Click for next page ( 44


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
CONCLUSIONS AND RECOM}NDATIONS In evaluating a group of materials that have similar applications, several criteria should be considered. In the current situation, where seven pesticides are registered for control of subsurface termites, the Committee has been asked to assess which, if any, of the materials are most appropriate for use in military rousing from the standpoint of heal th risks. Table 4 summarizes some of the important factors that the Committee has considered. Health-effects information of prime ry important e includes the critical responses in humans ant acute and chronic effects in animals. The environmental end points in Table 4 affect the hazard of these materisle by influencing the possible extent of exposure. Vapor pressure, persistence in the environment, and amount o f material that needs to be appl fed for optimal ef fectiveness al 1 influence the potential airborne concentrations in residences. Because chlordane is the most widely used termiticite in military housing, the following comparisons of risk are made with reference to it. COMPARIS ON OF CARC INOGENIC RISK There are insufficient data to determine whether carcinogenicity is the critical biologic end point in humans exposed to these pesticides, but available animal data allow some useful comparisons of risk. Chlo~dane, heptachlor, aldrin, dieldrin, and 1 indane have been tested in the National Cancer Institute carcinogenes is bioassay screening program, as discussed earl ier. Similar experimental protocols were used for each. Both sexes of Osborne-Mendel rats were fed the chemicals in the diet for up to 80 wk and observed for approximately 30 wk more before sacrifice. Both sexes of B6C3F1 mice were exposes to the chemicals in the diet for up to 8~) we and then observed for another 10-13 wk. The only consistent tumorigenic response was the occurrence of nepatocellular carcinomas in the male mice. This end point was used for the comparison of the carcinogenic activity of these cnemicale. The experimental data are summarized in TaDle 5. The EDlo (dosage producing a 1 iver tumor incidence 10 percent above background) was calculated for each termiticide with a procedure given by Crump et al. (1977~. The EDlo was chosen because it is the lowest effective dosage that can be estimated with satisfactory precision, owing to the size of the experimental groups. Estimates of the ED~,o. and their 95 percent confidence limits are in Table 6. Chlordane had an EDlo of 16 ppm. Heptachlor, aldrin, and dieldrin had approximately the same carcinogenic activity and were more potent in this bioassay than chlordane, on the basis of their EDlos. Lindane abowed the lowes t carcinogenic activity. On the basis of the EDlo, chlordane is approximately 6 times ( 103:16) as carcinogenic as 1 indane. Lifetime cancer risk and upper 95 percent confidence bounds on lifetime cancer risk at low toses have also been statistically 43

OCR for page 43
estimated for these pesticides (NRC, 1977b) . The upper bounds on lifetime cancer risk are shown in Table 7 and are expressed as probabilities of cancer after lifetime daily consumption of 1 L of water containing 1 fig of the compound. They have been corrected for species conversion on the basis of dose per unit surface area; other conversion factors, such as dose per kilogram of body weight, 8180 are available. me estimates of cancer risk were derived by using a probabilistic multistage model. At low doses, this model is often mathematically equivalent to the linear or single-hit model. If the precise mechanism of carcinogenesis is represented by a threshold or log-normal dose-response relationship, the multistage model may considerably overestimate the risk at low doses. However, this possibility cannot be reasonably quantified. Use of different risk-assesement procedure. could yield quite different estimates, and they rely to some extent on subjective judgment about the nature of the dose-response curves beyond the experimental data and on assumptions about the unknown effects of potential species differences in metabolism, physiology, and carcinogenic processes. Because of these uncertainties, some are of the opinion that the magnitude of cancer risk cannot be ret iably es timated (NRC, 197 7a) . An additional 1 imitation in interpreting the care inogenicity data is that the route of exposure in the bioassays is ingestion, whereas inhalation is the primary route of exposure of humane from pesticide use in termite control. However, within this limitation and with the EDlo. ant upper 95 percent confidence bounds on lifetime cancer risk, one can aseese The relative carcinogenic risk of these pesticides. The ranking from greatest risk to least risk would be: aldrin, dieldrin ~ heptachlor > chlordane ~ lindane. The carcinogenic potential of chlorpyrifos was investigated in CD-1 mice to which the compound was administered in the diet for 2 yr. There were no tumors relates to administration of chlorpyrifos. Direct comparison of the chlorpyrifo. data with chose on the chlorinated hydrocarbon tenmi~icides tested in the NCI bioassay ze not possible, because different strains of mice were used and a D - Xi~ tolerated dose was not used. One could make some comparison by esei~ting the hepetocellular~carcinoma rates above background in B6C3F1 mice for these compounds at the highest dosage of the chiorpyrifo. study. At 15.S ppe, which did not produce tenure in CD-1 mice exposed to chlorpyrifos, the following estimates would be obtained: Estimated proportion of 95 percent confidence Compound animals with tumors ~ imies Lindanc 0.016 0-0.03 Chlordane 0.10 0.06-0.14 Heptachlor 0.6 0.4-0.8 Aldrin and dieldrin are not included, because 15.8 ppm is above the maximum tolerated dose. At 15.8 ppe, heptachlor would be expected to yield a high rate of hepatocellular carcinoma. The estimated tumor rate of 10 percent for chlordane would not be expected to be - 44

