Can Carbs Give You A Heart Attack?

Feel like giving up eating?

Talk to almost anybody these days about healthy food and chances are you’re going to get an exasperated response:

“I don’t know what to eat! One day something is good for me, the next it’s bad for me! I don’t know what to do. Maybe I should stop eating all together.”

For years we were told that “fat makes you fat”. We were told that eating foods that people have eaten for millennia — like eggs — will kill you.

Now — oops! — you’re being told that fat isn’t really that bad for you and eating eggs can help you lose weight.

It turns out that things aren’t so cut and dried. Not all fat is bad for you.

Then we were told “Sorry, it’s actually carbs that make you fat. No carbs for you!” We sadly gave up our bread and pasta. We looked suspiciously at fruit.

Now, again, we’re learning that we can’t demonize an entire food group.

In All About Carbohydrates we learned that there are different kinds of “carbs”, and Mother Nature has packaged them in various formats — with different chemical configurations, with different nutrients, with different amounts of fibre and fats that change absorption, and so forth. A cup of butternut squash, for instance, isn’t the same as a cup of M&Ms, even though both are carbohydrate sources.

Two ways that we distinguish between carbohydrate types are by glycemic index and glycemic load.

Glycemic index versus glycemic load

Glycemic index

Glycemic index (GI) [1] is a relative measure of how fast your body can convert a carbohydrate to glucose and absorb it.

GI is relative because pure glucose (50 g) is used as the benchmark (with glucose absorption being 100% or 100). Everything else is measured in relation to glucose. All GI measures are done with 50 grams of whatever is being tested.

“]

Figure 1 – The original glycaemic index graph from Jenkins and colleagues 1981 [1

Figure 2 - Glycemic Index of different foods

When you eat a meal that contains carbs, your blood sugar goes up (as carbs are converted to glucose) and then down (as your body responds to and absorbs the nutrients).

How high your blood sugar rises and how long it remains high both depend on the quality of the carbs (GI) and the quantity. High-GI carbs shoot your blood sugar up quickly, then drop it when the body’s insulin response kicks in. Low-GI carbs typically release glucose in a slower, more controlled way — blood sugar doesn’t get too high, nor does it suddenly spike, then plummet.

Glycemic load

However, as you’ll see, GI doesn’t tell us the whole story. For example, spaghetti and apples have the same GI — 40. But yet we know that they act differently in the body. We needed another way to measure this.

In 1997, the GI of a given food plus that food’s total amount of carbs was combined into one number, glycemic load (GL). This helps us figure out the impact of foods on your blood glucose more easily and realistically.

Math alert! (Though compared to your taxes this is really easy.)

To figure out GL you multiple GI by the amount of carbs and divide by 100.

For example: if both spaghetti and apples have a GI of 40, how do you calculate the difference?

An apple has a GI of 40 and has 15 grams of carbohydrate. Spaghetti, same GI, but 52 g of carbohydrate per cup.

Apple: 40 x 15 / 100 = 6

Spaghetti: 40 x 52 / 100 = 20.8

Thus, the apple has a GL of 6, while a cup of spaghetti has a much higher GL of 20.8.

Carbohydrates and blood lipids

Not only do carbohydrates increase your blood glucose, they also increase plasma triglyceride levels and reduce your high density lipoproteins (HDL or good cholesterol) [2].

Gee, after years of low fat and high carbs, with people gaining weight and having more and more coronary heart disease, could it be the white bread and soda pop — not the bacon and eggs — causing all the problems?

This week’s review looks at glycaemic load and whether it increases the risk of coronary heart disease.

Sieri S, et al. Dietary glycemic load and index and risk of coronary heart disease in a large Italian cohort: the EPICOR study. Arch Intern Med. 2010 Apr 12;170(7): 640-7.

Methods

This study, which looks at the effect of GI and GL on heart disease, is just a small part of a huge study — appropriately called EPIC (short for European Prospective Investigation into Cancer and Nutrition).

Starting in way back in 1993, over half a million volunteers from all over the European Union took part in the large study. This created a bunch of other findings (over 700 scientific publications) that are mostly about food and cancer. For more on that, see here.

Why do I mention this? Sometimes more isn’t better, but in epidemiological studies like this one, more (people) is always better. This study has a nice big sample.

This particular, smaller study analyzed only the Italian subset of the total volunteer questionnaire data.  There were 47, 849 people altogether (15, 171 men and 32, 578 women).

Questionnaires

Questionnaires were either written or interviews that asked about food frequency from the past year (here’s the link to the questionnaires).

Right away you can see a small methodological problem: Interviewers asked people how often they ate particular foods last year.

