Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Applicator Exposure to Pesticides JOHN E. COWELL Monsanto Agricultural Company In handling chemicals for laboratory, agricultural, or domestic use, there is always an element of risk. The eminent toxicologist John Doull once said, "The safest chemical in existence can be handled dangerously, and conversely the most dangerous chemical can be handled safely." The issue of pesticide use reflects this statement. Pesticides are of great benefit in the production of foodstuffs and are widely applied. But while some very toxic insecticides are routinely applied with no adverse affects to the applicators, chronic exposure to pesticides and potential long-term health effects are always of concern. This paper discusses techniques for measuring exposure to the applicator during the agronomic use of pesticides. Since the 1950s, scientists have been concerned about the mea- surement of worker exposure to pesticides. In 1962, Durham and Wolfe reviewed the methods for measurement of the exposure of workers to pesticides. In 1985, Turnbull edited the book "Occupa- tional Hazards of Pesticide Use" which reviews the literature on the subject from 1951 to 1984. There are two basic approaches to measure exposure. One ap- proach is to measure the external deposition of pesticides on the worker's body. This dermal deposition approach is utilized pri- marily when little information on the pharmacokinetics (absorption, metabolism, and excretion) of the chemical is available. The second approach is biological monitoring for the pesticide or its metabolites 156
157 in the worker's body fluids, usually urine or blood. Discussed be- low are the ways that these two approaches address measurement of exposure through the respiratory, dermal, and oral routes. Also considered are other related studies and their use in the calculation of an estimated worker body dose. PASSIVE DOSIMETRY STUDIES Dermal absorption is the major route of pesticide exposure dur- ing the agricultural application sequence. Dermal exposure is usually estimated by the "patch technique" originated by Durham and Wolfe which involves attaching 10 to 12 absorbent pads at various locations on the worker's body. Patch techniques typically measure deposition on a small 100 square centimeter gauze pad, and the amount of pesticide deposited on the pad is multiplied by an appropriate fac- tor to represent the larger skin surface area where it was affixed. These pads tend to overestimate exposure by trapping some liquid which undoubtedly would be displaced from the skin surface during physical activity. In addition, actual pesticide deposition is highly variable and estimates depend upon whether splashes hit or miss the pad. If a splash or a few droplets hit the pad, then the extrapolation to the whole areaâwhich amounts to as much as 35.5xâcould be significantly overestimated. Conversely, if a splash or droplets miss the pad, the extrapolation could be underestimated. Thus, this der- mal deposition technique requires many replicates to provide reliable estimates of exposure. An alternative approach to measuring dermal deposition is the disposable coverall. Some researchers have removed the coverall garment and sectioned it into parts such as the legs, above and below the knee; the arms, above and below the elbow; and the front and back of the torso. A patch still needs to be used on the head. Contamination in the removal and sectioning process is a concern, and the analyses of these large sized samples cause numerous analytical problems. To assess the protective value of clothing, some patches are placed inside the clothing and compared to patches placed in ad- jacent areas outside the clothing. Protection factors estimated in this manner are highly variable depending upon the type of clothing worn, resulting in order-of-magnitude differences in similar studies. Pesticide exposure to the forearms and hands account for ap- proximately 70-90 percent or more of total dermal exposure. Hand
