August 6, 2007

Why the Glycemic Index Fails for Many People with Diabetes

As simple as it seems, most doctors and dietitians still don't tell people with diabetes that the carbohydrate content of the food they eat is what raises their blood sugar and that lowering their carbohydrate intake will lower their blood sugar.

Instead, they recommend the so called "good carbs" which are those which are low on the "Glycemic Index," chief of which are what they call "Healthy whole grains," like whole wheat bread, brown rice, pasta, and oatmeal.

If you look any of these foods up in your handy carb counter--you DO have a carb counter, I hope!--you will see they all contain a lot of carbohydrate. Two ounces of whole wheat bread--one thin slice--generally contain around 29 grams of carbohydrate and how many people only eat one slice?

A single ounce of dry oatmeal contains 18 grams, but what most people consider a full serving is at least twice that size. Two ounces of low glycemic pasta contain around 56 grams of carb, but again, two ounces is a very small amount--about 1/3 of what most people consider a normal portion of pasta.

If you measure your blood sugar for several hours after eating a normal serving of whole wheat bread or oatmeal you will see a spike, possibly a very high spike well over 200 mg/dl. You may have to measure your blood sugar 4 hours after eating pasta to see the spike it causes because of how slowly it digests, but eventually it does digest, and if you keep testing you will see it cause a spike, too.

So what's going on here?

The answer is that the glycemic index only works for people who have a normal second phase insulin response. If you don't produce very much insulin, or if your insulin resistant body isn't responding to the insulin you produce, it won't work for you.

All the Glycemic index does is measure the speed with which the carbohydrate in a given food is digested and hence the speed with which the carbohydrate it contains hits the blood sugar.

Foods that are low glycemic usually contain a lot of fiber or, like pasta, contain complex starch molecules that take a while to digest.

This is very helpful to normal people because they have a robust second phase insulin response. When they start to eat, their body immediately dumps a load of insulin into the bloodstream to take the edge off any spike. That's the first phase insulin response--which people with any kind of diabetes rarely have. Then, as the food digests, if their blood sugar rises over 120 mg/dl (6.7 mmol/L) their beta cells dump a second load of insulin into the blood stream. This is the "second phase insulin response" and for many people including some people with mild type 2 diabetes it is much stronger than their first phase response.

For a normal person, this second phase insulin response kicks in at half an hour after eating and is done shortly after one hour after eating. If your carbohydrate is slowed down just a bit, a normal second phase insulin response can mop it up completely avoiding any spike in blood sugar response. So for a normal person with normal insulin response, the Glycemic index does point to foods that are healthier.

But people with diabetes don't usually have a normal second phase insulin response!

Here's a simple graphic from a page called "Secretion of Insulin and Glucagon" which shows you the amount of insulin secreted by normal people, people with impaired glucose tolerance and people with the kind of Type 2 that is characterized by Insulin resistance but who still produce some insulin.

Many people with Type 2 diabetes have little or no second phase insulin response left. And those that do usually are very insulin resistant, so although they may still secrete normal levels of insulin, their body has stopped responding to it, making that second phase ineffective for controlling blood sugar. And of course, most people with Type 1 diabetes whose beta cells are dead have NO natural insulin response at all.

When a person who doesn't have a robust second phase insulin response eats a "low glycemic" "healthy whole grain" there's no second phase insulin response waiting to mop up the carbs released from these foods. So every single gram of those carbs will hit the blood stream over the next couple hours. And since insulin isn't there to greet it or isn't able to do the job thanks to Insulin Resistance those grams of carb are going to raise the blood sugar. And because those "healthy whole grains" have a lot of carb in them, they're going to raise it a lot.

If a person is injecting insulin, slowing the carbs down a bit might allow these slower carbs to meet the insulin. Or it might not, because it is very tough to predict exactly when the low glycemic food will digest completely. If you try to dose for 4 ounces of pasta, for example, the insulin you injected to cover those 112 grams of carbs may well hit your blood stream before the pasta does, resulting in a nasty low.

But many so-called "low glycemic foods" actually hit the blood stream very fast--oatmeal certainly does for me. So your "low glycemic" oatmeal may raise your blood sugar before your injected insulin can get there to process it.

Unfortunately, doctor and nutritionists often don't seem to understand that you need an intact second phase insulin response for the glycemic index to be helpful. Nor do they realize that the research used to establish the "glycemic index" ratings of various foods was all done in normal people with robust second phase insulin responses.

If you are a Type 2, you can easily determine if you have an intact second phase insulin response. Eat a serving of whole grain bread with nothing else and then test your blood sugar at 1 hour 2 hours and 3 hours after eating it. If you don't see a blood sugar value over your target blood sugar, you have still got a second phase insulin response and you can use the Glycemic Index to help you choose what to eat. If you see a large spike after eating, you don't. It's as simple as that.