OCR for page 43
significantly different from that in controls, if there were 50 animals per exposure group. Lind8ne would possibly not produce more tore than in controls at 15.S ppm. Therefore, under the test conditions used for chlorpyrifos, chlordane and lindane might also have given negative results for carcinogenicity. Obviously, one cannot predict from these data the behavior of chlorpyrifos in B6C3F1 mice or in CD-1 mice at nigher dosages. Data on the carcinogenicity of pen~cachlorophenol are not adequate for comparison with the other termiticides. OVERALL ASSESSMENT OF RISKS The four cyclod~ene termiticides are similar in overall health risk. Exposure of humans to cyclodienes produces effects on the central nervous system, characterized by se inures, tremor, and incoordination. CNS effects have been observed in cases of acute poisoning, as well as after exposure for longer periods. But data on chronic exposure are sparse, and there is no information on effects of long-term exposure at 1 ow airborne concentrations. A recent epidemiologic study of workers producing chlordane suggested that exposure has no long~eenn ef facts. However, there were shortcomings in the study, and more complete data are needed before firm conclusions with regard to long-term health risks can be reached. Each of the four cyclodienes has produced hepatocellular carcinomas in mice. lone carcinogenic rise varies to some extent; NCI mouse bioassays have suggested that Aldrich, dieldrin, and heptachlor have a greater carcinogenic risk than chlordane. However, the dif f erenc e is no t suf f ic lent to make one termit ic id e more des irabl e than another solely on the basis of health risk. In addition, these three pesticides are no more effective, and may even be less effective, for control of termites than chlordane. Another factor to be considered in comparing tenniticides is the expected airborne concentration after application. Chlordane is the only one of these tenniticides on which published analytic data were available to the Committee, although some preliminary data were available on aldrin and dieldrin. Vapor pressure can provide some additional indication of expected airborne concentration. Aldrin and dieldrin are leas volatile than chlordane. Therefore, although chlordane has a smaller carcinogenic risk, the possibility of greater airborne concentration might result in a greater haze rd than would be expected from health data alone. Lintene also has toxic effects on the central nervous system in hens. Sign. of poisoning include tremors, ataxia, convulsions, and prostration. Violent tonic and clonic convulsions have occurred in severe cases. In animals, exposure to lindens teas produced diarrhea, hypothermia, hyperirritability, incoortination, and convulsions. Extended exposure has produced nervous symptoms and fatty degeneration of the liver. Carcinogenicity data on lintane have been equivocal. On the Dasis of the EDlo estimated from the NCI mouse bioassay, the carcinogenic risk of lindens is considerably less than that of 45

OCR for page 43
chiordane. However, it is much more Probative than chiordane and persists in the environment only about half as long as chlordane. l~nerefore, lindane woult have to be applied more often to be as effective, and this would increase the potential for human exposure to it. In humans, pentachlorophenol also affects the central nervous system. Effects include lose of appetite, respiratory difficulty, anesthesia, hyperpyrexia, sweating, dyspnea, and coma. Animals exhibit symptoms associated with uncoupling of oxidative phosphorylation. In addition, pathologic changes in the 1 iver and kidneys have been observed. Unlike the other chlorinated hydrocarbons discussed here, pentachlorophenol has not been shown to be carcinogenic (although it was not tested under the same conditions). However, it has produced embryotoxicity and f etotoxicity in rate. Pentachlorophenol is not as persistent as chlordane and therefore would have to be applied more of Iced to be as effective. This would increase the likelihood of toxic effects, particularly from acute exposure. Culorpyrifos presents a somewhat different ~ ituationO Being an organopbosphate, it is in a different chemical class from the other pesticides discussed here. The principal effect in humans and animal. exposed for short periods is a reduction in plasma and red~cell cholinesterase activity. There is no information on effects of long-term exposure of humane. Ibe only observed effect in animals exposed for long periods was a decrease in cholinesterase activity. Carcinogenicity of chlorpyrifo. has not been observed in various animal species tested (although it was not tested under the same conditions as the other tenniticides). Although the rick of chronic effects may not be 88 great as thee for chlorine, a potential for acute effects exists. Because chlorpyrifos is not as persistent as chlordane and needs to be applied more often to be effective, there is a potential for higher airborne concentrations, which could increase the 1 ikel ihood of acute ef fecte. RECO - ENDAT IONS GUIDELINES FOR A _ ISIS 11~1~111" The airborne exposure li~te suggested here are intended to provide guidance in estimating the health risks of the pesticides in military housing. These are not standards 1 ike those suggested by the Occupetio~1 Safety and Health Administration, and they do not guarantee absolute safety. Given the available data and the fact thee under conditions of prolonged exposure of families in military housing there may be persons, such as young children, who in general are more susceptible to environmental insults, the Committee concluded that it could not determine a level of exposure to any of the termiticides below which there would be no biologic effects. The exposure lifts were derived on the basis of health considerations ant reflect the combined judgment of the Committee members; the feasibility of 46

