June 12, 2014 - Blogs
Editors note: Fructose and its effects on health has been a hot button in public debate for several years now. So it was interesting when two contradictory studies were published recently about the fructose content of beverages sweetened with HFCS. These types of scientific yin and yang always fascinate me, but I never seem to have the time to analyze the merits of each one to see if the conclusions of one or the other might carry more weight.
Concerning these studies, I had a short discussion with an acquaintance, a nutrition student, who pointed out the differing analytical techniques (as well as the accusations of scientific fraud being leveled). Analytical chemistry not being my area of expertise, I mentioned I might find someone that could explain pros and cons, and why the numbers for fructose content were so different, especially since the same lab was involved in each of the studies. A few days later, he mentioned this blog by another nutrition student who had already gone that route. Best of all, the author, Kevin Klatt kindly allowed me to rerun it here.
-Lynn A. Kuntz
Dubiously Different Fructose Contents of HFCS-sweetened Beverages
The ever-growing focus on industry influence on the scientific method certainly fits the modern focus distrusting the government, assuming we live in a 1984 Orwellian dystopia. I am far from totally okay with how much money plays into politics nowadays, and Capitalism + Science isn't a love story for the ages. However, I believe that, in science, we should be thorough in calling out biases, and not let our own distrust bias our accusations.
I have long followed Marion Nestle's Food Politics, and have always enjoyed her commentary on the food industry, though I often find myself seeking out the reasoning of the alternative position without blindly following her opinions. One of her recent posts, "Guess who funded the contradictory fructose study?" was particularly problematic to me.
In the post, Nestle notes that two conflicting studies came out about the relative content of fructose in drinks. (And if you're interested in the fructose debate, I encourage you to read David Despain's blog post featuring an interview with John Sievenpiper.) The first study was funded by NIH (1), and followed up on a study by Ventura et al (2) that found the mean fructose content of beverages sweetened with high-fructose corn syrup (HFCS) to be higher than 55 percent—the value reported by the Corn Refiner's Association (CRA). The other study, entirely funded by the International Society of Beverage Technologists (ISBT), and authored by John White, Ph.D, previously known for defending HFCS's similarity to sucrose (4), found that the fructose content of HFCS-sweetened beverages randomly obtained from retail store shelves was concordant with the commonly accepted 55-precent fructose proportions.
Nestle doesn't go into the methodologies of the study, but states in the blog, "Really, these kinds of results are so predictable that all I have to do is see the results to guess who must have funded the study. Coincidence? I don’t think so."
Is there something dubious about the ISBT study? There are a couple important issues here related to different methodologies:
As noted in the discussion portion of the NIH study, the original LC data was limited because it didn't include maltose in their measurements—maltose, being a disaccharide of glucose, could affect the relative fructose:glucose. In “table 1: formulas," we can see a major limitation (noted by commenter Henry Chin in the comments of the food politics post): total sugar was measured as sucrose + lactose + maltose + galactose + fructose + galactose. Looking back at CRA’s site, HFCS is composed of 55 percent fructose, 42 percent glucose and 3 percent sugars and other polysaccharides. As noted in the ISBT study introduction, the remaining 3 percent is composed of maltose, maltotriose, maltotetraose and higher polysaccharides—all polymers of glucose. The NIH study appears to have missed out on measuring these, skewing the relative ratios to appearing to have higher fructose contents.
The ISBT study measured fructose, glucose and these other polysaccharides (referred to as DP2+) in 80 samples randomly pulled from retail store shelves. The measurement was taken by two independent laboratories that verified the separation and detection methodology and precision, using the accepted ISBT protocol. The authors go on to note that there are limitations to gas chromatography with acid catalysis, which can degrade oligosaccharides but also degrades fragile sugars, which has limited other (unpublished) attempts to report that HFCS contained more carbohydrate than listed (5). White even goes a bit further to suggest that the Ventura et al's concerns regarding sucrose in their previous findings (2) were a failure to consider acid-catalyzed sucrose inversion to free fructose and glucose that readily occurs in the low-pH environment of carbonated drinks, and that this concern "appears to based on incomplete understanding of sugars chemistry in carbonated beverages"—Oy.
For more on the analyses used for measuring saccharides, CRA openly lists its protocols.
Sometimes, industry association leads to a biased study. Other times, industry means that they are the experts and have hired the best scientists to determine the best methodologies to produce the products they state. Industry being involved with a study should not automatically mean that its results are to be discredited. I'm no analytical chemistry expert as it applies to sugar (I've used HPLC in relation to amino acids in breastmilk), and not many are. Those who are are employed by CRA and ISBT to figure this stuff out. It's one thing to ask questions about methodology and ask for transparency, it's another to accuse an entire industry of funneling more fructose into the food supply.
-Kevin C. Klatt
Kevin C. Klatt is currently a graduate student in nutritional sciences and is pursuing his PhD/RD. His research interests are in the realm of nutrigenomics and maternal-infant nutrition. He thoroughly enjoys scientific writing and communication, and is a big proponent of evidence-based nutrition. Follow him on Twitter @nutrevolve or read more of his blogs on Evidence-Based Nutrition+Genetics+Evolution.
References
1. http://www.sciencedirect.com/science/article/pii/S0899900714001920
2. http://onlinelibrary.wiley.com/doi/10.1038/oby.2010.255/abstract
3. http://www.nature.com/ijo/journal/vaop/ncurrent/pdf/ijo201473a.pdf
4. http://ajcn.nutrition.org/content/88/6/1716S.full
5. http://www.fasebj.org/cgi/content/meeting_abstract/24/1_MeetingAbstracts/562.1
Editors note: Fructose and its effects on health has been a hot button in public debate for several years now. So it was interesting when two contradictory studies were published recently about the fructose content of beverages sweetened with HFCS. These types of scientific yin and yang always fascinate me, but I never seem to have the time to analyze the merits of each one to see if the conclusions of one or the other might carry more weight.
