II. HAZARD IDENTIFICATION
This section evaluates the available data on the toxicologies of sarin and cyclosarin. Additionally, it assesses the potential for human health risk from a possible low-level chemical warfare agent exposure after the destruction of chemical-warfare-agent-filled munitions at Khamisiyah on March 10, 1991. In this report, a low-level exposure means one that produces no signs or symptoms, including cholinergic depression.
B. Toxicological Profile of Sarin Nerve Agent
Sarin is a toxic, relatively nonpersistent, volatile nerve agent developed in Germany during World War II and, subsequently, produced and tested in the United States. It is readily absorbed by most materials, but readily desorbed within a few hours after removal from the source of delivery. It evaporates at about the same rate as water. It exerts its effects by powerfully inhibiting the enzyme acetylcholinesterase (AChE), which is required for proper nerve and muscle function. This inhibition adversely affects skeletal muscle, parasympathetic end organs, and central nervous system operation, thereby interfering with the fundamental mechanisms required for normal function of the central and peripheral nervous systems. The type and severity of clinical symptoms of sarin exposure depend on this agents route of entry into the body and the magnitude of exposure.
When the eyes and nasal passages are exposed to sarin vapor, the earliest signs expected are local effects, specifically miosis (pinpoint pupils) and rhinorrhea (runny nose), rather than systemic effects. However, if exposure is through the skin or by ingestion, the pupils may be approximately normal in size and rhinorrhea absent, while other symptoms (e.g., localized sweating, muscular fasciculations (twitching), and/or gastrointestinal disturbances) may be present. Decreases in cholinesterase in red blood cells (RBC-ChE) and plasma (serum or Butyrylcholinesterase (BuChE)) often can be detected at low-level exposures to sarin and, depending on exposure route and duration, before other symptoms, including miosis, appear.
Sarins effects may be cumulative if the time between two or more exposures is less than approximately 120 days, because it takes this long to restore cholinesterase activity to its normal level. Thus, repeated asymptomatic exposures to small amounts of sarin, if somewhat close together, can result in symptoms. An exposure dosage, Ct (concentration (C) in milligrams per cubic meter (mg/m3), multiplied by the exposure time in minutes (t)) of 0.5 mg-min/m3 of sarin is deemed the no-effect dosage in humans; less than 1% of a working population would show miosis (McNamara and Leitnaker, 1971). No systemic effects are expected below 4.0 mg-min/m3, no deaths are expected below 10.0 mg-min/m3, and the effective Ct for red blood cell cholinesterase depression in 50% of humans is 20 mg-min/m3. The estimated 50% lethal concentration (LCt50) for sarin in a 70-kilogram man breathing at the rate of 15 liters per minute is approximately 70 mg-min/m3. Chronic toxicity studies are not available; however, scientists infer the long-term health risks associated with low-level exposures to sarin from several human studies that used dosages causing acute (short-term) effects, several unintentional high-dose human exposures, and animal studies.
Sarins principal effects are on the nervous system. However, if a victim recovers from acute sarin poisoning, no long-term effects are expected unless the associated lack of oxygen to the brain and convulsions go unchecked so long that irreversible central nervous system damage occurs. One concern is organophosphate-induced delayed neuropathy (OPIDN) and its possible subtle effects on behavior, psychological state, memory and cognition, and altered electroencephalographs (EEGs.) Augerson (2000) reviewed numerous studies of OPIDN focusing on organophosphate insecticides that did not produce acute symptoms; the resulting information is insufficient to determine whether low-level exposure to chemical warfare nerve agents causes long-term effects.
