HFCS study authors defend work against attacks
Photo: BoekeMarion Nestle, along with other nutritionists have joined the Corn Refiners Association in criticizing the recent Princeton study on High Fructose Corn Syrup. Indeed the very title of Nestle’s post on the subject — “HFCS makes rats fat?” — seems to question the well-established practice of using rats to test hypotheses regarding human nutrition.
While I personally found her objections unpersuasive, she raised several issues with the researchers work that have been echoed far and wide. She questioned the methodology of the study as well as whether there was indeed a direct comparison made between HFCS and sucrose. She also claimed the following:
[A]s summarized in Table 1 in the paper, the researchers did only two experiments that actually compared the effects of HFCS to sucrose on weight gain, and these gave inconsistent results. Their other experiments compared HFCS to chow alone.
…Although the authors say calorie intake was the same, they do not report calories consumed nor do they discuss how they determined that calorie intake was the same. This is an important oversight because measuring the caloric intake of lab rats is notoriously difficult to do (they are messy).
In response, Jennifer LaRue Huget of the Washington Post, in an otherwise unsympathetic blog post, contacted the study’s lead author, Princeton’s Dr. Bart Hoebel, to give him a chance to respond to Marion Nestle’s (and others’) criticisms.
In sum, he basically declared Nestle incorrect. According to Hoebel, the researchers did compare rats which consumed HFCS with those which consumed sucrose directly and they did carefully record and present in the paper precise caloric intake for each rat, as well as the source. He also observed that they put anti-drip guards on the liquid dispensers and picked up any spilled chow to eliminate the risk of overestimating caloric consumption. As Hoebel put it a bit puckishly: “This is all explained in the article for anyone who wants to study it carefully.”
It appears that Nestle didn’t review the study rigorously. It’s certainly true that she has long been skeptical of finger-pointing studies on particular ingredients. She feels strongly that all processed sweeteners are bad (which they are) but her reaction to this study 1) suggests that in this case her bias has gotten the better of her objective review of new research and 2) inadvertantly aids the very corporate behemoths that she has spent much of her career combating. The food industry is already touting Nestle’s opposition to the study as an indication of its lack of merit.
Tom Philpott summed up the issue well in his recent post concerning the idiocy of our obsession with corn:
All processed sweeteners add empty calories to food; but calorie for calorie, HFCS appears to be even worse than white sugar. Although the two sweeteners have roughly the same fructose/glucose ratio, we mammals seem to metabolize the HFCS differently than we do cane sugar.
Still, I have reached out to the study author myself to see if he can clarify some of the other objections that have arisen. There is clearly hostility in surprising quarters to evidence that HFCS may accelerate weight gain and cause increased risk of metabolic disorders in humans. And yet it’s a crucial piece of the obesity epidemic puzzle. I don’t understand why Nestle and others who care about addressing obesity don’t see it that way.
The full exchange between the WaPo’s LaRue Huget and Dr. Hoebel is below:
JLH: Can you explain, for the nonscientific folks out there, how reliably rat-study results translate to humans?
BH: The rat is much like the human in regard to feeding behavior and basic metabolism. Rats evolved eating a diet something like ours; in some cases our own garbage. Thus the rat is a reasonable “animal model” of human feeding behavior in many cases. There are some cases in which rats taste things differently and metabolize food differently than people. Rat studies need to be reproduced in people whenever possible, after the critical variables have been established using laboratory animals…. To see a review of all fructose studies in animals and people, I recommend [George Bray’s article in the February 2010 issue of Current Opinion in Lipidology, “Soft drink consumption and obesity: it is all about fructose.”] However, note that many human studies compare pure fructose with other sugars, which is not what we did. We offered rats high-fructose corn syrup or sucrose, which are the common ingredients in soft drinks. The results do not apply directly to humans, but strongly suggest the need for more experiments along these lines.
JLH: Marion Nestle criticized the study for not being clear about the rats’ calorie consumption. Can you speak to that?
BH: She says: Although the authors say calorie intake was the same, they do not report calories consumed. We say: Caloric intake was reported in the Results section for Experiment 1. She says: nor do they discuss how they determined that calorie intake was the same. We say: In the Methods section we explain that we measured HFCS, sucrose and chow intake daily. We computed the calories consumed, which is also described in the Methods section. She says: This is an important oversight because measuring the caloric intake of lab rats is notoriously difficult to do (they are messy). We say: We do not see any oversight. The drinking tubes had an anti-drip device built in and we collected spillage for the food pellets for accuracy. We reported the caloric intake and the standard error, which shows the variability in intake for a given group. This is reported in the Results section.
JLH: Also, many critics felt the study was designed poorly, never pitting HFCS directly against sucrose and testing for what seemed to be almost arbitrary time periods. Again, anything you could offer to help us better understand those issues would be helpful.
BH: Not true. In male HFCS was compared directly to sucrose in Exp 1. In Exp. 2, we compared HFCS to sucrose in females, and we describe in the paper our reasons for using 12-h and ad libitum access to sucrose vs. HFCS. We were interested in comparing the effects of limited vs. continuous access.
JLH: Finally, something about the way the study was written sounds as though researchers set out to link HFCS to obesity, not to determine whether such a link exists. It feels as though the dots aren’t all connected very clearly. Can you address that, please?
BH: The study was designed to explore the comparison of HFCS with sucrose and at the same time to compare 24 h access with 12 hr access, plus males with females. Keep in mind that this is research, and one does not know which will be the key results when starting out. We discovered that male rats drinking HFCS were heavier than the matched sucrose controls in Exp 1 and heavier than the other groups in Exp 2. Regarding females, again the HFCS rats were the heaviest; although the sucrose control group had sugar available for less time per day, we had reason to think this was not a critical variable based on our prior publications in which 12 and 24 hr access to sucrose leads to the same body weights. Moreover the females, like the males, showed elevated triglycerides and increased fat deposits. This is all explained in the article for anyone who wants to study it carefully.