APPENDIX B Hazard Identification
In reviewing possible causes of Gulf War veterans health concerns, the Department of Defense (DoD) identified a possible low-level exposure to the chemical warfare agents sarin (GB) and cyclosarin (GF). This Appendix reviews and evaluates the information and assesses the human risk from this possible low-level chemical warfare agent exposure.
US servicemembers from the 37th Engineer Battalion destroyed munitions in bunkers and an adjacent pit at the Khamisiyah Ammunition Storage Depot in southern Iraq on March 10, 1991. DoD subsequently obtained and reviewed evidence indicating 122mm chemical warfare agent-filled rockets containing GB and GF were destroyed or damaged there. US forces proximity to and the prevailing winds at the demolition site indicated a chemical warfare agent exposure may have occurred; in 1997, DoD and the Central Intelligence Agency (CIA) identified this release as a priority for further study (CIA and DoD, 1997; OASD(PA), 1997).
US servicemembers detected no chemical warfare agents before, during, or immediately after the demolition. The nearest US forces were approximately three miles from the rockets when they were destroyed. Forces in the area at the demolition reported no illnesses, signs, or symptoms consistent with chemical warfare agent exposure. Medical personnel near Khamisiyah found no evidence of health effects related to nerve agent exposure (Perrotta, 1996). However, modeling results suggest some forces may have been exposed to very low levels of sarin and cyclosarin too low to activate chemical alarms or cause any noticeable symptoms (e.g., pinpointed pupils (miosis), runny nose, or tightness in the chest (CIA and DoD, 1997)).
In response to concerns about forces possible exposure to low levels of chemical warfare agent, this toxicity assessment and risk evaluation review toxicological information about sarin and cyclosarin and discuss the anticipated effects and possible risks from low-level exposure to these agents.
II. SARINS TOXICOLOGICAL PROFILE
Sarin (GB; isopropyl methylphosphonofluoridate, CH3PO(F)OCH(CH3)2) is an acutely toxic, relatively nonpersistent nerve agent developed in Germany during World War II and subsequently produced and tested in the United States. More recently the agent was used in terrorist attacks in an apartment complex in Matsumoto, Japan, in 1994 and in a Tokyo subway in 1995 (Sidell and Hurst, 1997). When pure, sarin is a colorless, odorless liquid whose vapor is colorless as well (US Army ERDEC, 1994). Tab B-I to this appendix lists these agents general chemical properties. Although data on chronic (i.e., long-term) human exposure is lacking, considerable acute (i.e., short-term, less than 24 hours) and animal chronic studies are available from which to infer the possible long-term effects.
A. Fate and Transport
Sarin is volatile, having a vapor pressure of 2.9 mm Hg at 25� C. Most materials readily absorb GB vapors; however, its high volatility results in almost complete desorption a few hours after removal from the source. Sarin also undergoes hydrolysis at a rate depending on the mediums temperature and pH, with a minimum rate between pH 4 and pH 6 (MacNaughton and Brewer, 1994). Hydrolyzing sarin initially produces two less toxic products, hydrogen fluoride and methyl phosphonic acid (Rosenblatt, 1996).
GB evaporates at about the same rate as water and undergoes fairly rapid loss from unconfined soils by evaporation and hydrolysis (Rosenblatt, 1996). Once released as a liquid into the soil in an arid environment such as that at Khamisiyah, scientists expect sarin to persist for several days (three or fewer, but probably in amounts smaller than would cause symptoms after one day). Any hydrolysis of GB that occurred would further reduce the vapors potential to migrate and/or evaporate (MacNaughton and Brewer, 1994).
B. Acute Effects
Similar to all G-agents, sarin exerts its effects by inhibiting the enzyme acetylcholinesterase (AChE), required for proper nerve and muscle function (Perrotta, 1996; Taylor, 1990a). Acetylcholine (ACh) is the neurotransmitter at postganglionic nerve endings in the parasympathetic nervous system, between preganglionic and postganglionic fibers of both sympathetic and parasympathetic neurons, at myoneuronal junctions of striated muscle (somatic nerves) and cholinergic synapses in the central nervous system (Sidell, 1992; Sidell and Hurst, 1997). AChE prevents ACh from accumulating after its release as a chemical messenger at the nervous systems cholinergic synaptic junctions (Perrotta, 1996; Lefkowitz et al., 1990; Sidell, 1992). Like most other nerve agents (e.g., soman and tabun), GB is a highly toxic organophosphate that binds irreversibly to AChE within an hour or less of their reaction (Grob and Harvey, 1957). Consequently, ACh accumulates at synapses depending on cholinergic transmission; also produces enhanced, continuous stimulation; and ultimately causes the loss of functions this cholinergic system mediates. Thus, AChE inhibition adversely affects skeletal muscle, parasympathetic end organ, and central nervous system function and interferes with a fundamental mechanism required for the central and peripheral nervous systems to function normally and transmit cholinergic nerve impulses (Perrotta, 1996).
Although the great majority of GBs known effects are due to its anticholinesterase action, nerve agents also may have noncholinergic actions, such as inhibiting enzymes other than cholinesterases and altering other neurotransmitter systems. In high concentrations, GB may directly affect receptors (Somani et al., 1992). The clinical significance of these noncholinergic actions, if any, and the doses at which they may occur are unclear.