158 contamination has been estimated from analysis of hand rinses and of thin cotton absorbent gloves worn during an operation. A crit- icism of the hand rinse technique has been that it only removes the unabsorbed residue of the deposition and may underestimate exposure. On the other hand, cotton gloves have been criticized for overestimating exposure due to absorption of more liquid than would normally adhere to the skin. Many studies have shown that most of the dermal exposure occurs to the hands in mixing, loading, or tank-filling operations. Rubber gloves are used to prevent hand exposure, and numerous studies have been aimed at determining glove protection factors. Also, glove material studies have been performed with diffusion cell apparatus to investigate which materials provide the most resistance to penetration. The results show that most rubber glove materi- als are adequate for the duration of worker exposure to pesticides. Indeed, most exposure to the hands results not from penetration but from the glove removal process. Data to support this contention stem from analyses of cotton gloves worn under rubber gloves. Signif- icantly higher exposures have been shown on one hand as one rubber glove is removed with a rubber gloved hand and the other rubber glove is removed with a bare (cotton gloved) hand. Thus, the U.S. Environmental Protection Agency (EPA) is recommending that the rubber gloves worn during pesticide use be rinsed with water before removal. Inhalation exposure estimates have been made in passive dosime- try studies by analyzing filter pads attached to respirators worn by workers. However, there are major disadvantages in monitoring respi- ratory exposure by pads: the pads themselves may not be an efficient trapping medium for the pesticide studied; the pads are easily con- taminated by the worker's hands; and the pads often become very wet due to exhaled moisture which could lead to hydrolysis of some chemicals. A preferred alternative approach has been the measure- ment of the airborne concentration of the pesticide in the breathing zone of the worker. The concentration is then multiplied by an esti- mated respiratory rate. This has been accomplished through the use of air sampling pumps which draw a known volume of air through a collecting material that traps the pesticide. Personal air-sampling pumps have been found to be most convenient for this sampling. The sampling media usually have been absorbents such as silica gel, activated charcoal, florisil, alumina, and polyurethane foam plugs. When the inhalation exposure is calculated with analytical data
159 from the trapping media, 100 percent human absorption is assumed. Oral exposure, whether by ingestion of air participate or dermally contaminated food, is not accounted for in the passive dosimetry studies. In the case of dermal exposure, 100 percent human absorption is assumed unless additional studies have shown otherwise. In vivo an- imal dermal penetration studies have been used by some researchers for this purpose. The hair of the back or abdomen skin of the test animal is lightly shaved to permit efficient application of the pes- ticide to the skin. The pesticide is applied and after an interval washed from the skin surface. Accountability is achieved by adding the amount of pesticide residue excreted in the urine and feces to the amount washed from the skin. Total accountability is often poor with these studies, so the amount of pesticide which is not accounted for is added to the amount absorbed for a conservative estimate. Some researchers have used in vitro human skin studies as an al- ternative to estimate the amount of percutaneous absorption. Since absorption through the membrane barrier (stratum corneum) of the skin is presumed to be a passive diffusion process, no elaborate conditions to maintain the physiological state are required. Freshly obtained abdomen skin is used in a two-chambered diffusion cell ap- paratus where similar fluid is placed on both sides of a membrane, and the diffusion of a compound from one side to the other is ob- served. A more representative apparatus developed by Franz has a one-chambered cell with the stratum corneum surface of the skin exposed to the environment. The receptor fluid is continually circu- lated and maintained at a physiological temperature. The in vitro technique also suffers from a lack of total accountability of the pes- ticide due to such physical parameters as volatility, solubility, and skin uniformity. In general, comparisons of in vivo and in vitro techniques have shown poor correlations. It is known that considerable variation in dermal absorption is a result of inconsistent penetration through skin at various anatomical sites. For instance, according to Maibach the penetration rate for foot (plantar) skin is approximately 300-fold less than for scrotum skin, and the rate for back skin is one-half that for palm skin. These studies provide the final piece of information that allows the calculation of the absorbed percentage of deposited pesticide which, when added to the inhalation exposure, estimates the total body dose.
