Toxic L-tryptophan: Shedding Light on a Mysterious Epidemic—Government Agencies Disagree on Cause of EMS

Toxic L-tryptophan: Shedding Light on a Mysterious Epidemic—Government Agencies Disagree on Cause of EMS

by William E. Crist

The FDA’s finding that control L-tryptophan itself was a contributing factor to the syndrome is surprising, because it is inconsistent with the epidemiological evidence linking production lots of implicated L-TRP to cases of EMS. Earlier studies by leading researchers, including those at the Center for Disease Control (CDC), clearly suggested that L-tryptophan itself was not responsible for EMS and that the disease was caused by contaminant(s) in a single manufacturer’s L-TRP.[1-3]

In August 1989, researchers headed by Edward A. Belongia, M.D., from CDC, reported in The New England Journal of Medicine on a study of EMS cases in Minnesota, “Our data suggests that the syndrome is not produced by tryptophan itself.”[2]

Steven Auerbach, M.D., and Henry Falk, M.D., at CDC’s Center for Environmental Health and Injury Control, reported, “Even at the time of the initial reports, the epidemiological pattern suggested that EMS was not caused by tryptophan per se, but was probably due to a contaminant. Further research linked LTCP’s [L-TRP containing products] ingested by EMS case-patients to specific lots of a single manufacturer of bulk L-tryptophan.”[4]

A CDC study by Mary L. Kamb, M.D., and associates clearly established a dose relationship in South Carolina patients who consumed varying dosages of one particular brand (i.e., Showa Denko’s). Eighty-four percent had definite or possible EMS from taking more than 4 grams/day of implicated L-TRP, “suggesting that a contaminant and not tryptophan itself was the etiological [casual] agent, since persons in this cohort who took [high doses of] non-implicated brands of tryptophan were not similarly affected.”[ 5, 6]

Gerald J. Gleich, M.D., at the Departments of Immunology and Medicine, Mayo Clinic and Foundation, summed it up: “The issues are twofold: 1. Tryptophan itself clearly is not the cause of EMS in that individuals who consumed product from companies other than Showa Denko did not develop EMS. 2. The evidence points to Showa Denko product as the culprit and to the contaminants as the cause.”[7]

But FDA interpreted these findings differently.

In policy statement letters and Congressional Hearings following the epidemic, FDA officials seemed to ignore the consistent epidemiological evidence, in favor of shifting the focus of blame away from Showa Denko’s product to include all sources of the food supplement L-tryptophan.

David Kessler, FDA Commissioner, in testimony before a Congressional subcommittee in July 1991, said that FDA was not convinced that the impurities in the bad batches of Showa Denko L-tryptophan were the total answer. Some cases of eosinophilia-myalgia syndrome occurred prior to the 1989 epidemic and had been linked to other batches and to other sources of L-tryptophan. “For that reason, the FDA continues to keep L-tryptophan off the market, although a number of other amino acids continue to be available….”[8]

An FDA letter on L-tryptophan-related EMS mailed to a U.S. Senator stated, “Both initial and subsequent epidemiological studies on the EMS epidemic have identified cases of EMS, and another related disease, eosinophilic fasciitis, that occurred before the 1989 epidemic that appear to be related to other batches or sources of L-tryptophan…. Taken together, these findings support previous suggestions that the L-tryptophan-associated EMS was caused by several factors and is not necessarily related to a contaminant in a single source of L-tryptophan….”[9]

Stuart L. Nightingale, M.D., FDA Associate Commissioner for Health Affairs, stated, “Althoughno other company’s product has been definitely linked to EMS [emphasis added], in some instances the source of LT could not be fully ascertained. However, several observations have caused FDA not to eliminate other brands of LT, or LT itself, as causal or contributing to the development of EMS:

  • Three to five percent of the EMS cases have not been definitely linked to Showa Denko’s product.
  • At least eight cases with EMS-like symptoms have been associated with 5-hydroxy- tryptophan a compound related to LT, but produced from botanical sources.
  • Surveillance for EMS has demonstrated that “non-epidemic” cases were occurring prior to the epidemic in 1989-90….”

