Research -------- From the article: "This small amount [of methanol] is readily metabolized by the body. This fact is illustrated by clinical studies in which aspartame failed to increase the levels of methanol (or formate, a metabolite) in blood or urine." [Stegink 1984a, Roak-Foltz 1984] First of all, the author cites a whole 650-page compilation of 33 articles for his first reference (Stegink 1984a). It would have been helpful if he had just cited one or two articles rather than the whole book. The second reference cited by the author to back up the above-quoted statement (Roak-Foltz 1984) is nothing more than a NutraSweet estimate of projected aspartame intake and has nothing to do with methanol or formic acid in the blood or urine. Rather than belaboring the point, I will address the clinical studies in the cited reference (Stegink 1984a) which does measure methanol and formic acid after aspartame ingestion. I will also address the recent chronic dosing study (Leon 1989) which the author cited earlier in the article. a. Frey (1976), page 496 of book. This G.D. Searle (NutraSweet)-funded study tested aspartame on 61 healthy children and sucrose on 65 healthy children (ages 2 - 21) for 13 weeks. Methanol blood and urine measurements were obtained after an overnight fast on 22 aspartame-ingesting subjects at the end of week 7 and week 13. No changes in blood or urine methanol were noted. Flaws ----- i. After an overnight fast, all of the ingested methanol would have either been eliminated or converted to formaldehyde and formic acid (DHHS 1993a, Liesivuori 1986, Stegink 1981a). It is ridiculous to test for methanol in the blood after an overnight fast. ii. Plasma phenylalanine measurements were also taken after the overnight fast. The fast allowed the plasma phenylalanine levels to return to normal or near normal levels. Another worthless test. iii. The subjects were all healthy. It is important to test healthy subjects, but the detrimental effects of methanol would likely be much greater in in subjects with particular health problems such as multiple chemical sensitivities, folic acid deficiency, or persons on pharmaceuticals that may increase the toxicity of methanol. iv. Test product was freshly prepared and not real-world aspartame products which would have a higher percentage of DKP, beta-aspartame, and possibly other breakdown products such as free methanol. v. Test product was given with food more often than normal. This would significantly delay the absorbtion of methanol and cut down on its toxicity. (Posner 1975). Conclusion ---------- A worthless methanol test and other major flaws in this industry-funded research project only proves that the researcher was completely unaware of how to properly test aspartame and its breakdown products. It is interesting to note that the eye complaints were limited to people ingesting aspartame as opposed to sucrose. b. Stegink (1981a), page 542 of book. While one may be able to imagine that the researcher in the Frey (1976) experiment didn't know how to properly take plasma methanol and amino acid measurements, this "experiment" contains so many ridiculous flaws that it seems to me to approach blatent dishonesty as opposed to a legitimate research project. This G.D. Searle (NutraSweet)-funded one-day experiment was performed with 30 healthy adults. Twelve subjects received a dose of 34 mg/kg of aspartame mixed in cold orange juice. The remaining 18 subjects were divided into three groups of six and received either 100, 150, or 200 mg/kg of aspartame. The blood methanol was measured for eight hours after ingestion. No significant blood methanol concentration increase was found at the 34 mg/kg dose. Significant increases in blood methanol concentrations were found in the 100, 150, and 200 mg/kg dose. The increases returned to near fasting levels eight hours after aspartame ingestion. Blood and urine formate concentrations were measured for the six subjects receiving 200 mg/kg of aspartame. The mean of the blood formate concentrations was listed for the first eight hours after aspartame ingestion. There were no statistically significant changes in the blood formate concentrations noted. There was a large, significant increase in urinary formate excretion during the first eight hours after aspartame ingestion. This increase was especially large (three times fasting levels) during the first four hours after aspartame ingestion. The authors claimed that "the rate of formate synthesis did not exceed the rate of format excretion, since blood formate levels were not elevated." Flaws ----- i. The aspartame was given with orange juice. This major flaw, by itself, renders the entire experiment worthless. The researchers claim that the "methanol content of fruit juices may range from 12 to 640 mg/l, with an average of 140 mg/l." Since the intent of the experiment was to measure methanol and formate levels, why in the world would they add what they claim to be a large amount of methanol in the form of fruit juice! By their own standards, this totally screws up their methanol and formate measurements. The important point to remember, though, is that the fruit juice almost certainly protects against methanol poisoning by keeping much of the methanol from being absorbed or converted to toxic metabolites such as formaldehyde and formic acid. Extremely high doses of aspartame such as 200 mg/kg will, of course, overwhelm the protective factors in orange juice. I believe that the researchers deliberately used fruit juice to lower the toxicity and skew methanol and formate measurements by changing its metabolism. They certainly didn't use it to add accuracy to their methanol and formate measurements. ii. The limit for methanol detection was 4 mg/l (or 0.4 mg/dl)! This very serious flaw renders their methanol tests completely worthless. Researchers who are actually interested in measuring methanol levels (as opposed to hiding negative results) from low- level exposure to methanol do not use tests that are incapable of measuring all but the largest spikes in plasma methanol levels. Cook (1991) used a test that was developed in 1981 that was capable