Anticholinesterase agents can produce all these effects: (1) stimulation of muscarinic receptor responses at autonomic effector organs; (2) stimulation followed by depression or paralysis of all autonomic ganglia and skeletal muscle (nicotinic actions); and (3) stimulation with occasional subsequent depression of cholinergic receptor sites in the central nervous system (CNS) (Taylor, 1990b). Consequently, GB intoxication signs and symptoms are of three types: muscarinic, nicotinic, and central.
Sarin exposures type and severity of symptoms depend on where the nerve agent enters the body and the magnitude and duration of exposure. When the eyes and nasal passages are exposed to the agent vapor, the earliest signs expected are local effects, specifically miosis and rhinorrhea, rather than systemic effects. Virtually all patients treated after the sarin attack in the Tokyo subway had localized symptoms including miosis and other eye effects, rhinorrhea, and frequently coughing and tightness in the throat (Masuda et al., 1995; Okumura et al., 1996; Yokoyama et al., 1996; Kato and Hamanaka, 1996). However, if a person is exposed through the skin or by ingestion, the pupils may be approximately normal in size and rhinorrhea absent. Diagnosis must then rely on other symptoms (e.g., localized sweating and muscular fasciculations and/or gastrointestinal disturbances (Sidell, 1992)). Decreases in cholinesterases (ChE) in the blood, either red blood cell cholinesterase (RBC-ChE) and plasma (serum) or butyrylcholinesterase (BuChE), are often detectable with very low-level exposures to sarin, particularly chronic. Depending on exposure routes and duration, either ChE decrease may appear before other symptoms (e.g., miosis). However, acute (i.e., short-term) exposures resulting in local symptoms may not always result in ChE depression. Several victims manifesting mild local symptoms (e.g., miosis) after the Tokyo attack did not have detectable ChE decreases (Masuda et al., 1995; Yokoyama et al., 1996; Kato and Hamanaka, 1996). RBC-ChE was a more sensitive predictor than BuChE, but one patient with severe symptoms did not even have reduced BuChE (Masuda et al., 1995; Yokoyama et al., 1996). However, all patients with decreased RBC-ChE also had miosis. No clinical symptoms of nerve agent exposure were detected in forces in the vicinity of the Khamisiyah demolition (Perrotta, 1996). However, RBC-ChE levels were not determined at the time due to lack of even the mildest symptoms associated with nerve agent exposure (e.g., muscarinic effects such as miosis or salivation), and because it was not suspected the munitions contained chemical warfare agents.
At Rocky Mountain Arsenal, Holmes et al. (1959) compared two GB exposure groups signs and symptoms (Group A included 635 cases and Group B, 356 cases) with the percent reduction in ChE. The subgroups analyzed included those with ChE reductions of 0 to 25%, 25 to 40%, 40 to 60%, and over 60% in both Group A and Group B; only Group A included a 0 to 10% subgroup. The authors separately analyzed symptoms and signs of different anatomical systems (e.g., eyes, respiratory tract, gastrointestinal tract, etc.) to gain a clearer picture of the incidence of such symptoms as miosis, salivation, sweating, etc. In cases considered mild, it was uncertain whether the individuals really had been exposed. Therefore, the authors compared cases with miosis to those without. Symptom complexes between the two mild exposure groups did not differ greatly except for a higher incidence of dizziness, headache, weakness, muscle twitching, sweating, and paresthesia (abnormal sensations, e.g. prickling or burning) in the miosis group. The most common symptoms in all groups were miosis, chest constriction, rhinorrhea, breathing difficulties, and coughing, which occurred in more than 50% of the cases in the mild (0 to 10 and 0 to 25%) ChE reduction groups; miosis, chest constriction, and rhinorrhea occurred in these two groups more than 75% of the time. Blurred or dimmed vision also occurred in the mild ChE reduction groups more than 50% of the time. Other common (i.e., 25% or more often) symptoms associated with mild ChE reduction were lacrimation, nausea or anorexia, increased sweating, fatigability, and common cold after exposure. These symptoms duration was commonly from 1 to 7 days, with a few cases (4.4%) persisting for 2 or more weeks. In groups showing 25 to 40% ChE reduction, as many as 41% of the cases persisted for 4 to 7 days and almost 18% persisted for 2 weeks or more. A summary of this report concluded severe GB exposure results in 60% depression of RBC-ChE and 100% exposed will have miosis; if miosis does not occur, the chance RBC-ChE will drop by less than 25% is 85%. Severe intoxication (presumed to mean high-level exposure effects) is expected as a result of ChE depression of more than 60%. The report stated eye symptoms other than miosis generally are absent in people receiving a mild GB exposure.
Respiratory exposure to sarin produces symptoms faster than dermal exposure, within seconds to minutes, depending on vapor concentration. The eyes and nose, the most sensitive organs, usually are the first affected. Death may occur within minutes of exposure to large amounts of vapor (Sidell, 1992; Sidell and Hurst, 1997). The time to onset of symptoms from dermal exposure to GB is delayed and depends on dose, as well as on factors such as location on the body, temperature, and humidity. Typical early effects of vapor exposure (miosis and rhinorrhea) are absent, and local sweating, muscle fasciculations, and gastrointestinal effects may be present. Death from skin absorption may occur from 15 minutes to hours after high-level exposure (Sidell, 1992; Sidell and Hurst, 1997). Sarin is more effective as an inhalant than a skin penetrant due to its rapid evaporation rate (Munro et al., 1994). Sarins effects are cumulative within the time necessary to restore cholinesterase activity, approximately 120 days for RBC-ChE (i.e., it takes red blood cells about four months to recover completely from sarin-induced ChE damage (Sidell, 1997; McNamara and Leitnaker, 1971)). Thus, repeated exposure to low sarin concentrations, if not too far apart (within a few hours, days, or even weeks) can result in symptoms characteristic of one or more effects previously described (Opresko et al., 2001; Somani, et al., 1992; McNamara and Leitnaker, 1971).
