United States Department of State Chemical Warfare in Southeast Asia and Afghanistan Special Report No. 98 Report to the Congress from Secretary of State Alexander M. Haig, Jr., March 22, 1982
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
United States Department of State
Chemical Warfare inSoutheast Asia andAfghanistan
SpecialReport No. 98
Report to the Congressfrom Secretary of StateAlexander M. Haig, Jr.,March 22, 1982
TO THE CONGRESS OF THE UNITED STATES:
The years from 1914 to 1918 were among the most destructive of human life inmankind�s history. Yet the sacrifice of millions brought no lasting peace. Of theelaborate structure for collective security, and the series of pacts outlawing war andcontrolling armaments which were negotiated in the aftermath of this First WorldWar, little remains today. The League of Nations, the Kellogg-Briand Pact, and theWashington Naval Agreement were all swept away in the tide of aggression whichculminated in a second global conflict. Almost the sole surviving monument, in thelaw of nations, to the twenty million dead of the First World War is the 1925 GenevaProtocol outlawing chemical and biological warfare.
Today this accord, among the oldest of arms control agreements still in force,along with another more recent such agreement banning biological and toxinweapons, is again in danger of being swept away by a new tide of aggression. Overthe past seven years chemical and toxin weapons have been used, on an ever-widening scale, in genocidal campaigns against defenseless peoples. These weaponsare being used for precisely the reason mankind has condemned and sought tooutlaw them-because of their indiscriminate action and horrific effects. Todayevidence of chemical and toxin warfare has accumulated to the point where theinternational community can no longer ignore the challenge.
The enclosed report on the use of chemical and toxin weapons by the SovietUnion and its Allies in Laos, Kampuchea, and Afghanistan has been prepared forsubmission to the Congress, to the United Nations, and to each member of theinternational community. The report is drawn from information made available tothe United States Government since 1975. It contains the most comprehensivecompilation of material on this subject available, and presents conclusions whichare fully shared by all relevant agencies of the United States Government.
The international community and the world public need not rely solely on thisreport to form their judgment, nor only upon the United States to provide theirinformation. Lethal chemical and toxin weapons are regrettably still in use in Laos,Kampuchea, and Afghanistan. New victims appear, new witnesses come forward,new scientific evidence is uncovered with increasing frequency. The great bulk ofthe information in the enclosed report could have been collected and analyzed byany interested government, international organization, or major news service. If theefforts of the United States Government to call attention to chemical warfare inAfghanistan and Southeast Asia stimulate others to discover for themselves, andjoin in efforts to expose the truth, this report will have served its most importantpurpose.
Sincerely,
Alexander M. Haig, Jr.
THE SECRETARY OF STATEWASHINGTON
Chemical Warfare in Southeast Asia and Afghanistan
CONTENTS
Page
Introduction .................................................... 4Key Judgments ................................................ 6Methodology .................................................... 6Discussion of Findings .................................... 7What Chemical Agents Are Being
Used? ......................................................... 7Laos .................................................................. 8Kampuchea ..................................................... 10The Soviet Connection in Southeast
Asia .......................................................... 13Afghanistan .................................................... 14
Motivation for Using ChemicalWeapons .................................................. 17
ANNEXES
Annex A: A Lao Pilot�s Account ... 18Annex B: Findings of U.S. Government
Investigative Teams: Useof Chemical Agents Againstthe H�Mong in Laos ................................. 19
Annex C: Medical Evidence .......................... 21Annex D: Analysis and Review of
Trichothecene Toxins ............................... 23Annex E: Overview of Natural
Occurrence and SignificantProperties of Trichothecenes ................... 27
MAPSLaos: Chemical Warfare Opera
tional Areas ................................................ 9Kampuchea: Chemical Warfare
Operational Areas .................................... 12Afghanistan: Chemical Warfare
Operational Areas .................................... 15
Page
TABLESTable 1: Laos: Summary of Re-
ported Chemical Attacks andAssociated Deaths, 1975-81.. 8
Table 2: Kampuchea: Summary ofReported Chemical Attacks andAssociated Deaths, 1978-81.. 11
Table 3: Afghanistan: Summary ofReported Chemical Attacks andAssociated Deaths, 1979-81 .. 14
Table B-1: Reports of ProbableChemical Agent Attacksin Laos 20
Table C-1: Comparison of Reported�Yellow Rain� Effects WithKnown Trichothecene Effects 22
Table D-1: Symptoms of ChemicalAttacks Reported in Laos,Kampuchea, and Afghanistan 24
Table E-1: Trichothecene-Produc-ing Fungi .............................................. 28
Table E-2: Historical TrichotheceneMycotoxicoses ..................................... 28
Table E-3: PhysicochemicalMethods for Detection of
Trichothecenesin Feedstuffs ............................................ 30
Table E-4: Spontaneous Occurrence of Trichothecene Mycotoxins ....................................................... 31
FIGUREFigure E-1: General Structure of
Trichothecenes ......................................... 29
This study presents the evidence available tothe U.S. Government on chemical warfareactivities in Laos, Kampuchea, and Afghani-stan through January 1982 and examines theSoviet involvement in those activities. It isbased on a massive amount of information,from a variety of sources, which has beencarefully compiled and analyzed over theyears. The paper is accompanied by annexesand tables that provide details of the medicalevidence and sample analyses, a technicaldescription of trichothecene toxins, andother supporting data.
INTRODUCTION
Nearly 7 years ago, reports of the use oflethal chemical weapons began to emergefrom Laos. In 1978, similar reports startedto come from Kampuchea, and in 1979from Afghanistan. Early reports wereinfrequent and fragmentary, reflecting theremoteness of the scene of conflict and theisolation of those subjected to such attacks.In the summer of 1979, however, the StateDepartment prepared a detailed compila-tion of interviews with refugees from Laoson this subject. That fall, a U.S. Armymedical team visited Thailand to conductfurther interviews. By the winter of 1979,the United States felt that it had suffi-ciently firm evidence of chemical warfareto raise the matter with the governments ofLaos, Vietnam, and the Soviet Union. Allthree governments denied that a basis forconcern over the use of chemical warfareagents existed.
Dissatisfied with these responses, andpossessing further reports that lethalchemical agents were in use in SoutheastAsia and Afghanistan, the U.S. Govern-ment in 1980 began to raise the issuepublicly in the United Nations, with theCongress, and in other forums. In Augustof that year, the State Department providedextensive documentation containingevidence of chemical weapons attacks tothe United Nations and also made thismaterial publicly available. In December,as a result of efforts by the United Statesand other concerned nations, the U.N.General Assembly voted to initiate aninternational investigation into the use ofchemical weapons. This investigation isstill underway. To date, the U.N. investigat-ing team has been denied admission to anyof the three countries where these weaponsare in use.
Despite the volume of information onchemical warfare in Southeast Asia whichhad become available by 1980,
there remained one major unresolved issue� the exact nature of the chemical agentsin use. Collection of physical samples washindered by the remoteness of the thenprincipal areas of conflict - as many as 6weeks by foot to the nearest internationalborder. Tests for known chemical warfareagents on those samples that were obtainedproved consistently negative.
In order to identify the chemical agentsin use, U.S. experts in late 1980 began togo back over all the reporting-as far backas 1975-looking for new clues. In particu-lar, they sought to match the reportedsymptomatology of victims-whichcommonly included skin irritation,dizziness, nausea, bloody vomiting anddiarrhea, and internal hemorrhaging- withpossible causes. As a result of this review,the U.S. Government in mid-1981 began totest physical samples from Southeast Asiafor the presence of toxins. These sub-stances are essentially biologicallyproduced chemical poisons. Although theyhave never before been used in war, thiswas a technical possibility, and it wasnoted that certain toxins could produce thesorts of symptoms observed in SoutheastAsian victims of chemical warfare.
In August 1981, unnatural levels andcombinations of lethal trichothecene toxinswere detected in the first sample to betested by the United States for such agents.This consisted of vegetation taken from avillage in Kampuchea where an attackoccurred in which people had died afterexhibiting the symptoms described above.In succeeding months, further samples,taken from the sites of attacks in bothKampuchea and Laos, yielded similarresults. So did samples of blood taken fromvictims of a chemical attack inKampuchea.
Despite a continued flow of reports,dating back over 7 years, of chemicalwarfare in Southeast Asia and morerecently Afghanistan, and despite the stillmounting physical evidence of the use oftrichothecene toxins as warfare agents,doubts as to the conclusive nature of theavailable evidence have persisted. Thesedoubts have arisen for several reasons. Forone, the evidence of the use of lethalchemical weapons has become availableover a period of several years and from avariety of sources. Few governments,journalists, or interested members of thepublic have been exposed to all of thisevidence, nor has it been available in anyone place. A second difficulty has been theinevitable need for the U.S. Government toprotect some of the relevant information,often gathered at personal risk to individu-als who secured it, or obtained through theuse of highly sensitive methods.
Chronology of Diplomatic/ InternationalActions on Chemical Warfare Use
October 1978
The United States called to the attention of the LaoCharge d�Affaires in Washington the press reportsalleging use of poison gas in Laos.
Assistant Secretary of State for East Asian andPacific Affairs Holbrooke traveled to Vientiane anddiscussed our concerns over H�Mong human rightsand other issues with Lao leaders.
Late 1978
The Department of State directed U.S.diplomatic missions in the Southeast Asia area toseek to develop information on the alleged use ofpoison gas against the H�Mong.
January 1979
The Department of State again informed theLao Embassy of U.S. concerns about reports ofpoison gas use in Laos, coupling this with a similardemarche in Vientiane. The Lao denied the validityof the reports.
March 1979
The U.S. Representative to the 35th session ofthe U.N. Human Rights Commission expressedU.S. concern about the plight of the H�Mong,specifically raising the poison gas use issue.
May 1979
A State Department representative went torefugee camps in Thailand to interview H�Mongclaiming to be eyewitnesses and/or victims ofpoison gas attacks in Laos.
A State Department representative visitedVientiane where he discussed the problem withvarious diplomatic missions and the senior U.N.representative in Laos. During that visit, he raisedU.S. concerns about the problem directly with theLao Foreign Ministry.
September 1979
A Department of Defense medical team wasdispatched to Thailand to interview and prepare areport on H�Mong refugees having knowledge ofgas attacks in Laos.
November 1979
Demarches were made to the Vietnamese inParis and to the Soviets in Moscow expressingU.S. concerns about reports of poison gas beingused against �resistance forces� in Laos. Both theSoviets and Vietnamese supported the Lao denialof the validity of the reports.
December 1979
State and Defense Department officialspresented evidence of gas attacks in Laos to theHouse Foreign Affairs Committee.
February 1980
A bilateral demarche was made to the Sovietsabout U.S. concerns regarding chemical warfareuse in both Laos and Kampuchea and about reportsthat chemical weapons were being used by theSoviets in Afghanistan. The demarche was made inGeneva in the context of the U.S./Soviet bilateralnegotiations on a comprehensive prohibition ofchemical weapons production, development, andstockpiling.
May 1980
An interagency team of U.S. Governmentpolitical, technical, and intelligence officers wasdispatched to Europe to brief the allies about theproblem and to stimulate support for having animpartial international investigation conducted.
July 1980
Another bilateral demarche was made to theSoviets in the context of the U.S./Soviet bilateralchemical warfare negotiations, concerning theproblem of the reported use of chemical weaponsin both Southeast Asia and Afghanistan.
The Inter-Parliamentary Union adoptedresolution calling for an impartial internationalinvestigation of reports of chemical weapons use.
August 1980
The United States circulated to U.N. memberstates a 125-page compendium of reports anddeclassified intelligence information pertaining tothe use of chemical weapons in Laos, Kampuchea,and Afghanistan.
The 40-nation Committee on Disarmamentincluded language in its Annual Report to the U.N.General Assembly on the need for an impartialinternational investigation of the problem ofchemical weapons use.
December 1980
With the full and active support of the UnitedStates, the West, and others, the U.N. GeneralAssembly adopted a resolution (A/35/144 C)establishing a U.N. investigation, under theauspices of the U.N. Secretary General and withthe assistance of qualified medical and technicalexperts, of reports of chemical weapons use. Thevote was 78 in favor to 17 opposed, with 36abstentions.
March 1981
In accordance with U.N. GeneralAssembly Resolution A/35/144 C and the re-quest of the U.N. Secretary General, the U.S.submitted detailed information pertaining tothe reports of the use of chemical weapons inSoutheast Asia and Afghanistan. The U.S.submission consisted of a letter summarizing
the U.S. submission, the U.S. compendium ofreports from August 1980, an update to thatcompendium covering the period throughJanuary-February 1981, the transcripts ofcongressional hearings held on the subject inDecember 1979 and in April 1980, and the texts ofHouse and Senate resolutions condemning the useof chemical weapons.
July 1981
The United States provided furtherdetails and written responses to questions from theU.N. Group of Experts concerning the U.S.submission of March 1981.
September 1981
Secretary Haig announced, in hisSeptember 13 speech in Berlin, that the UnitedStates had obtained physical evidence of the use oflethal mycotoxins in Southeast Asia, discovered inthe analysis of a leaf andstem sample obtained from the site of a chemicalattack in Kampuchea.On September 14, the United States submitted areport on the new evidence pertaining to the use ofmycotoxins to the U.N. Group of Expertsinvestigating reports of chemical weapons use.
Under Secretary of State for PoliticalAffairs Stoessel held a press conference inWashington on September 14 and provided adetailed press backgrounder on the newevidence.
Secretary Haig raised U.S. concerns about thenew evidence pertaining to the use of lethalmycotoxins in Southeast Asia and about the 1979Sverdlovsk anthrax incident with Soviet ForeignMinister Gromyko duringtheir bilateral consultations at the United Nationsin New York.
October 1981
Following up the Haig/Gromyko discussions,detailed bilateral demarches were made to theSoviets in Washington by Acting Arms Control andDisarmament Agency Director Grey, and afollowup in Moscow by the U.S. Deputy Chief ofMission, on the general subject of SovietBiological Warfare Convention compliance andspecific U.S. concerns regarding the 1979Sverdlovsk anthrax incident and the evidence ofthe use of trichothecenemycotoxins in Southeast Asia. The Soviets rejectedU.S. concerns once again in their formal responsein November.
An interagency team of political, technical,and intelligence officers was dispatched to Europeto brief the allies about the new evidence of the useof lethal mycotoxins in Southeast Asia.
A delegation of U.S. Government politi-cal, technical, and medical experts appeared
before the U.N. Group of Experts to respond toquestions pertaining to the U.S. submission onSeptember 14 of new evidence concerning the useof lethal mycotoxins in Southeast Asia.
November 1981
The U.N. Group of Experts investigatingreports of chemical weapons use traveled toThailand to visit refugee camps and interview andexamine survivors and eyewitnesses of chemicalattacks in Laos and Kampuchea. While there, theexperts also obtained samples from allegedchemical attacks and samples of vegetation andblood from refugees exposed to chemical attacks.
Richard Burt, Director of the Bureau ofPolitico-Military Affairs, in testimony before theCongress, announced the results of analyses ofadditional samples of chemical warfare userevealing the presence of high levels of mycotoxinsand the results of analyses of control samples fromSoutheast Asia which were found to contain nomycotoxins.
The United States submitted a report on itsanalyses of chemical warfare use samples fromboth Kampuchea and Laos, which were found tocontain high levels of mycotoxins, to the U.N.Group of Experts investigating reports of chemicalweapons use.
Demarches were made to the Vietnamese inNew York and to the Lao in Vientiane regardingthe evidence of the use of lethal mycotoxins in theconflicts in Kampuchea and Laos. Both theVietnamese and the Lao rejected the evidence anddenied the validity of U.S. concerns.
December 1981
The U.N. Secretary General submitted theReport of the U.N. Group of Experts investigatingreports of chemical weapons use (A/36/613). Thereport was inconclusive and stated that the grouphad been unable to carry out all the actions it hadintended (i.e., on-site visits to Afghanistan, Laos,and Kampuchea) due to the refusals to cooperateof the countries concerned, and that it had beenunable to complete some of the actions it hadplanned (e.g., on-site visits to Pakistan, analysis ofthe samples obtained in Thailand) in the timeavailable.
With the full and active support of the UnitedStates, the West, and others, the U.N. GeneralAssembly adopted a resolution (A/36/96 C)extending for another year the mandate of the U.N.Secretary General�s Group of Experts investigatingreports of chemical weapons use. The vote on theresolution was 86 in favor to 20 opposed, with 32abstentions.
5
This report represents an effort of theU.S. Government to correct the first defi-ciency and to ameliorate the second to theextent possible. In preparation of this report,all of the information available to the U.S.Government on chemical weapons use inLaos, Kampuchea, and Afghanistan wasassembled in one place. This information wasagain reviewed, analyzed, cross-indexed, andorganized in a coherent fashion. Based uponthis comprehensive analysis, a set ofconclusions were drawn, conclusions whichhave since been reviewed and agreed onwithout qualification by every relevantagency of the U.S. Government.
The evidence upon which this report isbased is of several kinds, including:
� Testimony of those who saw experi-enced, and suffered from chemical weaponsattacks;
� Testimony of doctors, refugee workersjournalists, and others who had the opportu-nity to question large numbers of those withfirsthand experience of chemical warfare;
� Testimony of those who engaged inchemical warfare or were in a position toobserve those who did;
� Scientific evidence, based upon theanalysis of physical samples taken from siteswhere attacks had been conducted;
� Documentary evidence from opensources; and
� Intelligence derived from �nationaltechnical means.�
These sources provide compellingevidence that tens of thousands of unsophisti-cated and defenseless peoples have for aperiod of years been subjected to a campaignof chemical attacks. Taken together, thisevidence has led the U.S. Government toconclude that Lao and Vietnamese forces,operating under Soviet supervision, have,since 1975, employed lethal chemical andtoxin weapons in Laos; that Vietnameseforces have, since 1978, used lethal chemicaland toxin agents in Kampuchea; and thatSoviet forces have used a variety of lethalchemical warfare agents, including nervegases, in Afghanistan since the Sovietinvasion of that country in 1979.
