Bread with Seeds and the Glycemic Index of Bread

Bread with Seeds

A couple more breads to stick in the freezer, against the coming kitchen renovations.  

These loaves were deliberately made to be dense.  I usually get my hydration up much higher than this.  But this -- and other recent experiments with low hydration, high density bread (aka "bricks") -- have been trials in preparation for some non-wheat naturally fermented doughs I want to play with.

1. 10% Rye Bread with Seeds

The seeds are 211g of sunflower and pumpkin seeds.  The hydration is 62%.  The rest of the ingredients are the usual: 90% whole wheat, 6.7% wheat germ, 20% sourdough starter, 2% salt.

This bread went right into the freezer.  No crumb shots.

2. Whole wheat bread with Amaranth Seeds

100% whole wheat, with 6.8% wheat germ, at 65% hydration.  I added 200g of amaranth seeds. Kneading this dough felt like petting a lizard.

I ate one of these breads, the other went into the freezer.  Probably neither of them were good enough to give away.

Thoughts on the Glycemic Index

I have been reading about "The Glycemic Index", (a scale developed in the early 1980s by Toronto researcher Dr. David Jenkins and now a fairly complete table of virtually all foodstuffs that you can buy that contain carbohydrates). The scale ranks carbs according to how they spike the blood sugar.  This is important information for diabetics, and for people who want to lose (or maintain*) weight.

And those unfamiliar with regular use of the scale, like me, are often surprised to see where bread falls on the scale.

The scale usually ranks foods against pure glucose, the sugar all cells of the body use for energy.  Glucose is given a value of 100.**  

Table sugar (sucrose), which is only half glucose (the other half being fructose), gets a value of about 65, but (using the table at glycemicindex.ca) bread can be even higher (depending on the type of bread, the index can range from 51 (pumpernickel, whole grain, 1 oz.) to 95 (French baguette).  As wikipedia states, the values on the Glycemic Index can be counter-intuitive.  As is frequently pointed out, candy bars can have lower numbers on the glycemic index than whole grain bread.

I wanted to know where my own sourdough bread -- not a purchased bread -- might fall on this scale.  So one night at work when we weren't too busy, I tested my blood sugar.  I had been fasting for 25 hours, and at midnight I broke my fast on a single 89g slice of bread (with nothing else on it).  This was my dense whole wheat sourdough bread, with a tiny bit of amaranth seed in it.  At midnight I measured my fasting glucose, and thereafter I tested at 15 minute intervals, for 2 1/2 hours.  The following table shows how my blood glucose levels spiked.  These are Canadian values (measured in mmol/L, unlike the US values, which are measured in mg/DL.  To get the comparable US value, you'd have to multiply the BG value by 18.  Normal fasting range for non-diabetics in Canada is 3.5-6.5).

During the 2 1/2 hour period, I ate nothing else, but I did drink a bit of black tea.

TIME	BG (mmol/L)
0000	3.7 (fasting baseline)
0015	4.2
0030	5.3
0045	7.4
0100	7.9
0115	7.2
0130	6.7
0145	5.9
0200	4.6
0215	4.3
0230	4.6

I was using 89g of bread, not the usual 50g of bread that the glycemic index is built upon, but still, you can see that the 1 hour peak of my blood sugar was 7.9, so the glycemic index would probably fall somewhere around 79.

Why is bread higher on the Glycemic Index scale than table sugar?

Wheat is roughly 12% protein, 12% fiber, 1.5% fat,  71% carbohydrate, and the rest is ash (iron and other minerals) -- so bread made with wheat is going to fall into this sort of breakdown too (assuming it hasn't had parts -- like the fats and fiber -- removed).  The carbohydrate, or starch, is made up mostly of amylose (20-30%) and amylopectin (70-80%).  Amylose is a sugar made up of glucose units -- as many as thousands, mostly strung together linearly -- but it turns out that it is difficult for humans to digest, probably because of the varied shapes the multiple loops make, and because it is water insoluble.  But your gut flora -- all those good bacteria in your bowels that keep you healthy -- can digest much of it as it passes through.  In return for giving them amylose, they give you other things -- like more probiotic bacteria, digestive enzymes, vitamins, and gas (nothing says "it's partytime!" more than happy gas).  Amylopectin, on the other hand, is soluble in water, and is considered a complex carbohydrate.  Also made of glucose, amylopectin is highly branched and tightly packed.  The molecule's complexity is nothing to the human body, though, because we produce amylase, both in the mouth and in the gut, and this enzyme very efficiently liberates the glucose and allows it to pass directly through the digestive tract lumen into the bloodstream.  The wheat plant has stored a lot of energy in the seed; it is designed to provide a young seedling with a big boost for rapid growth -- kind of like what breast milk does for infants.  And when we eat it, we consume that concentrated energy that we can immediately use.