In mild nerve agent exposures, primarily muscarinic effects occur and may consist of signs such as miosis, salivation, lacrimation, rhinorrhea, tightness in chest, and gastrointestinal complaints. Except for miosis, these symptoms may be misdiagnosed as a cold or influenza.
The first noticeable effect (FNE) from exposure to low levels of sarin vapor is usually miosis. Dimmed vision is often present, presumably resulting from the reduced amount of light entering the eyes. Blurred vision and eye pain also may occur. At these same low levels, other mild symptoms may include runny nose, chest tightness, slight bronchoconstriction and secretions, and slight dyspnea (Sidell, 1992; McNamara and Leitnaker, 1971; Sidell and Hurst, 1997).
The product of the concentration of a chemical to which an organism is exposed (C) and the length of time of that exposure in minutes (t) is Ct; McNamara and Leitnaker (1971) report a Ct of 0.5 mg-min/m3 as the no-effect level for sarin in humans, at which less than 1% of a working population would show miosis, ChE depression, rhinorrhea, or other mild symptoms. The Ct does not reflect the amount of chemical retained in the lungs and thus systemic absorbtion is not represented in the Ct (Sidell and Hurst, 1997). The 50% effective Ct (ECt50) for miosis and rhinorrhea (time and concentration for half the exposed people to show an effect) is reported as less than 2 mg-min/m3 (Mioduszewski et al., 1998), with an exposure of 3 mg-min/m3 producing miosis in most of the population (Sidell, 1992). Some investigators suggest an estimated exposure of 1 mg-min/m3 is the first noticeable effects level (e.g., runny nose, tightness in the chest, dimmed vision, miosis) (McNamara and Leitnaker, 1971; CIA and DoD, 1997), so this was the FNE level used to investigate possible agent exposure at Khamisiyah. Mild dermal exposure to liquid sarin may increase sweating and muscular fasciculations at the site (Sidell, 1992); however, this review does not consider dermal exposure a significant pathway, as servicemembers exposure would have been to a vapor cloud (CIA and DoD, 1997). Mild dermal exposure to sarin vapor probably would not cause any effect.
In moderate nerve agent poisonings, symptoms may include bradycardia and hypotension, abdominal pain, bradypnea, dyspnea, headache, nausea, vomiting, miosis, diarrhea, generalized weakness, mild shock, perspiration, pallor, tight or painful chest, tachycardia, hypotension, and fasciculations of fine skeletal muscles. Symptoms of moderate exposure to sarin vapor include miosis, rhinorrhea, bronchoconstriction, secretions, and moderate to marked dyspnea (Sidell, 1992). McNamara and Leitnaker (1971) reported no systemic effects below 4.1 mg-min/m3, no deaths below 10 mg-min/m3, and an ECt50 of 20 mg-min/m3 for ChE depression. Effects from moderate dermal exposure to liquid sarin may include increased sweating and muscular fasciculations at the site, nausea, vomiting, diarrhea, and feelings of generalized weakness that may occur after a 4- to 18-hour asymptomatic period (Sidell, 1992).
In severe nerve agent poisonings, symptoms may include marked generalized muscle weakness, muscular twitching, involuntary jerky movements, convulsions, semistupor, coma, respiratory depression or failure, and death. High-level GB exposure may result in any of the symptoms of milder exposures, plus loss of consciousness, convulsions, generalized fasciculations, flaccid paralysis, apnea, and involuntary micturition and defecation (Sidell, 1992). Sarins estimated 50% lethal concentration (LCt50) for a 70-kg man breathing at the rate of 15 liters per minute (l/min) is approximately 70 mg-min/m3 (Mioduszewski et al., 1998), or a dose of approximately 15 µg/kg, assuming sarin is completely absorbed through the lungs. Sarins estimated 50% incapacitation concentration (ICt50) is approximately 35 mg-min/m3 (Mioduszewski et al., 1998). The effects of severe dermal exposure to sarin include those observed at smaller amounts and also may include loss of consciousness, convulsions, generalized fasciculations, flaccid paralysis, apnea, involuntary micturition and defecation, and generalized secretions (Sidell, 1992). The percutaneous 50% lethal dose (LD50) for liquid sarin is estimated at 1700 mg/70 kg man (Mioduszewski et al., 1998).
C. Chronic Effects
The largest obstacle to evaluating the possible health risks from short-term, low-level exposure to sarin is the lack of controlled human studies in which persons were exposed to doses of sarin calculated to avoid any symptoms and then monitored over extended periods. However, scientists can infer the long-term health risks associated with low-level exposure to GB from several studies using doses that caused acute (short-term) symptoms, several unintentional high-dose exposure investigations, and animal studies (Perrotta, 1996). In addition, scientists can draw on the extensive literature on organophosphate pesticides and other anticholinergics, but must interpret the results cautiously, as risks from nerve agents (e.g., ability to induce organophosphorus-induced neuropathy) may differ significantly from those of these other compounds.