The implications of chemical warfare inAfghanistan and Southeast Asia are painful tocontemplate but dangerous to ignore. Thisactivity threatens not only the peoples ofthose isolated regions but the internationalorder upon which the security of all depends.Those who today suffer chemical warfareagainst their homelands are powerless to stopit. The prohibitions of international law andsolemn agreement are not self-enforcing.
Only an alert and outspoken world commu-nity, intent to maintain those standards ofinternational behavior it has so painfullyachieved and so tenuously established, canbring sufficient pressure to bear to halt theseviolations of law and treaty. It is hoped thatpublication of this report will be one step inthis process, the end result of which will bethe cessation of chemical warfare and thestrengthening of the rule of law �in the affairsof nations.
KEY JUDGMENTS
Laos. The U.S. Government hasconcluded from all the evidence that selectedLao and Vietnamese forces, under directSoviet supervision, have employed lethaltrichothecene toxins and other combinationsof chemical agents against H�Mong resistinggovernment control and their villages since atleast �1976. Trichothecene toxins have beenpositively identified, but medical symptomsindicate that irritants, incapacitants , andnerve agents also have been employed.Thousands have been killed or severelyinjured. Thousands also have been drivenfrom their homeland by the use of theseagents.
Kampuchea. Vietnamese forces haveused lethal trichothecene toxins on Demo-cratic Kampuchean (DK) troops and Khmervillages since at least 1978. Medical evidenceindicates that irritants, incapacitants, andnerve agents also have been used.
Afghanistan. Soviet forces in Afghani-stan have used a variety of lethal andnonlethal chemical agents on mujahidinresistance forces and Afghan villages sincethe Soviet invasion in December 1979. Inaddition, there is some evidence that AfghanGovernment forces may have usedSoviet-supplied chemical weapons against themujahidin even before the Soviet invasion.Although it has not been possible to verifythrough sample analysis the specific agentsused by the Soviets, a number of Afghanmilitary defectors have named the agentsbrought into the country by the Soviets andhave described where and when they wereemployed. This information has beencorrelated with other evidence, including thereported symptoms, leading to the conclusionthat nerve agents, phosgene oxime, andvarious incapacitants and irritants have beenused. Other agents and toxic smokes also arein the country. Some reported symptoms areconsistent with those produced by lethal or
sublethal doses of trichothecene toxins, butthis evidence is not conclusive.
The Soviet Connection. The conclusionis inescapable that the toxins and otherchemical warfare agents were developed inthe Soviet Union, provided to the Lao andVietnamese either directly or through thetransfer of know-how, and weaponized withSoviet assistance in Laos, Vietnam, andKampuchea. Soviet military forces are knownto store agents in bulk and move them to thefield for munitions fill as needed. Thispractice also is followed in South, east Asiaand Afghanistan, as evidenced by manyreports which specify that Soviet technicianssupervise the shipment, storage, filling, andloading onto aircraft of the chemicalmunitions. The dissemination techniquesreported and observed evidently have beendrawn from years of Soviet chemical warfaretesting and experimentation. There is noevidence to support any alternative explana-tion, such as the hypothesis that the Vietnam-ese produce and employ toxin weaponscompletely on their own.
METHODOLOGY
The judgments of this study were arrived atthrough a rigorous analytical process.
� Every relevant piece of information onreported chemical warfare incidents wasreviewed, recorded, and tabulated. Numbersof attacks and deaths were screened forpossible duplication. Extensive data on theSoviet chemical and biological warfareprogram also were reviewed.
� All the test data on physical evidenceavailable to the U.S. Government - includingenvironmental samples and backgroundcontrols-were re, viewed.
� A scientific report on toxins, whichconcluded that trichothecenes probably wereamong the agents used in Southeast Asia, wasprepared.
� The medical evidence was analyzed,drawing on all available information fromSoutheast Asia and Afghanistan and incorpo-rating the findings of a Department ofDefense medical team, which concluded thatat least three types of agents were used inLaos.
� Extensive consultations were held withgovernment and nongovernment scientistsand medical authorities, many of whom wereasked to review the evidence. Experts fromother countries also were consulted.
6
After the data were organized to permitcomparative analysis, the study focused onthree separate questions.
� Have lethal and other casualtyproducing agents been used in Southeast Asiaand Afghanistan?
� What are these agents, and how and bywhom are they employed?
� Where do these agents originate, andhow do they find their way to the field?
Although the evidence differs for eachcountry, the analytical approach was thesame. Testimony of eyewitnesses - date,place, and type of attack-was matched againstinformation from defectors, journalists,international organizations, and sensitiveinformation that often pinpointed the timeand place of chemical attacks. In addition,information on military operations in theareas where chemical attacks had beenreported was examined to establish whetherair or artillery strikes took place or whetherthere was fighting in the areas wherechemical agents reportedly were used. In allthree countries, instances were identified inwhich eyewitness accounts could be corre-lated directly with information from othersources on military operations in progress.
There is no evidence of any systematicpropaganda campaign by either the H�Mongin Laos or the Afghan resistance forces topromote the allegation that chemical agentshave been used on their people. On the otherhand, there were early indications that PolPot�s Democratic Kampuchean resistance didengage in an organized propaganda campaignon chemical agent use. These indicationsmade U.S. Government analysts cautiousabout accepting DK allegations, whichincreased markedly after the chemical attacksin Laos were publicized. For Kampuchea,therefore, special efforts were taken toconfirm such allegations by analyzing sourcesof information that in no way could beconsidered part of a propaganda or deceptioncampaign.
DISCUSSION OF FINDINGS
In September 1981, the U.S. Governmentdeclared publicly that toxins- poisonouschemical substances extracted from biologicalmaterial-probably were the mysterious lethalagents used for many years in Laos andKampuchea. The statement was prompted bythe discovery of high levels of trichothecenetoxins in a vegetation sample collectedshortly after a March 1981 Vietnamesechemical attack in Kampuchea. This con-
clusion, however, rested on a much broaderbase of evidence than analysis of one sample.
By April 1980, the U.S. Government hadalready concluded that lethal agents almostcertainly had been used against H�Mongtribespeople in Laos. There was less certaintythen about the use of lethal agents inKampuchea, mainly because of the alreadymentioned suspicions about the propagandacampaign of Pol Pot�s DemocraticKampuchean forces, although their claimssubsequently were shown to be valid. It wasalso concluded that chances were about eventhat lethal agents had been used in Afghani-stan. There was little doubt by April 1980 thatriot-control agents and some form ofincapacitants had been used in all threecountries. Since that April 1980 assessment,additional evidence has allowed a muchfirmer conclusion. There is now no doubt thatcasualties and deaths have resulted fromchemical attacks in all three countries.
What Chemical Agents Are Being Used?
As soon as it was determined that chemicalagents had been used, an effort was made toidentify the specific agents. To do this it wasnecessary to collect and analyze at least oneof the following environmental samplescontaminated with agents, the munitions usedto deliver agents, or biological specimensfrom victims of an attack. A study bymedical-toxicological experts of symptomsexhibited by individuals exposed to toxicagents provides a good indication of thegeneral class of chemical agent used. Thus,the range of clinical manifestations fromchemical agents, as reported by a U.S. Armyinvestigative team in Thailand, resulted in thedetermination that nerve agents, irritants suchas CS, and highly toxic hemorrhagicchemicals or mixture of chemicals were usedin Laos.
Other medical-toxicological person-nel who reviewed the evidence andconducted their own investigation reachedthe same conclusion. They further in-dicated that toxins such as the trichothe-cenes were a probable cause of the lethalhemorrhaging effect seen in Kampucheaand Laos. In many cases, symptomsreported by the Democratic Kampu-chean forces in Kampuchea and the mu-jahidin in Afghanistan were similar tothose reported by the H�Mong in Laos.Moreover, symptoms reported fromAfghanistan and Kampuchea indicatedthat a highly potent, rapid-acting, in-capacitant �knockout� chemical also wasbeing used. Mujahidin victims and wit-
nesses to chemical attacks reported otherunusual. symptoms, including a blackening ofthe skin, severe skin irritation along withmultiple small blisters and severe itching,severe eye irritation, and difficulty inbreathing-all of which suggests thatphosgene oxime or a similar substance wasused.
Collecting samples possibly contami-nated with a toxic agent during or after achemical assault is difficult under anycircumstances but particularly when theassault is against ill-prepared people withoutmasks or other protective equipment.Obtaining contaminated samples that willyield positive traces of specific chemicalagents depends on many factors. Theseinclude the persistency of the chemical, theambient temperature, rainfall, wind condi-tions, the medium on which the chemical wasdeposited, and the time, care, and packagingof the sample from collection to laboratoryanalysis.
Many traditional or known chemicalwarfare agents are nonpersistent anddisappear from the environment within a fewminutes to several hours after being dis-persed. Such agents include the nerve agentssarin and tabun; the blood agents hydrogencyanide and cyanogen chloride; the chokingagents phosgene and diphosgene; and theirritant phosgene oxime. Other standardchemical warfare agents-such as the nerveagents VX and thickened soman and theblistering agents sulfur mustard nitrogenmustard, and lewisite - may persist for severaldays to weeks depending on weatherconditions.
The trichothecene toxins have goodpersistency but may be diluted by adverseweather conditions to below detectableconcentrations. To maximize the chances ofdetection, sample collections need to be madeas rapidly as possible after a chemical assault;as with many agents, this means minutes tohours. Under the circumstances of SoutheastAsia and Afghanistan, such rapid collectionhas simply not been possible. Although manysamples were collected, few held any realisticprospect of yielding positive results. It isfortunate that trichothecenes are sufficientlypersistent and in some cases were not dilutedby adverse weather conditions. Thus we wereable to detect them several months after theattack.
Samples have been collected fromSoutheast Asia since mid-1979 and fromAfghanistan since May 1980. To date,about 50 individual samples-of greatlyvarying types and usefulness for analyti-cal purposes-have been collected andanalyzed for the presence of known
7
TABLE 1
Laos: Summary of Reported Chemical Attacks and Associated Deaths, 1975-81
Time Period Area Attacks a Deaths b
Summer 1975 Vientiane 2 25Fall 1976 Phou Bia 8 10
Savannakhet 1 10Winter 1976-77 Phou Bia 2 16Spring 1977 Phou Bia 6 66
Khammouan 2 1Summer 1977 Phou Bia 6 95Fall 1977 Phou Bia 1 25Winter 1977-78 Phou Bia 10 1,328
Savannakhet 6 224Spring 1978 Phou Bia 34 969Summer 1978 Phou Bia 22 664Fall 1978 Phou Bia 19 572Winter 1978-79 Phou Bia 5 15Spring 1979 Phou Bia 36 257Summer 1976 Phou Bia 5 239Fall 1979 Phou Bia 10 56
Xaignabouri 2 24Winter 1979-80 Phou Bia 4 10Spring 1980 Phou Bia 3 24Summer 1980 Phou Bia 6 187Fall 1980 Xaignabouri 1 12
Phou Bia 7 88Savannakhet 3 1
Winter 1980-81 Xaignabouri 2 57Phou Bia 4 82Vientiane 1 1
Spring 1981 Houaphan 2 ?Phou Bia 7 218Vientiane 1 -
Summer 1981 Phou Bia 1Fall 1981 Phou Bia 4 500
Khammouan 3 534226 6,310
a This tabulation omits 35 attack sites, accounting for 194 deaths, which could not begeographically located in the reports. The totals overall were 261 attacks and more than 6,504deaths. b A plus sign indicates that the report(s) of deaths gave a minimum figure. In some cases(shown with a question mark) deaths were reported, but no number was given. Other reports(signified with a dash) gave no information on fatalities.
chemical warfare agents, none of which hasbeen detected. Based on recommendations bymedical and toxicological experts andfindings of investigators from the U.S.Army�s Chemical Systems Laboratory,several of the samples have been analyzed forthe trichothecene group of mycotoxins. Foursamples, two from Kampuchea and two fromLaos, were found to contain high levels oftrichothecene toxins. In addition, preliminaryresults of the analysis of blood samplesdrawn from victims of an attack indicate thepresence of a trichothecene metabolite of T-2,namely HT-2.
A review of all reports indicates the useof many different chemical agents, means ofdelivery, and types of chemical attacks. Theuse of trichothecene toxins has beenidentified through symptoms and sampleanalysis. In some cases, however, thesymptoms suggest other agents, such as nervegas, which have not been identified throughsample analysis. Significant differences aswell as similarities have surfaced in thereports from the three countries. Theevidence from each country, therefore, isdescribed separately, with attention drawn tosimilarities where appropriate.
Laos
Reports of chemical attacks against H�Mongvillages and guerrilla strongholds in Laosdate from the summer of 1975 to the present(see Table 1). Most of the reports wereprovided by H�Mong refugees who wereinterviewed in Thailand and the UnitedStates. More than 200 interviews were carriedout variously by U.S. Embassy officials inThailand, a Department of Defense team ofmedical-toxicological experts (see Annex B),U.S. physicians, Thai officials, journalists,and representatives of international aid andrelief organizations. According to theinterviews, Soviet AN-2 and captured U.S.L-19 and T-28/41 aircraft usually wereemployed to , disseminate toxic chemicalagents by sprays, rockets, and bombs. Insome cases, Soviet helicopters and jet aircraftwere said to have been used.
The reports describe 261 separateattacks in which at least 6,504 deathswere cited as having resulted directly fromexposure to chemical agents. The actualnumber of deaths is almost certainly muchhigher, since the above figure does not takeaccount of deaths in attacks for which nospecific casualty figures were reported. Thegreatest concentration of reported chemicalagent use occurred in the area where the three
provinces of Vientiane, Xiangkhoang, andLouangphrabang adjoin (see map). Thistriborder region accounted for 77% of thereported attacks and 83% of thechemical-associated deaths. Most of thereported attacks took place in 1978 and1979. Since 1979, the incidence of chemicalattacks appears to have been lower, butreported death rates among unprotected anduntreated victims were higher. Only sevenchemical attacks were reported in the fall of1981, for example, yet 1,034 deaths wereassociated with those incidents.
The medical symptoms reportedlyproduced by the chemical agents are� varied.According to knowledgeable physicians, thesymptoms clearly point to at least three typesof chemical agents - incapacitant/riot-controlagents, a nerve agent, and an agent causingmassive hemorrhaging. The last-named waspositively identified as trichothecene toxins.This was announced publicly by SecretaryHaig in September 1981.
In a number of the refugee reports,eyewitnesses described attacks as consistingof �red gas� or a �yellow cloud.�
8
9
Red gas was considered the more lethal. Aformer Lao Army captain stated that the �redgas� caused the H�Mong to die within 12hours. An employee of an internationalorganization interviewed victims of aSeptember 15, 1979 attack in which nonlethalrounds preceded an attack by five or six �redgas� bombs that covered a 500-meter area.Persons within 30-100 meters of the circledied in 10 minutes after severe convulsions.Others had headaches, chest pains, andvomiting but did not die.
Every qualified interrogator whosystematically interviewed the H�Mongrefugees concluded that they had beensubjected to chemical attacks. A U.S.Government medical team returned fromThailand in 1979 convinced that severalunidentified chemical warfare agents hadproduced the symptoms described by therefugees. This evidence was expanded bytestimony from a variety of sources, includingthat of a Lao pilot who flew chemical warfaremissions before defecting in 1979. Hisdetailed description of the Lao, Vietnamese,and Soviet program to use chemical agents todefeat the H�Mong resistance helped dispelany lingering suspicions that the refugees hadfabricated or embellished the stories. The Laopilot described the chemical rocket he hadfired as having a more loosely fitting warheadthan a conventional rocket. (His accountappears in Annex A.)
In 1977, a H�Mong resistance leaderfound a U.S. 2.75-inch rocket* with amodified Soviet warhead that fits the Laopilot�s description. Other sources reportedthat U.S. 2.75-inch rockets were fitted withSoviet-supplied lethal chemical warheads bySoviet and Vietnamese technicians atfacilities in Vientiane as well as inXiangkhoang and Savannakhet Provinces.Munitions storage facilities suitable forstoring chemical agents and weapons havebeen identified in each of these provinces.The aircraft types-AN-2s, L-19s, and T-28/41s-most often reported by the H�Mongrefugees as being used to deliver chemicalagents have been identified as based onairfields in northern Laos throughout thisperiod. A special Lao Air Force unit isresponsible for chemical rockets. The unit iscommanded by a Soviet-trained Lao and hasa Soviet rocket expert attached as an adviser.
* During withdrawal of U.S. forces fromVietnam, thousands of these fell intoVietnamese hands.
Obtaining additional data for Laos hasbeen difficult because of the nature of thefighting there. There have been few majoroperations. The reports reflect numerousminor engagements between the opposingforces. In nearly all cases, the chemical usereported has been directed against villages, inthe absence of obvious combat operations.This lends support to the Lao pilot�s claimthat the Vietnamese and Lao militarycommands were engaged in a �H�Mongextermination� campaign.
Of particular interest are the circum-stances surrounding the collection of twophysical samples found to contain lethaltoxins. The first was collected after a March13, 1981 attack on a village between thevillages of Muong Chai and Phakhao in thePhou Bia region. In this case, a largetwo-engine plane reportedly sprayed a mist ofa moist, yellow, sticky substance; twovillagers and all village animals died. Thesecond sample is from Ban Thonghak,another village in the Phou Bia region,collected following an April 2, 1981 attack inwhich a jet aircraft ,reportedly sprayed ayellow substance; 24 of the 450 villagersdied. In the spring of 1981, seven separatechemical attacks, resulting in 218 deaths,were reported to have occurred in this region.