of measuring plasma methanol levels lower than 0.26 mg/l. Cook showed that after exposure to methanol, the plasma levels rose in the subjects from a low of 0.26 mg/l to a high of 3.2 mg/l. (Averages increased from 0.57 mg/l to 1.88 mg/l.) Even the highest methanol spike in Cook's experiment would not show up using the worthless methanol test from this NutraSweet- sponsored experiment! d'Alessandro (1994) measured methanol levels below 1 mg/l in test subjects exposed to methanol. Davoli (1986) showed that only 8 mg/kg of aspartame caused a statistically significant rise in plasma methanol levels. It is obvious that a 34 mg/kg dose of aspartame will increase plasma methanol levels substantially. It appears that these researchers deliberately used an outdated, worthless methanol testing procedure so that they could claim that "no increase in blood methanol concentration was detected...." While they were clear in the publication what the limits of the methanol detection was, most people would not know that they used a bad test. iii. It may take as much as 12 to 16 hours after exposure to methanol for plasma formate to reach its maximum levels (McMartin 1975, Liesivuori 1987). The researchers in this project tested formate at intervals up to only 8 hours after exposure and then at the 24-hour mark. Therefore, it seems that they may have missed the period of time when plasma formate levels would be expected to spike to their highest level. iv. The average base blood formate concentration was 19.1 mg/l. This is an unusually high base blood formate level. This is more than twice as high as found by d'Alessandro (1994), Baumann (1979), and Heinrich (1982). Buttery (1988) measureed the plasma formate levels of 30 subjects and found that they ranged from approximately 4.8 to 11.2 mg/l. Osterloh (1986) noted an average value of 4 mg/l and an upper limit of 12 mg/l. While it is possible to find some people with that high a base formate level, for that to be an average seems a little bit ridiculous -- unless the subjects were breathing formaldehyde fumes before the test. As Kavet (1990) points out, the extremely high base formate levels and the variability in the formate measurements during this experiment would keep what might otherwise be a significant change in plasma formate levels from being noticed. It is also quite possible that there were errors in the plasma formate measurements or the assay itself may be faulty. Liesivuori (1986) mentions that analytical methods for measuring formic acid (including the method used in this industry experiment) are "notoriously inaccurate." v. The charts and numbers listed were averages for all the subjects for each time period. As discussed earlier, this technique tends to hide significant changes since following the administration of aspartame or methanol, each person's peak methanol and formate levels would likely occur at a different time. If Subject A's methanol level is sky-high at 2 hours, yet Subject B's level at two hours is still low and doesn't reach it's maximum until 4 hours, this would significantly bring down the average. In addition, any unusual increases in a small proportion of the subjects' measurements tend to not be seen in presentations of average values. vi. Each experimental group ingesting the higher doses of aspartame had a ridiculously small number of subjects, making statistical significance much more difficult to obtain. There were only 6 subjects in the group that had formate measurements taken. Given the wide variability in formate levels from one time period to the next and the tiny number of subjects in this group, obtaining statistical significance was highly unlikely. vii. This experiment was simply a one time ingestion of aspartame. As discussed earlier, short, single exposures to methanol or formaldehyde in the air often does not significantly raise formate levels. However, regular or longer-term exposure does show an increase in formate levels. While it can be helpful to conduct a few quality acute dosing studies, such experiments prove very little as far as the "safety" of aspartame or methanol when ingested regularly or even when ingested from time-to-time. It is important to keep in mind that conducting hundreds of these acute dosing studies shows us very little about the affect of aspartame as it is used in the real world-- regular, long-term ingestion. The very poor quality of these acute-dosing studies simply means that instead of showing us very little, they show us only that research on aspartame must be conducted independently in order to study the adverse health effects of aspartame. viii. The subjects were all healthy. It is important to test healthy subjects, but the detrimental effects of methanol would likely be much greater in subjects with particular health problems such as multiple chemical sensitivities, folic acid deficiency, or persons on pharmaceuticals that may increase the toxicity of methanol. ix. There were numerous useless tests for an experiment of this length. There was an ophthalmological examination before and after the aspartame ingestion. Eye damage occurs gradually after chronic methanol exposure, e.g., months or years after aspartame ingestion begins. There were blood tests for total protein, albumin, Ca, inorganic P, lactate dehydrogenase, total bilirubin, glutamic-oxaloacetic transaminase, Na, K, Cl, and CO2. All of these tests are worthless for such a short experiment which had so many experimental errors. Once again, the researchers measured the plasma amino acids levels after the subjects had fasted overnight. The plasma amino acid levels would return to normal or near normal after that length of time. Unfortunately, the large number of tests allow the researchers to claim that this proves the "safety" of aspartame. x. The aspartame given to the subjects was fresh aspartame and not nearly the same product that many unfortunate people are taking on a regular basis. The product sold at the stores have often degraded due to the length of time they have sat on the shelves and/or due to the exposure to high temperatures. Conclusion ---------- This experiment appears to be quite