As blood glucose levels go up, insulin is released by the pancreas, which takes glucose out of the bloodstream and puts it into cells that need it.  What happens to excess blood glucose that the body can't immediately use?  The liver scoops up some of it and makes glycogen, which is a packed form of glucose, even denser than amylopectin.  This can be later taken apart and dispersed to the other various organs as required.  The muscles also keep a store of glycogen nearby for their own rapid use.  But what happens when we have enough glucose for cells, and enough glycogen stored?

If we've still got more glucose than we need, insulin levels continue to rise in the bloodstream. The liver takes notice and begins to make fatty acid synthase, which begins the cycle of converting glucose into pyruvate, then acetyl CoA, than fatty acids. These circulate as triglycerides, which are taken up by adipose tissue, packing it as lipids.  If all goes well, the fat cells send out leptin, a hormone that suppresses appetite (they are saying, "enough already!")  In conditions of low carbohydrate input, the stored fatty acids can be broken down again to glucose for energy -- but this metabolic pathway is expensive, and it always seems easier to just eat more carbs to get energy, and that is one reason why fat is so difficult for many people to get rid of.

Fatty acids are ingested mostly from meat and dairy; but if you are not eating these things, your triglyceride levels might still be high if you get too much energy from any source, including carbs, especially wheat.

The Glycemic Index is a good tool for diabetics who need to learn to control the rapid rise of glucose in the bloodstream.  Eating low on the scale will keep the body's blood glucose at a nice, healthy level.  Bread may have its place in a diabetic's diet, but the danger is, it can raise the blood glucose level very high very quickly, without the pancreas' ability to safely modulate the levels.

But the GI Index is not the complete story for safe eating practice.  Satiety plays a large part in keeping one's energy intake high enough, but not too high: I've mentioned the hormone leptin which can modify our appetite.  Fiber may also play a role in keeping one from overeating carbs (and if we take the fiber out of our grain and eat only the starch, this can have dangerous consequences, and is one of the reasons why we are told that whole grains -- as opposed to grains with germ and bran removed -- are important).  Furthermore, bread is not often eaten as I did for this experiment, alone and without anything else.  When we add proteins or fats to our bread, or even other carbs, it will modify somewhat the rather sharp glucose rise that we see (sometimes because of delayed gastric emptying, sometimes for other reasons) -- unless we are diabetic, of course.  For diabetics -- for whom this glycemic index was developed -- it becomes even more important to eat low on the GI scale, and to make sure we don't consume more calories than we really need.

Others have proposed a scale that rates fullness, or satiety, in concert with the Glycemic Index.  Still others propose a scale that ranks foods with a better antioxidant profile.  In all of these scales, vegetables are at the end of the list that are best, fruits are towards that end too, and everything else is not.  Most diets (except perhaps the extreme high protein diets) will advise you to eat more vegetables and fruit.  

I'm glad I read about bread's glycemic index.  It shows me that bread is a high-energy food source.  It also reminds me to find more of my carbs from vegetables and fruits, whenever possible, and not to be so worried about eating bread with fats like butters, oils, eggs and cheeses.

Bread Results
This wasn't a particularly good bread, certainly not one of my favourites.  As I indicated, it was pretty dense, and although it was okay tasting, I've certainly made better.  The amaranth seeds were pretty much invisible to taste and mouth feel, so they seem to have added just about nothing to this loaf.