It is significant that these attacks tookplace following a period of escalation inoverall resistance activities in the Phou Biaarea in the winter of 1980-81. During thatperiod, joint suppression operations by LaoPeople�s Liberation Army and VietnameseArmy forces had achieved only limitedsuccess, perhaps spurring both forces on togreater effort. The more intense use ofchemical weapons may have been part of thiseffort.
Evidently the fact that chemical agentswere being used in Laos was not widelyknown among units of the Lao Army. In June1981, a group of refugees from a village inVientiane Province reached Thailand anddescribed attacks against them carried out amonth earlier by helicopters �droppingpoison� into their water supply. Lao fieldunits subsequently entered the village andwere surprised at the sight of many villagersstill suffering from symptoms of acutepoisoning. According to a villager, when theLao military personnel saw the �small yellowgrains� spread around the village, they wereconvinced that toxic chemicals had been usedon the village and requested medicalassistance for those villagers still sufferingfrom nausea and bloody diarrhea.
In a December 15, 1981 press conferencein Beijing, former Lao Health MinistryBureau Director KhamsengkeoSengsathit-who had defected toChina-confirmed that chemical weapons werebeing used �in the air and on the ground� inLaos, killing �thousands.� He asserted thatthe Vietnamese alone were using suchweapons, keeping the matter secret from theLao. He also stated that 3,000 Soviet adviserswere in Laos and �have taken control� of theLao Air Force, while 40,000-50,000Vietnamese troops had �reduced Laos to thestatus of a colony.�
Kampuchea
Since October 1978, radio broadcasts, pressreleases, and official protests to the UnitedNations by the Democratic Kampuchealeadership have accused the Vietnamese andthe Hanoi-backed People�s Republic ofKampuchea regime of using Soviet-madelethal chemical agents and weapons againstDK guerrilla forces and civilians. DKallegations for a time were the only source ofinformation concerning chemical warfareattacks in Kampuchea. In November 1979,however, the guerrilla forces of the KhmerPeople�s National Liberation Front reportedthat the Vietnamese had attacked them with atear gas which, from their description,resembled the riot-control agent CS.Subsequently, Thai officials, DemocraticKampuchea informants and refugees,Vietnamese Army defectors, U.S. and Thaimedical personnel, officials of internationalaid and relief organizations, and Canadianand West European officials also haveimplicated the Vietnamese in the offensiveuse of lethal and incapacitating chemicalagents in Kampuchea.
There are reports of 124 separate attacksin Kampuchea from 1978 to the fall of 1981in which lethal chemicals caused the deathsof 991 persons (see Table 2). The mortalityfigure represents a minimum because somereports state only that there were deaths anddo not provide a number. The earliest reportscite attacks in Ratanakiri Province, in thenortheastern corner of the country (see map).Reports from 1979 to the present show theuse of lethal chemicals primarily in theprovinces bordering Thailand. The greatestuse of chemical agents apparently has beenin Battambang Province, with 51 reportedincidents; Pursat Province has experiencedthe next highest frequency, with 25
10
TABLE 2
Kampuchea: Summary of Reported Chemical Attacksand Associated Deaths, 1978-81
Time Period Area Attacks Deaths1978 Ratanakiri 5 ?Summer 1979 Kompong Speu 4 37Fall 1979 Siem Reap 1 -
Battambang 4 22Pursat 2 1Koh Kong 2 6Kampot 1 3Kompong Chhnang 2 118
Winter 1979-80 Battambang 12 64Pursat 5 21Koh Kong 2 4
Spring 1980 Battambang 3 20Pursat 8 24Koh Kong 5 13
Summer 1980 Siem Reap 1 82Battambang 3 23Pursat 2 7Koh Kong 3 -
Winter 1980-81 Battambang 8 -Pursat 2 3
Spring 1981 Preah Vihear 1 -Battambang 12 163Pursat 3 42Koh Kong 1 -Kampot 1 -
Summer 1981 Battambang 3 7Kompong Thom/Cham 1 -
Fall 1981 Siem Reap 16 305Battambang 6 16Pursat 3 -Koh Kong 1 -Kampot 1 -
124 981
a A plus sign indicates that the report(s) of deaths gave a minimum figure. In somecases (shown with a question mark) deaths were reported, but no number was given. Otherreports (signified with a dash) gave no information on fatalities.
reported incidents. These numbers areconsistent with the overall high level ofmilitary activity reported in the borderprovinces.
A review of information from allsources provides direct and specific supportfor 28 of 124 reported attacks. There is, inaddition, some evidence that in all reportedinstances some form of attack took place.This evidence includes reports of troopmovements, supply transfers, operationalplans, postoperation reporting, and airactivity. It indicates that military activitytook place at the time and place of everyincident reported to involve lethal chemicalagents. In some cases, it provides strongcircumstantial evidence that the action
involved chemical substances-for example,the movement of chemicals and personalprotection equipment into the area.
There is no doubt that in late 1978 and1979 the Vietnamese, and what later becamethe People�s Republic of Kampuchea forces,made at least limited use of riot-controlchemicals and possible incapacitating agentsagainst both Communist and non-Communistguerrilla forces in Kampuchea. The chemicalsused probably included toxic smokes,riot-control agents such as CS, and anunidentified incapacitating agent that causedvertigo and nausea and ultimately renderedvictims unconscious with no other signs orsymptoms.
In March 1979, during Vietnameseoperations against Khmer Rouge forcesin the Phnom Melai area, a Vietnamese
Army private, who later defected, observedthe following activities related to chemicalwarfare. During the fighting, all regiment(740th) troops were issued gas masks.However, the 2nd Battalion, a �borderdefense unit,� was not issued masks. Thisunit was in the Phnom Melai area and wasvirtually surrounded by Khmer Rouge forces.At another point in the battle, the regiment�stroops were ordered to don masks. TheVietnamese Army private reported that hesaw two Soviets (Caucasians) fire a DH-10 (ahand-held weapon identified by the private�scomrades). He was about 50 meters from thefiring point. The weapon at impact, which hewas able to observe from his position, gaveoff clouds of white, gray, and green gas/smoke. His signal unit subsequently passed amessage reporting that there were 300 dead,including the unprotected Khmer Rouge andVietnamese of the border defense forces� 2ndBattalion. The corpses reportedly had tracesof white and green powder on their faces andclothes. Their faces were contorted, with eyeswide open. No blood was seen. (A H�Mongresistance leader described an incident in1981 in which two Soviet soldiers fired ahand-held weapon that dispersed a similarlethal agent.)
Starting in February 1980, reportsrevealed that the Vietnamese were using 60mm mortars, 120 mm shells, 107 mm rockets,M-79 grenade launchers filled with chemicalagents, as well as munitions delivered byT-28 aircraft. According to the DK, thechemicals used were green and yellow andpowderlike in appearance. In some instancesthe gas was described as yellow or white. Thesymptoms described were tightening of thechest, disorientation, vomiting, bleeding fromthe nose and gums, discoloration of the body,and �stiffening� of the teeth. In July 1980, theDK described artillery attacks that produced ablack smoke causing itchy skin, weakness,skin lesions, and in some cases decaying skinand blisters. In December 1980, the Vietnam-ese were once again firing chemical artilleryshells, and it was believed that poisonchemicals were being brought into Thailand�sborder region. By March 1981, the Demo-cratic Kampuchea forces had reportednumerous attacks directed against them withlethal chemical agents and the poisoning offood and water.
U.S. analysis of contaminated vege-tation samples collected within hours ofa March 1981 attack showed high levelsof three trichothecene toxins in a com-bination that would not be expected tobe found in a natural outbreak in this
11
12
environment. At the levels found on thevegetation, the three trichothecenes wouldproduce vomiting, skin irritations and itching,and bleeding symptoms. Water samples takenfrom the area of the same attack alsocontained trichothecene toxins. Controlsamples from nearby areas confirmed thatthese toxins were not indigenous to thelocale. (Details on the sample analysis appearin Annex D.)
There also is ample evidence of militaryactivity at the place and time of the acquisi-tion of the samples. Vietnamese Armydefectors described plans for multiregimentalsweep operations to be conducted along theborder in northwestern Battambang Provincebefore the end of the dry season in May.Actual fighting, however, continued to becharacterized by guerrilla tactics on bothsides, including, according to a VietnameseArmy defector, �staging ambushes, layingminefields, and use of deception.� Indeed,Democratic Kampuchean resistance forceswere ordered to avoid large-scale operationsand to limit combat operations to scatteredsapper attacks. Such information is consistentwith other reports of Vietnamese Army forcesspreading toxic chemicals in streams, alongroadsides, and around villages and firingtoxic gas shells against enemy positions. ThePhnom Melai sector, where Phnom MakHoeun is located, was described as an �anthillof DK activity,� and actions reported duringMarch were �sporadic firefights� aroundPhnom Mak Hoeun involving the VietnameseArmy�s 2nd Battalion, 2nd Border SecurityRegiment.
In Kampuchea, as in Laos, the period oflate 1980 through spring 1981 was one ofintensified Vietnamese operations to suppressthe resistance and break the will of theopposing forces. In July 1981, trucks loadedwith blue sacks filled with white powder werebeing moved by the Vietnamese into thePailin, Battambang, and Siem Reap areas.Vietnamese soldiers told villagers that thechemicals caused blindness, hemorrhaging,and vomiting.
Additional evidence was derived fromblood samples drawn from victims ofVietnamese chemical use that occurred onSeptember 19, 1981 in the Takong area.Takong is in the same general area asPhnom Mak Hoeun-that is, the centralregion of Battambang Province near theThai border. Although there is no indepen-dent confirmation of the accounts of theattack, American medical personnelvisiting a DK field hospital examined thevictims and obtained the blood samples.Analyses of these samples suggested the
use of trichothecenes. (Blood analysis resultsalso appear in Annex D.)
According to the DK soldiers affected,the chemicals used in the September 19Takong attack were dispersed as a gas orpowder and as a poison to water. The gas orpowder was released from containers bytripwires in the area of the rear forces. Thisdescription is consistent with the otherreporting for this area and time.
Thailand also has been concerned aboutchemical attacks against its own forces andcivilian population. In March 1981, one Thaidied from poisons placed by Vietnamesetroops, and others became ill after sufferingbleeding from the nose and mouth. In May1981, Thai forces captured two Vietnamese asthey were attempting to poison the watersupply in a Kampuchean relocation camp inThailand. The poison was analyzed by theThai and found to contain lethal quantities ofcyanide. Many reports indicate that it iscommon practice for Vietnamese units topoison water and food used by-the DKforces.
The Soviet Connectionin Southeast Asia
Much of the Soviet interest in Southeast Asiais dictated by their rivalry with China andtheir close alliance with the Vietnamese.Regional Communist forces have beenstrengthened to contain Chinese influenceand deter military incursions. The area ofnorthern Laos between Vientiane and theChinese border-where the H�Mong hill tribeshave stubbornly resisted and harassedVietnamese forces-is strategically significantto the Vietnamese because it adjoins a hostileChina. In the last few years the Vietnamesehave expanded their military construction andstrengthened their forces in Laos which nownumber 50,000.
Initially there was a tendency to interpretthe Soviet role as strictly advisory. Now,however, there is considerable evidence tosuggest that the Soviets are far more involvedin the Lao and Vietnamese chemical warfareprogram than was assumed earlier. Anestimated 500 Soviet military advisersprovide maintenance assistance and technicalsupport, actually running the Lao Air Force,and give advanced training to Lao personnelin conventional as well as chemical warfare.
The Soviets have had advisers andtechnicians working in Vietnam andLaos for many years and in Kampucheasince 1979. However, it was not until
early 1979 that evidence surfaced on theSoviets� direct involvement in chemicalwarfare activities. For example, the Lao Armychemical section in Xiangkhoang preparedSoviet-manufactured chemical items forinspection by a Soviet military team onFebruary 7, 1979. A seven-man team ofSoviet chemical artillery experts, accompa-nied by Lao chemical officers, inspectedchemical supplies and artillery rounds at theXeno storage facility in Savannakhet on June1, 1979. One report stated that the Sovietswould be inspecting the same chemicalexplosives used to suppress the H�Mong inthe Phou Bia area.
In addition to this information, H�Mongaccounts have described Soviet advisers andtechnicians participating in the preparation ofthe chemical weapons for the attacks on theH�Mong villages. H�Mong eyewitnessesclaim to have seen �Caucasian pilots� inaircraft, and one H�Mong report states that adowned Soviet aircraft was discovered in thejungle along with a dead Soviet pilot. InNovember 1981, a H�Mong resistance leaderdescribed how Soviet soldiers fighting withthe Lao Army fired handheld weapons thatdispersed a lethal agent over a 300-meterarea. Several Lao defectors have reportedseeing Soviet advisers present when aircraftwere loaded with chemical-agent rockets.
In July 1981, a Soviet shipment ofwooden crates filled with canisters describedby the Vietnamese as �deadly toxic chemi-cals� was unloaded at the port of Ho ChiMinh City. This incident further corroboratesthe judgment that the Soviets have beenshipping chemical warfare materiel toVietnam for some time. During the unload-ing, Vietnamese soldiers were caughtpilfering the wooden crates containing thecanisters. The soldiers dropped one of thewooden cases and intentionally broke it open;they wanted to determine if its contents wereedible or valuable for pilferage. When asoldier broke the nylon seal and attempted topry open a canister, special security personnelisolated the area and told the soldiers that thecanisters contained deadly toxic substancesfrom the U.S.S.R. The wooden crates, eachweighing 100 kilograms, were loaded onmilitary trucks and taken under special guardto the Long Binh storage depot.
This incident is only one in a seriesinvolving Soviet chemical warfare materieldating back several years. In 1975, forexample, a Soviet captain of a diving supportcraft engaged in salvaging a sunken ship inthe Black Sea, which had been transportingSoviet military supplies to Vietnam, said that
13
TABLE 3
Afghanistan: Summary of Reported Chemical Attacks and Associated Deaths, 1979-81
Time Period Province Attacks a Deaths b
Summer 1979 Badakhshan 1 2,000 c
Parvan 1 8Bamian 1 -
Fall 1979 Konarha 1 350Farah 1 ?Herat 1 ?Badghisat 1 ?
Winter 1979-80 Badakhshan 5 130+Takhar 1 -Konarha 2 10+Nangarhar 1 ?Bamian 1 ?
Spring 1980 Badakhshan 1 1 +Konarha 2 ?Oruzgan 1 -Qandahar 1
Summer 1980 Nangarhar 2Vardak 1 3Herat 2 300+Kabul 2 -
Fall 1980 Konarha 1 ?Lowgar 1 4Ghazni 1 100
Winter 1980-81 Lowgar 2 ?Spring 1981 Parvan 2 -
Lowgar 3 -Ghazni 2 ?Qandahar 1 -
Summer 1981 Nangarhar 2 ?Qandahar 2 16Herat 1 119
47 3,042
a This tabulation omits some attacks described in the text because they could not be dated or locatedwith high confidence.
b A plus sign indicates that the report(s) of deaths gave a minimum figure. In some cases (shown witha question mark) deaths were reported, but no number was given. Other reports (signified with a dash)gave no information on fatalities.
c The quality of reporting for this period is not as good as the information that became available afterthe Soviet invasion. We are concerned that this unusually high figure may reflect an accumulation ofdeaths from several incidents and not the single attack indicated. For example, reports were receiveddescribing over 1,000 deaths in Bamian Province in June-July 1979. An Afghan military officer reportedseeing the bodies of many mujahidin in Panjsher Valley in August 1979 after a chemical attack and statedthat many had been killed. An Afghan civil engineer reported hearing that many deaths resulted from achemical attack in the Jalalabad area, also in the summer of 1979. Because we could not obtain support-ing evidence, these reports were not included. Although sufficient evidence exists to conclude that AfghanGovernment forces used chemical weapons, Mainly bombs, from June to December 1979, no survivors oreyewitness accounts of these attacks are available to determine the type of agent and symptoms.
his divers came in contact with toxicchemicals, and a special Soviet salvage unittook over the operation after the diversbecame very ill. The salvage operations,conducted by the ASPTR-12 Salvage,Rescue, and Underwater Technical ServicesGroup based in Odessa, were monitored byhigh-ranking Soviet naval officers.
The operation began with the removalof tractors and helicopters which cluttered thedeck of the ship and prevented access to holdhatches. Once the surface clutter wasremoved, the divers attempted to enter thehold-. At this point, however, operations hadto be suspended temporarily because of aviolent outbreak of chemical poisoningamong the divers. Contact with the unidenti-fied chemicals resulted in reddish welts 1-3centimeters in diameter on exposed skin andwas accompanied by severe headaches,nausea, and a general feeling of fatigue. Thesymptoms disappeared on their own after 3-5days of rest. At this point, military authoritiestook over from the ASPTR-12 divers, whowere temporarily withdrawn from the project.Soviet naval divers were sent down anddetermined that the source of poisoning waschemical seepage from an open hatch of oneof the holds. The hatch was promptly sealed,and the salvage operation was once moreassigned to ASPTR-12 divers who resumedwork and retrieved ammunition and anassortment of other equipment. Once this wasdone, the military took over permanently. Theship was raised without removing thepoisonous chemicals and towed to an Odessashipyard where the chemicals were unloadedby military personnel. The ship was thenbroken up and scrapped. The entire operationtook about 3 years to complete.
As another example of Soviet involve-ment, two Vietnamese corporals, from the337th and 347th Vietnamese Army divisions,have stated that Soviet-supplied chemicalweapons were stored in eaves near Lang Sonin February 1979. Although their Vietnameseunits were issued gas masks, they were toldthat Soviet-supplied chemical weapons wouldnot be used unless the Chinese initiatedchemical warfare. As late as February 1981, ateam of uniformed Soviet military adviserswas attached to the corps headquarters. Theteam leader was a senior Soviet colonel. TheSoviets were involved in training corpspersonnel in the use of Soviet-suppliedweapons and equipment, including chemicalartillery shells and gas masks. The Sovietteam often inspected defensive positions andobserved training maneuvers.