C. Toxicological Profile of Cyclosarin Nerve Agent
Cyclosarin is a toxic, relatively nonpersistent nerve agent similar to but much less volatile than sarincyclosarin evaporates 20 times more slowly. However, it is unlikely to persist in the environment more than a few days. There is less information in the scientific literature on cyclosarins chemistry and toxicology than on sarins. Except for acute exposures, cyclosarin is less toxic than sarin by inhalation and slightly more toxic by dermal exposure. The combined toxicity of sarin and cyclosarin does not indicate synergism. Although chronic toxicity data for cyclosarin are sparse, this agent is expected to behave like sarin. Both cause neural lesions in nonhuman primates, but the LCt50 for cyclosarin is as much as five times higher than LCt50 for sarin (i.e., cyclosarin is less lethal). No data exist to suggest cyclosarin causes OPIDN at the low level exposure concentrations bounded by the GPL for this modeling. No studies are available on cyclosarins behavioral and psychological effects, although at high doses, the mechanisms of action for these possible effects are expected to be similar to sarin. Likewise, no reports of cyclosarin-induced EEG changes exist, but the possibility of producing effects similar to those of sarin cannot be dismissed. Cyclosarin-induced cardiomyopathy, convulsions, and skeletal muscle lesions have been observed in monkeys at LCt50 concentration levels. The current literature suggests that cyclosarin, like sarin, does not have carcinogenic or mutagenic properties and, although no human toxicity data on cyclosarins reproductive and developmental effects are available, its effects, like those of sarin, are unlikely to occur even at maternally toxic doses.
D. Recommended Exposure Guidelines for Sarin
Since none of the existing exposure guidelines were developed for the exposure and health outcome situations of concern in this assessment, the Department of Defense (DoD) reviewed the exposure guidelines developed by the US Army and the CDC as well as the critical toxicological studies used to develop the guidelines.
The existing occupational and general population guidelines for sarin were derived from no-adverse-effect levels and levels known to produce the mildest of detectable effects. The dosage of sarin vapor that produces first noticeable effects is approximately 1 mg-min/m3 or about 70 to 100 times lower than the lethal exposure dosage of sarin (Mioduszewski et al., 1998). The air concentration for the occupational guideline was calculated from data from experiments with human volunteers and kinetic studies of RBC-ChE and BuChE depression and recovery. The guidelines are based on studies (McNamara and Leitnaker, 1971) that determined a no-effect dosage of 0.5 mg-min/m3, the amount at which less than 1% of a human population would be expected to show even the mildest symptoms, such as miosis. The general population guideline for sarin was derived by applying to the occupational guideline another safety factor to protect the most sensitive individuals. These exposure conditions and occupational and general population limits have been extrapolated to chronic (lifetime) exposures, and the guidelines incorporate considerable safety margins.
III. DOSE-RESPONSE ASSESSMENT
This section describes some of the relationships between exposure dose and effect, and explains the effect time or exposure duration has on exposure effects. It discusses the rationale the CDC used to derive the limits for workers and the general population for exposure to nerve agents. Finally, it describes the exposure limit used in this reconstruction and how DoD arrived at that limit.
The dose-response assessment is a critical part of the hazard characterization process: it further evaluates the relationship between a toxic chemical and its effects on humans (EPA, 1995; NRC, 1994), emphasizes the quantitative relationship of dose and its toxic response, and includes factors affecting or modifying the response.
In most cases, the dose-response assessment involves extrapolating animal data to human response. In the case of nerve agents, data is available on human response to nerve agents, as described in the previous section. For non-cancer endpoints, the dose-response assessment can use existing data or extrapolate existing data to estimate a dose at which no effects are observed, called the no-observed-adverse-effects-level. Similarly, the lowest dose at which effects are observed is called the lowest-observed-adverse-effects-level (Goldsmith, 1995).
B. Dose-Response Relationship
A specific amount of a toxic material will produce a specific type and intensity of response. Dose-response relationship is a fundamental concept in toxicology and the basis for measuring a chemicals relative harmfulness. Dose-response is the quantitative relationship between the dose of a chemical and an effect it causes.