The National Center for Toxicological Research conducted a subchronic oral toxicity test of Types I and II sarin in CD-strain rats to obtain a lowest observed adverse effects level (LOAEL) of 0.075 mg/kg/d for statistically significant decreases in RBC-ChE. Researchers administered Type I (GB containing tributylamine) or Type II sarin (GB containing diisopropylcarbodiimide as stabilizers) to the animals by gavage (stomach tube) once per day, 5 days per week for 13 weeks at doses equivalent to 0, 0.075, 0.15, or 0.3 mg/kg/day. RBC-ChE in male rats administered 0.075 mg/kg/day Type II GB was depressed by 38% in Week 1, 23% in Week 3, 37% in Week 7, and 16% in Week 13, with only Weeks 3 and 7 reaching statistical significance. Females in the 0.075 mg/kg/day dose group showed smaller, statistically insignificant decreases. No other agent-related signs of toxicity appeared at any dose of GB Type II (Bucci and Parker, 1992; Opresko et al., 2001). The LOAEL for RBC-ChE depression with GB Type I also was observed as 0.075 mg/kg in female rats. In addition, two high-dose and one low-dose females in the GB Type I group had brain lesions. The low-dose females lesion was considered treatment-related, due to the presence of other signs of nerve agent toxicity. Neurotoxic esterase (NTE) levels were reduced in females in the high-dose GB Type I group, but no histological signs of delayed neuropathy were present (Bucci et al., 1991; Opresko et al., 2001).
In a subchronic inhalation study, Fischer 344 rats were exposed to 0.0001 or 0.001 mg/m3 of sarin for 6 hours per day, 5 days per week for up to 24 weeks. Researchers observed no signs of toxicity at either dose (Weimer et al., 1979; Opresko et al., 2001). However, this is a low dose compared to that of the Centers subchronic oral study, assuming an inhalation rate of 0.29 m3/day, an average body weight of 0.35 kg for rats, and 100% pulmonary absorption, the highest concentration on the subchronic inhalation study is only 0.15 µg/kg/day (Opresko et al., 2001).
It is not clear how the subchronic inhalation studies results correspond to servicemembers exposures at Khamisiyah because the servicemembers exposure duration was short, corresponding to an acute dose. On the other hand, the DoD considered the servicemember exposure concentration very low, more on the order of concentrations used in chronic or subchronic studies. Therefore, evaluating possible effects of low-dose exposures is essential.
Chronic exposure of L.C.R. Swiss and Strain A mice; Sprague-Dawley, Wistar, and Fischer 344 rats; and purebred beagle dogs to 0, 0.001, or 0.0001 mg/m3 of GB for 6 hours per day, 5 days per week, up to 52 weeks resulted in no dose-related, statistically significant changes in RBC-ChE in any species at any sampling time. Five of 20 dogs had abnormal electrocardiograms at the time of sacrifice; however, the heart showed no evidence of enlargement or physical abnormalities. The absence of pre-exposure data precluded considering this effect as agent-related. The colony rats showed an increased incidence of tracheitis, but investigators could not determine whether the tracheitis was nerve agent-related (Weimer et al., 1979; Opresko et al., 2001). The Fischer rats also showed atrophy of seminiferous tubules, but the animals had undergone heat stress and were genetically susceptible to this condition (Weimer et al., 1979; Opresko et al., 2001; Munro et al., 1994).
As sarin principally affects the nervous system, we must address the potential for sarin exposure to cause neurotoxicity. The acute effects of GB exposure, described previously, are readily reversible except in severe poisonings. Brain lesions in animals may occur with doses sufficiently high to cause convulsions and also have been reported in limited studies in the absence of convulsions. However, the latter results have not been reproduced (Bucci et al., 1991; Sidell and Hurst, 1997; Hartgraves and Murphy, 1992; Singer et al., 1987). Diazepam, a benzodiazepene anticonvulsant, apparently significantly reduces soman-induced brain damage in several species (soman is a nerve agent in the same chemical family as sarin) (Sidell, 1992). Kadar et al. (1995) administered a single LD50 of sarin intramuscularly to rats and performed histological and morphological studies on their brains at intervals of 4 hours to 90 days after exposure. Brain lesions were observed in 70% of the animals and were most pronounced in those exhibiting continuous seizures. Rats having only short seizure episodes showed minimal or no changes to their brains. These data suggest brain lesions are related to convulsive activity and would not occur with exposure to the low doses of nerve agent under consideration in this evaluation.
Historically, a victim who recovers from acute sarin poisoning should not experience chronic sequelae unless anoxia and convulsions have gone unchecked so long that irreversible CNS damage has occurred (DA, 1990). However, recent concerns about long-term neurological as well as psychological effects in workers accidentally poisoned by organophosphate pesticides and nerve agents are spurring investigations into the possible chronic effects of poisonings and the effects of low-level, chronic exposures (Ecobichon and Joy, 1994).