deceptive. Giving aspartame with a product that you know will screw up your measurements it quite a mistake! Using an outdated methanol test that can only measure very high levels of methanol is another particular serious mistake. Add to that, all of the other flaws: ridiculously high base formate levels, possible inappropriate times at which formate measurements were taken, use of average values for each time period, small numbers of subjects in each group, only a single ingestion of aspartame, healthy subjects, numerous useless tests, and fresh aspartame used. This experiment is a perfect example of just how bad "research" can get when G.D. Searle or the NutraSweet Company is connected in any way with the project. One wonders what good, if any, is the "peer review" process if a study like this can get published. The fact that a number of editors of this publication worked for G.D. Searle (the company that developed aspartame), may have something to do with how this study got published. It also may help explain why so many flawed industy- connected aspartame studies were published in this journal (Journal of Toxicology and Environmental Health). Even with all of its major flaws, this study contradicts the author's statement that "aspartame failed to increase the levels of methanol (or formate, a metabolite) in blood or urine." c. (Stegink 1983a), page of book. This General Foods Corporation-funded study was similar to the previous study, except that it was conducted on infants (8-14 months of age). Ten infants were given a single 34 mg/kg dose of aspartame mixed in a freshly-prepared, cherry-flavored beverage. Eight infants were given a dose of 50 mg/kg, and 6 received a dose of 100 mg/kg. Only the blood methanol concentrations were measured from the 1 ml of blood withdrawn at 4 different intervals within 3 hours following aspartame ingestion. Important measurements of blood formate and urinary excretion of formate were not done because, according to the researchers, there was too small an amount of blood available in the samples. The results of the blood methanol test were similar to that in the experiment discussed immediately above. Blood methanol concentrations showed no significant increase for infants ingesting the 34 mg/kg dose. Significant increases were seen in doses of 50 and 100 mg/kg. Flaws ----- i. It is two years later and these researchers are still using the same outdated, worthless methanol testing procedure that was used in the experiment discussed above. Therefore, any claim by the researchers that "no increase in blood methanol concentration was detected after administration of aspartame [34 mg/kg]..." is blatently deceptive. ii. Stegink's comparison of methanol from juices and methanol from aspartame seems particularly ridiculous. Stegink shows that a dose of 100 mg/kg of aspartame raises an infant's plasma methanol levels approximately 10 times above its base level. Stegink assumes that an infant weighs 10 kg. This means that they gave the infants approximately 1000 mg of aspartame. This is equivalent to 100 mg of methanol. The researchers cite the Francot (1956) study which claims that black current juice contains over 600 mg/l of methanol. One hundred milligrams (100 mg) of methanol (the equivalent amount to what the infants were given with aspartame) would be found in approximately six ounces of black current juice according to Francot's study. Stegink uses an example of an infant drinking six ounces of liquid and thus he must believe that this is a reasonable volume that an infant can ingest. Therefore, according to Stegink's reasoning, 6 ounces of black current juice given to an infant should raise that infant's blood methanol level by nearly 10-fold. I would love to see them try to show a 10-fold increase in blood methanol levels and a corresponding increase in blood formate levels after 6 ounces of black current juice were given to a 10kg infant. iii. This experiment did not test blood and urinary formate levels. Therefore, we do not know much about what happened to the methanol after it was absorbed into the bloodstream. iv. This experiment was simply a one time ingestion of aspartame. While it can be helpful to conduct a few quality acute dosing studies, such experiments prove very little as far as the "safety" of aspartame or methanol when ingested regularly or even when ingested from time-to-time. It is important to keep in mind that conducting hundreds of these acute dosing studies shows us very little about the affect of aspartame as it is used in the real world-- regular, long-term ingestion. v. The aspartame given to the subjects was fresh aspartame and not nearly the same product that many unfortunate people are taking on a regular basis. The product sold at the stores have often degraded due to the length of time they have sat on the shelves and/or due to the exposure to high temperatures. vi. The charts and numbers listed were averages for each time period. As discussed earlier, this technique tends to hide significant changes since following the administration of aspartame or methanol, each person's peak methanol and formate levels would likely occur at a different time. If Subject A's methanol level is sky-high at 2 hours, yet Subject B's level at two hours is still low and doesn't reach it's maximum until 4 hours, this would significantly bring down the average. In addition, any unusual increases in a small proportion of the subjects tend to not be seen in presentations of average values. Conclusion ---------- These researchers are still using a worthless methanol testing procedure. One wonders how many years this same "test" will be used to "prove" that aspartame does not increase plasma methanol levels. We will find out later in this document. They also repeated many of the same mistakes they made in the previous experiment. Finally, this experiment was of less use since formate levels were not measured at all. Even with all of its major flaws, this study contradicts the author's statement that "aspartame failed to increase the levels of methanol (or formate, a metabolite) in blood or urine." d. Leon (1989). From the article: "Results of this study showed no significant differences between [experimental and placebo] groups with regard to urinary excretion of formate (a metabolite of methanol)." In this G.D. Searle (NutraSweet)-funded study, 108 subjects were divided into two groups. Fifty-three (53) subjects ingested capsules containing 75 mg/kg per day of aspartame in three equal doses along with meals. The 55 subjects in the placebo group ingested cellulose capsules along with meals. Each group took their doses for 6 months. A large number of laboratory tests were performed throughout the study. In addition, plasma amino acid tests were performed to test for phenylalanine and LNAA levels. (These tests will be discussed in later sections.) Blood methanol and blood formate measurements were taken at the start of the experiment and after weeks 6, 12, 18, and 24. Urinary formate measurements were taken at the start of the experiment and after weeks 6, 12, and 24. Most of the blood methanol readings were below the detectable levels. Only at the week 18 test was there a much greater proportion of persons in the aspartame group with a detectable blood methanol level. No data was given, but the investigators stated that there was no significant difference in the measured blood formate, urinary formate excretion (or formate to creatinine excretion ratio) between the aspartame and the placebo group. Flaws ----- i. The aspartame used was fresh, dry aspartame administered in capsules. Sixteen years after the FDA pointed out the lack of information on the difference in bioavailability between real-world aspartame and capsule administration (Freeman 1973), six years after the National Soft Drink Association pointed out the enormous chemical differences in real-world aspartame as compared to fresh aspartame (NSDA 1983), four years after Tsang published detailed data proving an enormous difference in the chemical composition between aspartame freshly prepared in capsules and real- world products, and two years after Stegink proved that aspartame dissolved in liquid has a much more extreme effect than capsule-enclosed aspartame (Stegink 1987a), these researchers had no legitimate excuse for using capsule administration of aspartame. One of the investigators on this project was from the University of Iowa and should have been intimately familiar with the work of Stegink (who is also from the University of Iowa), especially Stegink's (1987a) study showing the difference in bioavailability between liquid and encapsulated aspartame. This obvious mistake of using encapsulated aspartame completely invalidates much of this experiment. The plasma amino acid measurements are worthless. Capsule administration reduces the toxicity of aspartame significantly for the following reasons: - Fresh aspartame is used instead of the more toxic soup that is found in real-world products. - The plasma amino acid (i.e., phenylalanine and aspartate) spikes are considerably lessened with capsule administration of aspartame. - Methanol absorption is delayed. This may reduce the methanol toxicity as pointed out earlier. The researchers may claim that they needed to use capsules for a double-blind study. The answer to such a claim is as follows: 1) Even if it were a good excuse to use capsule administration, the lack of bioequivalence between capsule and real-world aspartame administration still invalidates much of this experiment; 2) Most of the experiment was simply performing blood tests which do not need a double-blind protocol; 3) It may be possible to create a double-blind experiment using real-world, liquid aspartame-containing products as long as the taste is disguised. This has been done with MSG experiments in the past. ii. The aspartame capsules were ingested with full meals. This is one of many mistakes which combine to invalidate this experiment. The effects from the free amino acids would not be nearly pronounced when aspartame is taken with full meals. (I will go into more detail about this in a later section.) As Posner (1975) points out, taking methanol with meals slows its absorption and may affect its toxicity. The combination of this flaw plus the use of capsules would render formate and other measurements nearly useless. The chance of finding an adverse reactions to fresh, encapsulated aspartame, taken with full meals is tremendously reduced. One wonders if these researchers deliberately designed this experiment to produce negative results. iii. The blood methanol tests were taken after a 12-hour fast. Anyone who has done any reading on the subject of methanol would know that a blood test for methanol would be worthless after a 12-hour fast (as discussed earlier in this section). Had the investigator who conducted the blood methanol test in this experiment, Thomas Tephly, Ph.D., taken the time to read another paper that he co- authored (Stegink 1981a), he would have known that methanol gets converted to formic acid in the body causing the methanol concentration in the blood to return to normal levels within twelve hours after aspartame ingestion. On top of the worthlessness of taking the methanol test at the wrong time, these researchers used the same worthless blood methanol test that was used in the experiments discussed previously. This time, however, these researchers claimed that the blood methanol test had a limit of detection of 0.31 mmol/liter or, using the conversion factor presented in DHHS (1993a), approximately 1.0 mg/dl (10 mg/l). This is three times less sensitivity claimed than the previous experiments dicussed and as much as 45 times less sensitive than the methanol test used by Cook (1991)! One wonders how these researchers could possibly have found any of the methanol measurements over the limits of detection let alone 10 to 20 percent of the limits of detection. iv. The blood methanol and formate tests were taken on weeks 0, 6, 12, 18, and 24. The urine formate and creatinine tests were taken on weeks 0, 6, 12, and 24. What happened to week 18? Is it typical to skip an important test for no particular reason? Is it possible that the test was done, but the data did not meet the researchers' expectations? This