AfghanistanAttacks with chemical weapons against themujahidin guerillas in Afghanistan werereported as early as 6 months before theSoviet invasion on December 27, 1979. Theinformation specifies that Soviet-madeaircraft were used to drop chemical bombs,with no clear identification of Soviet or Af-ghan pilots or of the specific agents used. On
November 16, 1979, chemical bombsreportedly were dropped along with conven-tional air munitions on targets in Farah Herat,and Badghisat Provinces by Soviet-suppliedIL-28 bombers based at Shindand. A numberof Afghan military defectors have stated thatthe Soviets provided the Afghan military withchemical warfare training
14
as well as supplies of lethal and incapacitat-ing agents.
For the period from the summer of 1979to the summer of 1981, the U.S. Governmentreceived reports of 47 separate chemicalattacks with a claimed death toll of more than8,000 (see Table 3). Of the 47 reports, 36came from Afghan Army deserters, mujahidinresistance fighters, journalists, U.S. physi-cians, and others. For 24 of the reportedattacks, there is additional independentevidence supporting allegations of chemicalattacks. In seven instances, further individualreporting exists. Evidence for 20 of thereported incidents comes from information onSoviet or Afghan Army combat operations inprogress in areas and at times approximatingthose of a reported chemical attack (see map).
The reports indicated that fixed-wing aircraft and helicopters usuallywere employed to disseminate chemicalwarfare agents by rockets, bombs, andsprays. Chemical-filled landmines were
also reportedly used by the Soviets. Thechemical clouds were usually gray or blue-black, yellow or a combination of the colors.
Symptoms reported by victims andwitnesses of attacks indicate that non-lethalincapacitating chemicals and lethal chemi-cals�including nerve agents, phosgene orphosgene oxime, possibly trichotriecenetoxins, and mustard�were used. Medicalexaminations of some of-the victims includereports of paralysis, other neurologicaleffects, blisters bleeding, and sometimesdeath. While none of the agents being used inAfghanistan has been positively identifiedthrough sample analysis, there is no doubtthat the agents being used are far more toxicthan riot-control agents such as CN and CSor even adamsite.
Several descriptions of the physiologicalaction of a chemical agent or of the conditionof the corpses of victims were particularlyunusual. In one, victims were rapidly renderedunconscious for 2-6 hours and had few
aftereffects. In another, the bodies werecharacterized by-abnormal bloating and andblackened skin with a dark-reddish tinge, andthe flesh appeared decayed very soon afterdeath. In a third incident, three deadmujahidin guerrillas were found with handson rifles and lying-in a firing position,indicating that the attacker had used anextremely rapid-acting lethal chemical that isnot detectable by normal senses and thatcauses no outward physiological responsesbefore death.
Shortly after the Soviet invasion, manyreports were received that both Soviet andAfghan forces were using various types ofchemical agents. Ten separate chemical attacks,resulting in many deaths, were reported in thefirst 3 months of 1980. These reports camefrom northeastern Afghanistan and provide thehighest percentage of reported deaths. Duringthe mid-January to February 1980 period,helicopter attacks were reported in northeasternAfghanistan in which a grayish-blue smokeresulted in symptoms similar to those
15
described by the H�Mong refugees from Laos(e.g., heavy tearing or watering of eyes;extensive blistering and discoloration of theskin, later resulting in large, sheetlikepeeling; swelling in the areas affected by theblister; and finally numbness, paralysis anddeath). Medical reports from examinations inPakistan of refugees from a large attack in theupper Konar Valley in February 1980described red skin and blisters containingfluid described as �dirty water.� Refugeesestimated that about 2,000 people wereaffected after contact with a dirty yellowcloud.
By spring and summer of 1980, chemicalattacks were reported in all areas of concen-trated resistance activity. Many reports fromdifferent sources strongly support the casethat irritants were used to drive the insurgentsinto the open to expose them to attack withconventional weapons and incapacitants torender them tractable for disarming andcapture. On several occasions in April 1980,for example, Soviet helicopter pilots dropped�gas bombs� on insurgents, evidently to drivethem from caves.
A Dutch journalist, Bernd de Bruin,published an eyewitness account of twochemical attacks occurring in the Jalalabadarea on June 15 and June 21, 1980 (Niewsnet,August 2, 1980. He filmed an MI-24helicopter dropping canisters that produced adirty yellow cloud. A victim with blackenedskin, discolored by extensive subcutaneoushemorrhaging, was photographed in thevillage 5 hours after the attack. Thejournalist evidently was exposed because hedeveloped blisters on his hands and a swollenand itchy face. He also was exposed in thesecond attack, and it took about 10 days forhim to recover from skin lesions, nausea,diarrhea and stomach cramps.
An Afghan insurgent provided aneyewitness account of a July 6, 1980 attackon a village 10 kilometers east of Darae Jelgain Vardak Province. He reported that aSoviet MI-24 helicopter gunship dropped abomb that, on explosion, released a lethalchemical. A separate report confirmed thatSoviet bombing attacks on villages in Vardakas well as Lowgar and Parvan Provinces weretaking place during this period. In August1980, information surfaced on a Soviet attackwith chemical bombs on the village of SyaWusan, 30 kilometers southeast of Heratleaving 300 dead. It was during this time thatthe Soviet chemical battalion at Shindand setup an operational decontamination station.
Reports of chemical weapons use in1981 essentially parallel 1980 reportingwith respect to frequency and location of
attack. Soviet helicopter units participated inchemical attacks from April 20 to April 29,1961, in areas east West of Kabul and in theKonar Valley, according to eyewitnessaccounts. These attacks were intended todrive personnel from sanctuaries such ascaves, in order to engage them with conven-tional fire. The munitions were described asSoviet 250-kilogram RBK cluster bombs. TheSoviets have such a munition, which can befilled with chemical agents. Other reportsdescribed similar operations by helicoptersnorth of Qandahar on April 24 and April 26,1981.A former Afghan MI-8 helicopter pilot saidSoviet forces had used chemical weapons inBadakhshan, Quonduz, and Konhara.Chemicals in canisters that contained toxicgas, tear gas, and anti-respiratory gas, whichhas an incapacitating effect by causingchoking and difficulty in breathing, weremanually pushed from the cargo compartmentof helicopters. The pilot said that there alsowas a specific gas that is absorbed by thebody and leaves the skin so soft that a fingercan be punched through it. In one case, therewas a wind shift, and Soviet and Afghanforces were seriously affected. Other sourcesalso have described an incident where Sovietand Afghan forces were victims of their owngas attack.The following sequence occurred in a smallvalley in Qandahar Province in early June1981. According to an Afghan exile, Sovietcombat groups engaged rebel forces in thatvalley during a 2-week period. The situationworsened for the Soviets, and an airstrike wasconducted. The exile stated that a Soviethelicopter delivered a single rocket, releasinga chemical that killed 16 insurgents. Nearlyall reports state that chemicals were deliveredby aircraft or helicopters a few reportsdescribe chemical artillery rounds.Before a sweep operation in the Konar Valleyin September 1981, resistance leaders weretold by an Afghan officer that the Soviets hadfour agents available but would use only theincapacitant which they could defend againstwith wet rags over the face. During theoperation, Soviet helicopters conducted gasattacks in 25 different areas, using cylindersabout 1.5 meters long and 60 centimeters indiameter that exploded 4-5 meters above theground, releasing the incapacitating gas.Some victims lost consciousness, wereparalyzed, and recovered, but others died,and unprotected areas of their skin turneddark green to blue-green.
An Afghan tribal leader recently de-scribed a Soviet chemical attack againsta large resistance force in October 1981
near Maruf, about 100 kilometers east ofQandahar. Soviet helicopters dropped greencylindrical canisters (18 inches long, 3-4inches in diameter) which, upon hitting theground, emitted a greenish-yellow gas.According to the report, victims felt faint anddizzy; later their skin began to itch, and manylost consciousness. About 300 persons wereaffected by the gas and many died. Sovietground forces captured many of the survi-vors. Other information on Soviet andmujahidin activities in the Qandahar areaduring this period confirms that this incidentdid in fact take place.
In February 1982,.a member of theresistance, with considerable knowledge ofSoviet weapons, told a U.S. official that theSoviets were using irritants, a hallucinogenicgas, and what he said was an apparent nervegas. He described the �nerve agent� as anoff-white powdery substance dispersed fromhelicopters generally during artillery orbombing attacks. Victims realize they havebeen exposed to chemical attack only whenthey become faint and dizzy. Subsequently,they begin to vomit and bleed from the eyes,nose and mouth. Death occurs within a shorttime. The corpses are extremely relaxed, withno evidence of rigor mortis. Flesh and skinfrequently peel off if an effort is made tomove to bodies.
According to this account, survivorssuffer aftereffects for about 6 months,including chest congestion and pain,dizziness, and mental agitation. The powder-like substance is more effective at loweraltitudes where there is less wind to dilute thepoison, and mujahidin groups have experi-ence fatality rates as high as 70%. Manysurvivors of chemical attacks in Laos andAfghanistan have exhibited the same long-term health problems described in thisaccount.
Chemical defense battalions�standard inall Soviet divisions�are deployed with thethree Soviet motorized rifle divisionsoperating in Afghanistan at Qonduz,Shindand, and Kabul. Soviet operationalpersonnel decontamination stations wereobserved at several locations, and chemicaldecontamination field units were deployedduring a sweep operation of the Konar Valleyin eastern Afghanistan and near Shindand inthe west in 1980. The operational deploy-ment of decontamination units for personneland equipment suggests that chemicalbattalions have supported offensive chemicaluse. In addition, Soviet personnel have beenobserved wearing chemical protectiveequipment. The Soviets have specificallytailored their forces in Afghanistan, in part
16
could successfully deny or counter chargesthat chemical weapons had been used,recognizing that it would be especiallydifficult to compile incontrovertible evidencefrom inaccessible areas of Southeast Asia andAfghanistan. With respect to Kampuchea,they may also have calculated that, in view ofthe lack of international support for Pol Pot�sresistance, chemical weapons could be usedon his troops without significant internationaloutcry.
In addition, the Soviet military verylikely considers these remote areas asproviding unique opportunities for theoperational testing and evaluation ofchemical weapons under various tacticalconditions. Years of aerial and artillerychemical dispersion have undoubtedlyprovided the Soviets with valuable testingdata. Southeast Asia has offered the Sovietsan opportunity to test old agents that hadbeen stockpiled for many years as well asmore recently developed agents or combina-tions of agents. This conclusion is supportedby information from foreign military officerswho have attended the Soviet MilitaryAcademy of Chemical Defense in Moscow.According to their Soviet instructor, threetypes of chemical agents may be used duringthe �initial stages� of local wars: �harassingagents (CS, CN, DM), incapacitants such aspsychochemicals (BZ) or intertoxins[sic-possibly enterotoxins], and herbicides.�During the �decisive phase, lethal agents canbe employed under certain circumstances.� Ina local war, �chemical weapons can be usedto spoil enemy efforts to initiate operations,even if the enemy has not used them first.�The foreign officers� accounts, includingdetailed descriptions of the Soviet chemicalwarfare program, support the conclusion thatthe Soviets consider chemical weapons aneffective and acceptable means of warfare inlocal conflicts.
Insight into the Soviet bloc militaryperspective on the use of toxins is provided inthe following passage from a 1977 EastGerman military manual entitled Textbook ofMilitary Chemistry.
Toxins are designated as toxic agents whichare produced by biological organisms such asmicro-organisms, plants, and animals, and cannotthemselves reproduce.
By the middle of 1960 the toxins selected formilitary purposes were included among thebiologic warfare agents. In principle, this wasunderstood to mean only the bacterial toxins.Today it is possible to produce various toxinssynthetically. Toxins with 10-12 amino acids cancurrently be synthesized in the laboratory. Toxinsare not living substances and in this sense arechemicals. They thus differ fundamentally from thebiological organisms so that they can be includedamong chemical warfare agents. As a result
jor irritant to be removed as quickly andcheaply as possible. The use of chemicalagents has played a major role in driving theH�Mong from their mountain strongholds,relieving Vietnamese and Lao ground forcesof the need for costly combat in difficultterrain. Much of the H�Mong population thatlived in the Phou Bia mountain region hasbeen driven into Thailand, killed, or resettled.
In the mountainous areas of Afghanistan,where rebels are holed up in eaves or otherinaccessible areas, conventional artillery,high-explosive bombs, and napalm are notparticularly effective. Many reports indicatethat unidentified chemical agents have beenused on such targets. Caves and ruggedterrain in Laos and thick jungles inKampuchea also have frustrated attempts tolocate and destroy the resistance forces.
Chemical clouds can penetrate the heavyforests and jungle canopy and seep into themountain caves. Persistent agents linger inthe area and cause casualties days andsometimes weeks after the attack. Unpro-tected forces and civilians have little or nodefense against lethal agents like toxins,nerve gas, or blister agents.
Trichothecene toxins, which are knownto have been used in Southeast Asia, have theadded advantage of being an effective terrorweapon that causes bizarre and horrifyingsymptoms. Severe bleeding, in addition toblisters and vomiting, has instilled fear in theresistance villages. Not only have thevillagers and their animals been killed in agruesome manner, but the vegetation andwater also have been contaminated. Survivorsare reluctant to return to their inhospitablehomes and instead make the long anddangerous trek to camps in Thailand.
There is no clearcut explanation of whytrichothecene toxins have been used inaddition to irritants, incapacitants, and othertraditional chemical warfare agents. Specula-tion suggests that they are probably cheaperto make and are readily available from Sovietstocks; they are probably safer and morestable to store, transport, and handle in aSoutheast Asian environment, and they mayrequire less protective equipment when beingprepared for munitions. They are difficult totrace as the causative agent after an attack-asdemonstrated by the length of time it took forthe United States to detect them. Fewlaboratories in the� world have the analyticalcapability to identify precisely the type andamount of trichothecene toxin in a sample ofvegetation, soil, or water.
The Soviets may well havecalculated that they and their allies
Motivation for UsingChemical Weapons
In the course of this analysis, the question hasbeen posed: Is there a military-strategic ortactical rationale for the systematic use ofchemical weapons by conventional forces inLaos, Kampuchea, and Afghanistan? Themilitary problems faced in thesecountries-viewed from the perspective of theSoviets and their allies-make the use ofchemical weapons a militarily effective wayof breaking the will and resistance ofstubborn Anti-government forces operatingfrom relatively inaccessible, protectedsanctuaries.
The Soviets have made a largeinvestment in insuring that Vietnam andits clients succeed in extending theircontrol over Indochina. For Vietnam,the H�Mong resistance in Laos is a ma-
because of logistical constraints; 5,000troops and �nonessential� combat equip-ment were withdrawn, but the chemicalbattalions remain.
A Soviet military chemical specialist,captured by the mujahadin, gave hisname as Yuriv Povarnitsyn fromSverdlovsk. During an interview he saidthat his mission was to examine villagesafter a chemical attack to determinewhether they were safe to enter or requireddecontamination. An Afghan pathologistwho later defected described accompanyingSoviet chemical warfare personnel intocontaminated areas to collect soil, vegetation,and water samples after Soviet chemicalattacks. According to firsthand experience offormer Soviet chemical personnel, theSoviets do not require decontaminationequipment in an area where chemicalbombs are stored or loaded onto aircraft.Thus deployment of this equipment inAfghanistan must be assumed to beassociated with the active employment ofcasualty-producing chemical agents.
Afghan military defectors have providedinformation on ammunition and grenadescontaining phosgene, diphosgene, sarin, andsoman and have described where and whensome of them have been used. They also haverevealed locations where these agents werestockpiled. The agents used, plus the timeand location of the attacks, correspond withthe refugee reports and recorded militaryoperations.
The Soviet Union has stocked a varietyof toxic chemical agents and munitions tomeet wartime contingencies. Weaponssystems capable of delivering chemicalmunitions available to Soviet forces inAfghanistan include artillery, multiple rocketlaunchers, and tactical aircraft.
17
airfield (Vientiane Province), where therockets were fitted onto racks of the L-19aircraft for missions in the Phou Bia area;later, the rockets from Phonsavan weretransported to Ban Xon by trucks. AllU.S.-manufactured rockets were stored withthe tip and canister kept apart; the two partshad to be joined before being fitted to theracks on the aircraft. The pilot observed,however, that all the rockets transported fromPhonsavan to Ban Xon were alreadyassembled.
As part of his routine flight activities, thepilot would check his aircraft and, in doingso, examine the tip portion of new smokerockets that had been transported fromPhonsavan. He said that most appeared�loose� in the portion where the tip andcanister joined, whereas the tip and canisterof the ordinary explosive-type rockets atLong Tieng were noticeably more tightlyconnected.
In late 1976, during preparation forairstrikes on Kasy (LouangphrabangProvince) and in new areas of Phou Bia, thepilot said he began carrying two or threeVietnamese Army staff officers, sometimesaccompanied by a Lao staff officer, in T-41aircraft for reconnaissance over the targetareas. When these airstrikes were launched,the defector pilot initially flew his L-19aircraft on missions with another pilot and aLao staff officer. After 2 or 3 weeks, however,Vietnamese staff officers, who spokeexcellent Lao, began alternating with the Laoofficers. Before each mission, the Vietnameseor Lao staff officer would go over target areasoutlined on situation maps-which then weretaken along-and would point out the targetsto be attacked. The defector pilot noted that atno time did the Vietnamese staff officercommunicate with Lao officers on theground, as did the Lao staff officers. A newVietnamese officer was assigned for eachairstrike mission in the H�Mong areas.