“[T]hese findings raise serious questions regarding the safety of high dose levels of ‘uncontaminated’ LT.”[10]

Let’s examine these observations and reasons used by FDA to justify its ban on over-the-counter L-tryptophan sales:

1. Three to five percent of the EMS cases have not been definitely linked to Showa Denko’s product—The 3-5 percent figure cited by FDA apparently represents EMS cases that were untraceable to an L-TRP manufacturer. However, because any epidemiological study involves a reconstruction of events that led to an outbreak, such a failure to determine all of the facts is hardly unusual. Three to five percent ambiguity is rendered even more insignificant by the FDA’s own admission, “No other company’s product has been definitely linked to EMS.”[11]

The combined results from three trace-back studies (in Oregon, Minnesota and New York) are compelling: Out of 189 EMS cases, pills from all but two traced unambiguously to Showa Denko L-TRP. Of those two, one patient’s pills were initially traced to another manufacturer but were later tested by high performance liquid chromatography (HPLC) and found to have a “signature” (pattern of contaminant peaks) similar to Showa Denko product and dissimilar to the L-TRP produced by the company to which it was initially traced. The other case patient whose L-TRP pills were traced to another manufacturer had consumed two different brands before onset of illness, and the second brand was untraceable—which meant it could have come from Showa Denko.[1]

“Taken together, the Oregon, Minnesota, New York, and FDA/CDC product-tracing studies provide strong evidence implicating the tryptophan produced by one company, Showa Denko, as the cause of the EMS outbreak. All of the exceptions (traces to other companies) are easily attributable to the ‘noise’ expected in following a product through the multiple levels of its distribution system,” stated Kilbourne, a CDC epidemiologist.[1]

Thus, in three major trace-back studies published on the EMS epidemic, 99.5 percent of the cases were traced to Showa Denko’s L-TRP, and only one case (out of 189) was not clearly traceable. So it is unclear what FDA’s source is for the 3-5 percent of EMS cases that it claimed didn’t trace to Showa Denko.

2. At least eight cases with EMS-like symptoms have been associated with 5-hydroxytryptophan a compound related to LT, but produced from botanical sources—If eight 5-hydroxytryptophan [5-HTP]-related cases with EMS-like symptoms are relevant to FDA’s policy questioning the safety of L-tryptophan, then why does FDA still allow 5-HTP to be sold over-the-counter as a food supplement?[11]

“No EMS cases have ever been proven to be caused by 5-HTP (or, for that matter, uncontaminated L-tryptophan),” reported Joshua H. Beisler, in Rutgers Law Journal.[11] 5-HTP is not synthetically produced, which is significant because the tainted batches of EMS-implicated L-TRP were attributed to changes in the manufacturing process.[11,2] As an extraction from the seed of an African plant (Griffonia simplicifolia), 5-HTP “avoids the contamination problem,”[12] which bacteriologically-produced L-TRP is subject to. “Consequently, the chance of 5-HTP ever being associated with an outbreak of EMS is extremely small, if not non-existent,” said Beisler.[11]

These isolated, non-referenced cases of 5-HTP are a separate issue. Otherwise, FDA would have to include the 20,000 cases in Spain of Toxic Oil Syndrome (TOS), which also exhibited clinical features similar to EMS.[2 ]; (TOS was linked to consumption of an aniline-derived contaminant, 3-PAP, in denatured canola oil. Implicated L-TRP also contained an aniline-derived impurity, 3-PAA.)[14]