seems rather strange to say the least especially since this happens to be the week that blood methanol tests somehow showed the aspartame group well above the placebo group. v. The exclusion criteria stated, in part, that "All women included in the study were required at entry to be postmenopausal, to have been surgically sterilized, to be taking oral contraceptives, or to have had an intrauterine device in place for at least 6 months to prevent the possibility of pregnancy." The problem with this exclusion criteria is that women make of the majority of reported adverse reactions to aspartame (Mullarkey 1992, page 70, CDC 1984). It is not inconceivable that women who have certain hormonal fluctuations may be more susceptible to acute adverse effects from aspartame. Therefore, the researchers may have excluded a large percentage of the population who would have actually experienced adverse reactions (had the experiment not been so badly flawed). There appears to be no good reason for excluding this population of women. If these researchers insist on not studying anything but a perfectly healthy population, the least they could do is use a reasonable representation of a healthy population rather than further limiting it to whatever their whimsical desires call for. If, in the very unlikely occurrance, one of the women had become pregnant and experienced an adverse reaction, they could have simply noted that in their discussion. Instead, they chose to take an unrepresentative sample of subjects (healthy persons) and make it less representative by excluding women who may be more susceptible to acute adverse reactions.. vi. One subject with a history of bronchial asthma was dropped from the study due to symptoms of headaches, nausea, and malaise after aspartame capsule ingestion. The subject was rechallenged for a much shorter period of time (9 days as compared to 6 months), with a much smaller dosage (9 - 25 mg/kg capsules over a nine day period as compared to 75 mg/kg per day) with customized meals (which may have included foods to further offset the effects of methanol and amino acids). The subject experienced fewer reactions to aspartame under these conditions. The subject was rechallenged at least one more time (the paper does not mention under what conditions) and no aspartame reactions were found. This is an abuse of double-blind studies. The researchers cannot simply take results that they don't like such as a serious adverse reaction and then keep challenging the subject and changing the protocol until the desired results are achieved. Why didn't the researchers change the protocol and rechallenge persons who did not experience reactions? vii. The subject that was dropped from the study had a history of bronchial asthma. The exclusion criteria stated that subjects should be excluded from the experiment if they had "any chronic disease detected by history, physical examination, or routine laboratory tests." What was this person doing in the experiment in the first place? He should have been excluded according to the protocol. If these researchers cannot even follow their own exclusion criteria, and change the protocol for rechallenges of this subject as described above, one wonders how many other times they deviated from the protocol and did not mention it in the publication. viii. Plasma phenylalanine and phenylalanine/LNAA measurements were taken after the overnight fast. The fast allowed the plasma phenylalanine levels to return to normal or near normal levels. Another worthless test. While fasting plasma phenylalanine levels in some persons may change over time and cause serious health problems after regular ingestion of aspartame (as per the example given near the start of this review), that would not have been seen here because the use of fresh, encapsulated aspartame taken with meals skewed the results so badly. ix. The formate test had only an 80% chance of detecting an average change of 0.13 mmol/L (5.2 mg/l) in blood levels. (Blood formate levels average approximately 0.1 mmol/l or 4 mg/l as discussed earlier.) Therefore, a significant increase could occur using this test without it being noticed. The blood formate measurements were not even presented. These values would have been much more relevant than the hematology values had the experiment not otherwise had so many major flaws. Even if the values had been presented, Liesivuori (1986) mentions that analytical methods for measuring formic acid (including the method used in this industry experiment are "notoriously inaccurate." x. Once again, the charts and numbers listed were averages for all subjects at each time period. As discussed earlier, this technique tends to hide significant changes since following the administration of aspartame or methanol, each person's peak methanol and formate levels would likely occur at a different time. If Subject A's methanol level is sky-high at 2 hours, yet Subject B's level at two hours is still low and doesn't reach it's maximum until 4 hours, this would significantly bring down the average. In addition, any unusual increases in a small proportion of the subjects tend to not be seen in presentations of average values. xi. The quality of the study was monitored by an employee of the NutraSweet Company. This, in itself, I believe is a flaw due to the extreme corporate bias. Conclusion ---------- Even with all of these major flaws, the aspartame group experienced approximately 50% more adverse reactions than the placebo group. This significant increase was played down in the text by breaking up the adverse reactions into categories such as "headaches." One wonders if the results would have even been more striking had the researchers used real-world aspartame products at a similar dose and not built so many other serious flaws into this experiment. A research project like this can make a person lose all faith in the scientific process. Only industry (e.g., NutraSweet) would fund and monitor a study this poorly designed. The study was presented at the 1988 Annual Meeting of the Federation of American Societies for Experimental Biology (FASEB). Such a presentation