- Email: [email protected]
Hazard identification and risk assessment of pyrethroids in the indoor environment
Toxicology Letters 107 (1999) 193 – 199
Hazard identification and risk assessment of pyrethroids in the indoor environment Ju¨rgen Pauluhn * BAYER AG, Institute of Toxicology, 42096 Wuppertal, Germany Accepted 31 January 1999
Abstract Household insecticide products raise several important considerations concerning safety. These are related to the use of insecticides by untrained individuals, the difficulty of controlling the use of these products once purchased by the consumer and the potential exposure of the very young and very old, possibly with or without pre-existing pulmonary disease. Exposure to pyrethroids contained in mats or vaporizers, being slow release systems, have particular potential for long-term low-level exposure whilst for foggers, spray-cans or sprayed formulations the short-term high-level exposures may be of more concern. According to the volatility of the active ingredient contained in the household insecticide, its persistence in a non-inhalable matrix, i.e. sedimented house dust, may be short or long for highly volatile or low volatile active ingredients, respectively. On the other hand, the potential of exposure is apparently just reciprocal. This demonstrates that the extent and duration of exposure may be highly product-specific. Accordingly, the extent of exposure has to be accounted for and for risk assessment both concentration-dependent (e.g. sensory irritation) as well as concentration×time (= dose) related effects have to be considered and addressed in adequate bioassays. The issue as to whether pyrethroids adhering to house dust is of concern has been addressed in a model study using carpets treated with pyrethroids. This study has demonstrated that the total mass of pyrethroid applied to the carpet and that brushed off within an 18-h period is too small to be of any relevance for risk assessment. Therefore, assessment of health hazards in the indoor environment based simply on methodologies of emptying the household vacuum cleaner and analysing its content, which addresses contamination only, rather than examination of the actual airborne concentration, including other relevant airborne materials, is prone to tremendous errors and misjudgments. Due to the many substances potentially present in house dust and indoor air, e.g. bioaerosols originating from animals, pests and microorganisms, volatile organic substances (VOCs) or metals, prudent expert judgment is needed to assess the relevance of analytical findings. The complex indoor exposure scenario makes it especially difficult to causally relate clinical and epidemiological findings to arbitrarily selected indicator substances contained in a matrix not readily available to inhalation exposure. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Hazard identification; Risk assessment; Pyrethroids; Indoor environment; House dust; Sensory irritation
* Tel.: +49-202-368909; fax: + 49-202-364589. 0378-4274/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 4 2 7 4 ( 9 9 ) 0 0 0 4 7 - 8
194
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
1. Introduction Synthetic pyrethroids are used in agriculture and to a large extent also in the domestic environment for numerous applications. The favorable properties of this class of insecticides have promoted widespread application in virtually all sectors of food protection and pest control. They have also been proven to be useful as space and surface sprays in common houses, airplanes, in active or passive evaporators, in incense products, or to make textiles insect resistant (e.g. uniforms, bednets, carpets). With regard to effectiveness and toxicity, synthetic pyrethroids appear to be the first choice insecticides for this type of use pattern because pyrethroids are much more effective against a wide spectrum of pests than the organochlorine, organophosphate, and carbamate insecticides. Synthetic pyrethroids act directly on the nerve membrane through interference with the sodium channel gating mechanism that underlies the generation and conduction of each nerve impulse. This mode of action is more pronounced in insects than in mammalian species and demonstrates a negative temperature coefficient, i.e. lowering of the ambient temperature causes or promotes the appearance of toxic signs. This means, especially in insects and in the coldblooded vertebrate nervous system the neuroexcitatory mode of action of pyrethroids is more pronounced than in homeothermic laboratory animals or humans. The further prolongation of sodium current by pyrethroids at a lowered temperature accounts for the prolongation of repetitive nerve impulse trains. In conclusion, pyrethroids prolong the open time of voltage-dependent sodium channels, resulting in a prolongation of the inward sodium current during excitation. This effect accounts for the repetitive activity of sense organs at cooler body locations, for example the facial skin (Aldridge, 1990; Vijverberg and van den Bercken, 1990; Ecobichon, 1991). In insects, the major metabolism is related to an oxidative attack, whilst in higher animals and humans detoxification is mediated via hydrolytic (carboxylesterases) and oxidative actions. This ex-