Concerning these studies, I had a short discussion with an acquaintance, a nutrition student, who pointed out the differing analytical techniques (as well as the accusations of scientific fraud being leveled). Analytical chemistry not being my area of expertise, I mentioned I might find someone that could explain pros and cons, and why the numbers for fructose content were so different, especially since the same lab was involved in each of the studies. A few days later, he mentioned this blog by another nutrition student who had already gone that route. Best of all, the author, Kevin Klatt kindly allowed me to rerun it here.
-Lynn A. Kuntz
Dubiously Different Fructose Contents of HFCS-sweetened Beverages
The ever-growing focus on industry influence on the scientific method certainly fits the modern focus distrusting the government, assuming we live in a 1984 Orwellian dystopia. I am far from totally okay with how much money plays into politics nowadays, and Capitalism + Science isn't a love story for the ages. However, I believe that, in science, we should be thorough in calling out biases, and not let our own distrust bias our accusations.
I have long followed Marion Nestle's Food Politics, and have always enjoyed her commentary on the food industry, though I often find myself seeking out the reasoning of the alternative position without blindly following her opinions. One of her recent posts, "Guess who funded the contradictory fructose study?" was particularly problematic to me.
In the post, Nestle notes that two conflicting studies came out about the relative content of fructose in drinks. (And if you're interested in the fructose debate, I encourage you to read David Despain's blog post featuring an interview with John Sievenpiper.) The first study was funded by NIH (1), and followed up on a study by Ventura et al (2) that found the mean fructose content of beverages sweetened with high-fructose corn syrup (HFCS) to be higher than 55 percent—the value reported by the Corn Refiner's Association (CRA). The other study, entirely funded by the International Society of Beverage Technologists (ISBT), and authored by John White, Ph.D, previously known for defending HFCS's similarity to sucrose (4), found that the fructose content of HFCS-sweetened beverages randomly obtained from retail store shelves was concordant with the commonly accepted 55-precent fructose proportions.
Nestle doesn't go into the methodologies of the study, but states in the blog, "Really, these kinds of results are so predictable that all I have to do is see the results to guess who must have funded the study. Coincidence? I don’t think so."
Is there something dubious about the ISBT study? There are a couple important issues here related to different methodologies:
As noted in the discussion portion of the NIH study, the original LC data was limited because it didn't include maltose in their measurements—maltose, being a disaccharide of glucose, could affect the relative fructose:glucose. In “table 1: formulas," we can see a major limitation (noted by commenter Henry Chin in the comments of the food politics post): total sugar was measured as sucrose + lactose + maltose + galactose + fructose + galactose. Looking back at CRA’s site, HFCS is composed of 55 percent fructose, 42 percent glucose and 3 percent sugars and other polysaccharides. As noted in the ISBT study introduction, the remaining 3 percent is composed of maltose, maltotriose, maltotetraose and higher polysaccharides—all polymers of glucose. The NIH study appears to have missed out on measuring these, skewing the relative ratios to appearing to have higher fructose contents.
The ISBT study measured fructose, glucose and these other polysaccharides (referred to as DP2+) in 80 samples randomly pulled from retail store shelves. The measurement was taken by two independent laboratories that verified the separation and detection methodology and precision, using the accepted ISBT protocol. The authors go on to note that there are limitations to gas chromatography with acid catalysis, which can degrade oligosaccharides but also degrades fragile sugars, which has limited other (unpublished) attempts to report that HFCS contained more carbohydrate than listed (5). White even goes a bit further to suggest that the Ventura et al's concerns regarding sucrose in their previous findings (2) were a failure to consider acid-catalyzed sucrose inversion to free fructose and glucose that readily occurs in the low-pH environment of carbonated drinks, and that this concern "appears to based on incomplete understanding of sugars chemistry in carbonated beverages"—Oy.
For more on the analyses used for measuring saccharides, CRA openly lists its protocols.
Sometimes, industry association leads to a biased study. Other times, industry means that they are the experts and have hired the best scientists to determine the best methodologies to produce the products they state. Industry being involved with a study should not automatically mean that its results are to be discredited. I'm no analytical chemistry expert as it applies to sugar (I've used HPLC in relation to amino acids in breastmilk), and not many are. Those who are are employed by CRA and ISBT to figure this stuff out. It's one thing to ask questions about methodology and ask for transparency, it's another to accuse an entire industry of funneling more fructose into the food supply.
-Kevin C. Klatt
Kevin C. Klatt is currently a graduate student in nutritional sciences and is pursuing his PhD/RD. His research interests are in the realm of nutrigenomics and maternal-infant nutrition. He thoroughly enjoys scientific writing and communication, and is a big proponent of evidence-based nutrition. Follow him on Twitter @nutrevolve or read more of his blogs on Evidence-Based Nutrition+Genetics+Evolution.
References
1. http://www.sciencedirect.com/science/article/pii/S0899900714001920
2. http://onlinelibrary.wiley.com/doi/10.1038/oby.2010.255/abstract
3. http://www.nature.com/ijo/journal/vaop/ncurrent/pdf/ijo201473a.pdf
4. http://ajcn.nutrition.org/content/88/6/1716S.full
5. http://www.fasebj.org/cgi/content/meeting_abstract/24/1_MeetingAbstracts/562.1
BRUSSELS—European Union states have agreed 
