The Mechanisms of the Glycemic Index: A Fatty Acid Roller Coaster?

This post will focus on some statements made by a researcher who I've dubbed "America's GI Man", Dr. David Ludwig, in this review: The Glycemic Index: Physiological Mechanisms Relating to Obesity, Diabetes and Cardiovascular Disease, JAMA, 2002.

A Note Regarding Definitive Statements in Review Papers

As I found myself getting sucked down the Pubmed rabbit hole that is the GI, it was really this paper that took me past that point of no return.  I have taken a screenshot of the top of the PDF format of the article to convey a few things, and at this point I'm beginning to understand how it is that doctors can get sucked into bad paradigms when their "prestigious journals" put out "SPECIAL COMMUNICATION"s from their "CLINICIAN'S CORNER"

Let's begin with the title of this review.  It may seem a nit-pick, but it titled in the affirmative.  What you are about to read, in the JAMA article, discusses the physiological mechanisms by which the GI works.  The abstract acknowledges the debate over GI in general, but the purpose of this review is to examine the mechanisms.  Now he could still have gone on to present these mechanisms as accepted, the subject of debate, or speculative, but he did not further clarify.

If someone makes statements in the affirmative and fails to include any so-called "qualifiers", they are conveying to the audience a certitude in those statements.  This is especially so in a review of this nature that does otherwise acknowledge a level of debate and controversy over the topic at hand.  In a way it's like saying, this over here is controversial, that over there is controversial, but this stuff here is just how it is.  The mechanisms are presented on a "this is how it is" basis ... there are no qualifiers such as "might" or "may be" or "there is much evidence to support".   Now, if adding those qualifiers at every single turn needlessly clutters a manuscript, this is easily remedied.  The insertion of the word "proposed" at the start of the subtitle along with a reminder here or there would suffice.   The matter is made worse by the fact that on some occasions, qualifiers are used, leading the reader to assume (consciously or not) that when such words are missing, the author means business!

The Mechanism of High-GI ~ 2002
by Davis S. Ludwig

Ludwig's review contains the following full page graphic, Figure 2:  Sequence of Physiological Events After Ingestion of a High-Glycemic Index Meal Compared With a Low-Glycemic Index Meal.   I've included a screenshot more for the full effect as I'll break up the phases into manageable graphics you can see while I intersperse them with the explanation in the text.   Note again that this is not presented as a hypothesis or proposed mechanism, but with a sense of certainty.

So now I've broken up the full-page figure and unfortunately it got cut off:   The Low-GI is on the left, High-GI is on the right.

Early Postprandial:  0-2 Hours After Meal

The rapid absorption of glucose following consumption of a high-GI meal challenges [] homeostatic mechanisms, complicating in effect the transition from the postprandial to the postabsorptive state. 

Within the first 2 hours after a high-GI meal, integrated incremental blood glucose concentration can be at least twice that after a low-GI meal containing identical nutrients and energy. This relative hyperglycemia, acting in concert with elevated concentrations of the gut hormones GLP-1 and GIP, potently stimulates insulin release from pancreatic beta cells and inhibits glucagon release from alpha cells. The resultant high insulin-to-glucagon ratio would tend to exaggerate the normal anabolic responses to eating, including uptake of nutrients by insulin-responsive tissues, stimulation of glycogenesis and lipogenesis, and suppression of gluconeogenesis and lipolysis.
All of this seems pretty fair although whether or not lipogenesis (whether he is referring to fatty acid synthesis -- de novo lipogenesis -- or triglyceride formation in adipose tissue) is increased significantly in this early phase seems unlikely.  I am not sure it has been studied per se, so we are probably in opinion realm here.

Middle Postprandial: 2-4 Hours After Meal

Between 2 and 4 hours after a high-GI meal, nutrient absorption from the gastrointestinal tract declines, but the biological effects of the high insulin and low glucagon levels persist. Consequently, blood glucose concentration falls rapidly, often into the hypoglycemic range. The physiological significance of this hypoglycemia is demonstrated by a greater fall in glucose oxidation rate after consumption of a high-compared with a low-GI carbohydrate during this interval.  Free fatty acid, the other major metabolic fuel, is more suppressed after a high-GI meal.
All emphasis is mine.  Is this true?  What exactly does Ludwig mean by "the biological effects ... persist"?  He appears to be referring to the hormone levels themselves.  Perhaps if we had a study wherein these factors were measured to point to in support of the assertions?   Oooh!  We're in luck.  And one conducted by Ludwig himself!  I am, of course, referring to this study:  High Glycemic Index Foods, Overeating, and Obesity.