160 To assure the quality of the data resulting from a passive dosime- try study, several measurement parameters should be included. Con- trol samples should be exposed for the expected test interval to the ambient environment at the test site prior to the operations. When analyzed, these control samples provide an ambient background level. Other control samples should also be fortified with the pesticide at expected levels and handled, transported, and stored in the same manner as the study samples. These fortified samples can then be analyzed simultaneously with the study samples to provide a cor- rection for transport and storage stability as well as for analytical recovery. There are several advantages to a passive dosimetry approach. The chemical analysis of the parent active ingredient of the pesticide is relatively simple. The tests are completed after the operation, and there is no need to continue monitoring the worker. Each opera- tion such as tank mixing and loading or application can be tested individually even if performed by the same worker. Finally, the pharmacokinetics of the pesticide need not be known. BIOLOGICAL MONITORING STUDIES Some biological monitoring studies have measured specific ef- fects of acute exposure such as the organophosphorus depression of cholinesterase. In these studies blood samples for cholinesterase are collected by standard venipuncture techniques. The most recent bio- logical monitoring studies consist of measuring urinary pesticide and metabolite levels. Biological monitoring by chemical analysis of the worker's urine, in contrast to passive dosimetry on the worker or ambient monitoring of the environment, directly evaluates the amount of chemical that is absorbed by the body as an internal dose. Biological monitoring takes into account many factors which have to be estimated in tangential studies or assumed when approached by the passive dosimetry tech- nique. For instance, biological monitoring automatically accounts for inhalation, oral exposure, clothing protection, and percutaneous absorption. However, sufficient metabolic and pharmacokinetic data are a prerequisite for providing a quantitative measure of body dose. The optimal preparation for a biological monitoring study is a human pharmacokinetic study when absorption, metabolism, and excretion of the pesticide are intensively studied. Most of the time, this is not a viable option because of ethical considerations. In lieu
161 of human data, three types of animal studies can be used to demon- strate the probable pharmacokinetic behavior of the chemical under study. First, an oral 14C-radio-labeled pesticide dosing study with rodents, which provides sufficient amounts of excretion products for metabolite identification purposes, is typically required for pesticide registration. These data then serve as a basis to develop analytical methods to monitor the biological specimens for pesticide residues. The second type of study involves an intravenous (IV) injection of 14C-labeled pesticides to an animal such as the monkey, which has shown to be a good model for man. It is used to determine excretory recovery and distribution of the total amount of the chemical between urine and feces. These data are used to provide the necessary correc- tion factors for interpreting the data developed in subsequent dermal monkey studies. The third type of study is a dermal application of 14C-labeled pesticides to monkeys, which provides percutaneous absorption data regarding the percent of residue in the monitored matrix from a dermal dose. It also confirms the IV excretion pro- file if enough pesticide is absorbed through the skin. These three types of studies form a good scientific basis for an accurate biological monitoring approach to applicator exposure. To conduct a biological monitoring study, several quality control measures are necessary to insure a scientifically valid study: ⢠Pre-exposure biological specimens should be collected from the test subjects and analyzed to obtain a reliable baseline which enables establishment of the Limit of Detection (LOD) and the Limit of Quantitation (LOQ). ⢠Pre-exposure biological specimens should be fortified with the pesticide active ingredient and representative metabolites at the field site prior to the testing. These fortified controls are frozen and shipped to the laboratory along with the study samples. When analyzed simultaneously with the study samples, the for- tifications are used to correct for transport, storage stability, and analytical recovery. ⢠Urine is the most commonly employed biological specimen for these studies. Based on pharmacokinetic studies, all urine speci- mens should be collected for the determined interval of excretion plus an additional baseline time period. To avoid an excessively large number of urine samples from this collection, specimens can usually be combined into 12- or 24-hour composites for analysis. ⢠Test subjects are requested to avoid subsequent exposure to the