3. Surveillance for EMS has demonstrated that “non-epidemic” cases were occurring prior to the epidemic in 1989-90 —As previously explained, the non-epidemic cases of EMS and eosinophilic fasciitis (EF) that occurred for several years prior to the epidemic also appear to be linked to batches of Showa Denko’s L-TRP, manufactured using genetically engineered strains II-IV (see Pre-epidemic Cases of EMS and Where Did the Contaminants Come From?). FDA has assumed that these cases were associated with other manufacturers’ L-TRP, but the scientific literature is quite clear that all traceable EMS cases, whether epidemic or pre-epidemic, were linked to Showa Denko’s product.[15,1 ];The trace-back data on pre-epidemic EMS cases is limited, but still no other manufacturer’s L-TRP was clearly associated with the disease,[15,16] and no other company was sued over the EMS tragedy.[11]

Regarding the L-TRP-related EF cases, surprisingly, no trace-back studies have been done. The scientific literature cites one case of L-TRP-related EF that was rediagnosed as having EMS after news of the epidemic.[1] This suggests that other pre-epidemic L-TRP-related EF cases may also have been misdiagnosed, which would not be surprising because at that time (before the clinical discovery of EMS in late October 1989), no medical definition existed for the syndrome, so physicians could not make a proper diagnosis.[13] To clarify this key issue of cases of pre-epidemic EF—and other pre-epidemic L-TRP related diseases that shared common symptoms with EMS, such as scleroderma, fibromyalgia, and perimyositis—I searched and identified eleven such early victims, including two which were EF cases, and all eleven had been rediagnosed as having EMS following the epidemic (see 1987 EMS Victim–Initially Diagnosed with Fibromyalgia, Summary Profiles of Eleven Pre-epidemic EMS Cases, andExcerpt from “EMS Lawsuits”).

FDA has assumed that these pre-epidemic L-TRP-related EF and other early cases of EMS were caused by other manufacturers’ L-TRP. However, the scientific data, albeit limited, combined with information on these early cases traced in legal proceedings, clearly suggests that they were also linked to Showa Denko product, which apparently was contaminated in much lesser degrees and/or in few batches for several years prior to the epidemic.[17]

A study by CDC researchers found varying levels of contaminant EBT in case-associated Showa Denko L-TRP production lots dating back to August 19, 1986.[17]

“The amount of EBT present in Showa Denko tryptophan varied markedly in the period 1987-1989, presumably reflecting alterations in the manufacturing process,” said Arthur Mayeno and Gerald Gleich, researchers at the Mayo Clinic and Foundation. “It is likely that all of the contaminants varied with time. These data are consistent with the hypothesis that a contaminant in tryptophan is responsible for EMS and the sporadic cases of EF between 1986 and 1988.”[18]

[T] hese findings raise serious questions regarding the safety of high dose levels of ‘uncontaminated’ LT—The epidemiological evidence clearly does not support this conclusion by FDA, because the South Carolina study of L-tryptophan users showed that people who consumed high doses (i.e. more than 4 grams/day) of uncontaminated L-tryptophan—from other brands than Showa Denko—were not similarly affected with EMS.[5,6]

In summary, FDA has presented no substantial or compelling evidence to support its position that uncontaminated L-TRP was a causal factor in EMS. All of FDA’s reasons and/or observations mentioned above are based on unverified assumptions, including the agency’s interpretation of finding in the Lewis rat animal study by Love, et al. (see Problems Identifying and Testing for Trace Contaminants).

On this latter point, Charles Yanofsky, at Stanford University, commented:

“If they [toxic products] were produced during purification then it is very unlikely that they would be produced in animals from tryptophan itself. In any event testing high doses of tryptophan (or anything) for long periods in experimental animals is hardly an adequate test. Tryptophan is metabolized by man and important products like serotonin, niacin, and others, are derived from it.