Two particular areas of concern in assessing the ability of low-level sarin exposure to cause chronic neurological effects are the agents potential to induce a specific syndrome, "organophosphate-induced delayed neuropathy" (OPIDN), and its subtle effects on behavior, psychological state, memory and cognition, and electroencephalographs (EEGs). Many studies are available on organophosphate insecticides neurotoxicity, but because of the differences between insecticides and nerve agents, these studies may not always be accurately predictive (Romano et al., 2001).
a. Organophosphate-Induced Delayed Neuropathy. Researchers have related human exposure to certain organophosphates to the development of OPIDN. Available data strongly suggest GB does not cause OPIDN in humans at asymptomatic dose levels. Organophosphate-induced delayed neuropathy causes weakness and ataxia in the lower limbs 8 to 14 days after acute poisoning, with occasional progression to paralysis. This delayed neuropathy is associated with axonal degeneration and demyelination of peripheral nerves and certain parts of the CNS, with severity related to the specific compound and amount absorbed (Perrotta, 1996). The OPIDN syndrome has been known for almost 100 years and has been studied extensively (Ecobichon, 1996). Because of its susceptibility, much of the basic OPIDN research has been conducted in the adult hen, which remains key to the Environmental Protection Agencys (EPA) neurotoxicity testing protocol (Ecobichon, 1996).
Researchers now think organophosphates ability to produce delayed neuropathy partly depends on their ability to phosphorylate and irreversibly inhibit the enzyme NTE (Johnson, 1975; Ecobichon and Joy, 1994; Anthony et al., 1996). However, although NTE inhibition is useful for preliminary evaluation of organophosphorus esters potential to induce OPIDN, the role of this enzyme in initiating the syndrome is not understood, and the EPA neurotoxicity protocol requires histopathologic evidence (Ecobichon, 1996). Indeed, several organophosphates, including some nerve agents, can inhibit NTE. Yet exposed animals do not develop OPIDN except at far beyond lethal doses (Ecobichon, 1996; Johnson et al., 1985; Lotti, 1992; Marrs, 1993). Ecobichon and Joy (1994) empirically related a decrease of more than 70% in chicken brain NTE activity 24 to 48 hours after exposure to subsequent development of OPIDN-like symptoms.
Sarin administered at doses 20 to 30 times the LD50 (25 µg/kg) has induced signs of OPIDN and high NTE inhibition in sensitive hens (Davies et al., 1960; Davies and Holland, 1972; Gordon et al., 1983). These researchers protected hens against the lethal effects of exposure by treatment with atropine and oximes (cholinesterase reactivators) (Davies et al., 1960; Davies and Holland, 1972; Gordon et al., 1983). In vitro studies by Vranken et al. (1992) reported 78% inhibition with 0.35 µM sarin (about 10 to 100 times the lethal concentration). In vivo studies by Willems et al. (1983) showed 75% inhibition of NTE after oral administration of 200 µg/kg sarin in hens. Studies by Bucci et al. (1992) found no ataxia or signs of OPIDN pathology in hens dosed with 70.2, 140.4, or 280.7 µg/kg GB Type II, administered acutely or in three divided doses, nor even significant dose-related changes in NTE with single GB Type II doses of up to 750 µg/kg. One small study in Swiss albino mice exposed to 100 mg-min/m3 daily for 10 days reported signs of OPIDN, neuropathology, and decreased NTE activity (Husain et al., 1993). However, Bucci et al. (1991) found rats receiving 300 µg/kg/day of GB Type I for 90 days had reduced NTE activity but no histopathological signs of OPIDN. The same study showed rats given similar doses of GB Type II developed neither histopathological symptoms nor NTE decreases (Bucci and Parker, 1992).
The sarin doses administered in all these studies were high and certainly orders of magnitude higher than those under consideration in this exposure evaluation. Sarin did not induce OPIDN in humans exposed to acutely toxic GB doses (Munro et al., 1994), nor in cats receiving several low GB doses for up to 10 days (Goldstein et al., 1987; Goldstein, 1989). Human subjects exposed to this agent during studies at Aberdeen Proving Ground (NRC, 1985) showed no signs of OPIDN, even though doses were sufficiently high for the subjects to develop symptoms. Based on available data, it appears GB does not cause OPIDN in humans at the low dosages this evaluation considers.
Baker and Sedgewick (1996) examined single-fiber electromyographic changes at the neuromuscular junction of men exposed to 15 mg-min/m3 of GB vapor inside a test chamber. This dosage was sufficient to cause miosis, mild dyspnea, and a 40% decrease in RBC-ChE. The authors found small changes in the single-fiber electromyographic consistent with the development of nondepolarizing neuromuscular blockade, without frank clinical expression. The changes were reversible; the researchers regarded the changes as clinically unimportant. Whether these effects would occur with asymptomatic exposures is unknown.
b. Behavioral and Psychological Effects. No data are available to suggest human exposure to very low, asymptomatic levels of sarin results in any psychological or behavioral sequelae. Existing data suggest chronic neuropsychological sequelae may occur with higher-dose, symptomatic exposure, and the effects are likely correlated with ChE inhibition.
Scientists have reported behavioral and psychological effects from exposure to organophosphates for more than five decades (Ecobichon, 1996), including, among other manifestations, impaired vigilance and reduced concentration, reduced information processing and psychomotor speed, memory deficits, linguistic disturbances, depression, anxiety and/or irritability, excessive dreaming, sleep disturbances, mood changes, easy fatigability, and degradation in performance measures (Ecobichon and Joy, 1994; Sidell, 1992; Somani et al., 1992; Ecobichon, 1996; Savage et al., 1988). Although the incidences of these disturbances were higher after severe exposure, they may occur in persons with few or no physical signs of exposure and may last from days to years (Sidell, 1997; Burns et al., 1996; Ecobichon and Joy, 1994).
