Guidelines for Chemical Warfare Agents in Military Field Drinking Water (1995)

Lewisite is an outdated and ineffective organoarsenical chemical warfare agent. However, it might still be encountered in battlefield situations. The formulation of lewisite (chlorovinyldichloroarsine) as a war gas was done by W. Lee Lewis in 1918. It was considered an alternative to sulfur mustard, which had become the main chemical warfare agent in use by the end of World War I. Lewisite was used by the Japanese army during the Manchurian campaigns of the late 1930s and the early 1940s.

Lewisite is a potent blister agent. Like other blister agents, it not only produces casualties but also restricts use of terrain, hampers troop movements, and requires cumbersome protective gear (Somani, 1992). It is a colorless, oily liquid at room temperature with a faint “geranium-like” odor. It is more volatile than sulfur mustard and, therefore, can be used as a vapor over greater distances. To achieve greater effectiveness in combat, lewisite has been mixed with sulfur mustard. Because of its freezing point, lewisite is effective over a wider temperature range than sulfur mustard. Lewisite dissolves very slowly in water. The dissolved lewisite hydrolyzes rapidly to hydrochloric acid and lewisite oxide. It can form a surface film and globules that fall to the bottom of the water layer (Daniels, 1990c).

Lewisite causes painful blistering of the skin and eyes. If decontamination does not occur within 1 min, lewisite produces severe damage to the cornea, and permanent loss of sight can result. Reddening of the skin begins within 30 min, and blistering appears about 13 hr after exposure. As a vesicant, lewisite is about four times as fast-acting as mustard and is much less persistent (Somani, 1992). Lewisite' s effects are similar to those of sulfur mustard's, except that it is absorbed through the skin and acts as a systemic poison. Exposure to lewisite leads to pulmonary edema, diarrhea, restlessness, weakness, subnormal temperature, and low blood pressure. Prolonged exposure leads to severe pain in the throat and chest (U.S. Army and U.S. Air Force, 1975).

Trivalent arsenic is considered the component of lewisite that is primarily responsible for its vesicant and systemic toxicity; thus, field drinking-water standards for lewisite are expressed in terms of the arsenic fraction. In addition, the water-quality test currently used by the military does not detect lewisite directly; instead it detects the arsenic component.

Trivalent arsenic exerts its toxic effect by binding to sulfhydryl-containing proteins, especially enzymes, thus inhibiting pyruvate oxidation —a critical step in carbohydrate metabolism. The inhibition kills cells. The lipid solubility of lewisite also contributes to its toxic effects; trivalent arsenic readily penetrates skin, exerting its toxic action systemically and causing painful localized blistering. As a systemic toxicant, lewisite produces pulmonary edema, diarrhea, restlessness, weakness, subnormal temperature, and low blood pressure. Vascular damage, induced by lewisite, is partly responsible for effects such as blistering, tissue perforation, and hemorrhaging. Edema and hemorrhaging associated with lewisite exposure can lead to shock and death.

Human data concerning the toxicity of lewisite via the oral route of exposure are not available. However, there are limited toxicity data on lewisite ingestion from three animal studies.

In animals, ingestion of lewisite can produce acute inflammation of the mucous membrane of the stomach or intestine, which is characterized by hemorrhage, necrosis of epithelium, and submucous edema. Developmental effects have been reported in pregnant rats and rabbits exposed

to lewisite by intragastric intubation. Rats were exposed for 10 days (on days 6-15 of gestation), and rabbits were exposed for 14 days (on days 6-19 of gestation). A no-observed-adverse-effect level (NOAEL) of 0.016 mg/kg/day in rabbits and 1.5 mg/kg/day in rats was identified (Hackett et al., 1987b). The NOAEL of 0.016 mg/kg/day in rabbits was selected for developing the proposed field drinking-water standards for lewisite.

There are no data on chronic toxicity resulting from the ingestion of lewisite.

Lewisite, as an arsenical, might be carcinogenic, although no specific studies were found in which the carcinogenicity of lewisite was evaluated. There is evidence that arsenic might act as a cocarcinogen and promote the carcinogenic process. It is capable of producing DNA damage; however, direct tests of its mutagenic potential have been inconclusive.

