.
With monounsaturated fatty acids which are in olive oil, canola oil, and are part of the key to the Mediterranean diet, you get a clear benefit. Polyunsaturated fatty acids which are largely from vegetable oils, and remember that is processed oil, is largely neutral.
You’ve got to think of the change in oils that have occurred in the world in the last 30 years. It was entirely industry driven. You know, it went from natural fats, which is animal fats, to vegetable fats, because they can produce it and therefore charge for it. And this was swallowed, hook, line, and sinker, by the AHA and the WHO just repeated it.
So, fundamentally, our fat story is: some fats are good, some fats may be neutral but it is carbohydrates that are the worst thing.”
… and Dr. Yusuf goes on:
“The other thing is; in the US there is this big swing to reduce milk consumption, and even if you consume milk, they want you to consume 2% or 1% of fat. What is the evidence? A big zero. Absolutely no evidence that low-fat milk is better for you. If anything, if you look at dairy sources of saturated fats, it is protective.
If you look at meat sources of fat, saturated fats, it’s neutral. And if you look at white meat such as chicken and fish, there is a trend towards benefit. So, red meat in moderate quantities is not bad and white meat may be moderately beneficial. But dairy fats such as cheese are probably good for you, and milk, there is really no data at all to reduce the fat content of milk.”
Saturated Fats, LDL-Cholesterol, and CVD
Those of you who have read my blog will know that I have a great interest in the role of lipids in CVD. I’ve written many posts addressing LDL-cholesterol (LDL-C), HDL-cholesterol (HDL-C), triglycerides (TGs), lipoprotein(a), apolipoprotein B (ApoB), apolipoprotein E, VLDL-cholesterol (VLDL-C), TG/HDL-C ratio, familial hypercholesterolemia, dietary fats, and saturated fat in particular.
One of the things that I have pointed out is that we seem to have overemphasized the importance of the association between saturated fat consumption and LDL-cholesterol (6,7,8,9,10). Interestingly, Dr. Yusuf addresses this issue as well, based on data from the PURE study.
Dr. Yusuf says:
“Now, why did we go wrong. We went wrong because of surrogate end-points.
The story on saturated fats vs. LDL is consistent. Our data shows that as you increase the amount of saturated fats, your LDL goes up. But first, look at how much LDL goes up. This is a 150 thousand people in the analysis – from about 2.85 (108 mg/dL) to just under 3 (116 mg/dL). About 0.1 mmol/L (3.9 mg/dL) increase over a huge range of percentage of saturated fat (consumption).
But CVD shows exactly the opposite end-point. This is what dominated the guidelines. This wasn’t achieved. There have been so many randomized trials of fat reduction. And, other than the old Norwegian study where saturated fats were very high, reducing them did lead to benefits, but the Women’s Health Initiative, where total fats as well as saturated fats were reduced showed no benefits. That is a huge study of 49 thousand women followed for seven years.”
Macronutrients and the ApoA/ApoB Ratio
Dr. Yusuf also addresses the effects of carbohydrate intake on more advanced lipoprotein measurements:
“Now, what about carbohydrate intake? If you look at LDL, there will be an inverse relationship. But if you look at ApoB/ApoA ratio, which today we know is the most sensitive marker for risk prediction there is a steep increase – from about 0.72 to nearly 0.85 with increasing carbohydrate intake. But with saturated fats, if anything, it is neutral or tending to go down.”
“So, depending on the surrogate marker you can make different kinds of extrapolations. If you look at monounsaturated fatty acids and ApoB/ApoA, look at the steep decrease from about 0.8 down to about 0.72. The ApoB/ApoA story tracks with the risk factors better than the LDL story. And with polyunsaturated fat, it is pretty flat.”
Dr. Yusuf’s final remarks include these words:
“Contrary to common beliefs, the current recommendations to reduce saturated fats have no scientific basis. I’m not the only one saying this. You must have heard of the book called ‘The Big Fat Surprise’ by Nina Teicholz. She shook up the nutrition world, but she got it right.
Did you know that the seven countries studies that actually had a straight line between fat intake and CVD is fudged. I’m using the word fudged because 23 countries participated in that study and they took the seven best that fitted that line, and that’s what’s there. If you go through the literature, you will find that they chose the seven that fitted the line. The nutrition field has been distorted.”
So, is there a time for a reappraisal of public recommendations regarding the relationship between diet and heart disease following the presentation of the PURE date on macronutrients. Obviously, this is a rhetorical question. Let’s get to work.