Notes to Myself

• * Using bread to maintain weight is something not to be scoffed at. Plenty of people with illness or malnutrition due to dietary insufficiencies or allergies can benefit from the high energy to be obtained from whole grains.  People who can't afford a diet rich in varied organic fruit and vegetables can still maintain their caloric intake using whole grains.  
• ** In practice, no one eats pure glucose, so this was felt to be a difficult scale for diabetics to put into practice in their everyday life. At one time, the glycemic index was therefore scaled to white bread at 100 (and on this scale, glucose gets a value of 140).
• Incidentally, after the 2 1/2 hour mark, when I quit testing my blood glucose levels, I decided to eat even more carbs: an apple (Glucose Index 38) and then a banana (Glucose Index 55) and finally an orange (Glucose Index 44). The effect on the carbs is not additive, but rather averages out, believe it or not. An example of what I mean is sucrose (GI 65), which contains half-fructose (GI 19), half-glucose (GI 100).
• I highly recommend this blog entry, authored by Chris Masterjohn and linked from the Weston A. Price Foundation's webpages, which discusses the book "Wheat Belly" -- including Davis' use of the Glycemic Index to show that wheat spikes the blood sugar and makes people fat.  In particular, Masterjohn says:
  
  "But there is no evidence that minor fluctuations of blood sugar within the normoglycemic range cause harm, and a little bit of fat will nearly flatten the glycemic index of a carbohydrate-rich food in healthy people.  Whether insulin makes people fat is currently a matter of vigorous debate and Wheat Belly would have benefited if Dr. Davis had presented this relationship as a hypothesis rather than stating it as a simple matter of fact."
  
  Incidentally, both Davis and Masterjohn mention the 'exorphin' thesis -- that grain contains morphine-like substances, and this can explain why hunter-gatherers with good health moved to a less healthy agricultural/community-based diet.  Masterjohn says that exorphins are now found in lots of different plants (and of course they were discovered in dairy products around the same time they were discovered in wheat) and he suggests that they are likely to be pretty much ubiquitous in our food.  He also says that their biggest effect in humans is to delay food's passage through the gut.
   I see this all the time in my work, as I try to alleviate constipation in my patients in palliative care who are experiencing this primary symptom of pain management using narcotics. Whole wheat, due to its fiber, will alleviate much of this potential problem (and not to minimize the other effects of exorphins, I am of the opinion that some people who have wheat/milk allergies, sensitivities and trouble metabolizing them may also experience some central nervous system effects, as seen in autism, schizophrenia and certain dementias.  It may also have some addictive and euphoria properties too, who knows?).
• UPDATE:
  
  
  
  
  
  T. Wolever, author of "The Glycaemic Index: a physiological classification of dietary carbohydrate" (2006) defines the glycaemic index in this way:
  
  

  The GI is defined as the incremental area under the blood glucose response curve elicited by a 50g available carbohydrate portion of a food expressed as a percentage of the response after 50g anhydrous glucose taken by the same subject.

  
  
  Therefore, I have not determined the GI of my bread, since I did not obey the rules -- which would involve my fasting the correct number of hours (I fasted too long, which may lower the baseline glucose level too far), taking the correct amount of food (I needed 50g of carbs, and I calculate there was about 64g of carbs in this 89g slice of bread), and determining the mathematical area under the curve (which of course I didn't do, see below for a hint about the formul), and doing the test several times (didn't do it but once), comparing it to my response to pure glucose (didn't) and averaging out the Blood Glucose response of several people (it was just me).  All I have done, so far, is to show my own Glucose Response for one slice of my bread, over a short period of time.
  
  
  Glucose Response is individual, and can be affected by changing other parts of the diet.  Wolever explains:
  
  

  It is known that fat and protein affect glycaemic responses, but these effects have nothing to do with the glycaemic response of the carbohydrate. In addition, the effects of added fat and protein on glycaemic responses differ in normal subjects, subjects with type 1 diabetes and subjects with type 2 diabetes… On the other hand, the GI of individual carbohydrate foods is the same in all of these different types of subjects….The terms ‘glycaemic index’ and ‘glycaemic response’ should also not be confused because these entities have different mathematical and statistical properties.

  
  
  
  
  On p. 17-18, Woelver discusses different methods for determining the Area Under the Curve (AUC) mathematically.  The simplest appears to be the trapezoid rule, but "polynomial interpolation of third and fourth degree, Simpson's integration and cubic interpolatory splines" gave slightly more accurate results (variation about 2%).  Since I messed up the test, and probably can't correct for the errors in procedure, I won't be doing the math.  But at least I have an idea, how to do it in the future.