plains why inhibitors of the oxidative metabolism increase the potency of pyrethroids in insects synergistically, whereas in mammalians appreciable effects had not been observed. The rapid metabolic degradation of pyrethroids in mammalian systems ensures against bioaccumulation or biomagnification. Also toxicokinetic mechanisms are described to enhance the selective toxicity of pyrethroids to insects. In laboratory animals and humans, due to the high lipophilicity of this class of chemicals, the toxicity is highly dependent on route of administration or exposure, which is highest by injection and lowest by the dermal route. Following inhalation exposure, major barriers which might prevent or protract absorption are absent thus making the inhalation route of exposure particularly important. One notable form of toxicity associated with synthetic pyrethroids has been a cutaneous paraesthesia and respiratory sensations mainly observed in workers spraying pyrethroids in crops or in occupational settings (Tucker and Flannigan, 1983; Moretto, 1991). This means that exposure to higher concentrations or doses of pyrethroids often causes hypersensitivity to sensory stimuli, and a number of compounds have been demonstrated to induce tingling sensation in the directly exposed skin. The time course of these sensations is usually immediate or within a few minutes after contact. The respiratory sensations are known to occur more frequently in context with the use of the more potent a-cyano pyrethroids, also known as type II pyrethroids. Thus, these respiratory sensations appear to occur particularly following brief, high-level peak exposures rather than longterm low-level exposure of humans. An approach to assess the acute respiratory tract sensory irritation potential of pyrethroids in a highly controlled and standardised animal model is described below. This bioassay provides a readily available means to assess whether irritation of the upper and/or lower respiratory tract is likely to occur (ASTM, 1984). Public perception and concern may trigger complex and demanding toxicological examinations and, as far they are considered to be part of the registration process of the product, standardisation of such complex studies with regard to the
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
generation of test atmospheres, mode and duration of exposure, and selection of adequate toxicological endpoints is challenging. However, in order to satisfy the concerns of the public as well as of the regulators who strive to achieve to protect the public from hazardous exposures from household insecticides, the described approaches can serve as a prerequisite for an up-to-date and scientifically sound risk assessment. In this instance, the toxicological assessment of pyrethroids adhering to house dust appears to be most challenging since the shuttle function of such particulates may facilitate exposure to non-volatile agents, at least on a hypothetical basis. In this context, the analysis of exposure levels causing sensory irritation, sensations causing discomfort or unpleasant odour perception are equally important. Thus, the objective of the studies presented is to analyse the suitability of various regimens to evaluate for what type of household insecticide which bioassay provides data most relevant for risk assessment. Particular emphasis is made to assess the extent of exposure originating from pyrethroids bound to house dust.
2. Test models
2.1. Upper respiratory tract sensory irritation potential of pyrethroids Stimulation of the trigeminal nerve endings located in the respiratory epithelium of the nasal mucosa by irritant airborne substances causes a variety of reflex reactions. Sensory irritation includes both nasal pungency and eye irritation. Sensory nerve endings respond to chemical as well as physical stimuli. These nerve endings respond to a variety of chemicals and stimulation can be dissociated from smell or taste. Reflex reaction originating from this stimulation are considered to be of a protective nature, i.e. to limit contact to a variety of potentially hazardous chemicals. Although other nerves are involved with the effects of sensory irritants, e.g. vagal nerves, the trigeminal nerve appears to be the only nerve directly involved with the decrease in respiratory rate in laboratory rodents. The decrease in respiratory
195
rate appears commonly within a few minutes, depending on the chemical. Thus, the sensory irritation potential of airborne substances is of crucial importance in regard to air quality, especially indoor air quality. These reactions to airborne materials are of great interest to toxicologists because they are indicative of a chemical’s likelihood to irritate and potentially damage the respiratory tract and are easy to measure in experimental animals. Using mice or rats, animal models have been developed based on their characteristic effect on breathing patterns. The concentration causing a decrease of respiratory rate of fifty percent (RD50) provides a means to compare the relative irritating potency of exposure atmospheres (Alarie, 1973; Bos et al., 1992). Utilising a 45-min single exposure regimen for rats, for cyfluthrin an RD50 value of approximately 50 mg/m3 was determined. The measurements made with aerosolised cyfluthrin used spontaneously breathing, conscious rats in modified nose-only exposure restrainers modified as plethysmographs. Prior to measurements of respiratory rate, tidal volume and apnea time during a 45-min to 1-h exposure period to the pyrethroid, the animals were allowed to acclimatise to the exposure conditions for 15 min, followed by an exposure and recovery period of short duration (Fig. 1). This experimental finding demonstrates that the a-cyano-pyrethroids especially appear to evoke functional changes in the breathing pattern (reflexively induced bradypnoeic response) which is a characteristic finding of agents acting as upper respiratory tract sensory irritants. Furthermore, from Fig. 1 it is apparent that the onset of changes is immediate and that no exacerbation of effects occurs following extended duration of exposure. This means that the change in breathing patterns is a concentrationrather that dose-dependent phenomenon. Based on analysis of the reflexively increased duration of the pause between inspiration and expiration (which leads to an increased apnea time), for this pyrethroid, the sensory irritant threshold concentration was estimated to be approximately 0.1 mg/m3 air (single 1-h exposure). Respiratory measurements were also made repeatedly during a subacute 4-week inhalation study.