should have caused the audience to begin crying over the death of the scientific process. It boggles the mind that a peer-reviewed journal would accept this study for publication without addressing some of the serious questions raised above. It simply proves that the peer- review process does not prevent worthless, industry- sponsored studies. I would prefer to say that the experiments discussed above are the worst research I have ever seen. Unfortunately, these are quite typical of industry-sponsored studies. Many of the pre-approval studies for aspartame which will be discussed later were much worse -- so bad that rolling dice would have been a more accurate method for determining the "safety" of aspartame. Yet, somehow, even the pre-approval studies met the FDA's "standards" for acceptable research upon which a safety determination could be made. NutraSweet/FDA Arguments ------------------------ Animal Studies -------------- Studies of methanol on rats and Rheusus monkeys cannot be extrapolated to humans. The differences between how these species handle methanol is described in detail by Roe (1982). Sturtevant (1985) of G.D. Searle & Co. claims that Martin- Amat (1978) created an animal model using Rhesus monkeys. However, Roe (1982) points out that, as other investigators have found, it is "very unlikely that in rhesus monkeys (blood pH about 7.2 (av.) and blood formate about 7.5 mmol/l) amaurosis and atrophy of the optic nerve can be caused by methanol." Roe also details the significant differences between methanol metabolism in rhesus monkeys and humans as well as the many times greater toxicity of methanol in humans. Finally, even if an animal model for acute methanol poisoning is created in the future, it is unlikely that the model will apply to low-level, chronic methanol poisoning since the mechanism is probably somewhat different (i.e., no severe acidosis) The NutraSweet Company's attempts to use the Rheusus monkey as a test animal for aspartame/methanol poisoning only proves that they do not understand or care about the researched differences of the effects of methanol between humans and that particular species of monkeys. Human Studies ------------- Sometimes, as "proof" that low levels of methanol does not cause damage, the NutraSweet Company or the FDA cite several references. One study that is sometimes cited is a 1952 study published in the British Journal of Industrial Medicine (Leaf 1952). The FDA states (Federal Register 1984, page 6677): "In fact, studies in human subjects given oral dosages of methanol of 71 to 84 mg/kg body weight showed no toxic effects with blood levels of methanol reaching 47 to 76 mg per liter 2 to 3 hours afterwards." The reality is: 1. The study was performed on only five men given a dose of 2.5 to 7.0 ml (29-84 mg/kg) not 71 to 84 mg/kg. 2. It was a one-day, single-dose study of methanol and did not test chronic ingestion over months or years as happens with aspartame ingrestion. 3. There was no detailed discussion of reactions that may have been caused by the single dose of methanol in this experiment. 4. The first part of the study where methanol was given by itself showed that a large percentage of methanol was converted to formic acid. In the second part of the study where ethanol was taken concurrently with methanol, the ethanol blocked the conversion of methanol to formaldehyde and formic acid. (Remember, aspartame contains no protective ethanol.) 5. The investigators summarize as follows: "Owing to the slow rate of elimination of methanol from the body, repeated exposure to the vapour or liquid may result in accumulation and under such conditions the use of methanol would constitute a toxic hazard. . . . The elimination of methanol after doses of 2.5 to 7.0 ml has been studied in five human subjects. At any time the rate of elimination was found to be proportional to the concentration of methanol in the body. The significance of this finding is discussed. Only a very small fraction of the ingested methanol (about 2%) was eliminated via the respiratory and urinary routes." 6. A significant portion of this publication addressed the important part ethanol plays in protecting against methanol poisoning. As was mentioned earlier, aspartame contains no ethanol or other protective factors. The FDA attempts to give further examples of the alleged lack of toxicity of methanol at low levels (Federal Register 1984, page 6677): "From estimates based on blood levels in methanol poisonings, it appears that the ingestion of methanol on the order of 200 to 500 mg/kg body weight is required to produce a significant accumulation of formate in the blood which may produce visual and central nervous system toxicity (Rowe 1982, Friedman 1980)." The references cited by the FDA are only discussing severe, acute toxicity (i.e., poisoning) and not long-term, chronic toxicity (i.e., slow poisoning). Unfortunately, this is the type of obfuscation that one can expect from the FDA and the NutraSweet Company -- keeping people busy chasing down references which are irrelevant, but which allow them to convince the uninformed scientist that aspartame is "safe." NutraSweet researchers will sometimes cite two more recent studies which they claim to show that methanol poisoning cannot occur from aspartame (Stegink 1989, Stegink 1990). Many of the same serious problems, bordering on outright deception, that were found in Stegink (1981a) are also found in these studies. Stegink (1989) tested six normal adult subjects ingesting eight 8-ounce servings of Kool-Aid each sweetened with 600 mg of aspartame. The servings were taken at one hour intervals. On another day, the subjects followed the same procedure, but the Kool-Aid was unsweetened (i.e., no aspartame). Stegink (1990) was nearly the same experiment, but the subjects were individuals heterozygous for Phenylketonuria (PKU). (PKU is a genetic disorder where the person lacks an enzyme necessary to metabolize the amino acid, phenylalanine. A PKU Heterozygote has a slightly impaired phenylalanine metabolism.) Both experiments showed a significant increase in plasma phenylalanine levels as well as plasma