Tuesday, June 11, 2013

Glucose vs. Fructose and Their Effects on Glucose, Insulin & Fat Oxidation in Men on Both Ends of the BMI Spectrum

"Fructose handles"? "Glucose handles"? "Saturated fat handles"? No, just the net result of a trashy diet.
I don't have to tell you that I don't buy into the "fructose is the devil" hysteria that's rampant in the blogosphere and certain parts of the scientific community. It's a matter of quantity and quality that determines the toxicity of a poison and in most of the "convincing" evidence on the detrimental effects fructose. I mean let's be honest, you don't have to be a rocket scientists to figure out that 5+ cans of Coke a day cannot be good for you (cf. "Fat Content Per Energy Drink 0g, Body Fat Gain Per Energy Drink 18g!"; read more) "194 Bananas in Three Weeks", on the other hand, are nothing to be afraid of (learn why).Have we been fooled again or is it just a high fructose corn syrup producer conspiracy?

Enough of the rants, let's get to the facts!

I guess that's enough for the "ranty" introduction. Let's now have a look at what a group of researchers from the School of Medicine in Portland has in stock for us: It's a paper titled "Change in postprandial substrate oxidation after a highfructose meal is related to body mass index in healthy men" that's about to be published in one of the future installments of Nutrition Research. As you will by now probably have figured out, the Anne C. Smeraglio and her colleagues had two things in mind, when they came up with the protocol that involved
  • What did the subjects eat? Participants were fed an egg omelet, bagel with cream cheese, and sweetened beverage breakfast consisting of one-third of their estimated daily caloric. The meal consisted of 30% fat, 15% protein, and 55% CHO (as % of energy). The CHO energy was further divided into complex and simple CHOs; 25% of the total calories were from complex CHOs and 30% of the calories were from either glucose or fructose added to the beverage.
    12 healthy men without diabetes, with a mean age of 25 (23-31) years and a BMI less than 30 kg/m²,
  • 2 visits at their labs that were separated by at least 1 week, but less than 1 month,
  • two meals that were high in glucose or fructose which were served in random order as a breakfast after an overnight fast, and
  • fasting for 7h after the ingestion of the standardized breakfast (sitting around watching TV or performing other, non-exciting quiet activities without the propensity to produce a catecholamine response)
During the experiment, the oxygen consumption and CO2 production were measured by indirect calorimetry to calculate resting energy expenditure and respiratory quotient (RQ; high RQ = burning predominantly glucose, low burning predominantly fat). The scientists also took blood samples at pre-defined intervals and collected the urine of their participants.
Figure 1: Insulin and glucose levels, as well as non protein respiratory quotient (high = carb oxidation; low = fat oxidation) 0-7h after the fructose and glucose breakfasts (Smeraglio. 2013)
The data in figure 1 is a summary of the the most "significant" results. In that the "quotation marks" enclosing the word "significant" is in my humble opinion the most significant information here - one that's encoded with irony, because after all, the only statistically significant effect the scientists observed were the ~2.5x higher insulin levels in the glucose group 60min after the ingestion of the test meal... yep, that's in the glucose group.

"There must be a mistake here!? Fructose is bad for you!"

The scientists have really done their homework as they did even take into account whether or not the amount of protein in the meals would have been responsible for differences in the respiratory quotient. The latter was not the case, the "baseline RQs between the fructose and glucose study visits were equivalent (0.82 ± 0.08 and 0.81 ± 0.10, respectively) and the p-value, indicating that there was a difference even rose from 0.72 to 0.75, when "when protein use was accounted for by evaluating NPRQ [non-protein respiratory quotient]" (Smeraglio. 2013)
Surprised? Well, I guess over all the lustig (=German for "funny") and unwarranted hoopla about how bad even small amounts of fructose are, you must have forgotten why scientists believed not too long ago that fructose could be the solution to, not the cause of the diabesity epidemic. After all, the paradigm of the mid to late 20th century was: Fructose does not spike glucose, so it should be the ideal sweetener for diabetics, because it is not necessary that your pancreas produces insulin to get rid of it.

I will not have to tell you, though that this assumption and the corresponding notion that totally replacing glucose with fructose would be a great idea is about as unwarranted, as the current fear of the "toxicity" of the small amounts fructose contained you'll be exposed to from a couple of pieces of fruit. I mean, let's take a peek at the data again.