“Over-administration of any natural metabolite is likely to generate byproducts, which may or may not be toxic. Unless you know exactly how a toxin works, and can perform a valid functional test, it is scientifically unsound to claim that an animal shows what could be the same symptoms from over-administration, and therefore conclude erroneously that the same toxin is produced from tryptophan in animals.”[19]

“Administering a natural metabolite at high doses for an extended period is not a valid test of safety. This procedure might be appropriate for foreign substances which are not known to be metabolized and for which there is no knowledge of the normal levels that are established in the body, but it makes no sense for natural substances that are metabolized.[20]

Any natural metabolite if taken in excessive doses or if improperly purified could conceivably be a source of toxic substances that cause serious health problems. In addition, high levels of a metabolite may influence other metabolic pathways. However, administering sensible levels of natural compounds based on known features of metabolism is very unlikely to cause any medical problem, otherwise that problem would exist naturally.”[20]

The EMS problem doesn’t exist naturally, evidenced by the epidemiological data clearly linking EMS only to contaminated Showa Denko L-TRP. Therefore the EMS pathological features caused by implicated L-TRP must be fundamentally different from the side effects produced in laboratory animals from over administration (i.e., high doses) of uncontaminated L-TRP, a natural metabolite. FDA appears to have confused the real issue, using the side effects from high doses of control L-tryptophan in animals to divert attention from Showa Denko’s contaminated L-TRP—and its genetically engineered bacteria.

Otherwise, if tryptophan, an essential amino acid, were unsafe, why would current FDA policy permit L-TRP usage for medical purposes and in infant formulas?

“It makes no sense at all,” said Hans Fisher, Ph.D., who has conducted several animal studies on L-TRP at Rutgers University. “However, if people consume large amounts of LT, the conversion products, serotonin, kynurenine, etc., could have harmful effects. Thus there is reason to suggest care as to LT administration. Too much aspirin causes stomach bleeding!”[21]

But aspirin is not banned from the over-the-counter market—and L-tryptophan is.

This has caused alarm among some public health policy organizations and professionals questioning FDA’s real motives.[11]

“Through efforts to increase its regulatory control over dietary supplements, the FDA has exploited this unfortunate [EMS] tragedy and often cites to the incident when recommending greater regulatory control over dietary supplements…. The FDA’s ban of L-tryptophan… illustrates that the FDA’s motivations for increasing its regulatory power include an unacknowledged political element that exists independently of any desire to protect and preserve the public health,” wrote Beisler. [11]

Dennis Mackin, an attorney who represented an EMS victim at trial, commented, “The FDA had a political agenda that truth was not going to deter. Dr. Kessler of the FDA wanted to get control of food supplements in addition to drugs. Therefore it would not behoove his situation to lay the blame on an impurity in Showa Denko’s LT product…[H]e testified before the U.S. Congress that it wasn’t proven that Showa Denko was the culprit because the ‘etiological’ agent had not been identified.” Mackin said that the judge at his client’s trial would not admit Kessler’s testimony as evidence on Showa Denko’s behalf because the judge agreed with the plaintiff that Kessler had an agenda in testifying before Congress and had not been subjected to cross-examination.[22]

“At bottom, the FDA public ban of safe, uncontaminated L-Tryptophan is uneven, expensive, and biased in favor of the pharmaceutical industry,” said Dean Wolfe Manders, in Social Policy. “It is time for appropriate congressional committees to review openly and aggressively the entire matter of L-tryptophan.”[23]

Uncontaminated L-tryptophan is safely used today in the United States in infant formulas, medical foods, weight loss products, and animal feed; has never been associated with EMS in any of the many other countries that allow its over-the-counter sale; and may be the only effective treatment for EMS. (It’s true: The US Government issued patent number 5185157 on February 9, 1993, to use L-tryptophan to treat and cure EMS.)[11]