162 pesticide during the urine sample collection period in order to maintain the integrity of the results of the test operation. One of the disadvantages of the biological monitoring approach is that the analysis of urine or blood specimens for pesticides and metabolites is a much greater analytical challenge than the parent analysis involved in the passive dosimetry technique. The urine matrix is especially variable from individual to individual and even within an individual's profile due to dietary inconsistency. Quite often extensive cleanup procedures are needed to remove interfering components and/or chemical treatments are needed to convert many metabolites of a particular pesticide into a common chemophore. Calculation of body dose from a biological monitoring study is achieved by multiplying the amount of chemophore quantitated in each sample by the volume of the composite specimen and totaling the composite results. Totals are then normalized for body weight of each individual and the amount of pesticide applied by the individual. They are then corrected for the amount present in the monitored matrix as determined by the pharmacokinetic studies. Several biological monitoring studies have demonstrated that there are undetectable pesticide residues in donor specimens. Many researchers feel that body dose estimates should be reported as a range rather than an average. In the case of low exposures of toxico- logically significant pesticides, it is important to obtain an estimate of body dose from an undetectable residue. There are three approaches used to estimate these undetectable residues. The first approach, and the most conservative, is to assume the detection limit value. However, this value, when multiplied by the urine volume and plot- ted, demonstrates an excretory pattern that quite often can be very misleading and usually correlates only with the volume of voided urine. The second approach is to assume zero for these undetectable residues, even though they are unlikely to be zero but could probably be considered insignificant. If there is a real need for close approx- imation of the actual body dose, then a third and perhaps more reasonable approach can be taken, whereby a typical excretion curve drawn from measurable human applicators or animal models from dermal penetration studies is applied to the undetected data with the maxima set at the lower limit of detection. Overall, science has contributed significantly to the improvement of measurements of applicator exposure to pesticides, particularly during the past ten years. However, most attempts to compare the two approaches of passive dosimetry and biological monitoring for
163 measuring applicator exposure have not proven fruitful. In general, most indications are that passive dosimetry tends to overestimate worker exposure while, if the pharmacokinetics of the pesticide are known, a biological monitoring approach presents the most complete picture for assessing the body dose and ultimately calculating the risk to the applicator. REFERENCES Chester, G. and R.J. Ward. 1984. Occupational exposure and drift hazard during aerial application of paraquat to cotton. Arch. Enviro. Contam. Toxicol. 13:551-563. Cowell, J.E., and S. Dubelman. 1986 (accepted). Study of applicator exposure to alachlor by biological monitoring technology. IUPAC Symposium Series Publication. Davis, J.E. 1980. Minimising occupational exposure to pesticides: personal monitoring. Residue Reviews 75:34-50. Durham, W.F., and H.R. Wolfe. 1962. Measurement of the exposure of workers to pesticides. Bui. Wld. Hlth. Org. 26:75-91. Dubelman, S., R. Lauer, D.D. Arras, and S.A. Adams. 1982. Operator exposure measurements during application of the herbicide diallate. J. Agric. Food Chem. 30:528-532. Freeborg, R.P., W.H. Daniel, and V.J. Konopinski. 1985. Applicator exposure to pesticides applied to turfgrass. ACS Symposium Series 273:283-295 Franklin, C.A. 1984. Estimation of dermal exposure to pesticides and its use in risk assessment. Can. J. Physiol. Pharmacol. 62:1037-1039. Franz, T.J. 1978. The finite dose technique as a valid in vitro model for the study of percutaneous absorption. Curr. Probl. Dermatol. 7:58-68. Gold, R.E., J.R.C. Leavitt, T. Holcslaw, and D. Tupy. 1982. Exposure of urban applicators to carbaryl. Arch. Environ. Contam. Toxicol. 11:63-67. Lavy, T.L., and J.D. Mattice. 1986. Progress in pesticide exposure studies and future concerns. Toxicology Letters 33;61-71. Maibach, H.I., R.J. Feldmann, T.H. Milby, and W.F. Serat. 1971. Regional variation in percutaneous penetration in man: Pesticides. Arch. Environ. Health 23:208-211. Marzulli, F.N., and H.I. Maibach. 1983. Dermatotoxicology. Second Edition. Washington, D.C.: Hemisphere Publishing Corporation. Putnam, A.R., M.D. Willis, L.K. Binning, and P.F. Boldt. 1983. Exposure of pesticide applicators to nitrofen: Influence of formulation, handling systems, and protective garments. J. Agric. Food Chem. 31:645-650. Reinert, J.C., C. Nielsen, C. Lunchick, O. Hernandez, and D.M. Mazzetta. 1986. The U.S. Environmental Protection Agency's guidelines for applicator exposure monitoring. Toxicology Letters 33:183-191. Robbins, A.L., D.F. Nash, and S.W. Comer. 1977. A monitoring study of workers in a central Washington orchard. Bull. Environ. Contam. Toxicol. 17:233-240. Turnbull, G.J. 1985. Occupational hazards of pesticide use. London: Taylor and Francis.
164 Webster, R.C. and H.I. Maibach. 1985. In vivo percutaneous absorption and decontamination of pesticides in humans. Journal of Toxicology and Envi- ronmental Health 16:25-37. Winterlin, W.L., W.W. Kilgore, C.R. Mourer, and S.R. Schoen. 1984. Worker reentry studies for captan applied to strawberries in California. J. Agric. Food Chem. 32:664-672. Wojeck, G.A., J.F. Price, H.N. Nigg, and J.H. Stamper. 1983. Worker exposure to paraquat and diquat. Arch. Environ. Contam. Toxicol. 12:65-70. World Health Organization. 1982. Field surveys of exposure to pesticides. Standard Protocol: Technical Monograph N 7. Geneva, Switzerland.