1. Dose Terms
In toxicology, the dose given to test organisms is expressed in terms of the quantity administered, which may be expressed as:
The period of time over which a dose is administered is generally specified. For example, 5 milligrams of chemical per kilogram of the subject's body weight administered over 3 days is specified as 5 mg/kg/3 D. For a dose to be meaningful, it must be related to the effect it causes. For example, 50 mg/kg of chemical "X" administered orally to female rats has no relevance unless the researcher reports the effect of the dose in all test subjects.
2. Dose-Response Terms
The National Institute for Occupational Safety and Health defines several general dose-response terms (DHHS, 1979) as:
Toxic dose low (TDLO): The lowest dose of a substance introduced by any route, other than inhalation, over any given period of time, and reported to produce any toxic effect in humans or carcinogenic, neoplastigenic, or teratogenic effects in animals or humans.
Toxic concentration low (TCLO): The lowest concentration of a substance in air to which humans or animals have been exposed [through inhalation] for any given period of time, and to produce any toxic effect in humans or carcinogenic, neoplastigenic, or teratogenic effect in animals or humans.
Lethal dose low (LDLO): The lowest dose (other than LD50) of a substance introduced by any route, other than inhalation, over any given period of time in one or more divided portions and reported to have caused death in humans or animals.
Lethal dose fifty (LD50): A calculated dose of a substance which is expected to cause the death of 50% of an entire defined experimental animal population. It is determined from the exposure to the substance by any route, other than inhalation, of a significant number from that population.
Lethal concentration low (LCLO): The lowest concentration of a substance in air, other than LCt50, which has been reported to have caused death in humans or animals. The reported concentrations may be experienced for exposures which are less than 24 hours (acute) or greater than 24 hours (subacute and chronic) [see Duration and Frequency in Section III. B.4].
Lethal concentration fifty (LCt50): A calculated concentration of a substance in air, exposure [through inhalation] to which for a specified length of time is expected to cause the death of 50% of an entire defined experimental animal population.
3. Determining the Relationship Between Dose and Response
Scientists plot the data of dose administered versus effects demonstrated to obtain a dose-response curve that shows the relationship between chemical dose and the number or kinds of health effects induced in a species. These curves fall into two general groups: those in which scientists observe no response until some minimal dose (threshold) is reached and those in which any dose produces some degree of risk of deleterious response (non-threshold).
Since it is impossible to determine effects at extremely low doses, scientists often assume the origin of the dose-response curve is zero and the shape of the curve is linear (i.e., increases in doses above zero increase adverse effects.) At some level, all members of the population demonstrate the effect under study. However, it is important to note that rarely, if ever, does a change in dose result in qualitatively versus quantitatively different results.
4. Factors Influencing Toxicity
Many factors affect an organisms reaction to a toxic chemical. The specific response a dose elicits varies depending on the species and variations among individuals of the same species. The following factors can affect toxicity:
- When combined, synergistic
chemicals cause a greater than additive effect. For example, exposure to
both ethyl alcohol and carbon tetrachloride enhances liver toxicity.
- In potentiation, a type of synergism, the potentiator usually is not toxic in itself, but can increase other chemicals toxicity. Isopropyl alcohol, for example, is not toxic to the liver by itself, but it increases the livers toxic response to carbon tetrachloride.
- When combined, antagonistic chemicals lessen a predicted effect. Types of antagonists include functional, which produce opposite effects on the same physiologic function; chemical, which react with toxic compounds to form less toxic products; dispositional, which alter absorption, metabolism, distribution, or excretion; and receptor, which either bind to the same receptor as the toxic chemical or block the receptors action, reducing the toxic effect.
Animal studies, epidemiological investigations of exposed human populations, and clinical studies or case reports of exposed humans produce information on chemical compounds toxic properties and the dose-response relationships. Epidemiological investigations are based on a human population exposed to a chemical and compared to an appropriate, non-exposed group to determine whether a statistically significant association exists between health effects and chemical exposure. Clinical cases involve individual reports of chemical exposure. Scientists then use all this information to determine whether a substance is toxic, its level of toxicity, and the dose-response relationship.