The effectiveness of lewisite as a chemical warfare agent depends in large degree on whether toxic doses can be produced in the battlefield. Field experiences indicate that doses large enough to affect military operations are probably not attainable with any reasonable expenditure of munitions (Gates et al., 1946). Neither saturation of fields nor delivery of thickened lewisite vapors through bomb and airplane spray has proved to be effective. This lack of sufficient exposure to lewisite might also account for the dearth of epidemiological data on the health effects of lewisite, although the literature on the toxic effects of arsenic in humans is abundant. Given the limited epidemiological data, the proposed drinking-water standards for lewisite (Daniels, 1990c) were derived from three animal studies—a sparse data base.

The shortcomings of those animal studies for predicting the effects in humans exposed to lewisite in water, and, in particular, in military personnel exposed to field drinking water, have been clearly recognized (Daniels, 1990c). For example, the obvious advantages that experimental studies offer, such as the ability to control genetics and diet and the opportunities for more intensive observations, are counterbalanced by the uncertainties of interspecies extrapolation. Therefore, biologically plausible assumptions, including best quantitative estimates, have been used to arrive at the proposed standards. In addition, the animal studies were evaluated in light of all relevant data on the health effects in question, the host, and the chemical itself. However, there are major data

gaps on the toxicity of lewisite, and the usefulness of most of the studies is limited by the lack of a satisfactory animal model. Little information exists on the reaction of lewisite with biologically important molecules, although it is reasonable to assume that, as with sulfur mustard, DNA is a major target (IOM, 1993).

There are no adequate data on the acute effects of lewisite following dermal exposure. Very little is known about its specific effects on skin, and data on its absorption, disposition, and excretion following dermal exposure are minimal. Microscopic examination of affected skin has not been pursued extensively.

The proposed Army standards for exposure to lewisite in field drinking water were derived from a rabbit study, in which the NOAEL was estimated to be 0.016 mg/kg/day (Hackett et al., 1987b). Converting the arsenic fraction of 0.0058 mg/kg/day to total consumption per day for a 70-kg human consuming 5 L or 15 L of water per day gives (0.0058 × 70)/(5 or 15 L) = 0.08 and 0.027 mg/L, respectively. No additional safety factors were assumed. Whether the rabbit is the species most sensitive to lewisite is not known, and the applicability of the rabbit data to the human situation requires further evaluation. It is recommended that future consideration should include a comparative analysis of pharmaco-kinetics, metabolism, repair mechanisms, and genetic constitutions.

Neurological effects following acute exposure to lewisite have not been documented in animals. Acute exposure to high concentrations of lewisite leads to a shock syndrome that is thought to result from increased capillary permeability (Goldman and Dacre, 1989). No direct evidence exists that lewisite might cause neurological problems in humans, although arsenic is considered a neurotoxin and peripheral neuropathy has been reported in humans following a single arsenic exposure (Le Quesne and McLeod, 1977).

Given the limited data on the acute health effects associated with exposure to lewisite through ingestion of water by animals and humans, the subcommittee concludes that the Army's proposed interim standards for lewisite in military field drinking water are sufficient to reduce the risk of lewisite-induced health effects or performance degradation.

The Army's proposed short-term drinking-water standards for lewisite are 220 and 75 µg/L (or 80 and 27 µg/L if expressed as the arsenic fraction), assuming a water consumption of 5 L/day and 15 L/day, respectively. Those standards represent an approximate 10-fold reduction from the previous drinking-water standard for lewisite of 2 mg/L (for 5 L/day consumption). It is further recommended that accompanying guidance for unit commanders underscores the following points:

• No human data are available on the health effects of lewisite following ingestion.

• These drinking-water guidelines are based on extrapolations from limited animal studies.

• Providing no other toxic compounds are present, acute effects are not expected at the assumed levels of water consumption.

• Some evidence suggests that lewisite might be carcinogenic (CDC, 1988); as a consequence, there might be some risk of developing malignancy later in life following lewisite exposure at the guideline levels.

At present, there are no field drinking-water monitoring capabilities that can reliably detect lewisite or elemental arsenic at the recommended concentrations. Therefore, the subcommittee recommends that field monitoring techniques be developed for detecting low concentrations of arsenic and lewisite.