The pilot related that before flyingL-19 airstrike missions with a full loadof rockets he was often warned by a Laocommander to fly at above-normal alti-tudes when firing rockets-to precludehazard to the occupants of the aircraft.For this reason the pilot surmised thatthe �smoke� rockets fired at the H�Mongwere unusual. He was able to observethat the �smoke� rockets detonated inthe air and that some produced whitesmoke, with a mixture of blue, whileothers produced red smoke, with a mix-ture of yellow. The ordinary explosive-type rockets detonated on impact. The
Kampuchea, and Afghanistan-has any knownlarge-scale facility or organization for themanufacture of chemical and biologicalmaterials. Nor are they known to haveproduced even small quantities of chemicalwarfare agents or munitions. The technicalproblems of producing large quantities ofweapons-grade toxins, however, are not sogreat as to preclude any of the four countriesfrom learning to manufacture, purify, andweaponize these materials. It is highlyunlikely, however, that they could masterthese functions without acquiring outsidetechnical know-how.
ANNEX A
A Lao Pilot�s Account
One of the most complete descriptions ofchemical warfare activities in the 1976-78period came from a Lao pilot who wasdirectly involved in chemical warfare. Thepilot, a former Lao People�s Liberation Army(LPLA) officer who defected in 1979,reported that he flew captured L-19 and T-41aircraft equipped to dispense toxic chemicalagents on H�Mong villagers in the Phou Biaarea of northern Laos. He said that the LPLA,in cooperation with the Vietnamese Army,had conducted chemical warfare operations inLaos since April or early May 1976. At thattime, two Lao H-34 helicopters were flownbetween Long Tieng and the Phonsavanairfield, both in Xiangkhoang Province, on aseries of flights to transport rockets toPhonsavan for storage.
Between June and August 1976, theLPLA launched attacks in the area ofBouanilong-in Xiangkhoang Province-astronghold for remnants of the forces offormer H�Mong Gen. Vang Pao. The LPLAused L-19 aircraft for rocket attacks in thatarea aimed at eliminating the H�Mongresisting government control. Lao crewsresponsible for loading rockets on the attackaircraft noted that they were not allowed touse the rockets that had been moved fromLong Tieng to Phonsavan, even thoughPhonsavan was much closer to theBouamlong target area than Long Tieng,where Lao aircraft had to rearm. The pilotsaid that, during nearly 3 months of flyingmissions against the Bouamlong area, he flewhis L-19 aircraft to Long Tieng to be armedwith rockets.
In late 1976, the pilot�s L-19 aircraftwas rearmed with rockets stored atPhonsavan. Initially, H-34 helicopters wereused to transport the rockets fromPhonsavan to a depot near the Ban Xon
of their peculiarities they are designated simply as�toxin warfare agents.� They would be used incombat according to the same principles and withthe same methods used for chemical warfareagents. When they are used in combat theatmosphere can be contaminated over relativelylarge areas-we can expect expansion depths of upto 6 kilometers before the toxin concentrationdrops below lethal concentration 50 ... the toxinwarfare agents can be aerosolized. They can beused primarily in micro-bombs which are launchedfrom the air or in warheads of tactical rockets.Toxin warfare agents concentrates can be appliedwith aircraft spray equipment and similardispersion systems.
The Soviet designation for severalpathogenic Fusarium products is �IIF(iskusstvennyy infektsionny fon), whichstands for �artificial infection background.�IIF devices are used in the Soviet Uniondeliberately to contaminate soil in experimen-tal agricultural test areas with spores ofdisease-producing fungi. We are not certain ifthe IIF compounds include trichothecenes.Nor are we certain as to the intent of thisagricultural research program. It is possiblethat these programs are designed to colonizesoil with pathogenic organisms either todetermine which crop varieties are mostresistant to disease or, alternatively, to testeradication and control methods in infectedsoils. Elsewhere in the Soviet agriculturalresearch program, however, it is known thatthere is widespread use of certaintrichothecenes, including sprays from lightaircraft. A capability exists within the SovietUnion for multi-ton production of lightaircraft spray-delivered microbial productssuch as those described above.
Evidence accumulated since World WarII clearly shows that the Soviets have beenextensively involved in preparations forlarge-scale offensive and defensive chemicalwarfare. Chemical warfare agents anddelivery systems developed by the Sovietshave been identified, along with productionand storage areas within the U.S.S.R. andcontinuing research, development, and testingactivities at the major Soviet chemicalproving grounds. Soviet military forces areextensively equipped and trained foroperations in a chemically contaminatedenvironment. None of the evidence indicatesany abatement in this program. The Sovietshave shown a strong interest in improving orenhancing their standard agents for greaterreliability and effect. Their large chemicaland biological research and developmenteffort has led them to investigate other kindsof chemical warfare agents, particularly thetoxins.
None of the four countries con-sidered in this report-Vietnam, Laos,
18
commander or his designated representativetold the pilot before every mission that theoperations-called Extinct DestructionOperations-were intended to �wipe out thereactionary H�Mong people.�
Before a mission involving �smokerockets,� the commander warned the pilots tokeep the operation secret. The Lao defectorsaid that, during the nearly 2 years in whichhe flew rocket missions he learned from theLao staff officers accompanying him thatthere were two types of rockets. The first,mostly �smoke� rockets, were to be fired attargets far away from Lao and Vietnamesetroops to avoid exposing them to the poisonsmoke. The second was of the ordinaryexplosive type, considered �close support�rocket that could be fire near Lao trooppositions. Initially, the L-19 aircraft carriedeight rockets-five �close support� and three�smoke� rockets. Later, only four rockets,mainly of the �smoke� type, were carried.
After each mission in which chemicalwarfare rockets were used, the pilot wasreturned to a �rest house� at Phonsavanwhere a Lao Army doctor and nurse wouldexamine him. He said that after his missions,especially in 1978, he was particularly welltreated by the examining doctor and watchedvery closely by the nurse. Those L-19 aircraftpilots assigned to missions utilizing chemicalwarfare rockets had special privileges,including additional flight pay and free mealsat the Phonsavan cafeteria. In October 1978,the Lao Army stopped using L-19 aircraft oncombat missions and began using SovietMiG-21s for chemical attacks on the PhouBia areas.
Several H�Mong reports corroborate thetestimony of the Lao pilot. A village chief, forexample, described attacks covering all 7days of the week of June 5, 1976 in theBouamlong area. He described L-19 aircraftfiring rockets that produced red and greensmoke: Ten villagers were killed by gas and30 by shrapnel. Most of the H�Mong reportsdocumented by a U.S. Foreign Service officerin June 1979 and a Department of Defensemedical team in October 1979 were consis-tent with the pilot�s testimony. H�Mongobservers familiar with military aircraftreported L-19s in use until late 1978. Afterthat time, reports described jets or �MiGs�and some accurately described Soviet AN-2s.
A review of information back to1975 shows L-19 and T-28 aircraftwere operating from airfields in
northern Laos-including the one atPhonsavan, where AN-2s were seen in 1978.Failure to observe chemical decontaminationequipment at the airfields does not rule outthe presence or handling of chemicalmunitions. The Soviets supervise thechemical warfare activities in Laos; it isassumed that chemical munitions are handledin about the same manner as in the U.S.S.R.According to former Soviet chemical warfarepersonnel, no protective clothing or specialdecontamination equipment is required forloading chemical bombs onto aircraft andhelicopters at chemical munitions test ranges.
The Lao pilot�s description of the rocketsused on the L-19 was corroborated by othersources. A H�Mong refugee, a formercommander of a 500-man resistance force,reported that in 1977 he found a rocketcanister and a separated warhead that hebelieved were the kinds used by the Vietnam-ese and Lao. The canister had authentic U.S.markings identifying it as a U.S.-manufac-tured 2.75-inch rocket and, reportedly, threelines of Russian writing which he could nottranslate. Another H�Mong resistance forceofficer, reportedly trained as a liaison officerand ordnance expert before the Communisttakeover of Laos, stated that he, too, believedthat the rocket canister was of U.S. manufac-ture and that the Soviet technicians in Laoshad modified the upper stage to contain apoisonous (i.e., lethal) chemical.
The diameter of the warhead wasreported to be 12.5 centimeters (5 inches),probably a measurement taken on a modifiedwarhead, because the United States does nothave a 5-inch warhead for the 2.75-inch�rocket motor.� During the Vietnam conflict,about 35 million U.S.-manufactured,conventional 2.75-inch rockets were sent tothe war zone, and many tens of thousands ofthese fell into North Vietnamese hands whenthe South Vietnamese forces collapsed. TheVietnamese may be using some of theserockets with existing loads, but modifiedwarheads for the 2.75-inch rocket motorcould easily be fabricated in Vietnam andfilled with a lethal or nonlethal agent in Laos,especially with Soviet assistance. Accordingto U.S. experts, fabrication of a warhead 5inches in diameter, necked down to fit the2.75-inch rocket, could be accomplished bytrained technicians in a small, well-equippedmachine shop and laboratory.
ANNEX B
FINDINGS OF U.S. GOVERNMENT INVESTIGATIVE
TEAMS: USE OF CHEMICAL AGENTS AGAINST
THE H�MONG IN LAOS
State Department TeamIn May 1979, State Department officialsvisited Thailand to interview H�Mongrefugees and investigate allegations of the useof chemical agents against H�Mong tribesmenin Laos (see Table B-1). From the signs/symptoms described and observed, it issuggested that at least two and possibly threedifferent chemical agents may have beenused, such as:
� A nerve agent (five or six individualsreported symptoms that could be attributed toa nerve agent);
� An irritant or riot-control agent(one-third of the interviews); and
� More than half of the interviewsindicated such a variety of signs andsymptoms that it is difficult to attribute themto a single known agent.
It is possible that in some cases two ormore agents were combined.
� Reported signs and symptomssuggesting a nerve agent include sweating,tearing, excessive salivation, difficulty inbreathing, shortness of breath, nausea andvomiting, dizziness, weakness, convulsions,and death occurring shortly after exposure.
� Reported signs and symptomssuggesting a riot-control or irritant agentinclude marked irritation or burning of theeyes, with tearing and pain; irritation andburning of the nose and throat; coughing;burning and tightness in the chest; headache;and nausea and vomiting in a few cases.
� Reported signs and symptoms notrelated to any known single agent include amixture of the above as well profuse bleedingfrom mucous membranes of the nose, lungs,and gastrointestinal tract, with rapid death ofthe affected individuals in some instances.
Estimates from the H�Mong interviewedindicate that approximately 700-1,000persons may have died as a result of the useof chemical agents and that many times thisnumber became ill. It was reported that onmany occasions entire villages were devas-tated by these agents, leaving no survivors.
In the episodes described, most ofthe animals exposed to the chemicalagents were killed. Generally, all
19
chickens, dogs, and pigs died and, to a lesserextent, the cattle and buffalo. On severaloccasions it was reported that where theseagents settled on tree and plant leaves, manysmall holes appeared in the leaves within 2or 3 days. Rarely did agent exposure resultin the defoliation or death of the plants.
Department of Defense Team
From September 28 to October 12, 1979, ateam from the U.S. Army Surgeon General�sOffice was in Thailand to conduct a similarseries of interviews.* The team visited thefollowing H�Mong refugee camps ofnorthern Thai land: the detention center atNong Kai, the large H�Mong camp at BanVinai, and two smaller camps at Nam Yaoand Mae Charim. As the great majority ofrefugees as well as the H�Mong leadershipare at Ban Vinai, most interviews wereobtained there.
The team was prepared to obtain bloodand skin samples (for cholinesterase activityand study of pathological changes, respec-tively) from those exposed to chemicalagents. For such samples to yield meaning-ful results they must be taken within 6-8weeks of exposure. Since the last reportedexposure was in May 1979, no blood or skinsamples were collected.
Interviews were conducted throughinterpreters; one was an employee of theU.S. Consulate at Udorn, and the remainderwere hired from among the refugees. Theinterpreters screened those refugees whovolunteered to talk to the team and selectedonly those who had been eyewitnesses to orhad themselves been exposed to an agentattack. Team members interviewed 40 men,2 woman, and a 12-year-old girl. Eachinterview took 1-2 hours. To insureconformity, a prepared questionnaire wasused as a guide.
The chemical attacks reportedlyoccurred between June 1976 and May 1979(Table B-1). The absence of reports ofattacks after May 1979 may be because
* The authors of the U.S. Army SurgeonGeneral�s report are Charles W. Lewis, M.D.,COL, MC, Chief, Dermatology Service, BrookeArmy Medical Center, Fort Sam Houston, Texas;Frederick R. Sidell, K.D., Chief, ClinicalResources Group, U.S. Army BiomedicalLaboratory, Aberdeen Proving Ground, Md.;William D. Tigertt, M.D. (Brigadier General, Ret.,USA), Professor of Pathology, University ofMaryland, Baltimore, Md.; Charles D. Lane, LTC,Southeast Asia Desk Officer, OACSI, Departmentof the Army, Washington, D.C.; and Burton L.Kelley, SP5, USA, Dermatology Technician,Brooke Army Medical Center, Fort Sam Houston,Texas.
TABLE B-1Reports of Probable Chemical Agent Attacks in Laos
Department of State Interviews Conducted in Summer 1979Method of Attack Material Used
Date Location by Plane (Smoke/Gas)
Oct. 1977 Phu Hay, S. of Phou Bia Rockets Yellow-gray1978 Pa Sieng, S. of Phou Bia Bomb YellowFeb. 1978 Ban Nam Luk, S. of Phou Bia Spray (?) Yellow/white
Feb. 1978 20 kms SE. of Phou Bia Spray YellowFeb. 1978 Ban Ko Mal Bomb Yellow
Mar. 1978 Pha Houei Sacks, burst in air BrownMar. 1978 Ban Na Pong YellowApr. 1978 Ban Phamsi White, green,
blood-colored
May-Apr. 1978 Ban Nong Po Cloud Yellow-brownlike rain
June 1978 Ban Nam Teng Rocket YellowJune 1978-May 1979 Ban Don area Spray YellowMid-1978 1-3 kms NE. of Phou Bia Rocket, air burst Red
Oct. 1978 Nam Kham Rockets, air burst YellowOct. 1978 6 kms N. of Phou Khao Rockets, air burst RedOct. 1978 3-4 kms N. of Phou Bia Rockets, air burst Yellow-gray
Nov. 1978 Phou Xang Noi Spray Yellow, blueNov. 1978 near Phou Bia Bomb, air burst YellowNov. 1978 NE. of Pha Khao Rocket, air burst YellowApr. 1979 Ban Nouia Pong Spray YellowMay 1979 Nam Po Spray YellowMay 1979 Pha Mai Spray, air burst Yellow
Department of Defense Interviews Conducted In Fall 1979Method of Attack Material Used
Date Location by Plane (Smoke/Gas)
June 1976 Pou Mat Sao Rockets Red, greenJan. 1977-Oct. 1978 Pha Khao Rockets Yellow, red, greenMar. 1977 Nam Theuna Rockets Red, yellowApr. 1977 Houi Kam Lang Rockets YellowMay 1977 Pha Khae Rockets RedMay 1977 Nam Moh Rockets YellowMay 1977 Pha Ngune Spray/rockets Yellow1977-1978 (3 attacks) Phu Seu Rockets Red, green, yellowJan� 1978 Houi Xang Rockets Red, greenFeb. 1978 Sane Mak Ku Rockets YellowFeb. 1978 Tham Se Sam Leim Rockets Yellow, blackFeb. 1978 Kio Ma Nang Rockets YellowMar. 1978 Mouong Ao Rockets WhiteMar. 1978 Khieu Manang Rockets GreenApr. 1978 Tha Se Rockets WhiteJune 1978 Pha Phay Rockets YellowJune 1978 Phou Seng Rockets Red, white, blackJuly 1978 Phou Bia Rockets RedJuly 1978 Ban Nam Mo Spray YellowJuly 1978 Phou Lap Rockets YellowAug. 1978 Pha Houai Rockets Red, greenAug. 1978 Ban Thin On Rockets Green, redAug. 1978 Bouamlong Rockets Red, green, yellowSept. 1978 Pha Koug Rockets YellowSept. 1978 Ban Nam Tia Spray/rockets Yellow, green, redSept. 1978 Pha Na Khum Rockets RedOct. 1978 Phou Bia RocketsOct. 1978 Ban Done Spray YellowOct. 1978 Phou Bia Rockets White, green, redNov. 1978 Phou Bia Rockets White, redFeb. 1979 Pha Mat Spray YellowFeb. 1979 Tong Moei Rockets Yellow, redMar. 1979 Pha Mai Spray YellowMar.-May 1979 Pha Mai Spray Yellow
(6 attacks)Apr.-May 1979 Pha Mai Spray Gray-white
(4 attacks)May 1979 Phou Bia Spray YellowMay 1979 Moung Phong Rockets Red
20
few refugees crossed the Mekong River afterthat time-as a result of heavy rains andflooding from June to September 1979. Mostof the early reports were of the use of rocketsreleasing the agent; beginning in the fall of1978, the majority of the attacks were carriedout by aircraft spraying a yellowish substancewhich �fell like rain.� The attack sites,concentrated around the H�Mong strongholdin the mountainous Phou Bia area, also arelisted in Table B-1.
The team was given a plastic vialcontaining pieces of bark, stained by a yellowsubstance, which several H�Mong refugeesclaimed was residue from an aircraft sprayattack in April 1979. Preliminary chemicalanalysis of the sample indicates that nostandard chemical agent (i.e., an agent listedin TH 8-285, U.S. Army, May 1974) waspresent.
ConclusionsThe conclusions of these teams, based uponinterviews obtained from H�Mong refugees,are as follows:
� Chemical agents have been usedagainst the H�Mong.
� The reported effects of these agentssuggest the use of a nerve agent, ariot-control agent, and an unidentifiedcombination or compound.
ANNEX C
MEDICAL EVIDENCE
Southeast AsiaSince 1975, many different sources-refugees,relief workers and medical personnel,including specially qualified physicians--consistently have detailed unusual signs andsymptoms of victims of �yellow rain.�Specifically, victims in Southeast Asiasubjected to a direct attack of the yellowpowder, mist, smoke, or dust would be seento begin retching and vomiting withinminutes. These effects and those describedbelow were not pronounced in individualseven 100 meters from the attack zone,indicating a relatively dense chemical/carriercombination that was effective in low windconditions.