Savage et al., (1988) examined the chronic neurological sequelae of acute organophosphate pesticide poisoning in 100 matched pairs of persons with no medical history of organophosphate poisoning within the three months before the study. Of more than 50 scores from the neurological examination, subjects demonstrated abnormalities only on measures of memory, abstraction, and mood, and on one test of motor reflexes. Differences between the two cohorts were much greater in the neuropsychological tests, including those in intellectual functioning, academic skills, abstraction and flexibility of thinking, and simple motor skills. Halstead-Reitan Battery summary scores in the range characteristic of people with cerebral damage or dysfunction afflicted twice as many cases as controls (24 vs. 12); these cases all had documented histories of organophosphate exposure and a physicians diagnosis of symptoms consistent with organophospate poisoning. Others have reported similar findings subsequent to organophosphate poisoning (Rosentock et al., 1990; Dille and Smith, 1964; Midtling et al., 1985).
Blick et al. (1986, 1987a) and Hartgraves and Murphy (1992) studied the behavioral effects of low-level subclinical soman exposure in primates using a primate equilibrium platform (PEP). In this test the animal is to compensate for an external signal producing unpredictable perturbations in the platform position by manipulating a joystick mounted in front of it on the platform to keep the platform level. This task depends on both fine motor control and the integrity of the sensorimotor system to maintain equilibrium and orientation in space. 2.5 µg/kg was determined to be the acute 50% effective dose (ED50) for PEP performance deficit. There was also a 65 to 70% inhibition of serum ChE. The multiple dose ED50 was 0.97 µg/kg/day (dose producing a 50% PEP performance deficit on or before the 5th day of exposure) and inhibited serum ChE 85 to 90% (Blick et al., 1987b; 1988). The authors reported reliably detectable decreases in PEP performance occurred at doses failing to produce extensive physical symptoms (e.g., miosis, salivation, fasciculation). When exposure was discontinued after the first detectable performance decrement, performance quickly returned to normal and no further symptoms were noted. In the single case in which an animal was exposed for three additional days after the first detectable performance decrease, clear, quite severe neurological symptoms developed, and the animal was incapable of performing the task for several days. The symptoms gradually abated and disappeared in two to three weeks (Hartgraves and Murphy, 1992). The authors suggest subtle performance deficits may indicate incipient neurotoxicity.
Sirkka et al. (1990) studied the effects of single inter-peritoneal sarin doses of 12.5 and 50 µg/kg on behavior, motor performance, and nociception (i.e., painful sensation from neuron stimulation or injury) in male Wistar rats. The doses used were significantly below the reported inter-peritoneal LD50 of 218-250 µg/kg (DHHS, 1987). At the higher dose, sarin decreased the proportion of entries made into the open arms of a plus maze and the amount of time spent to reach the safe platform on an elevated horizontal bridge, but there was no impairment of learning frequency in the one-trial passive avoidance test, rotarod performance, or nociception in the hot plate test. At the lower dose of 12.5 µg/kg the authors observed no significant effects on any of the parameters examined. These results suggest acute doses of sarin at 50 µg/kg alter rats behavior (anxiety) and motor coordination and balance.
Sheets et al. (1997) conducted subchronic neurotoxicity screening studies with six organophosphate insecticides and assessed behavior and morphology relative to cholinesterase inhibition. The organophosphates were orally administered in four amounts through diet to Fischer 344 rats. Animals were evaluated using a functional observational battery, automated measures of activity, and detailed clinical observations, with half the animals undergoing microscopic examination of neural and muscle tissues. Satellite groups were used to measure the effects of each treatment on plasma, erythrocyte, and brain ChE activity. All treatment-related neurobehavioral findings were ascribed to cholinergic toxicity and occurred only at dietary amounts producing more than 20% inhibition of plasma, RBC, and brain ChE activity. ChE inhibition measures were the most sensitive indices for exposure to all six compounds; in general, RBC-ChE was inhibited to a greater extent than plasma or brain ChE. No additional cumulative effects on ChE activity or neurobehavioral endpoints appeared after four or eight weeks of treatment, respectively. The authors suggest employing a general standard of more than 20% ChE inhibition for cholinergic neurotoxicity screening.
c. Electroencephalographic Changes. No data are available to suggest very low, asymptomatic sarin doses can induce possibly long-lasting changes in EEGs in humans. Available data suggest there is an as-yet-undefined threshold dose for reported agent-induced EEG changes. The subtle EEG abnormalities observed at higher doses are difficult to identify; researchers have not correlated these abnormalities with clinically significant effects (e.g., altered behavioral, psychological, or performance parameters).
Organophosphate compounds are known to potently affect the nervous system, including inducing EEG changes. While early studies documented short-term changes in EEG (Grob and Harvey, 1953), Metcalf and Holmes (1969) suggested organophosphate poisoning may lead to long-term EEG changes (Perrotta, 1996). Subsequent studies evaluated EEG changes from exposure to known doses of sarin in monkeys and unknown but symptomatic doses in accidentally exposed workers.