Now, I don’t know about you, but I can barely remember what I ate for lunch yesterday. And I’d bet that I also have a fairly selective memory. I probably would like to recall all the carrot sticks and forget all the cookies I ate.

Despite the problems with false reporting and selective memory, nutritional questionnaires are the only realistic way of getting this information. After all, the alternative would be to either follow someone for a year, writing down everything they ate; or to keep them in a lab where you only feed them a controlled menu.

Since most people would probably find this unbelievably annoying and tedious, and might object to being locked in a lab for 365.25 days while people with clipboards dispense TV trays of prepackaged food, we’re stuck with a questionnaire for now. We just recognize that this method has its flaws.

Categorizing by GI and GL

Once they got all the questionnaires in, the researchers used software to figure out the glycemic index and glycemic load of the foods that people ate. The researchers divided the foods into high and low GI foods, using 57 as the midpoint (thus, over 57 was high; under 57 was low).

Results

Figure 3 shows a breakdown of the main sources of carbohydrate in the volunteers’ diets over a year. (12% from cake? So much for that “healthy Mediterranean diet” we hear about.)

Figure 3 - Main sources of dietary carbohydrates

Comparison grouped by glycemic load

Since the big findings happened in the group of women I’m going to focus on them, though it’s important to keep in mind that there were sex differences.

After nearly 8 years the researchers followed up, found 158 women with coronary heart disease, and reanalyzed the original data with this new information.

To figure out the relative risk of coronary heart disease, the women were sorted into four groups (1^st quartile, 2^nd quartile, 3^rd quartile and 4^th quartile) based on either carbohydrate consumption, carbohydrate from high GI food, carbohydrate from low GI food, starch, sugars, glycemic index, and glycemic load.

The table below shows the increased relative risk of coronary heart disease, comparing the lowest to the highest consumers.

I’ve highlighted the measures that seem to contribute notably to increased risk:

• overall carbohydrate consumption;
• a higher intake of high-GI carbs; and
• the glycemic load of the foods.

Yep, it looks like eating more carbs increases the risk of coronary heart disease by 2 times.

But before you jump on your low-carb soapbox, let’s take a closer look. If you check out carbs from low GI food, you see no increased risk. This tells us that the problem isn’t carbs, per se; it’s high-GI carbs.

And the biggest factor is the glycemic load — it increases disease risk by 2.24.

You may be thinking Big deal! 2.24 times greater risk of me getting heart disease — who cares?

Well, let’s put this in context.

• Smoking increases heart disease by 2-3 times [3].
• Getting no physical activity at all increases your risk by 1.5-2.4 times [4].

Huh. Suddenly, 2.24 times increased risk with glycemic load seems like a lot. If this study is repeated with the same results, you might start seeing glycemic load listed on the “Don’t Go Like That” poster at your doctor’s office, right next to smoking and inactivity.

Summary

Sorry Sophia, you'll have to quit eating all that cake.

Nearly 8 years after figuring out what a group of over 45,000 volunteers from all over Italy ate, researchers found that 463 of these volunteers had coronary heart disease. Women that ate the most carbohydrates had the biggest risk of coronary heart disease, but eating carbs made no difference with the men.

Taking a closer look at what these women ate, researchers found that high glycemic foods and glycemic load were the real risk factors, since low glycemic foods didn’t increase the risk of coronary heart disease.

Sex differences

Now, again, remember, this was in women. Carbohydrate intake didn’t seem to be as important for men — perhaps because in men, heart disease emerges from slightly different causes, or because carbohydrate intake trumps women’s other protective mechanisms.

For example, women have a lipoprotein (like low density lipids, LDLs) metabolism about twice as fast as men. Women also have higher estrogen levels (which appear to be cardioprotective for women), while obviously men have higher androgen levels.

Bottom line

You can look at this in two ways if you’re not an Italian female.

• Fantastic! I can eat whatever carbs I want since I’m not in the risk group!
• Hmm, I could cut back on bread, sugar, honey, jam, pizza and white rice, since there are healthier, lower glycemic-load alternatives like quinoa, fruit, legumes and vegetables.

Don't be fooled by the shape... this probably isn't good for your heart.

References

1. Jenkins DJ, Wolever TM, Taylor RH, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981;34:362–6.
2. Liu S, et al. Dietary glycemic load assessed by food-frequency questionnaire in relation to plasma high-density-lipoprotein cholesterol and fasting plasma triacylglycerols in postmenopausal women. J Clin Nutr. 2001 Mar;73(3):560-6.
3. Tobacco-related mortality, Fact Sheet. www.cdc.gov/tabacco
4. Pate RR, et al. Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA. 1995 Feb 1;273(5):402-7. Review.