Following the victim�s exposure toyellow rain, the initial induced vomit-ing-unlike that caused by a traditionalriot-control nausea agent-was pro-tracted over hours to days. It was oftenaccompanied by dizziness, rapid heart-beat and apparently low blood pressure,chest pain, loss of far-field vision, and afeeling of intense heat and burning onthe skin, although not described as being
most acute in the groin and axillae. Thus, theacute signs and symptoms match some effectsof traditional vomiting and blister agents butclearly not all.
Within the first hours after the attack,many victims also reported intense red eyes,bleeding gums, convulsions or more oftentrembling, and vomiting of blood, with orwithout production of copious amounts ofsaliva-lasting many hours to days, apparentlydepending on the exposure level. Thickmucous, pinpoint pupils, respiratory collapse,prolonged spasticity, and involuntaryurination or defecation were never reportedafter a yellow rain attack; the absence of thesesymptoms helped to rule out organophos-phate nerve agents in the minds of chemicalwarfare experts. Many medical and environ-mental samples also ruled out these and othertraditional agents such as DM, DS, andothers.
Many observers of �yellow rain� effectsreported formation within several hours ofsmall (1 centimeter) homogeneous, hard,fluid-filled blisters over only exposed areas ofskin, frequently including the victim�s hands,arms, entire throat, and face-wherever skinwas uncovered. In most cases the vomit, after2-8 hours, contained blood and, in manycases, large amounts of it. About half of thosereceiving the most concentrated doses ofyellow material-those who had, been directlyunder the spray-were observed within severalhours to cease vomiting temporarily. Thisinterval was often followed in 5-15 minutesby a period of great pain when the victimwould hold his abdomen and emit a gush ofblood from mouth and nose. These individu-als usually died within minutes afterward.
Close questioning by physicians ofwitnesses to these final moments leaves nodoubt that the effects resulted from severegastrointestinal bleeding, significantpulmonary bleeding, temporary compressionof accumulated blood in the stomach, and,finally, projectile vomiting of as many asseveral hundred milliliters of blood. Thesefindings were consistent with animal andhuman autopsies.
Many victims of the yellow materialreceived less than the full brunt of aspray, entered the attack zone severalhours to 2 days later, or consumed foodor water contaminated by the material.These individuals- often within the next24 hours-developed signs and symp-toms similar to those more directly af-fected but often without pronouncedskin effects if they had not contacted thepowder residue directly. In addition to
attacks of intense vomiting five or six times aday, they also had diarrhea, with bloodystools passed up to eight times a day.Bleeding under the fingernails and around theskin of the eyes and severe bruising of theskin also were commonly reported. Opiateshelped the fluid loss in adults, but in childrenor young persons unable to tolerate thetreatments of raw opium and water, deathoccurred after 10 days to 2 weeks in abouthalf the cases. On the basis of reported signsand symptoms, the cause of delayed deathalmost certainly was dehydration.
In many cases, chemical attacks arereported to produce symptoms other thanthose described here. However, there hasalways been a direct association of the abovesymptoms with reports of yellow rainattacks-that is, when yellow material is usedthese symptoms appear; other agents maygive rise to other symptoms. Although it ispossible to exhibit one or even several ofthese symptoms associated with traditionalchemical warfare agents, no expert has beenable to fit the sequence, severity, andconsistency with any of them. In many cases,victims and observers were examined,histories taken, and interviews conducted byseveral health professionals weeks apart.Remarkable consistency has been observed.
From the beginning of the yellow rainepisodes in 1975, autopsies occasionally havebeen reported anecdotally. Some have beendone inexpertly, some by nonphysicians, andsome were performed on animals rather thanon human victims. However, the consistencyof the early reported �putrefaction� or�rottenness� of the digestive tract within12-48 hours after death led many forensicmedical experts to suspect that one effect ofthe poison-whatever it was- was to causenecrosis (cell death) of rapidly dividingmucosa (mucous membranes), especially inthe stomach and upper small intestine. Otherautopsy findings included hyperemia(engorgement with blood) of digestivemucosal linings and remarkably intensecongestion and swelling in the lungs, liver,spleen, and sometimes the kidneys. These andother findings often led experts in toxicologyand pathology, on the basis of clinical andpathological data alone, to suggest mycotoxinor even trichothecene intoxication.
Trichothecene effects have beenreported in the forensic, oncological, andtoxicological literature for several years.Unpublished findings often have beendiscussed in symposiums. In severaldozen cases, toxic effects in humans and
21
TABLE C-1
Comparison of Reported �Yellow Rain� EffectsWith Known Trichothecene Effects
Yellow Rain Reports* Effects of Trichothecenes1. Nausea, vomiting�severe, immediate 1. Nausea, vomiting�severe, immediate
2. �Failing down, world turning� 2. Dizziness
3. �Burning of skin�-small blisters 3. Generalized erythema with a burningsensation of skin
4. �Shaking all over, flopping like fish 4. Ataxia (failure of muscular coordination),out of water� occasional tremors and convulsions
5. �Bleeding eyes� 5. Congestion of the sclera (white outercoat of eyeball) and blood in tears
6. �Pounding� chest, rapid heartbeat, 6. Hypotension (abnormally low bloodweakness pressure) with secondary rise in
heart rate
7. Severe pain in center of chest 7. Angina (substernal chest pain)
8. Sleepiness, �not able to talk� 8. Somnolence, central nervous systemsymptoms
9. Bleeding gums and profuse salivation 9. Stomatitis (inflammation of oral mucousmembranes) and ptyalism (excessivesalivation)
10. �Can�t breathe� 10. Shortness of breath
11. �Skin and body hot with cold� 11. Fever and chills
12. Diarrhea with blood 12. Diarrhea with blood
13. Loss of appetite, inability to eat 13. Anorexia
14. Bleeding into skin and fingernails 14. Thrombocytopenia (decrease in numberof platelets, white blood cells involvedin clotting of blood) and purpura (skindiscoloration caused by hemorrhage intotissues)
15. Drop in white blood cell count 15. Leukopenia and anemia
16. �Rotten esophagus, stomach, intestines; 16. Rapid necrosis of linings of gastro-soft spleen and liver� intestinal tract; lymphoid necrosis in
spleen and liver
17. Swelling of all organs 17. Congestion of all organs
* Effects are immediate at levels near to or above a rough estimate of 500-1,000 mg total body burden foran adult. Although inhalation data are pending, the levels are consistent with reported lethal and sublethal doses.Trichothecenes in combination, when directly ingested or inhaled, or in purified form, are more toxic in lowerconcentrations, and the order of signs and symptoms and timing varies.
animals have been carefully recorded, andthey match those of yellow rain with goodprecision (see Table C-1). There are noadditional signs or effects of knowntrichothecene intoxication not frequentlyreported by victims, nor are there anyreported yellow rain symptoms that cannot beexplained by the effects of the four specifictrichothecene toxins found in the samples.
There are no significant medicaldifferences in the reporting from Laos andKampuchea. Although the timing anddelivery systems have sometimes varied, theeffects of the chemical agent, clinically andpathologically, are identical. In some cases, aseries of blood samples from Kampucheanvictims also showed a trend toward leukope-nia (reduction in the number of white bloodcells) and the presence of a trichothecenemetabolite (HT-2) consistent withtrichothecene intoxication (see Annex D).Dose-response effects that were observed androutes of administration were both consistentwith effects of trichothecenes.
An early hypothesis (1978-79) was that asignificant number of deaths, especially inLaos, could be explained by the heavy use ofriot-control agents such as CS, CN, DM, andagents which cause itching and/or blistering.This hypothesis was rejected quickly on twogrounds. First, trace contaminant analysisfailed to show the presence of any of thesecompounds in samples; several samples did,however, contain a trichothecene precursor.Second, contrary to commonly held views,the epidemiology of diseases endemic to thecentral highlands of Laos and the publichealth situation of the H�Mong do notsupport the view of malnourished, disease-ridden, and weak persons who wouldsuccumb easily to riot-control agents. Also,many studies have shown the opposite: arelatively low incidence of pulmonarydisease, lower than what could otherwiseaccount for certain effects; better nutritionalstates than could otherwise account for deathin 10 days to 2 weeks from water loss(dehydration) and calorie depletion; and adeath rate of nearly zero from causes otherthan infection, old age, and trauma.
Afghanistan
Some deaths associated with bleedinghave been described in the accountsfrom Afghanistan. In one set of cases, aphysician examined persons who hadbeen exposed to sublethal doses of ayellow smoke/black smoke combinationattack and one man near death after aseries of attacks. Hemoptysis (nasal
bleeding)�but not hematemesis (bleedingfrom the gastrointestinal tract)�was reportedin about half of these cases.
Several features of at least one of thechemical agents�an incapacitant�used inAfghanistan defy explanation at this time.One possibility is that the agent(s) are highlyselective for the central nervous systemrather than the autonomic nervous system.As yet, no good candidate agent has beenidentified which will selectively inhibit thecentral nervous system so as to causeunconsciousness for several hours. Anotherfinding has been the presence of a der-
mal anaesthesia, affecting only exposed areasof skin.
Postattack Medical SurveyThere is evidence that after some attacks inLaos and Afghanistan, Lao Communist orSoviet forces entered the attack zones toconduct surveys. Several reports indicatethat survivors from a toxin attack on a Laovillage were taken several kilometers fromthe village and injected with a small volumeof a clear solution said by their captors to bea �new� medicine to assess the gas. Theinjections, given intramuscularly in theupper arm, reportedly did nothing to
22
alleviate the weakness, nausea, vomiting, ordiarrhea suffered by the survivors. One victimreported the drug caused an immediatesensation of warmth throughout his body.Only the use of opium later eased thediscomfort. It is probable that this procedurewas a test either of a new antidote or of adrug developed to reduce incapacitation fromthe nausea and vomiting.
Similarly, in a few cases in Afghanistan,Soviet troops reportedly disembarked fromhelicopters or armored personnel carriers atthe edge of an attack site. Three or four,dressed in full anti-contamination gear,walked among the dead, examined thecorpses and, opening them with a crudeincision examined the organs in the abdomi-nal and thoracic cavities. In one case, asolution was poured into the incision. Whenthe corpses were later recovered by themujahidin, the body cavity contents had beendestroyed beyond recognition. These and afew additional reports support the hypothesisthat the perpetrators of some of the attackswere interested in studying aftereffects,lethality, or some other quasi-experimentalaspect of the use of a new chemical weapon.Recent indications from Afghanistan indicatethat one purpose of the field surveys andbody examinations is to determine levels oftoxic materials still present in the attack zonebefore Soviet troops occupy it.
ANNEX D
ANALYSIS AND REVIEW OFTRICHOTHECENE TOXINS
Sample Analyses for Trichothecenes
The Trichothecene Hypothesis. Since 1975,the U.S. Government has received remark-ably consistent reports detailing chemicalattacks in Southeast Asia. Some of thesereports described the use of lethal agentswhich produced symptoms that could not becorrelated with those produced by known ortraditionally recognized chemical warfareagents or combinations of them (see TableD-1). It is readily apparent that the symptomsmost frequently described in Laos andKampuchea correspond most closely withthose produced by a group of mycotoxins-thetrichothecenes. A review of the scientificliterature revealed not only that thesecompounds had physical and chemicalproperties indicating potential as chemicalagents but also that they were the subject ofintensive investigation by Soviet scientists atinstitutes previously linked with chemical andbiological warfare research. In the fall of
1980, the trichothecenes were added to thelist of agents suspected to have been used inSoutheast Asia and Afghanistan. Othercandidates under consideration includedphosgene oxime, arsines, cyanogen chloride,nerve agents, riot-control agents, andcombinations of these agents.
Many samples from chemical attacks inLaos and Kampuchea were examined at theU.S. Army�s Chemical Systems Laboratory(CSL) for the presence of traditional chemicalwarfare agents and were reported to benegative. In March 1981, CSL reported thepresence of an unusual compound (C
15H
24) in
the vapor analyses from several clothing andtissue samples taken from the victim of achemical attack. The compound was closelyrelated in structure to the simpletrichothecenes. This finding sparked therequest for analysis of all future samples forthe presence of trichothecene mycotoxins.
The Kampuchean Leaf and Stem Sample:The First Analysis for Trichothecenes. OnMarch 24, 1981, a number of samples werereceived from the U.S. Embassy in Bangkok.Two were reported to have been collectedfrom the site of a chemical attack thatoccurred in the vicinity of TV 3391, an areajust south of Phnom Mak Hoeun. Avegetation sample and a water sample werecollected within 24 hours of the attack.Examination of bodies of victims of victimsof this attack by medical personnel revealedhighly unusual degeneration of the mucosallining of the gastrointestinal tract. The effectsdescribed paralleled those known to beproduced by the trichothecenes. The sampleswere submitted to the Chemical SystemsLaboratory for analysis for the presence ofchemical warfare agents. With the exceptionof the unusual presence of high levels of CN-,Cl-, and F-ions, no evidence of knownchemical warfare agents was found. An initialtest for the trichothecenes by thin layerchromatography was inconclusive because ofsevere problems with interfering substancesand the lack of appropriate standards.
The trichothecenes are difficult todetect even under ideal circumstances, andthe presence of interfering substances inthe sample may make identification andquantification by thin layer chromatogra-phy inconclusive. A review of the limita-tions and potentials of the analyticalmethods for trichothecenes led to theconclusion that the computerized gaschromatography/mass spectroscopymethod in the selected ion-monitoringmode enabled precise identification andquantification of these compounds incomplex mixtures. A comparison of the
currently available methods suitable fortrichothecene analysis and an assessment oftheir utility and limitations is presented inTable E-3.
A portion of the leaf and stem samplewas furnished to the U.S. Army MedicalIntelligence and Information Agency forfurther analysis. This sample, a positivecontrol sample to which T-2 toxin was added,and a negative control sample of similarvegetation were forwarded to Dr. Chester J.Mirocha of the Department of Plant Pathol-ogy, University of Minnesota. Dr. Mirochawas given no information concerning thehistory or content of the samples and wasrequested to analyze the three unknowns forthe presence of trichothecene toxins using thebest methods at his disposal.
The analysis involves a series ofextractions followed by ferric gel separation,selected ion monitoring on a computerizedgas chromatograph/mass spectrometer, and afull mass spectral scan for comparison withknown standards. The methods used areamong the most sensitive and specific fordetection of these compounds; also, falsepositives are rare. Toxins can be identified bytheir mass spectra and quantified with a highdegree of accuracy. The vegetation sampleallegedly exposed to a chemical warfare agentwas found to contain 109 parts per million(ppm) of nivalenol, 59.1 ppm ofdeoxynivalenol, and 3.15 ppm of T-2 toxin;each is a potent toxin of the trichothecenegroup. No trichothecenes were detected in thenegative control sample, and 35 ppm of T-2toxin were detected in the sample to whichT-2 toxin had been added. It was Dr.Mirocha�s assessment that a mixture of theseparticular toxins in the high levels detectedcould not have occurred as a result of naturalcontamination.
The possibility that the identified toxinswere produced by natural fungal contamina-tion was discounted on the basis of theclimatic conditions required for production ofT-2 toxin, the high levels of toxins detected,the unusual mixture of toxins found, and theresults of surveys of Southeast Asia for thepresence of these toxins. This conclusion wassupported by the analysis of normal florasamples from Kampuchea described below.
Analyses of Control Samples FromKampuchea for the Presence ofTrichothecenes. On September 20, 1981, theU.S. Army Medical Intelligence andInformation Agency received nine controlsamples from U.S. Army personnel inBangkok for the purpose of conductinglaboratory analyses for background
23
TABLE D-1Symptoms of Chemical Attacks Reported in Laos, Kampuchea, and Afghanistan
% of ReportsMentioning Tricho- Nerve Phosgene Incapacitant Control
Symptom Symptom thecenes Agents Arsines Oxime Cyanogens (BZ) Agents
Laos
Multiple deaths 84.6 X X X XVomiting 71.4 X X X XDiarrhea 53.1 X X XHemorrhage 52.0 X Xa
Breathing difficulty 47.95 X X X X X X XItching and skin irritation 43.9 X X X XNausea 42.8 X X X X XAnimal death 41.8 X X X XBlurred vision 39.7 X X X X X X XHeadache 36.7 X X X - X XFatigue 35.7 X X - XNasal excretion 34.7 X X X X - XRash or blisters 32.6 X X X - XTearing 30.6 X X X X X XCoughing 28.6 X X X X X
XEffect on vegetation 26.5 X X XDizziness and vertigo 25.5 X X X X XFacial edema 20.4 X X X XThirst and dry mouth 20.4 X XSkin color change 16.3 X XTachycardia 12.3 X X X X X XTemporary blindness 9.18 X X X X XRapid loss of consciousness 9.18 Xb X X XSalivation 6.12 Xc XHearing loss 5.1 XTremors or convulsions 4 X X X XSweating 3 XParalysis 3 X X XLoss of appetite 3 X X XFrequent urination 2 X X
Continued on p. 25
Note: This table is a compilation relating the signs and symptoms reported in the threecountries to symptoms associated with certain chemical agents. The frequency with which aparticular symptom was reported is expressed as a percentage of the total number of attacks.
levels of trichothecene toxins. Thesamples were collected from an areanear TV 3391 that had not been sub-jected to any reported chemical attacks.The samples were collected by U.S. per-sonnel under instructions to reproducethe sampling conditions, handling,packaging, and transfer conditions of theoriginal sample as closely as possible.The same species of plant was sampled,and four other vegetation samples alsowere collected. A water sample and twosoil samples were recovered. Corn andrice samples from the area also weretaken. These grains provided an idealsubstrate for growth of toxin-producingfungi and would, therefore, be a sensi-
tive indicator of any natural occurrence.The nine samples were forwarded undercode to Dr. Mirocha for trichotheceneanalysis. A portion of each sample alsowas submitted to Chemical SystemsLaboratory for background determina-tions of CN-, Cl-, and F-levels. Notrichothecenes were detected in any ofthese samples, indicating that nivalenol,deoxynivalenol, T-2, and diacetoxyscir-penol are not prevalent in the geographi-cal area from which the alleged chemical,warfare-exposed sample was collected.The appearance of these trichothecenesin high levels and unique combinations ina sample associated with a chemical at-tack�which produced symptoms typicalof trichothecene exposure�indicates
that these toxins may have been used aschemical weapons. This conclusion is fur-ther supported by the evidence providedby analysis of additional alleged chemicalwarfare samples from Laos and Kampu-chea as described below.