Burchfiel et al. (1976) exposed monkeys to GB at two dose schedules: (1) a single high-level dose of 5 µg/kg intravenously that produced overt signs of toxicity and convulsions requiring artificial respiration to prevent secondary brain anoxia, or (2) a series of ten smaller doses of 1 µg/kg intramuscularly given at one-week intervals that did not produce any major clinical signs (the report did not specify exactly what signs were present). EEGs were recorded before drug administration, 24 hours post-treatment, and one year post-treatment. Both high- and low-dose animals exhibited statistically increased temporal lobe beta voltage at 24 hours and one year after administration, relative to control animals (Burchfiel and Duffy, 1982).
In a related study, researchers selected human subjects based on sarin exposure history, resultant clinical signs and symptoms consistent with overexposure, and RBC-ChE reduction of at least 25% below the subjects pre-exposure baseline. The overexposed group was further divided into "exposed" (one or two exposures within the last six years) and "maximally exposed" (three or more exposures within the last six years). Computerized spectral analysis of the human EEGs showed probable differences in the exposed group and highly significant differences (increased beta activity) in the maximally exposed group relative to the control group. Investigators also found increased rapid eye movement during sleep in both the exposed and maximally exposed group compared to the controls. When the authors performed computer analyses to derive classification rules based on combinations of EEG variables in an effort to identify a person as belonging to a specific treatment group, results were inconclusive. However, the results suggested the exposed population was bimodal, with one mode coinciding with the mode of the control distribution. Based on these data, the authors suggested the EEGs of some workers exposed to sarin were not affected and there is a threshold for this type of sarin neurotoxicity (Burchfiel and Duffy, 1982).
The Burchfiel and Duffy (1982) study reporting long-term sarin-induced EEG changes in humans clearly involved dose levels large enough to cause symptoms. The low-dose monkey study suggests exposure to reasonably small doses of sarin may result in long-term EEG changes, although it was unclear whether these animals were asymptomatic or that RBC-ChE levels were not significantly depressed. Grob and Harvey (1957) reported that after small, asymptomatic sarin oral doses (~0.0005 to 0.002 mg/kg acutely and 0.007 to 0.044 mg/kg over three days), EEGs did not change. Slight alterations in EEGs occurred with the appearance of mild symptoms (0.022 mg/kg acutely and 0.088 mg/kg over three days), and increasing irregularities or abnormalities occurred with moderate symptoms (0.028 mg/kg acutely and 0.102 mg/kg over three days). The EEG changes were detectable for four to eight days after the symptoms disappeared. The appearance of symptoms correlated with increased doses and decreased ChE activity.
Together these data suggest sarin can induce possibly long-lasting, subtle changes in EEGs in both animals and humans, but there likely is a threshold for this effect. The reported EEG abnormalities did not correlate with changes in behavioral, psychological, or performance parameters. In their 1988 recommendations for long-term exposure limits to GB, the Centers for Disease Control and Prevention (CDC) concluded, "The EEG changes reported after intoxication with GB were considered to be of questionable significancegiven the difficulty of demonstrating such changes and the absence of clinically significant effects even when EEG changes are present" (CDC, 1988). However, the existence of known psychological and behavioral effects reported in organophosphate pesticide and nerve agent poisonings and the potential for long-term alterations in EEGs from exposure to sarin suggest exercising the utmost caution to prevent any exposures resulting in even mild symptoms or significant ChE reduction.
Available data suggest low-dose (well below the convulsive level) exposures to sarin do not cause cardiomyopathy. In fact, only convulsive doses have induced cardiomyopathy in any species examined (Bucci et al., 1991; Bucci and Parker, 1992; Weimer et al., 1979). Singer et al., (1987) reported generally mild, randomly located, reparable myocardial lesions in Sprague-Dawley rats exposed to high doses (111 to 221 µg/kg) of sarin. Only 31 of 71 animals survived 24 hours. Of these survivors, 15 convulsed, and 9 convulsing and 1 nonconvulsing rats had brain lesions. Four of the five animals with cardiac lesions also had brain lesions; all had convulsed. The authors suggest high doses of sarin may cause general hypoxia and/or neurogenic cardiomyopathy in rats.
3. General Health Effects
No reports of any general adverse health effects from low-level exposure to sarin are available. An examination of several follow-up measurements conducted on men exposed to anticholinesterase chemicals (including GB) during experiments at Aberdeen Proving Ground in Edgewood, Maryland, did not find significant increases in hospital admissions, self-reported medical problems, impairments, malignancies, or other adverse health outcomes (NRC, 1985). The authors observed the expected healthy soldier effect in some of the standardized morbidity (or mortality) ratios calculated in this effort. The authors admit to low-to-moderate statistical power to identify differences; therefore, only large differences likely would be uncovered (Perrotta, 1996; NRC, 1985).
Relatively little is known about organophosphates effects on the immune system. Burns et al. (1996) associated occupational exposure to organophosphates with decreased polymorphonuclear white blood cell chemotaxis and increased upper respiratory infection. Prolonged or high-dose exposure to the organophosphate insecticides malathion, parathion, or methyl parathion has been shown to alter immune function. No reports of immunotoxicity associated with exposure to cyclosarin have been identified. Kassa et al. (2000) have recently reported asymptomatic doses of sarin decrease lymphoproliferation and N-oxides production in rats 12 months after exposure.