And if you don't have a second phase insulin response, the only difference between a high glycemic and a low glycemic food is that one hits you sooner and the other hits you later. But every single gram of carbs in those foods will eventually raise your blood sugar. And If that's the case, for any food that raises your blood sugar over the target you've set, the carbs it contains are not "good carbs" or "healthy carbs" no matter how healthy they might be for people who do not have diabetes!


Robin said...

You presented a test for seeing if we have 2nd phase insulin response. How about doing an essay with a variety of those kinds of tests, for those of us who have doctors who don't believe in doing the tests Bernstein recommends? Don't ask me why my doctors haven't done the tests - I won't but tantrums and rages want to emerge!

renegadediabetic said...

I find that several "low glycemic index" foods, such as whole grains or oatmeal, still jack up my blood sugar. The amount of carbs also is a big factor. Every diabetic needs to use a glucose meter to see how different foods affect blood sugar.

Anonymous said...

I am a diabetic resulting from pancreatitus 9 years ago. Your explainations of good vs. bad carbs, the glycemic index, and whay "healthy" foods like oatmeal raises my sugars over 200 points is the best I have come accross. Your explainations are quick, simple, complete, easy to read, and contains examples. Great Job!! Please do more on this subject.

Anonymous said...

Just read your article on Glycemic index failure. I'm not diabetic--instead I have hypoglycemia--but many of the same precautions apply. I have a blood sugar "crash" some time after a breakfast of "low-GI" oatmeal, and have long wondered why.
Your article lets me know that I also can't base my diet on recommendations aimed at people with "normal" blood sugar situations. I will eat differently from now on. Thanks for the info.

Joe To You said...

I like the information presented and I find it useful. I do object to all the quote marks. There's a bias implied and it tends to come across as sarcasm. I'm type 2 diabetic and oatmeal does wonders for me. Testing two hours after eating yields an almost non-existent spike, maybe ten point on my meter. I've also read that there are some supplements that can modulate the glucose spike. Also, one must factor in the insoluble carbs as opposed to the soluble carbs. Don't want to throw out the baby with the bath water.

Jenny said...

Joe, Those foods work for YOU, but there are many of us for whom they don't work. Back when the newsgroup was active this topic came up for discussion a lot, and that was where I learned how variable people's responses are to all carbs including the so called low glycemic ones.

It is because the index is an average response that this variability is ignored, and the nutritionists who advise people with diabetes don't understand that these fiber rich foods are only suitable for people who, like you, still have a robust second phase insulin response. Many people with diabetes do not.

Harold said...

I wonder if Joe has tested at 3,4,and 5 hours after eating oatmeal to be sure that he is not getting a late rise in bs?

Grey Beard said...

The pancreas secretes insulin on a 3-6 minute cycle, so I think your characterization of a 1st and 2nd stage of insulin is overly simplified.

Those without the ability to produce insulin will have no 1st, 2nd nor nth response, so that group really should be excluded from the current post's discussion. It clouds the point of your post.

Finally, it's important to remember what the glycemic index is - a recording of blood glucose levels at 15 minute intervals for 2-3 hours.

FWIW, when exercising hard, provided that adequate insulin is available, venous and capillary glucose levels are dramatically different, because the large skeletal muscles quickly burn venous glucose, while it remains quite high for several hours in small capillaries found in places like finger tips. As a result, the readings you get when you test your blood glucose can be radically different, depending on where you draw your blood sample from if you are exercising.

Marie @ Myculinarycalling said...

If you actually look at a glycemic index chart (university of Australia has a good one) oatmeal ( all kinds), breads and pastas (whole grain or not) are either barely low(45-55) or not low at all! I follow a low glycemic diet, and eat only heavy 100% rye bread that can hardly be classified as bread like, and steel cut oats as an ocassionaly treat with cinnamon, yogurt and fruit for dessert. ( pairing with a bit of protein can lower the gi a bit) The glycemic index does work, if you actually follow the scientific version, not just diabetic books. I have read a few, all of which have said different things. I also do not eat any sorts of flours or grains as they are basically all above the acceptable range. Almond, bean and coconut flours are good, low glycemic options that totally work!

Jenny said...

Marie, To echo your rude tone, if you'd actually read my post you'd see that I don't say it doesn't work for everyone. Just most people with diabetes. Before I was on my current medications, one portion steel cut oats would send my blood sugars soaring into the 200s as would that European heavy rye made with the whole kernals that you cite. I have heard the same from many people with diabetes.

And, of course, I have looked very closely at several glycemic index charts including Jennie Brand-Miller's.

No one can know how a food will work for them by depending on anyone else's tests. That's why the main thrust of everything I write is that people have to test their own blood sugar to know what works for them.