Analysis of Additional ChemicalWarfare Samples From Laos andKampuchea for the Presence of Tri-chothecenes. The U.S. Army Medical Intelli-gence and Information Agency receivedfrom the Chemical Systems Laboratorythree additional suspected chemical war-fare samples for analysis for trichothe-cenes. The first sample consisted of 10ml of water taken from the same chemi-
24
TABLE D-1 (continued)Symptoms of Chemical Attacks Reported In Laos, Kampuchea, and Afghanistan
% of Reports RiotMentioning Tricho- Nerve Phosgene Incapacitant control
Symptom Symptom thecenes Agents Arsines Oxime Cyanogens (BZ) Agents
Kampuchea
Multiple deaths 72.4 X X X � X � �Hemorrhage 62.06 X � � Xd � � �Dizziness and vertigo 51.7 X X � � X X XVomiting 41.3 X X X � � � XNausea 34.5 X X X � � X XSkin irritation 27.6 X � X X � � XRapid loss of consciousness 24.1 Xb X � � X X �Fever 20.68 X � � � � � �Headache 17.2 X X X X XTearing 13.8 X X X X X X XBreathing difficulty 13.8 X X X X X X XFatigue 13.8 X X � � � X �Paralysis 10.3 X X � � X � �Numbness 6.9 X X � � X X �Blurred vision 6.9 X X X X X X XDry throat and thirst 6.9 X � � � � X �Edema 6.9 X � X X � � �Salivation 3.4 Xc X � � � � �Vegetation affected 3.4 X � X � � � �Diarrhea 3.4 X X X � � � �Cough 3.4 X � X X X X XNasal discharge 3.4 X X X X � � XRash or blister 3.4 X � X X � � XChills 3.4 X � � � � � �Hearing loss 3.4 X � � � � � �
Afghanistan
Rapid loss of consciousness 47.9 Xb X � � X X �Skin irritation and itching 31.5 X � X X � � XMultiple deaths 30.1 X X X � X � �Nausea 20.5 X X X � � X XVomiting 19.1 X X X � � � XTearing 17.8 X X X X X � XDizziness and vertigo 16.4 X X � � X X XBlisters or rash 15 X � X X � � XDifficulty breathing 13.7 X X X X X X XParalysis 13.7 X X � � X � �Headache 12.3 X X � X � X XTemporary blindness 8.2 X � X X � X XSalivation 6.8 Xc X � � � � �Loss of appetite 6.8 X X X � � � �Effects on vegetation 5.5 X � � � � � �Fatigue 5 X X � � � X �Confusion 4.1 X X � � � X �Hemorrhage 4.1 X � � Xa � � �Change in skin color 2.8 X � � X � � �Diarrhea 2.8 X X X � � � �Coughing 1.3 X X X X X X X
a Bloody frothing.b Only at very high doses.c Depending on which trichothecenes.d Blood flecked frothing.
25
cal attack site in Kampuchea as the leafand stem sample previously examined. Thesecond sample came from the site a�yellow rain� attack occurring on March13, 1981, in the village of Muong Cha (TF9797) in the Phou Bia region of Laos. Theagent was sprayed from a twin-enginepropeller aircraft at about noon, local time.The falling substance was described as�like insect spray� and sounded likedrizzling rain. Quite sticky at first, it soondried to a powder. Symptoms described byvictims include nausea, vomiting, anddiarrhea. A sample of the agent scrapedfrom the surface of a rock by a victim andcarried in to Thailand was turned over toU.S. Embassy personnel. The third samplewas taken from the site of a �yellow rain�attack that occurred at 2:00 p.m. on April2, 1981, at Ban Thong Hak (T 9177).Twenty-four people reportedly died in thisattack; there were 47 survivors. Symptomsincluded severe skin irritation and rash,nausea, vomiting, and bloody diarrhea. Asurvivor of the attack scraped this samplefrom the surface of a rock with a bambooknife. Although the individual tookprecaution (that is, cloth mask), a severeskin rash and blisters developed.
These three samples were submitted toDr. Mirocha for analysis. The water samplefrom Kampuchea contained 66 ppm ofdeoxynivalenol and a trace amount ofdiacetoxyscirpenol. A trace quantity of thesecond sample was screened as strongpositive for trichothecenes. Furtheranalysis of that sample confirmed thepresence of high levels of T-2 toxin (150ppm) and diacetoxyscirpenol (25 ppm).Nivalenol and deoxynivalenol may also bepresent but are being masked by interfer-ence from phtalate compounds (leachedfrom the plastic packaging). An effort tomodify the extraction process is beingmade in order to overcome the interferenceso that nivalenol and deoxynivalenol canbe measured more easily. Interestingly,examination of the petroleum ether fractionfrom the sample revealed the presence of ayellow pigment almost identical to thatpreviously identified by Dr. Mirocha incultures of Fusarium roseum, indicatingthat the yellow powder probably consistedof the crude extract of a Fusarium culture.
There was little of the third samplecontained in the vial received for testing.The quantity was too small to be weighedaccurately, and inspection of the vialrevealed only a small speck estimatedto weigh much less than 0.1 mg. Thatspeck contained 10 ng of diacet-oxyscirpenol, a level equivalent to
100 ppm at the very least and probably muchhigher. The sample size was too small toallow adequate analysis for the other threetrichothecenes of interest.
These results support the hypothesis thattrichothecenes have been used as chemicalwarfare agents in Laos and Kampuchea. Thepresence of these high levels of trichothecenetoxins in water and in yellow powder scrapedfrom rocks argues against natural occurrence,since neither water nor rock is a suitableenvironment for growth of the fungi requiredto produce the toxins.
Differences between the analyses of theKampuchean leaf and stem sample and thewater sample collected from the same attacksite raise additional questions. Failure to findT-2 toxin in the water sample is probably dueto the relative insolubility of T-2 toxin inwater. The presence of diacetoxyscirpenol inthe water might be the result of biotransfor-mation or breakdown of T-2, as they are sostructurally similar, differing only in thesubstitution on carbon 8. While this hypoth-esis cannot be entirely ruled out, it is unlikelyon the basis of known biotransformation ofT-2 in the laboratory. The initial vegetationsample was not screened fordiacetoxyscirpenol, although the mass spectrafrom the initial analysis will be reexaminedfor trace amounts of it.
The absence of nivalenol in the watersample is more difficult to explain becausenivalenol is water soluble. The effect ofenvironmental conditions and microorgan-isms on the stability of these compounds mayvary widely for each of the specific com-pounds and may explain the analyticalresults. Further scientific investigation ofthese factors is needed.
Analysis of Blood Samples FromChemical Attack Victims
Blood samples drawn from. victims of recentchemical attacks in Kampuchea have beenreceived by the U.S. Army Medical Intelli-gence and Information Agency for analysisfor indications of trichothecene exposure.Little is known concerning the rate ofmetabolism of trichothecenes in humans; it isdifficult, therefore, to estimate the probabilityof detecting trichothecenes or their metabo-lites in blood samples. T-2 is rapidly clearedfrom the blood in animals, and 25% of thetotal dose is excreted within 24 hours afterexposure; it is unlikely that trichothecenescould be detected unless blood samples wereobtained within 24-48 hours after an attack.Other blood parameters are affected by
the trichothecenes, however, and may proveto be useful markers. The trichothecenesinduce a severe leukopenia (decrease inwhite cell count which can persist for severalweeks following exposure. In addition, thetrichothecenes affect some liver and kidneyfunction marker enzymes which can bemonitored in the blood.
On October 11, 1981, four whole bloodsamples and four blood smears werereceived from the U.S. Embassy Bangkok.The blood was drawn from four KhmerRouge soldiers on October 7, 1981 at aKhmer Rouge hospital inside Kampuchea.Detailed medical histories as well asdescriptions of the attack were recorded oneach individual from whom a blood samplewas taken. All four men were victims of agas attack occurring near Takong onSeptember 19, 1981. Symptoms experiencedincluded vomiting, blurred vision bloodydiarrhea, difficult breathing, dry throat, lossof consciousness, frontal headache, tachycar-dia, and facial edema. Unfortunately, thesamples could not be refrigerated until 48hours after collection. Thus, it was impos-sible to obtain data concerning white cellcounts and blood chemistry. The four wholeblood samples were submitted to Dr.Mirocha for analysis for trichothecenemetabolites because of the possibility,admittedly remote, that some of themetabolites might bind to blood proteins andmight still be detectable even 3 weeks after aattack.
On October 22, 1981, additional bloodsamples were received. These ha( beendrawn from nine victims from the September19 attack and from four control individualsof similar age and background who had notbeen exposed to a chemical attack. Thesamples had been properly refrigerated andwere accompanied by complete and detailedmedical histories taken by trained medicalpersonnel who examined the individuals.Included in the package were blood smearsand heparinized and nonheparinized,samples from each individual. The sampleswere submitted for blood assay to the U.S.Army Medical Research Institute ofInfectious Diseases.
The above results show no statisticallysignificant differences between exposed andcontrol groups (students T-test). In eightindividuals exposed to chemical agent, atrend toward depressed white cell countswas observed. Such an observation would be compatible with the clinical pictureof toxin exposure; however, it is alsocompatible with a number of other medicalproblems, and a larger control sample would
26
be required before such results could beadequately interpreted. Abnormal liver andkidney functions were not indicated by thesedata.
Portions of the blood samples wereanalyzed by Dr. Mirocha for the presence oftrichothecenes and/or trichothecene metabo-lites. The results of the analyses are consis-tent with trichothecene exposure in at leasttwo of the gassing victims and tend tosupport the hypothesis that atrichothecene-based agent was used in thisattack.
Using the selected ion-monitoring gaschromatography/mass spectroscopy analysistechnique, Dr. Mirocha was able to identifytentatively a metabolite of T-2 toxin (that is,HT-2) in the blood of two alleged victims.The compound was identified on the basis ofits selected ion masses and gas chromato-graphic retention times.
The tentative identification of HT-2 inthe blood of two victims, and the trendtoward depressed white cell counts in thesesame victims, cannot be taken as conclusivescientific proof of toxin exposure because thetrace amount of the compound presentprecluded unequivocal identification andquantification and because many othermedical problems in addition to toxinexposure can cause a decrease in white cellcounts. It is interesting to note that theindividual who showed the greatest amountof the compound tentatively identified asHT-2 in his blood reportedly received thegreatest exposure to the agent. He wasexposed to contaminated water for more than30 minutes and was the only victim who felldown in the water and actually swallowedsome of it. However, the description byvictims of symptoms correlating exactly withthose associated with trichothecene poisoningprovides strong circumstantial evidence thattrichothecenes were used as chemical agentsin yet another chemical attack in SoutheastAsia.
Trichothecenes have been identifiedpreviously in environmental samples takenfrom several other chemical attacks in Laosand Kampuchea. Analysis of controlvegetation, water, soil, corn and rice samplesfrom these areas, as well as reviews ofpublished scientific literature, indicates thatthe particular toxins that have previouslybeen identified are not known to occurnaturally in the combinations found and atthe levels detected in Southeast Asia. Thelatest analysis results contribute anotherpiece of evidence to the growing body ofdata supporting the charge thattrichothecenes have been used as chemical/biological agents in Southeast Asia.
ANNEX E
OVERVIEW OF NATURAL OCCUR-RENCE AND SIGNIFICANT PROPERTIESOF TRICHOTHECENESHistorical Trichothecene MycotoxicosesThe trichothecenes are members of a largegroup of naturally occurring toxins known asmycotoxins. The word �mycotoxin� isderived from the Greek �mykes� meaningfungus and the Latin �toxicum� meaningpoison. It refers to a metabolite produced by amold that is toxic to man and animals.Mycotoxicoses have been described as the�neglected diseases,� and before 1960English language literature concerning thediseases caused by mycotoxins was scarce.Soviet scientists have been involved inresearch with some of these compounds foralmost 30 years longer than their Westerncounterparts. The Soviet Union has hadserious problems with mycotoxin contamina-tion of food and has suffered several severeoutbreaks of disease in humans. The firstcomprehensive studies of mycotoxin diseaseswere conducted in the Soviet Union in thelate 1930s.
Since the 1940s, the group of mycotoxinsfiguring most prominently in Soviet scientificliterature are the trichothecenes, a class ofchemically related, biologically active fungalmetabolites produced primarily by variousspecies of Fusarium. Table E-1 lists some ofthe toxins in this group and producing fungi.The fungi are well-known plant pathogensthat frequently invade many agriculturalproducts.
Trichothecene toxins, perhaps more thanany other mycotoxins, have been associatedwith acute disease in humans. Most of thehuman intoxications have occurred in theSoviet Union (Table E-2). The earliestrecognized outbreak occurred in 1891 in theUssuri district of eastern Siberia. Humanswho consumed contaminated grain exhibitedheadache, chills, nausea, vomiting, vertigo,and visual disturbances. Dogs, horses, pigs,and domestic fowls reportedly were affected.
The most extensive mycotoxicosisoutbreak reported to have caused multi-ple fatalities in man also occurred in theSoviet Union. In 1944, 30% of thepopulation of Orenburg district, nearSiberia, was affected by alimentary toxicaleukia (ATA), a disease later shown tobe caused by ingestion of trichothecenetoxins. More than 10% of the entire
population of the district died of the disease.Many other outbreaks of ATA occurred in theSoviet Union, mainly during the 1942-47period. The contamination was traced tooverwintered millet, wheat, and barleyinfected with Fusarium. Symptoms of thedisease included vomiting, skin inflamma-tion, multiple hemorrhaging (especially of thelung and gastrointestinal tissue), diarrhea,leukopenia, and suppression of bone marrowactivity.
In 1939, Premier Joseph Stalin dis-patched Nikita Khrushchev to the Ukraine toorganize and improve agricultural operationsand to identify the disease causing the deathsof many horses and cattle. The problem wastraced to hay and straw contaminated withStachybotrys atra. The disease, later referredto as stachybotryotoxicosis, occurred afteringestion or contact with the contaminatedgrain. Symptoms included ulcerativedermatitis, peroral dermatitis, blooddyserasias, hemorrhagic syndromes, abortion,and death. The greatest economic impact wasdue to loss of horses, although cattle, sheep,poultry, and humans also were affected.
Other disease outbreaks in whichsimilar symptoms were present occurredin 1958 and 1959 among horses and cat-tle in the Soviet Union and EasternEurope; thousands of animals were lost.Other intoxications were reported later
Soviet Scientists Involved inMycotoxin Research
A. Kh. Sarkisov�All Union Scientific Research Institute of Experimental VeterinaryScience, MoscowV. I. Bilay (also spelled Bilai)�UkrainianS.S.R. Institute of Microbiology andVirology, KievV. A. Tutel�yan�U.S.S.R. Academy of MedicalSciences Nutrition Institute, MoscowM. A. Akhmeteli�U.S.S.R. Academy ofMedical Sciences Institute of Epidemiologyand MicrobiologyL. Ye. Olifson A. M. KoganM. F. Nesterin D. T. MartynenkoK. Z. Salomatina N. A. KostyuninaYe. P. Kozhevnikova V. V. YerinakovN. D. 0sadchaya I. A. KurmanovL. F. Mikhaylova V. V. SemenovSh. M. Kenina Z. K. BystryakovaV. L. Kartashova Z. Z. OrlovaL. R. Filonova L. S. L�vovaT. Ye. Tolcheyeva L. I. LozbinaKn. A. Dzhilavyan T. A. ShevtsovaI. S. Yelistratov I. Yu. MakedonN. S. Tishkova N. S. ProskuryakovaV. I. Kaplun A. V. BorovkovYe. P. Kozhevalkova M. N. NazypovS. M. Gubkin L. I. LozbinL. I. Il�ina M. S. MarovaP. A. Il�in
27
TABLE E-1Trichothecene-Producing Fungi
Type T-2 Type Nivalenol-Type MacrocyclicTrichothecenes T-2 Toxin Nivalenol Roridins
HT-2 Toxin Monoacetyl- VeirucarinsNivalenol
Diacetoxyscirpenol Diacetyl-Nivalenol SatratoxinsNeosolaniol Deoxynivalenol Vertisporin
Fungus F. tricincturn F. nivale Myrotheciumverrucaria
F. roseum F. opisphaeriaM. roridum
F. equiseti F. roseumF. sporotrichioides Stachybotrys atra
Verticimonosporiumdiffractum
F. lateritiumF. poaeF. solaniF. rigidiusculumF. sernitectum
TABLE E-2Historical Trichothecene Mycotoxicoses
Districts andToxicosis Affected Species Symptoms
�Taumelgetreide� Toxicosis U.S.S.R.: man, farm animals Headache, nausea, vomiting,vertigo, chills, visualdisturbances
Alimentary toxic aleukia U.S.S.R.: man, horse, pig Vomiting, diarrhea, multiplehemorrhage, skininflammation, leukopenia,angina
Stachybotryotoxicosis U.S.S.R., Europe: horse Shock, stomatitis, hemorrhage,dermal necrosis, nervousdisorders
Bean-hull toxicosis Japan: horse Convulsion, cyclic movement
Dendrodochiotoxicosis U.S.S.R., Europe: horse Skin inflammation, hemorrhage
Moldy corn toxicosis United States: pig, cow Emesis, hemorrhage
Red mold toxicosis Japan, U.S.S.R.: man, horse, Vomiting, diarrhea, congestionpig, cow and hemorrhage of lung and
intestine
in Japan, Europe, the Soviet Union, and theUnited States, affecting various domesticanimals and�in the case of �red moldtoxicosis��man. All of these diseases havenow been shown to be due to ingestion oftrichothecenes rather than to an infectiousagent. In earlier outbreaks, the levels oftoxin present in the contaminated grainwere not measured; however, the levels ofnivalenol and/or deoxynivalenol measuredin toxic grains implicated in more recentoutbreaks (i.e., �moldy corn toxicosis� and�red mold toxicosis�) typically werebetween 2 and 8 ppm.
