Flawed aspartame study

A couple of months ago I posted my take on Kevin Trudeau’s book Natural Cures “They” Don’t Want You To Know About. I had occasion to look up Mr. Trudeau’s book on Amazon.com recently, and I ran across a review by one Robert P. Beveridge containing the following paragraph that I found interesting.

I have a hard-and-fast rule for diet books that, after reading this, I am going to expand to all health books: you can judge the veracity of the author’s beliefs by said author’s feelings about aspartame [the artificial sweetener in Equal]. I’ve never had to raise this past the first level of disbelief, as most authors will just categorically state that aspartame is bad for you, at which point you can dismiss them from your open-mindedness and rest assured you’re dealing with a quack.

It appears that Mr. Beveridge’s definition of open-mindedness is anyone seeing things the same way that he does.
I’m going to risk having Mr. Beveridge think me an un-openminded quack because I don’t particularly like aspartame. I don’t like the stuff for a number of reasons. First, a molecule of aspartame breaks down in the body releasing a molecule of aspartic acid, a molecule of phenylalanine, and a molecule of methanol. The first two are amino acids that we typically consume in our diets and are harmless; the latter – methanol – is transformed into formaldehyde, a not particularly benign substance when taken orally. So, one reason I don’t like aspartame is because it metabolizes into a toxic chemical albeit it pretty tiny amounts. But then, I can’t think of any amount of formaldehyde that I would willingly consume.
Second, in my years of clinical practice I’ve come across a number of patients who have had bizarre reactions to aspartame that have run the spectrum from short term memory loss to severe unremitting abdominal pain.
And, third, aspartame is difficult to work with. It’s not heat stable so you can’t cook with it. It has a funny (to me, at least) aftertaste, especially if used in large amounts.
Having said all this, I can tell you that I have not done an exhaustive medical literature search on aspartame, so I can’t quote chapter and verse about all the problems it supposedly causes. Why haven’t I done an exhaustive search? Because I’m just not that interested. I would much rather spend the time and whatever brain power I have on other subjects that interest me more. If I were desperate to use aspartame, I would undertake such a search, but since I don’t use it myself and I don’t recommend it to others for the above reasons, I pretty much ignore it.
Aspartame has about 200 times the sweetening power of sugar, which means that one gram of aspartame has the sweetening power of 200 grams or one cup of table sugar. One gram of aspartame is a lot of aspartame. It’s about 2 1/2 to 3 times the Acceptable Daily Intake in the United States, but I can tell you that despite the supposed problems with aspartame and despite the fact that I don’t use it myself I, believe that almost anyone would be better off consuming one gram of aspartame per day instead of one cup of sugar.
This long preamble is a lead in to the real meat of this post, which is about an atrocious Italian study purporting to show that aspartame is a menace that should be removed from the shelves. Even thought I’m not a fan of aspartame, I’m less of a fan of shoddy science. And just because shoddy science supports a position I believe in doesn’t make it any less shoddy.
The study in question is published in Environmental Health Perspectives, an online journal published by the U.S. Government.
This paper (click here for full text of study) starts by detailing what the authors consider major shortcomings of previous studies of the carcinogenic potential of aspartame. They point out that in the animal studies used to obtain FDA approval for aspartame the animals consuming aspartame were sacrificed and examined for tumor development long before their lives would have run their normal course. The authors contend that the animals should have been allowed to live out their natural lives before being evaluated for cancer, and that the premature sacrifice of the animals allowed them to be examined for tumors before the cancers had enough time to develop. These authors also contend that in the previously done studies the groups of animals studied were not large enough to provide meaningful results. The study in question was designed answer these doubts.
The researchers went to great lengths to randomize large numbers of Sprague-Dawley rats (100-150/sex/group) into groups consuming varying daily doses of aspartame in their feed and a control group consuming none. The study began when the rats were 8 weeks old and ended at 159 weeks when the last rat died. All the rats in the study were examined by autopsy for cancerous changes.
After analysis of the data the researchers determined that the groups of rats consuming aspartame developed both malignant and non-malignant tumors at a much larger rate than did the control (non-aspartame consuming) rats. What’s more, they found that the rats developed cancer in a dose-response manner, i.e., the rats consuming the most aspartame got the most cancer, those consuming moderate amounts developed less, and those consuming small amounts had even fewer tumors.
All in all the study was a pretty strong indictment against the continuing use of aspartame. The authors concluded that

Our study has shown that aspartame is a multipotential carcinogenic compound whose carcinogenic effects are also evident at a daily dose of 20 milligrams per kilogram of body weight (mg/kg), notably less than the current acceptable daily intake for humans.

They call for

an urgent reevaluation of the current guidelines for the use and consumption of aspartame.

It all seems pretty cut and dried.
But is it?
Let’s look a little deeper.
First, it must always be borne in mind that rats are not just furry little humans. Things that cause problems in rats don’t always cause problems in humans and vice verse.
Second, Sprague-Dawley rats, the ones used in this study, are a strain of rats that are extremely prone to develop cancer. Just because something causes a Sprague-Dawley rat to develop cancer does not mean that the same thing will cause a human to do so. Far from it, in fact. The authors as much as acknowledge this fact when they point out in their criticism of an earlier rat study in which Wistar rats showed no propensity to develop cancer due to aspartame consumption that

it cannot be disregarded that this strain is more resistant than Sprague-Dawley rats to developing cancer.

In other words, if we want to find cancer, let’s use animals that develop it at the drop of a hat.
A third problem with this study is that the statistical significance of the numbers of rats developing cancers were given as p values that, although showing statistical significance, were not accompanied by CI (confidence interval) limits. If the CI were only 90% CI values, then the p values are meaningless. Scientists typically use 95% CI values and list them – since these authors didn’t see fit publish their CI values along with their p values my suspicions are raised that they lowered the CI to get the p they wanted. This might seem a persnickety observation on my part, but it really isn’t because if the p values aren’t valid, then there is no statistical significance in the study. (If anyone is interested I’ll be happy to post on p values and confidence intervals and what they all mean in simplified terms. Just let me know.)
The last couple of issues I have with the study are the most problematic as far as “urging a reevaluation of the current guidelines” goes. Although the animals consuming the aspartame had much higher rates of tumor formation as shown by autopsy than did those consuming no aspartame, they lived just as long. Buried deep within the paper is this sentence:

No substantial difference in survival was observed among the groups.

Hmmm. Must not have been very malignant cancers.
Looking at the mortality curves (located on the next to last page of the study) one can see that there is a shift to the right with increasing consumption of aspartame, which means that the rats consuming the most aspartame actually lived the longest.
Of course the authors don’t spend a lot of time dwelling on these findings because they detract from their fairly shrill conclusions.
Once again, a careful reading of a study gives us an entirely different picture than the authors and the press want us to have. (Click here to see a press report on this study, and you’ll see what I mean.)
After reading this paper I still don’t like aspartame, I still won’t use it myself, and I won’t recommend it to others. But this paper as presented is crap, and shouldn’t be used by anyone to frighten people who have made the decision to use aspartame. I especially resent having my tax dollars promote this kind of nonsense.