Next section: Unanswered Letters to Government Officials


  1. Edwin M. Kilbourne, “Eosinophilia-myalgia syndrome: Coming to Grips with a New Illness,” Epidemilogical Reviews, (1992), Vol. 4, pp. 16-36.
  2. Edward A. Belongia, M.D., et al., “An Investigation of the Cause of the Eosinophilia-Myalgia Syndrome Associated with Tryptophan Use,” The New EnglandJournal of Medicine (August 9, 1990), Vol. 323 (6): pp. 357-365.
  3. Laurence Slutsker, et al., “Eonsinophilia-Myalgia Syndrome Associated with Exposure to Tryptophan from a Single Manufacturer,” Journal of the American Medical Association (1990), Vol. 264 (2): pp. 213-217.
  4. Steven B. Auerbach and Henry Falk, “Eosinophilia-myalgia Syndrome: CDC Update,” Cleveland Clinical Journal of Medicine (May-June 1991), Vol. 58 (3): pp. 215-217.
  5. Lee D. Kaufmann and Rossanne M. Philen, “Tryptophan: Current Status and Future Trends for Oral Administration,” Drug Safety (1993), Vol. 8(2): pp. 89-98.
  6. Mary L. Kamb, et al., “Eosinophilia-myalgia syndrome in L-tryptophan exposed patients,Journal of the American Medical Association (January 1, 1992), Vol. 267(1): pp. 77-82.
  7. Gerald J. Gleich, Position Statement, National Eosinophilia-Myalgia Syndrome Network, courtesy Sharron Lobaugh, president, email communication, May 25, 2000.
  8. Newspaper article by Barbara Brachtl, American Journal (in Seattle area), Sept. 15, 1993.
  9. Letter by Diane E. Thompson, Associate Commissioner For Legislative Affairs, FDA, to The Honorable Howell Heflin, U.S. Senate, March 24, 1994.
  10. Letter by Stuart L. Nightingale, Associate Commissioner for Health Affairs, FDA, September 3, 1992.
  11. Joshua H. Beisler, “Dietary Supplements and their Discontents: FDA Regulations and the Dietary Supplement Health and Education Act of 1994,” RutgersLaw Journal, Winter 2002, pp. 511-550.
  12. Michael T. Murray, N.D., email communication, April 25, 2005.
  13. Frank Silvestri and John Massicot, “EMS Lawsuits,” National Eosinophilia-Myalgia Syndrome Network Newsletter (June 2001), Vol. II (2), pp. 5.
  14. Arthur N. Mayeno, Ph.D., et al., “3-(Phenylamino)alanine, a Novel Aniline-Derived Amino Acid Associated with the Eosinophilia-Myalgia Syndrome: A Link to Toxic Oil Syndrome?” Mayo Clinic Proceedings (1992), Vol. 67: pp. 1134-1139.
  15. Samuel P. Caudill, et al., “Important Issues Affecting Eosinophilia-Myalgia Syndrome Investigations Based on Analyses of L-Tryptophan Samples,” Journal of Occupational Medicine and Toxicology (1993), Vol. 2 (1): pp. 41-52.
  16. Edwin M. Kilbourne, M.D., et al., “Tryptophan Produced by Showa Denko and Epidemic Eosinophilia-Myalgia Syndrome,” Journal of Rheumatology (1996), Vol. 23, Supplement 46, pp. 81-92.
  17. Rossanne M. Philen, M.D., Ph.D., et al., “Tryptophan Contaminants Associated with Eosinophilia-Myalgia Syndrome,” American Journal of Epidemiology (1993), Vol. 138 (3): pp. 154-159.
  18. Arthur N. Mayeno and Gerald J. Gleich, “Eosinophilia-myalgia syndrome and tryptophan production: a cautionary tale,” TIBTECH (September 1994), Vol. 12, pp. 346-352.
  19. Charles Yanofsky, Ph.D., Stanford University, email communication, June 17, 1999.
  20. Yanofsky, email communication, April 6, 2001.
  21. Hans Fisher, Ph.D.,Rutgers University, email communications, April 27&30, 2001.
  22. Dennis S. Mackin, J.D., personal interview, May 13, 2005; and email communication, March 6, 2001.
  23. Dean Wolfe Manders, “The FDA Ban of L-Tryptophan: Politics, Profits and Prozac,Social Policy (Winter 1995), Vol. 26 (2): pp. 55-57.

© Copyright 2005 William E. Crist. All Rights Reserved