Compared to the same amount of glucose, the consumption of the fructose equivalent of 5-6 medium sized (185g) apples (50-70g fructose, which is the amount of fructose the subjects in the study consumed) produces lower insulin levels and does not change either the leptin, triglyceride or glucose concentration in the blood or the ratio of glucose to fatty acid oxidation in healthy non-obese volunteers...

...apropos, non-obese, there was another thing to the headline wasn't there?

You are absolutely right, the research question involved (a) finding out what happens if you ingest a realistic breakfast where the carbohydrate content comes from (i) glucose or (ii) fructose and (b) determining whether the reaction would depend on the body weight / height² (BMI) ratio of the participants. So what about that, then? Let's see...
Did you know that there is a catalytic dose of ~40g of fructose per day (=6 normal size bananas) that will improve your glucose metabolism? (learn more)
"Although the absolute values for fat and CHO oxidation were not different between the fructose and glucose study visits, we did find a correlation between BMI and change in fat oxidation as a result of consuming the high-fructose meal compared with the high-glucose meal. The difference in fat oxidation (fat oxidation after the fructose meal minus fat oxidation after the glucose meal) was negatively correlated with BMI at the 4- and 7-hour time-points (Fig. 3; r =−0.59 [P= .04] andr=−0.59 [P= .04] for 4- and 7-hour time points, respectively) but not the 1-hour time point (Fig. 3; r=−0.52, P< .09). Nonprotein RQ displayed these same trends but did not reach significance." (Smeraglio. 2013)
As the scientists rightly point out, this suggests that the postprandial fat oxidation after the fructose meal was less than the fat oxidation after the glucose meal only among subjects with a higher BMI, and that the correlation with body weight, but not the difference itself was statistical significant.

Figure 2: Correlation of BMI with change in fat oxidation pearson correlation with linear regression trend lines between BMI and change in fat oxidation (fat oxidation after the fructose study visit minus fat oxidation after the glucose study visit; Smeraglio. 2013)
What's more, if you take a peek at the linear regression in the graph on the right hand side (figure 2) you will realize that that this does also mean that the fatty acid oxidation in lean individuals is actually increased after the ingestion of fructose. If I intended to drive my message "fructose from real foods is not your problem, folks!" home at all costs (which is what the "fructose is the devil" advocates like to do), I could seize on this observation and tell you: "Look folks, as long as you are already lean fructose will help, not impair your effort to get ridiculously shredded." I would yet hope that you are clever enough to see through this tactics and realize that neither the effect on the left hand side (=more fatty acid oxidation in the leaner folks with fructose vs. glucose), nor the one on the right hand side of figure 2 is physiologically relevant.

And that's not just because it's simply too small, but also because the ratio of glucose to fatty acid oxidation, i.e. the respiratory quotient (RQ) is not determining whether you store or lose body fat - if it were, you'd better be training in the "fat burning zone" for the rest of your (in that case) miserable lives.



Figure 3: Adding 7.5g of fructose ( to a 75g glucose load will improve not detoriate the glucose metabolism and that without increasing the amount of insulin that's released in response to the glucose load (Moore. 2000)
Bottom line: I am confident that that even without resorting to extreme interpretations of cherry picked data, the main message of today's article is clear. The comparatively small amounts of fructose you'll get right with the appropriate polyphenols & other cofactors from fruit and other fructose containing whole foods in your diet is not your enemy (Other items? Yeah, you know that even onions have 2g fructose, right?).

On a related note, you are aware that small amounts of fructose, like the 7.5g of fructose scientists added to the 75 g of glucose their 11 healthy subjects ingested during an oral glucose tolerance test had an up to 31% lower glucose response (these were the values for the 6 subjects with the highest level on the regular test) in the absence of concomitant increases in insulin response (see figure 3; Moore. 2000)!? You did not know that? Well, I guess it was about time to take a mental note, then ;-)

References:
  • Moore MC, Cherrington AD, Mann SL, Davis SN. Acute fructose administration decreases the glycemic response to an oral glucose tolerance test in normal adults. J Clin Endocrinol Metab. 2000 Dec;85(12):4515-9. 
  • Smeraglio AC, et al. Change in postprandial substrate oxidation after a high-fructose meal is related to body mass index in healthy men. Nutr Res.2013 [epub ahead of print]