Natural Occurrence of TrichotheceneMycotoxinsPublications concerning the occurrence oftrichothecenes are relatively scarce becauseof the lack of convenient detection methodsand the complexity of the trichothecenefamily of compounds. Only recently havescientists developed methods capable ofdistinguishing between close structuralderivatives and accurately quantifying thelevels of toxin present (see Table E-3 forcomparison of analytical methods).Extreme care must be taken when review-ing the scientific literature on naturaloccurrence of these compounds becauseerroneous conclusions can be drawn on thebasis of results obtained with inadequateanalytical techniques. Misidentification ofcompounds and gross over-estimation ofconcentrations have occurred usingtechniques such as thin layer chromatogra-phy.
Table E-4 lists the reports of naturaloccurrence of T-2 toxin,diacetoxyscirpenol, and nivalenol that wereobtained from a literature search of morethan 3,000 citations concerned withtrichothecene toxins. Levels that arequestionable on the basis of techniquesused are indicated. It is immediatelyapparent that the levels of toxins found inthe various samples from Laos andKampuchea are highly unusual, even if oneaccepts the questionable reports in TableE-4 as valid. The levels of these toxins (150ppm of T-2 toxin, 109 ppm of nivalenol,more than 100 ppm of diacetoxyscirpenol,and 66 ppm of deoxynivalenol) aremarkedly higher than those reported tooccur in nature. It should also be noted thatthe incidences recorded in Table E-4concern levels of toxin produced whenFusarium is growing on its ideal substrate,while the Laos
and Kampuchea samples were takenfrom surfaces�rocks and water�thatwould be extremely unlikely to supportFusaria growth and toxin production.Higher levels of toxin production can, ofcourse, be induced when the moldspecies is grown in pure culture underideal laboratory conditions; for instance,the Soviets have succeeded in producing4 grams of T-2 per kilogram of sub-
strate. In a natural environment, however,the Fusaria species cannot compete wellwith other molds such as species ofAspergillus and Penicillium, and levels oftoxin produced are orders of magnitudelower.
The conclusion that the levels of toxinsfound in the Southeast Asia samples couldhave occurred only by means of anunnatural mechanism is also strength-ened by surveys of the area conducted
28
FIGURE E-1
General Structure of Trichothecenes
T2 Toxin Nivalenol Deoxynivalenol
R1=OH R
1=CH R
1=OH
R2=OA
cR
2=CH R
2=H
R3=OA
cR
3=OH R
3=OH
R4=H R
4=OH R
4=OH
R5=OCOCH
2CH(CH
3)
2R
5= =O R
5= =O
by various researchers. Surveys of thetoxigenic fungi and mycotoxins naturallypresent in Southeast Asia conducted by theMahidol University in Bangkok and theMassachusetts Institute of Technology havenot revealed the presence of T-2, nivalenol,deoxynivalenol, or diacetoxyscirpenol,although other mycotoxins such as aflatoxinwere identified. These results were confirmedby our analysis, using our own methodology,of normal flora samples of vegetation, soil,water , corn, and rice from Kampuchea thatfailed to reveal the presence oftrichothecenes.
Skeptics have formulated theoreticalexplanations for the analytical results tosupport a hypothesis of natural occurrence ofthese toxins. It was postulated that thetrichothecenes found were absorbed throughthe roots of a plant, translocated. to theleaves, and exuded and washed onto thesurface of a rock and into water where theywere found. A 1981 publication by Jarvis etal. reported a Brazilian shrub that appeared toabsorb, translocate, and chemically alter amacrocyclic trichothecene produced by soilfungi. While this citation is used to support ahypothetical mode for natural deposition inSoutheast Asia, it should be noted that theplant reported in this publication did notexude the toxin, that the toxin was extremelyphytotoxic to all other plants assessed, andthat the plant was not capable of de novotrichothecene synthesis. No othertrichothecenes have been found to beabsorbed and translocated in any other plantin this manner. Control samples of soil andvegetation from Southeast Asia do notsupport endemic presence of these toxins.The appearance of these particulartrichothecene toxins in these high levels inenvironments generally inhospitable to theirformation cannot reasonably be attributed toa natural contamination.
Chemical and Physical Properties of theTrichothecenesWhen considering the suitability oftrichothecenes as agents, factors such asstability, solubility, and ease of productionmust be considered. The general structure forthe trichothecene group is shown in FigureE-1. There are more than 40 currently known,naturally occurring, 12 to 13epoxytrichothecenes. The R groups may behydroxyls, acylated hydroxyl groups oresters. The R group for the toxins detected inthe sample is shown below the generalstructure. All of the compounds have incommon an olefinic double bond at car-
bon atoms 9 and 10 and an epoxy group atcarbon atoms 12 and 13. These compoundsare stable, especially in the solid form. Theymay be stored for years at room temperaturewith no loss of activity. They are heat stablewith no loss of activity noted after heating for1 hour at 100° centigrade. The solubilitydepends on the R groups; highly hydroxy-lated derivatives are more water soluble. Thecompounds are also quite stable in solution.Detoxification can be accomplished
by treatment with strong mineral acid,which will open the 12 to 13 epoxide bondand abolish all biological activity. Most ofthe toxins are well absorbed throughmucous membranes and some through skin;this property is also a function of the Rgroup.
Some of these compounds have beensynthesized chemically; however, biosyn-thesis employing Fusarium species is themost effective way to produce largequantities. In a preliminary search of re-cent Soviet literature, 50 articles dealing
29
TABLE E-3Physicochernical Methods for Detection of Trichothecenes in Feedstuffs
Trichothecenes Detection Required Use andMethod Detected Limits Standards Limitation
Thin-layer All 0.1 microgram/spot Reference Qualitativechromatography Standard Interference1 -dimension Not confirmatoryThin-layer All 0.1-1.0 microgram/ Reference Qualitativechromatography spot (H
2SO
4) Standard Less interference
2-dimension ConfirmatoryGas-liquid Nonhydroxy- 0.03-0.05 Reference Quantitativechromatography lated or TMS microgram/ Standard Monoglyceride
derivatives microliter injection interferenceEquivocable
identificationGas chromato- TMS 0.02-0.05 Reference Serniquantitativegraphy/mass derivatives microgram/ Standard or Less interferencespectrometry- microliter Spectrogram Unequivocablenormal scanning injection identificationmodeGas chromato- TMS 0.007-0.02 Reference Quantitativegraphy/mass derivatives microgram/ Standard or Best for complexspectrometry- microliter Spectrogram mixturesselection ion injection Unequivocablemonitoring identificationNuclear- All Reference Confirmatorymagnetic- Standard or Purified toxinresonance Spectrogram structure
elucidationRadio- T-2 toxin 1-20 nanogram Rabbit anti- Sensitiveimmunoassay T-2 toxin Low interference(developmental anti-bodystage)
HT-2 toxin Relative structuralspecificity
with the trichothecenes were reviewed. Ofthese, 22 dealt with defining optimumconditions for biosynthesis of the com-pounds. N.A. Kostyunina has reportedproduction of T-2 toxin at levels of 4 gramsper kilogram of substrate (normally wheatgrain or rice). Many industrial microbiologyplants have been identified in the SovietUnion. Some are involved in production ofsingle-cell protein for fodder additives, othersproduce antibiotics, and the function of stillothers is unknown. Fusaria are produced inthe Soviet Union at a facility long reported inthe open literature as being a suspectedbiological warfare agent production andstorage facility. This facility, Berdsk Chemi-cal Works, is near the science city ofNovosibirsk in Siberia. The only differencebetween an antibiotic and mycotoxin is theirtarget specificity. Both are produced by fungi,but the mycotoxins are relatively more
toxic to man than to microorganisms.Mycotoxins can be produced in good yieldemploying the same techniques used toproduce some antibiotics. Thus it may beconcluded that the Soviets could producetrichothecenes in large amounts. Theyproduce an antibiotic that is a trichothecenederivative, which would provide an idealcover for agent production facilities.
Medical Effects of the Trichothecenes inHumansThe most prominent symptoms associatedwith trichothecene poisoning are listed inTable E-2. Striking among these is the rapidonset of vomiting, along with severe itchingand tingling of the skin. Hemorrhage of themucous membranes and bloody/diarrheafollow, The symptoms shown in Table E-2are similar to those reported by victimsof trichothecene attacks in Laos, Kampu-
chea, and Afghanistan. The correlation isstriking.
The LD50
�S (dose required to producedeath in 50% of a test population) of thetrichothecenes in laboratory animals rangefrom 0.1 mg/kg to greater than 1,000 mg/kg,depending on the particular toxin, species,and route of exposure. The LD
50 of T-2 toxin
in a cat is 0.5 mg/kg. However, the ED50(dose required to produce a desired physi-
ological effect in 50% of a test population) ismuch lower. The ED
50 to produce a vomiting
reaction is 0.1 mg/kg; for skin irritation it isin the tenths of microgram range.
Most of the data concerning thetoxicological effects of the trichothecenesare derived from animal data in which purecompounds were administered by oral,subcutaneous, intraperitoneal, or intravenousroutes. Unfortunately, there are no reportsconcerning the effects of inhalation ofmixtures of the compounds. Therefore, it isdifficult to speculate concerning the effectsthat would be expected in humans exposedto an aerosol of mixtures of these potenttoxins. The most useful data concerningexposure in humans were obtained in aphase I clinical evaluation of anguidine(diacetoxyscirpenol) as an anticancer drug.Diacetoxyscirpenol was administered byintravenous infusion. Doses of 3 mg/m2/ daycaused immediate onset of nausea, vomiting,diarrhea, somnolence and/or mentalconfusion, fever, chills, a generalizederythema with a burning sensation, hypoten-sion, dyspnea, stomatitis, hives, and ataxia.Because of the side effects, the treatmentwas discontinued. The properties whichmake the use of diacetoxyscirpenol poten-tially useful as an anticancer drug are thesame as those responsible, in part, for itsextreme toxicity. It and the othertrichothecenes cause extensive damage torapidly dividing cells such as tumor cells.Unfortunately, the cells of the lining of thegastrointestinal tract and bone marrow arealso rapidly dividing, and the effects of thetrichothecenes on these cells result in severe,rapid degeneration of these tissues. Thecompounds also have direct effects on theclotting factors in the blood (that is, aprimary effect on Factor VII activity and asecondary effect on prothrombin), whichresult in excessive hemorrhage followingtrauma.
The other useful body of clinical dataconcerning the effects of trichothecenes inhumans is drawn from descriptions of thecourse of the disease in the natural
30
TABLE E-4Spontaneous Occurrence of Trichothecene Mycotoxins
Concentration(parts Per
Toxin Country Source million) Referencea FOOTNOTES
T-2 Toxin U.S. Mixed feed 0.08b 15 a References:U.K. Brewer�s grains NDc 19India Sweet corn 4b, d 5 1. Balzer et al. (1977)Canada Corn ND 4 2. Ciegler (1978)India Sorghum NDd 22 3. Eppley et al. (1974)Canada Barley 25d 20 4. Funnel (1979)India Safflower seed 3-5d 6 5. Ghosal et al. (1978)U.S. Corn stalks 0.1lb 16 6. Ghosal et al. (1977)U.S. Feed supplement ND 7 7. Hibbs et a/. (11974)U.S. Corn 2 8 8. Hsu et al. (1972)U.S. Mixed feed 0.3 14 9. Isshi et al. (1975)France Corn 0.02b 10 10. Jemmail et al. (1978)U.S. Corn ND 2 11. Marasas et al. (1977)
12. Miller (1976) 13. Mirocha (1979)
Diacetoxy- U.S. Mixed feed 0.5 15 14. Mirocha (1979)scirpenol U.S. Mixed feed 0.38 15 15. Mirocha et al. (11976)
India Safflower seed 3-5d 6 16. Mirocha et al. (1979)India Sweet corn 14d 5 17. Mirocha et al. (1979)Germany Corn 31.5d 23 18. Morooka et al. (1972)
U.S. Corn 0.88 21 19. Petrie et al. (1977)20. Puls and Greenway et al. (1976)
Deoxynivalenol U.S. Corn stalks 1.5b 16 21. Romer, T., Ralston Purina,U.S. Corn 1.8b 15 St. Louis, MO (personalU.S. Corn 1.0b 15 communication)U.S. Corn 0.1b 15 22. Rukmini and Bhat (1978)U.S. Mixed feed 0.04b 15 23. Siegfried (1979)U.S. Mixed feed 1.0b 15 24. Vesonder and Ciegler (1979)U.S. Mixed feed 1.0b 15 25. Vesonder et al. (1976)U.S. Corn 7.4 9 26. Vesonder et al. (1978)U.S. Corn 0. 1 -25d 21 b Zearalenone (F-2 toxins) also detectedU.S. Corn trace-25d 2,21 in the sample.U.S. Corn 1.1-10.7 26 c N D - toxin concentration was not deterU.S. Corn 4125 mined.
1.0b d Levels that are questionable on theU.S. Corn 17 basis of techniques used.U.S. Oats 5 17Japan Barley ND 18U.S. Corn 1.0b 13U.S. Corn 0.06b 13U.S. Mixed feed 0.07b 13France Corn 0.6b 10South Africa Corn 2.5 11Zambia Corn 7.4 11U.S. Corn N D 2Japan Barley 7.3 18Austria Corn 1.3 24Austria Corn 7.9 24Canada Corn 7.9 24
Nivalenol Japan Barley ND 18France Corn 4.3b 10
Partially U.S. Corn ND 25characterized India Safflower seed NDd 6trichothecenesSkin irritant U.S. Corn 93 positiveb 3factors�not of 173analyzed U.S. Corn Multiple positive 21chemically samples
Yugoslavia Corn 16 positive 1
United States Department of StateWashington, D.C. 20520 31
Bureau of Public AffairsUnited States Department of StateWashington, D.C. 20520
Official Business
If address is incorrectplease indicate change.Do not cover or destroythis address label.
Postage and Fees PaidDepartment of State
STA-501
outbreaks that occurred in the Soviet Union.The effects produced are similar to radiationpoisoning, and there is a latent phase similarto that seen in radiation poisoning, in whichthe overt symptoms disappear.
The clinical picture may be divided intofour stages.
The first stage occurs within minutes tohours after ingestion of toxic grains. Thesymptomatology described was produced byoral exposure to low doses. In exposure byinhalation, the symptoms may be morepronounced or the time course accelerated.The characteristics of the first stage includeprimary changes, with local symptoms, in thebuccal cavity and gastrointestinal tract.Shortly after ingestion of toxic grain, thepatient experiences a burning sensation in themouth, tongue, throat, palate, esophagus, andstomach as a result of the toxin�s effect on themucous membranes. The tongue may feelswollen and stiff, and the mucosa of the oralcavity may be hyperemic. Inflammation ofthe gastric and intestinal mucosa occurs,along with vomiting, diarrhea, and abdominalpain. In most cases excessive salivation,headache, dizziness, weakness, fatigue, andtachycardia accompany the initial stage.There may be fever and sweating, but
the body temperature normally does not rise.The leukocyte count may begin to decrease inthis stage, and there may be an increasederythrocyte sedimentation rate. This firststage may last from 3 to 9 days.
The second stage is often called thelatent stage or incubation period because thepatient feels well and is capable of normalactivity. It is also called the leukopenic stagebecause its main features are disturbances inthe bone marrow and the hematopoieticsystem, characterized by a progressiveleukopenia and granulopenia and a relativelymphocytosis. In addition, anemia and adecrease in erythrocytes, in the platelet count,and in hemoglobin occur. Disturbances in thecentral nervous system and autonomicnervous systems may occur as well asweakness, vertigo, fatigue, headache,palpitations, and mild asthmatic conditions.Visible hemorrhagic spots (petechiae) beginto appear on the skin, marking the transitionto the third phase. The second stage may last3-4 weeks. The transition to the third stage issudden, and symptoms progress rapidly.
In the third stage, petechial hemor-rhages occur on the skin of the trunk, arms,thighs, face, and head. They can vary from amillimeter to a few centimeters in size.Capillaries are fragile, and any slight traumaresults in hemor-
rhage. Hemorrhages of membranes of themouth, tongue, soft palate, and tonsils occur.Nasal, gastric, and intestinal hemorrhages canbe severe. Areas of necrosis begin to appearon the lips, fingers, nose, jaws, eyes, and inthe mouth. Lymph nodes are frequentlyenlarged, and the adjoining connective tissuecan become so edematous that the patient hasdifficulty opening his mouth. Blood abnor-malities previously described are intensified.Death may occur from hemorrhage, strangu-lation due to swelling, or secondary infection.
The fourth stage is convalescence.Three or 4 weeks of treatment are requiredfor disappearance of necrotic lesions andhemorrhagic effects. Two months or moremay elapse before the bloodformingcapability of the bone marrow returns tonormal.
Published by the United States Department of State� Bureau of Public Affairs � Office of PublicCommunication � Editorial Division � Washington,D.C. � March 1982 Editors: Norman Howard andColleen Sussman � This material is in the publicdomain and may be reproduced without permis-sion; citation of this source is appreciated.
Related Documents