phenylalanine/LNAA ratios. (This will be discussed in more detail in the Phenylalanine section.) No statistically significant difference was seen in the blood methanol levels, blood formate levels, or urinary formate levels on the day where aspartame was ingested compared to the day it was not ingested. Flaws ----- i. These "researchers" are still using the same outdated (1969), non-sensitive, worthless methanol test that they were using in 1981! With all the money being poured into this lab from the NutraSweet Company, one would think that they could afford equipment for accurate and useful tests. Stegink (1990) took it one step beyond a mere deceptive, worthless methanol test. A graph was shown of blood methanol levels over 24 hours. The "apparent" aspartame day methanol levels were shown as the exact same as the Kool-Aid (alone) day. This is extremely deceptive for two reasons. The levels shown were more than three times below the detection limit of the methanol test used and therefore the figures on the graph were a wild guess (at best). Davoli (1986) had already shown a significant increase in blood methanol levels after a single dose of 500 mg of aspartame and therefore Stegink, et al. must have known that the methanol levels rose significantly and that the graph was complete nonsense. ii. It may take as much as 12 to 16 hours after exposure to methanol for plasma formate to reach its maximum levels (McMartin 1975, Liesivuori 1987). The researchers in this project tested formate at intervals up to only 8 hours after exposure and then at the 24-hour mark. Therefore, it seems that they may have missed the period of time when plasma formate levels would be expected to spike to their highest level. iii. The average base blood formate concentration was again ridiculously high -- 21.0 mg/l in Stegink (1990). This is two or three times higher than in almost any independent experiment (d'Alessandro 1994, Baumann 1979, Heinrich 1982, Buttery 1988, Osterloh 1986). While it is possible to find some people with that high a formate level, for that to be an average seems ridiculous -- unless the subjects were breathing formaldehyde fumes before the test. As Kavet (1990) points out, the extremely high base formate levels and the variability in the formate measurements during this experiment would keep what might otherwise be a significant change in plasma formate levels from being noticed. It is also quite possible that there were errors in the plasma formate measurements or the assay itself may be faulty. Liesivuori (1986) mentions that analytical methods for measuring formic acid (including the method used in this industry experiment are "notoriously inaccurate." iv. The charts and numbers listed were averages for all the subjects for each time period. As discussed earlier, this technique tends to hide significant changes since following the administration of aspartame or methanol, each person's peak methanol and formate levels would likely occur at a different time. If Subject A's methanol level is sky-high at 2 hours, yet Subject B's level at two hours is still low and doesn't reach it's maximum until 4 hours, this would significantly bring down the average. In addition, any unusual increases in a small proportion of the subjects' measurements tend to not be seen in presentations of average values. v. Each experimental group ingesting the higher doses of aspartame had a ridiculously small number of subjects, making statistical significance much more difficult to obtain. There were only 6 subjects in the test groups. Given the wide variability in formate levels from one time period to the next and the tiny number of subjects in this group, obtaining statistical significance was highly unlikely. vii. This experiment was simply a one day ingestion of aspartame. As discussed earlier, short, single exposures to methanol or formaldehyde in the air often does not significantly raise formate levels. However, regular or longer-term exposure does show an increase in formate levels. While it can be helpful to conduct a few quality acute dosing studies, such experiments prove very little as far as the "safety" of aspartame or methanol when ingested regularly or even when ingested from time-to- time. It is important to keep in mind that conducting hundreds of these acute dosing studies shows us very little about the affect of aspartame as it is used in the real world--regular, long-term ingestion. The very poor quality of these acute-dosing studies simply means that instead of showing us very little, they show us only that research on aspartame must be conducted independently in order to study the adverse health effects of aspartame. viii The aspartame given to the subjects was fresh aspartame and not nearly the same product that many unfortunate people are taking on a regular basis. The product sold at the stores have often degraded due to the length of time they have sat on the shelves and/or due to the exposure to high temperatures. Conclusion ---------- These two studies, while slightly better than the single dose experiments, were still only one day long. Numerous other flaws as well as deceptively faulty testing procedures make it impossible to draw any useful conclusion from the experiment. Sometimes NutraSweet researchers will try to argue that aspartame cannot possibly increase formate levels more than 9 mg/l (for 3 liters of diet cola) and is therefore within "normal" blood formate range -- meaning that it could not possibly be harmful. This argument was put forth by Shahangian (1984). The problem with this argument is that people have their own unique blood chemistry. To arbitrarily add a significant amount of formate to the blood and say that it is okay simply because they can find someone else in the world with that level of formate (meaning that the level is still "normal") does not make sense. An increase in formate of 9 mg/l would be considered by many health professionals to be a dangerous jump in formate levels if it occurred regularly. Chronic methanol poisoning does not necessarily mean that there are large jumps in the blood and urine methanol and formic acid levels. Small increases