OCR for page 43
achieving the suggested airborne concentrations was not taken into account. However, every effort should be made to minimize exposure to the greatest extent feasible. In deciding which, if any, of the termiticides are most appropriate for use in military housing, one should take into account not only the toxicity and suggested airborne exposure limits, but also other fac tore that would influence the extent of exposure and hazard. Some of these factors are discussed briefly in this report; they include vapor pressure, persistence in the environment, and amount of material that needs to be applied. The suggested guidelines for airborne exposure should be reviewed again as soon as additional health-effects data become available. en lordane In 1979, the Committee on Toxicology (NRC, 1979) suggested an interim guideline for airborne chlordane of 5 ~gJm3. This was a pragmatically derived concentration that was based on known concentrations of chlordane in Air Force housing, on a review of reported health complaints of residents of contaminated homes, and on a comparison with the acceptable daily intake derived from long-tenm anima1-feeding studies. The Committee stressed that additional data were needed to assess fully the human health risks of exposure to chlortane. In particular, the Committee recommended in 1979 an epidemiologic study of persons who had resided in military housing that had been contaminated in several previous episodes. The Committee recommended that attempts be made to correlate health effects with blood concentrations of chlordane. From an extensive review of the literature, the Committee has concluded that there are no new data that justify a change in the currently suggested airborne concentration of chlordane. The one recent epidemiologic study ~ Shindell and Associates, 1980) had 1 imitations, including a lack of reporting of airborne concentrations of chlordane and a sample size that limited the ability to detect small responses in the population. Al though this study suggested a trend in standard mortality ratio. for teethe due to cancer in workers with increasing duration of employment and chlordane was shown to produce hepatomas in mice (NCI, 1977a), these reports do not provide information on the health risks in humans and animals associated with various degrees of exposure to airborne chlordane. Because of the shortcomings of current data and in view of the request thee more definitive data be developed, the airborne concentration for chlortane of 5 ~g/~3 should be regarded as an interim guideline for exposures not exceeding 3 ye. This- 3-yr period is suggested with the expectation that it will provide adequate tin for the needed health data to begin developing. Heptachlor, Aldrin, Dieldrin On the basis of the EDlo. estimated from the NCI mouse bioassay data, hep~cachlor is approximately 3 ties as carcinogenic ant altrin and dieltrin 5 times as care inogenic as chlortane. For quantitative 47

OCR for page 43
comparison of health risks, these appear to be the best available data. Using these data, the Committee suggests interim guidelines for airborne heptachlor and airborne aldrinJdieltrin of 2 and 1 ug/D`3, respectively, for exposures not exceeding 3 yr. If is is assumed that all this inhaled material would be absorbed into the blood, the guidelines for airborne exposure compare favorably with the blood concentrations associated with no observed adverse effects. For example, with 100 percent absorption and assuming a blood volume of 5 L and a daily volume of inspired air of 20 m3, exposure of dieldrin at 1 ug/m3 would result in a blood concentrat ion of 4 ~ g/ L. Th is is probate ~ y a conservative es t imate, in that it is not 1 ikely that 100 percent of the inhaled material would be absorbed in the blood. Even with the possibility of accumulation from repeated exposure taken into account, Hunter et al. (1969) reported that, in volunteers given HEOD (pure dieldrin) at 50 ug/d for up to 2 yr, the blood concentration was only 5.0-8.6 ~g/L. A blood concentration of 4 ~ gJL is cons iderably lee s than the concentration of 105 ug/L reported to be a no-adverse~effect concentration for humans ~ Jager, 1970) . Li ndane Because the care inogenicity data on 1 indane are equivocal, the Committee does not believe that this information should be used as a basis for suggesting a guideline for airborne exposure to lindane relative to exposure to chlordane. In the absence of other data for estimation of the risks of exposure to airborne 1indane and because lindane is not now used to control termites ire military housing, the Committee does not suggest a guideline for airborne exposure. Pentachl orophenol Pentacnlorophenol is rarely used to control subsurface termites through soil application or injection, but rather is applied directly to wood. Because of the wide use of pentachlorophenol in ways other than as a eermiticite, because it is not now used to control termites in military housing, because of its complex toxicity, and in the absence of definitive information on effects of long-term exposure to airborne pentachloropenol, the Committee does not suggest a guideline for airborne exposure. Ch lorpyrifo. The Commit tee pre~riouly suggested a guideline for airborne chlorpyrifo. of 100 ~g/~3, applicable for 9 - d continuous exposure of Navy personnel in submarines (NRC, 1978~. That guideline was based on data from ingestion. In the absence of data on effects of long-term exposure to airborne chlorpyrifos, the committee concludes that the ingestion studies of fer the best available information from which to derive a guideline. However, because the population z~ military housing is more heterogeneous than that in submarines, the - 48