Summary (skip if familiar!):  The relevant part of this study are the glucose, FFA, insulin and glucagon profiles reported after the morning test meal.  Keeping in mind that these were obese teenage boys fed a 400 calorie breakfast.  The Low GI was an egg omelet and fruit, 30P:40C:30F, while the mid and high GI meals were oatmeals 16P:64C:20F.  Mid-GI was steel-cut oats sweetened with fructose with milk and cream, while the High-GI was instant oats sweetened with dextrose (glucose) with lactase treated milk and cream.  As I discussed in The Glycemic Index ~ It Was Supposed to Be About Carbs!, the "Low GI" meal was inappropriate here, as macros varied, and had less to do with glucose digestion/absorption/bloodstream appearance rate than net amount of carb.  
I have adapted the below from Figure 1 from this study.
High-GI = Instant Oatmeal, Low-GI = Steel Cut Oats, Low-GL = Omelet & Fruit
For high vs. low GI comparison, compare red squares and black circles
Postprandial Phases:   Lavender = Early, Aqua = Mid, Peach = Late (note readings ended at 5 hr)
Apologies for the stretch, an attempt to line up vertically and still have a "squarish" graphic.
Before chopping it up I wanted you to have the full effect.  Now, I'll separate them out, because I think it is helpful to look at what is going on in a particular "phase" without being prejudiced by what happens before or after.  I don't think it is worthwhile to discuss glucagon, as the High vs. Low-GI just bounce around there below baseline and return.  This is normal, and if anything one might say the suppression was a bit extended vs. normal (see here, from this study).  It is the rise in glucagon for the Low-GLoad meal, likely attributable to the 2X protein & 2/3rdsX carb, that is "off" (though likely normal).  So as I chop up, let's look.

Below-left is the Early Postprandial Period.  Yes the glucose (top) and insulin (bottom) do vary widely, and keeping your eye on red vs. black, let's all agree that there is a difference in these responses for nutrient matched (64% carb) High-GI vs. Low-GI.  Please look at the serum fatty acids (aka FFA, aka NEFA) in the middle there.  Click on the image to enlarge as much as you like or scroll up.  There is no discernible difference in "fat getting locked away" here.


Above-right is the Mid Postprandial period.  The glucose levels are dropping, the High-GI dropping faster because they were higher.  But at 3 hours, everything is almost identical across the board (blue vertical sliver).  Glucose going down, insulin going down, fatty acids coming back.  After that?  Glucose and insulin leveling off, fatty acids rising.  If you really want to make something of a non-significant difference, from the 3-4 hour mark, the High-GI group has a teensy lag in rebound of fatty acids from levels virtually identical for all diets regardless of composition, at the 2 hour mark lasting for the two GI diets through 3 hours.

Returning now to Ludwig's GI Review ...

Late Postprandial: 4-6 Hours After Meal

Approximately 4 to 6 hours after a high-GI meal, the low circulating concentrations of metabolic fuels trigger a counter-regulatory hormone response that restores euglycemia by stimulating glycogenolytic and gluconeogenic pathways and elevates free fatty acid concentration to levels well above those observed after a low-GI meal. This combination of elevated counterregulatory hormone and free fatty acid levels resembles a state of fasting normally reached only after many hours without food.  After a low-GI meal, by contrast, hypoglycemia and its hormonal sequelae do not occur during the post-prandial period owing to continued absorption of nutrients from the gastrointestinal tract and rising hepatic glucose output.
Thus, consumption of meals containing identical energy and nutrients can produce markedly different physiological responses throughout a 6-hour period.

Through the 6-hour period.  OK ... well, it would be nice, if one is making such claims, that there be some studies/citations wherein these were actually measured.  In the oatmeal study, readings ended at 5 hours.  There they are on the right.

Ahh yes ... glucose and insulin leveling out with ever so slightly lower (about 10 mg/dL) glucose levels for the High-GI.  The fatty acids do appear to be on the upswing, while there is one data point making the Low-GI levels appear to possibly be leveling off.  You cannot conclude anything there though, certainly not to write such a mechanism into a review paper for a period extending to 6 hours.  In case you might be tempted, note that the Low-GLoad meal fatty acids are tracking right along with the High-GI meal.