occurring regularly, over many months or years may be enough to cause damage a little bit at a time, especially in persons who 1) do not take the ethanol protective factor with aspartame ingestion, 2) have a folic acid or other nutrient deficiencies, 3) have a sensitivity to methanol or its toxic metabolites such as the growing number of persons with Multiple Chemical Sensititives (MCS), and 4) are taking pharmaceuticals which may react negatively with methanol ingestion. Finally, we have to keep in mind that the mechanism(s) for acute methanol toxicity and chronic methanol toxicity may be different in some ways and similar in others. Any highly technical argument by NutraSweet trying to convince readers that low levels of methanol is safe will have to address the following issues: a. Low levels of chronic methanol exposure in industry has caused adverse reactions after a relatively short period of time. The technical argument will have to show how and why low levels in industry cause adverse reactions, but levels not much lower than that from aspartame would somehow miraculously not cause any reactions even though it is proposed for a lifetime of use in even the most chemically sensitive individuals. b. Given that environmental researchers agree that there is a lack of data on long-term exposure to low-levels of methanol, how can NutraSweet, with any conscience, conduct a clinical trial of low levels of an extremely dangerous, human-specific poison on much of the population. This is especially true since prelimary experiments of short-term exposure to low levels of methanol show minor changes in brain function. Recent Case History Samples --------------------------- 1. I'm a 29 year old atheletic female victim of deadly NutraSweet poisoning. I began to use it in the spring of 1990, and soon afterwards my condition began to deteriorate. I never imagined my strange symtoms were from this chemical, after all its approved by the FDA and on restaurant tables. My eyesight began to fade until I was almost blind in one eye. Next my hearing became dull and my legs, feet and torso lost sensation and became numb. I was dizzy most of the time, irritable and depressed and had constant terrible headaches. I became very clumsy, dropped and bumped into things. I used to skate but having lost my balance I had to give it up. My mind was affected and I thought I had a brain tumor. Reading comprehension and memory drastically decreased. I though I was dying. Two different doctors made the diagnosis of Multiple Sclerosis. One was an eye specialist I saw about my vision loss. This spring a friend explained to me that all my symptoms were coming from aspartame (NutraSweet) and that I had methanol toxicity which mimics Multiple Sclerosis. I was only drinking three Diet Cokes a day. I could hardly believe the horrible nightmare I was living was sold in 6 packs at the grocery store, but I would try anything so I stopped. Right away I began to feel better and today most of my symptoms have completely disappeared. My friend, Betty, was right, I was suffering from methanol toxicity. My hearing and vision have returned and so has my memory. I'm no longer clumsy and have regained my balance, and feeling has returned to my lower extremities. 2. I became a big drinker of Diet Coke, drinking more and more because I liked the taste. When I had vision disturbances my physician said it was optic migraines. There were silvery lines in my eyes and I would get dizzy. Sometimes I would have these optic migraines twice a day, and up to 6 times a week. One day I baked cookies with NutraSweet and really got worse. Never did I associate my problems with this substance. About six months ago I was given some information on aspartame (NutraSweet) and immediately stopped using it in any form. Within a week I realized I wasn't having any optic migraines or dizziness. In six months I've only had two small episodes, and feel they are related to having accidently consumed something with aspartame in it like the ice cream I had recently. I was not aware aspartame was in it until after I had eaten it. .... P.S. Since we've been notifying others of the dangers of aspartame they, too, have abstained, many with the disappearance of symptoms, including my husband, Bob. It turns out this is a very serious problems and the average person does not consider the association of this toxin with their medical problems. It is easily established this is the culprit when symptomatology does not return after elimination. Conclusion ---------- Methanol ingestion by aspartame users is much greater than what they would naturally ingest with fruit juices. The lack of natural ethanol protective factor makes methanol from aspartame even more dangerous. It is likely that toxic methanol metabolites are not produced in significant amounts after ingestion of food products as they are after ingestion of aspartame. Persons with certain health conditions or who are taking certain pharmaceuticals may be even more susceptible to chronic methanol poisoning. Clinical experience has shown that many persons who have ingested aspartame regularly for an extended period have suffered from symptoms which are common in methanol poisoning. NutraSweet-funded studies usually do not test chronic ingestion of aspartame, are usually hopelessly flawed, at best, and, in my opinion, represent a deliberate attempt to deceive. The question we have to ask ourselves is whether it is fair to sell a product which essentially provides the equivalent of part-time employment for much of the population, from the healthiest to the sickest, from children to pregnant women, at methanol-laden printing shops and chemical plants. Given that more and more people are becoming sensitive to chemical exposure and that an increasing number of people are developing chronic immunological and neurological problems, is it fair to significantly add to the toxic load without warning the consumers? Not only is it not fair to experiment with low levels of poison on the population, but contributing to the destruction of so many people's health simply to make money borders on criminality in my opinion.