5. Occupational Exposure Thresholds
For non-cancer symptoms, scientists usually extrapolate the dose-response relationship from high dose results and data to expected low dose effects. Scientists use available data to determine thresholds below which they either observe no effects or just begin to observe effects. Once determined, scientists adjust these levels by safety (or uncertainty) factors and modifying factors (Goldsmith, 1995) to obtain a reference dose or concentration, which can be an occupational or general population limit.
Nerve agents are among the most toxic compounds made. When it began its demilitarization program to destroy these compounds, DoD needed to determine the air concentrations to which the general population could be exposed, even indefinitely, without showing any effects of that exposure. As DoDs executive agent, the Army followed the approach the American Conference of Governmental Industrial Hygienists uses to develop threshold limit values (TLV) for toxic materials (McNamara and Leitnaker, 1971).
The Army developed two different TLVs for the chemical nerve agent sarin: one for workers in an occupational setting and one for the general population. As noted in Section II, a complete toxicological data set does not exist for sarin exposure in humans, especially for the type of chronic exposure on which the TLVs are based. However, the Army performed several extrapolations and derivations using chronic and acute animal exposure data and human acute exposure data. The process began with two assumptions based on the experimental data examined:
Based on several human inhalation exposure studies, the Army estimated sarins no-death dosage as 10 mg-min/m3 (based on a subjects breathing rate of 15 liters per minute for a 2 minute exposure time), since none of those exposed developed severe signs. In a further evaluation of human exposure data, the Army determined humans did not display neuromuscular effects when exposed to sarin dosages below 6 mg-min/m3, then calculated the "no neuromuscular effects" dosage to be a Ct of 4.0 mg-min/m3. Scientists consider this a very conservative estimate of a threshold for neuromuscular effects. Finally, scientists determined that red blood cell (RBC) cholinesterase in humans is depressed by 10% for each 1 µg/kg of exposure to GB vapor. The human and canine recovery rates correspond to the RBC production rate, but this RBC cholinesterase recovery does not correlate with the recovery from toxic effects. Thus, a person exposed to sarin may regain normal resistance to the nerve agent (measured by plasma cholinesterase) while RBC cholinesterase still is depressed. If the plasma cholinesterase recovery rate corresponds to the sarin detoxification rate, then the constant daily dose would be about 10% of the "no effects" dose or 0.05 mg-min/m3. Thus, the Army and CDC set the maximum safe concentration for unmasked workers exposed for up to one hour at 0.001 mg/m3, which is equivalent to a Ct of 0.06 mg-min/m3 (0.001 mg/m3 times 60 min) or approximately the no effects dose. But the 8-hour exposure limit is based on cholinesterase inhibition. The expected plasma cholinesterase inhibition resulting from the no effects dose (0.05 mg-min/m3 per day) is the equivalent of a single exposure of 0.5 mg-min/m3, which equals an exposure of 0.075 µg/kg; thus, an 8-hour exposure to 0.0001 mg/m3 (0.0001 mg/m3 x 8 hr x 60 min/hr = 0.048 mg-min/m3) would inhibit less than 1% of plasma cholinesterase. This difference is impossible to quantify because of daily individual variation and analytical accuracy (McNamara and Leitnaker, 1971).
These occupational exposure limits are based on a healthy worker population. To accommodate the greater sarin sensitivity possibly occurring in women, the very young, the very old, and other special populations at risk with congenital conditions or disease, the Army applied an uncertainty factor of 0.1 to the occupational limit to derive the general population limit. Thus, the maximum concentration to which the general population could be exposed must fall within 0.0001 mg/m3 for up to one hour and 0.000003 mg/m3 for 72 hours.