Look again at the Mid Postprandial.  Do you see any significant differences in the circulating metabolic fuels?   

What I did leave out of my enhanced plots were the ones for growth hormone and epinephrine.  I'll include these (non-enhanced, squares=High-GI, dots=Low-GI, triangles=Low-GL) here.  So GH and epinephrine do "spike" at the 5 hour mark, but what was the effect?  Glucose is leveling out and fatty acids are on the rebound.  Too bad they didn't take the test out to 6 or 7 hours!  But then the boys would have eaten breakfast at 7 am and not gotten lunch until 2 pm ...  In other words, it is fairly usual to have a snack or meal around the 5 hour mark.

Metabolic Fuel Roller Coasters   

The sugar roller coaster has been beaten to death,
Early:  Spike
Mid:  Crash ►►►► Hunger!!
Late:  Rebound to baseline or slightly above.
Eat more sugar, rinse, repeat.

The fatty acid roller coaster is usually not presented as such.  Low carb advocates seize on the normal suppression of fatty acids in the postprandial period following a mixed meal, and claim your fatty acids remain locked away starving your cells for long after insulin and glucose have returned to normal (because in this version, they remain high).    Ludwig's fatty acid roller coaster is a bit different, it goes like this:
Early:  Plummet
Mid:  Stay low
Late:  Rebound to well above baseline ►►►► Hunger!! 
Eat more sugar, rinse, repeat.

Ludwig's claim is that the "combination of elevated counterregulatory hormone and free fatty acid levels resembles a state of fasting normally reached only after many hours without food." ... in other words you "feel" like you haven't eaten the whole day even though you ate six hours before.  This is because you have NEFA running amok through your veins!!  Wait, WUT?   I know, it sounds contradictory to what some may intuitively think, and to the mechanism by which the sugar roller coaster makes you hungry, but this is what Ludwig presented with a degree of certitude in a review entitled "The Glycemic Index: Physiological Mechanisms Relating to Obesity..."

Before I move on ... here are just the glucose (top), fatty acid (middle) and hunger ratings (bottom) for the 5 hours after the breakfast test meal.  Keep your eyes on the red and black in each phase to compare glycemic index.  The green triangles represent a meal with twice as much protein as the macro-matched High and Low GI meals.

This result should have raised questions with the researchers regarding any metabolic fuel hypothesis.  An unbiased person would be hard pressed to find a consistent relationship in any phase.  Further, the inclusion here of the high protein meal -- tracks in glucose with the Low-GI after about 1 hour, tracks in fatty acids with both GIs at virtually every time point -- which blows the low protein meals out of the water on the hunger scale, counters any circulating metabolic fuel effect.  (Hormones not much better, except glucagon.  A topic for another day, perhaps.)   (Again apologies for the distortion, I had to do some additional playing with proportions to get the time axis to match up on the hunger plot.)

Summary of How High GI Carbs Make Us Fat ~ 2002

  • Glucose and insulin spike higher in the early postprandial period.  Insulin remains high in the mid period as glucose falls.  Insulin remains in the late period causing hypoglycemia.
  • Fatty acids are suppressed in the early and midpostprandial periods.  This trips counterregulatory hormones and fatty acids climb rapidly in the late period to levels above baseline, which are reminiscent of a lengthy fast.
  • Thus low blood glucose and more elevated FFA signal a lengthy period without food and hunger in the late postprandial period.
Only One Problem ...  The results of Ludwig's own 1999 study bring these mechanisms into dispute.


Jane Karlsson said…
This is astonishing. Ludwig seems to be just as muddled as Lustig, who thinks insulin causes leptin resistance when the evidence says the opposite.

I'm starting to find all this muddle in the heads of people who are supposed to understand these things a bit scary.
carbsane said…
Check out this guy who is a PharmD and supposedly a cardiovascular researcher. He has apparently taken his toys and blocked me. Reading some of his "scientific" papers is truly scary. These couldn't have passed for term papers when I was in high school.

This crap making the rounds "current dietary guidelines that allow up to 25% of total cals as added sugars". WHERE?— Evelyn (@CarbSane) February 7, 2015
carbsane said…
BTW, note that I just asked WHERE? ... because it seemed outrageous to me. As it turns out, there's a Canadian organization that does say that. :(