OCR for page 43
Committee BuggeSt8 an interim guideline of 10 ~g/~ for exposures not exceeding 3 yr. The Coomiittee recognizes the 1 imitations of extrapolating from ingestion to inhalation exposure, such as the potential wide variability of the respiratory rate in a heterogeneous population ant the absence of data with which to estimate the absorption factor for inhales chlorpyrifos, but believes it to be the best available approach. RESEARCH RECOM}NDAT IONS The Committee strongly recommends that more definitive hum~n-healtb data be developed for a better assesament of the risks o f exposure to these termiticides. With increasing reports of Truman exposure to some of these termiticites in military and civilian housing, a clearer understanding of the potential risks becomes even more important. To provide a stronger data base on which to compare these materials more ful ly, the f al lowing research agenda is recommended: i. The primary route 0 f human exposure to the seven termiticides is inhalation, and there are only minimal hewn or animal data on this route. It is possible That there will be variation between Ingestion and inhalation, e.g., in pharmacoKinetice, dose to critical organ, and toxicologic end points. D i f f erenc es b e tween routes o f adminis tra t ion c out d modi fy the re 1 at ive or absolute risks of these materials. Therefore, the Committee recommends 1 ong-term inhalation studies of these compounds. Biologic end points to investigate include neurotoxicity, carcinogenicity, effects on blood-forming tissues, and teratogenic and reproductive effects. Studies on the mechanisms of care inogenicity ( part icularly the cyclodienes) and neurotoxicity of these compounds should also be undertaken. In addition, the role of metabolism in influencing the toxic effects of the termiticides needs to be examined further. If resources are insufficient for the immediate testing of all seven materials, the Committee suggests that the first studies be done with chlordane, aldrin, and some of rbe noncyclodiene compounds, such as 1 indane and chlorpyrifos. Heptachlor and dieldrin would be expected to yield results qualitatively similar to those of chlordane and aldrin. The Committee also suggests investigation of the possibility of undertaking these studies jointly with other government agencies that are responsible for testing environmental chemical ~ . Airborne Monitoring. An important consideration in assessing the rip ides is knowledge of the airborne concentrations in residences after application. Published quantitative analytic data were available to the Committee only on chlordane; some preliminary data were available on aldrin and dieldrin. The Committee suggests a program to determine the airborne concentrations of the termiticite. under conditions ~ imilar to those now found in military housing. Monitoring should be conducted over an extended period to delineate the effects of time, temperature, and other variables on airborne concentrations of the termiticites. 49

OCR for page 43
Epidemiology Data. The Committee (NRC, 1979) has previously suggested an epidemiologic tudy of chl ordane in mil itary hous ing. persons who have been exposed to T.nis group can be followed more readily than the general population, because their heal Ah records and whereabouts are more easily traced. The Committee now reco~ende that, at a minimum, a biologic monitoring program be undertaken as a first step in a comprehensive analysis of the human health effects of chlordane. Concentrations of chlordane and its metabolites in fat, blood, and urine of persons known to have lived in military housing where culordane was applied should be measured. If data are available, these concentrations before and after exposure should be compared, to learn whether there ze appreciable accumu~ac~on or chlordane in the body. Observational data on the health status of these persons should also be obtained. Ld particular, in~restigatore ^~..1 ~ 1 ^^k Ens oi one of neurotoxicit~-8uch as seizures ~ movement disorders, tremors, and chorea--and for ~ igns of anemia and diseases of blood-forming tissues. Neurologic symptoms appear to be the most sensitive indicator of exposure in humane. It might be possible to develop a retrospective case~control study of those with suggestive necrologic symptoms and appropriate matched controls (neighbors, unaffected siblings, etc.), including correlation with concentrations of the termiticides in tissues and in indoor air. Any of the other termiticides for which there are sufficient data on exposure in military housing should be investigated ire a similar fashion. cam ~~ _ ~ _ ~~ .~ 50