Both the Department of the Army (DAMS, 1990) and CDC (1988) recommend the same limits for workers occupational exposure without respiratory protection (i.e., worker population limit (WPL)): a maximum air concentration of 0.0001 mg/m3, averaged over an 8-hour work day. A person exposed for eight hours at the occupational limit would receive a calculated dosage of sarin of 0.048 mg-min/m3. Even a worker exposed to this maximum allowable concentration 40 hours per week for a working lifetime of 40 years should not show any adverse effects.
Both the US Army (Sidell, 1997) and CDC (1988) recommend the same general population limit: a maximum averaged air concentration of 0.000003 mg/m3 (averaged over 72 hours). If a person was exposed for 72 hours at this limit, the calculated dosage of sarin would be 0.01296 mg-min/m3. Even if a person was exposed to this maximum allowable concentration for 72 hours, there should not be any adverse effect (e.g., miosis or cholinesterase depression). This exposure limit applies specifically to non-workers, for example, members of communities located near chemical weapon incinerators. CDC reviewed the evidence for this limit in 1988, concluding that human health will be adequately protected from long-term exposure to sarin at the general population limit and that even long-term exposure to these concentrations would not create any adverse health effects (CDC, 1988).
6. Militarily Significant Exposure Concentrations
On the battlefield, military commanders expect nerve agent concentrations will be high enough to cause immediate effects, ranging from miosis to death, in the affected population. In those situations, commanders most immediate concern is to prevent or mitigate the effects of that exposure. Thus, the military requires nerve agent detectors to have sufficient sensitivity to detect smaller amounts of nerve agents than the amount causing miosis. The Ct for miosis effects for 50% of the population is approximately 1 mg-min/m3; the threshold dosage is 0.5 mg-min/m3, as described above.
People exposed to these and slightly higher concentrations of sarin vapor also may exhibit other mild, local ocular and respiratory symptoms, including runny nose, chest tightness, and dimmed vision with eye pain (Sidell, 1997). While these effects are not lethal, they are incapacitating; from a military point of view, a person exhibiting these signs and symptoms would be combat-ineffective. Recovery from exposure at low levels depends on the amount of sarin absorbed. Miosis can persist several days before pupils return to normal size, while maximum visual accommodation to darkness can take several weeks. At higher vapor exposures, more severe symptoms appear, including increased difficulty in breathing and increased mucous secretions (McNamara and Leitnaker, 1971). The exposed military member will have deteriorated from combat-ineffective to casualty status and may require medical treatment to survive. At lethal exposure concentrations (above the LCt50), intoxicated persons experience severe symptoms including unconsciousness, convulsions, flaccid paralysis, cessation of breathing, and, if not treated promptly, death.
C. Exposure Dose and Time of Exposure Relationship
Vapor exposure to toxic chemicals is expressed as the concentration-time product or Ct. This is known as Habers Law. The relationship does not hold for very short or very long exposure times (Sidell, 1997). Scientists determined that the LCt50 for sarin in laboratory animals (mice, rats, monkeys, dogs, cats, guinea pigs) increased by a factor of six as exposure time increased from 0.1 to 60 minutes; that is, it took a higher concentration of sarin as the time of exposure increased (McNamara and Leitnaker, 1971). Those authors suggested that the Ct relationship is not linear, especially for minimal effects, for exposures over 20 minutes. They also cite a report showing less cholinesterase inhibition in dogs exposed to sarin for 240 minutes than dogs exposed for 20 minutes to the same Ct (10 mg-min/m3).
Other scientists (Mioduszewski, et al., 1998) describe the variation in the 50% effective concentration (ECt50) with time. For a 2 minute exposure to sarin, the ECt50 is 15 mg-min/m3; as the exposure time is increased to 30 minutes, the ECt50 also increases to 23 mg-min/m3. This means that more sarin is required to produce the same effect than is calculated by simply dividing the Ct product by the time (15/30 = 0.5 mg/m3); dividing the increased ECt50 by the exposure time (23/30 = 0.76 mg/m3) shows the increased quantity of sarin required to produce the same effect.