Organophosphate nerve agents are not recognized as carcinogens. No human studies are available to suggest GB is carcinogenic in humans (Opresko et al., 2001; Perrotta, 1996). In a follow-up study of approximately 995 US Army volunteers who participated in anticholinesterase studies at the US Army laboratories, Aberdeen Proving Ground, Edgewood, Maryland, during 1955 to 1975, the National Research Council (NRC) found no consistent pattern of increased cancer risk (NRC, 1985). The study, of relatively low statistical power, was able to identify only large differences. Based on these findings and the 10 lifetime studies of carcinogenicity of organophosphates sponsored by the National Cancer Institute, the investigators concluded that anticholinesterase compounds did not induce cancer among the Edgewood subjects (NRC, 1985).
Weimer et al. (1979) conducted an inhalation study exposing I.C.R. Swiss and Strain A mice; Sprague-Dawley, Wistar, and Fischer 344 rats; and purebred beagle dogs to sarin concentrations of 0.001 or 0.0001 mg/m3 for six hours per day, five days per week. No agent-related tumors were found in any species. Pulmonary adenocarcinomas did occur in Strain A mice. However, Strain A mice have a high propensity to form pulmonary tumors, with an incidence of approximately 53% in animals 12 months and 90% at 18 months of age (Heston, 1942; Opresko et al., 2001).
Information on sarins genotoxicity in humans is not available. Goldman et al. (1987) reported GB is not genotoxic or mutagenic based on both in vivo and in vitro assays. Negative results were obtained in the Ames Salmonella bacterial gene mutation assay using five revertant strains (TA135, TA100, TA98, TA1537, and TA1538) and a range of concentrations of sarin. GB Types I and II did not induce a significant increase in forward mutations in mouse L5178Y lymphoma cells in concentrations ranging from 50 to 200 µg/ml. Increases in sister chromatid exchange did not appear in Chinese hamster ovary cells exposed in vitro to 200 µg/ml of sarin, nor did sister chromatid exchange in splenic lymphocytes significantly increase in mice receiving a maximally tolerated inter-peritoneal dose of 360 µg/kg of GB. Exposure of rat hepatocytes to sarin at molar concentrations as high as 2.4 x 10-3 M decreased deoxyribonucleic acid (DNA) repair synthesis. The authors conclude that sarin does not damage DNA directly but might inhibit DNA synthesis after non-agent-induced DNA damage already had occurred (Goldman et al, 1987; Opresko et al., 2001; Perrotta, 1996).
6. Developmental and Reproductive Effects
In their study of toxicity in dogs exposed to chronic GB, Jacobsen et al. (1959) bred the male animals after 25 weeks of daily moderate doses of GB; the offspring were normal (Perrotta, 1996). Weimer et al. (1979) conducted a one-year, low-dose GB inhalation exposure study in a variety of animals and found no abnormalities in reproduction and fertility, fetal toxicity, or teratogenesis in Sprague-Dawley and Wistar rats. Testicular atrophy was noted in the Fischer rat, but the authors did not consider the effect agent-related, since later experiments using a different exposure route did not replicate the finding.
In developmental toxicity studies conducted on GB Types I and II in CD rats and New Zealand rabbits, rats were dosed orally with 0, 100, 240, or 380 µg/kg of GB on Gestation Days 6 to 15. Animals were observed for clinical signs of toxicity, and at autopsy gravid uteri were weighed and internal or external malformations were noted. Maternal toxicity and mortality (8 of 29 for GB Type I and 13 of 29 for GB Type II) occurred in the high-dose group. Neither the incidence of resorptions nor the average body weight of live fetuses per litter differed significantly among treatment groups. No dose-related morphological anomalies were observed. No fetal toxicity, teratogenicity, or dose-related morphological anomalies were reported in similar studies of New Zealand rabbits administered 0, 5, 10, and 15 µg/kg/day of GB on Gestational Days 6 to 19. Maternal toxicity was observed at the highest dose (LaBorde and Bates, 1986; Opresko et al., 2001).
Denk (1975) evaluated the developmental toxicity of sarin in Sprague-Dawley rats exposed to GB vapor (0.1 and 1 µg/m3) for six hours per day, five days per week for varying time periods. In one series, male rats were exposed for one week to one year and then mated with unexposed females. The females were sacrificed 19 days after mating and examined for numbers of corpora lutea, deciduomata, number of fetal deaths, and number of live fetuses. In a second series, mated matched pairs of rats were exposed to GB for one, two, or three weeks or until all the pups were whelped. Intrauterine deaths were observed, and all fetuses were examined for abnormalities. In the third series, males and females were exposed to GB for 10 months and then mated, as were the F1 and F2 generations. The number and sex of offspring, number of preweaning deaths, number weaned, and pup weights at various ages were recorded. The author reported that at the dose and by the exposure route used, GB had no adverse effects on reproductive performance, fetal toxicity, teratogenesis, and dominant lethal mutations (Denk, 1975; Opresko et al., 2001).
Human toxicity data on GBs reproductive and developmental effects currently do not exist. However, organophosphates are not generally considered to significantly affect reproduction, and the laboratory animal studies discussed above suggest such effects are unlikely, even at dose levels that are maternally toxic (Opresko et al., 2001; Perrotta, 1996).