In his review of nerve agents, Augerson (2000) describes animal data that show high amounts of rapidly acting agents like sarin are not cleared effectively by the body, but lower amounts produce metabolic processes that are capable of detoxifying sarin. This is offered as one possible explanation for the increased amount of sarin required to produce effects at low exposure concentrations.
D. Low-Level Toxicity Thresholds Used in this Report
The thresholds used to establish low-level exposure for the modeling in this report are: the First Noticeable Effects (FNE) level, the dosage below which no visible effect, including the most sensitive, miosis, becomes manifest; and the General Population Limit (GPL) concentration, the amount below which any member of the general population could be exposed (i.e., inhale) continuously for 24 hours a day, every day, for a lifetime (70 years) without experiencing any adverse health effects.
1. Threshhold Values
Sarins FNE dose was estimated as 1.0 milligram-minute per cubic meter of air (mg-min/m3) for short-term exposures. In 1988, the CDC approved a GPL for sarin defined as a concentration of 0.000003 milligrams per cubic meter of air (mg/m3) (CDC, 1988), a value various Army regulations and guidance documents have implemented (DAMS, 1990; Mioduszewski, et al., 1998). In 1998, the US Army ERDEC reviewed and confirmed this amount in a scientific re-evaluation of existing chemical warfare agent air standards (Mioduszewski, et al., 1998). The CDC and the Army did not originally establish airborne exposure limits for the agent cyclosarin. However, the 1998 scientific review calculated the GPL for this nerve agent (USACHPPM, 1999) by applying the ratio of relative toxicities between sarin and cyclosarin. Scientists consider sarin one-third as toxic as cyclosarin at low levels equal to the GPL (ratio 1:3), resulting in a GPL for cyclosarin of 0.000001 mg/m3. The DoD modeling team used these values in the 2000 modeling of the combined hazard area of these agents. At exposure levels that would produce an FNE, scientists consider GB one-half as toxic as cyclosarin (ratio 1:2); therefore, cyclosarins FNE is estimated as 0.5 mg-min/m3.
2. Calculation of Dispersion Model Input Values
Because the possible exposures that servicemembers experienced from the Khamisiyah Pit release involved much briefer periods than those for which the GPL was designed, the DoD team applied an adjustment factor more closely representing a single days exposure limit to the GPL to accommodate the difference in exposure duration. To use this adjusted value (AdjGPL) in the dispersion modeling, which divides time into five-minute increments, the concentration was converted to an exposure dosage rate (Ct) in units of mg-min/m3, even though the relationship is not linear for more than about 20 minutes. Assuming a linear relationship greatly enlarges the predicted hazard area. Table 1 presents the calculations that resulted in the dispersion model input values.
Exposure rate per time (daily "Ct")
3. Differences from 1997 Report Toxicity Values
In the 2000 modeling effort, the modeling team revised some toxicity values used in the 1997 CIA and DoD modeling report (CIA and DoD, 1997), using recently published information and the existing toxicological data more correctly. The 1997 modeling used a GPL assumed to be a 72-hour time-weighted average, which was how existing documents had described the GPL (NRC, 1997; DAMS, 1990). However, the 1998 re-evaluation study (Mioduszewski et al., 1998) clarified this value as a 24-hour time-weighted average. The 1998 report also clarified the GPLs toxicological basis that resulted in our applying an adjustment factor to account for lifetime assumptions. The resulting 2000 adjusted GPL Ct value, therefore, is higher than the previous estimate. Specifically, for sarin the 1997 Ct value was calculated as:
GPL (mg/m3) x exposure time (72 hrs x 60 min/hr) = 0.000003 x 4320 min = 0.01296 mg-min/m3
versus the 2000 calculation of:
AdjGPL (mg/m3) x exposure time (24 hrs/day x 60 min/hr) = 0.00003 x 1440 min/day = 0.0432 mg-min/m3
Finally, the 1997 modeling focused efforts exclusively on sarin and did not include cyclosarin.