196
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
Fig. 1. Average respiratory rate, minute volume and apnoea time measured in flow nose-only plethysmographs on male rats (n=8) during a 1-h exposure period. Prior exposure to 80 mg cyfluthrin/m3; base-line data were collected during an air-exposure period of 15 min ( = 100%).
From this study it is apparent that repeated exposure of rats was tolerated without time-related alterations in responsiveness (Fig. 2) even when exposed to highly effective RD50 concentration. This means that from the data shown in Fig. 2 it emerges that the single 45-min to 1-h and 5 ×6-h/
Fig. 2. Repeated nose-only exposure of rats to actual concentrations of 0 (air), 0 (vehicle), 0.44, 6 or 47 mg pyrethroid/m3 air (exposure 6 h/day, 5 days/week for 4 weeks). In each group measurements of respiratory rate were made on four rats/ group on 1 day per week during weeks 2, 3 and 4. WP10%: single 45-min exposure of rats to a dust containing 10% cyfluthrin. The respective airborne concentrations were 5.9 mg cyfluthrin/m3 and 76 mg TSP/m3 (TSP: total suspended particulate matter). The decrease in respiratory rate as a result of exposure to the vehicle was taken into account for the calculation of the sensory irritant threshold concentration.
day 4-week studies yielded virtually the same NO(A)EL (no observable adverse-effect level) and susceptibility did not change during the course of study. In an ancillary study rats were exposed for 45 min to dust containing 10% cyfluthrin (WP 10%); the resultant concentration of the pyrethroid in the vicinity of the rats’ breathing zone was approximately 6 mg/m3 air. As illustrated in Fig. 2, the magnitude of response following this single exposure was similar to that observed following 5× 6-h/day 4-week exposure. This finding demonstrates that the carrier-matrix of the pyrethroid does not appear to play any significant role. Taking into account the accumulated experimental evidence from single 45-min to 1-h, repeated 4-week and 13-week (6-h/day for 5 consecutive days/week) inhalation studies it can be demonstrated that the respective NO(A)ELs are almost identical (range: 0.09–0.44 mg/m3 air) and, accordingly, appear to be determined by local rather than systemic effects.
2.2. Analysis of exposure potential to pyrethroid-contaminated house-dust More recently, in the western world public concern has been raised that pesticides bound to house dust are likely to accumulate in this matrix. In this context, carpets are considered to be the most important sink for dust bound pyrethroids. Such dust can readily be sampled by vacuum cleaning and appreciable amounts of pyrethroids in that matrix are collected by analytical chemists, despite the absence of standardised sampling, sighting, and balancing procedures. Furthermore, house dust is considered to be a poor passive sampler, because neither the most effective subfraction of dust nor its true duration of contact with the incriminated agent is known. Moreover, house dust is a heterogeneous mixture and a number of sources contribute to this mix, including tracked-in or re-suspended soil particles, clothing, atmospheric deposition of particulates, hair, fibers (artificial and natural), molds, pollen, allergens, bacteria, viruses, arthropods, ash, soot, animal fur and dander, smoke, skin particles, cooking and heating residues, and building components among others. In Germany, for example,
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
197
Fig. 3. Dislodgment of permethrin contained in carpet wool from carpets. The carpet was dusted with kaolin as house dust substitute (3 g/m2). The contact time of kaolin was 1 day or 6 weeks. 0–2 h: time weighted average concentration following onset of brushing; 3–18 h: time weighted average concentration during the remaining brushing period.
concentrations exceeding 1 mg pyrethroid/kg house dust have often been misconstrued as ‘heavily contaminated’ and decontamination measures are communicated or mandated. The fear of the public is that a redispersion of such sedimented dust containing pyrethroids may cause short-term peak exposures in dwellings. This allegation prompted complex and costly inhalation studies with carpets and were designed to arrive at conclusions relevant for real-life human exposure, i.e. whether sedimented dust (vacuum cleaner sampling) can indeed be used to predict airborne concentrations likely to occur following mechanical stress to surfaces. For this purpose wool carpets made moth-resistant with permethrin were subjected to continuous brushing for approximately 18 h. Following treatment of carpet with an aqueous (EW) or dry powder formulation (WP) containing a type II pyrethroid, a series of well controlled measurements was made. Analytical characterisation of exposure atmospheres revealed that airborne pyrethroids were solely associated with the dust particles which is consistent with the physicochemical properties of this class of chemicals. The relative
percentage of pyrethroid applied to the carpet in relation to the average mass brushed off the carpet was used to calculate the dislodgment factor for each trial and was compared with the timeweighted average (TWA) of airborne total suspended particulate matter. Furthermore, a carpet made moth-resistant with permethrin (the wool contained approximately 150 ppm permethrin), was dusted with 3 g kaolin/m2 as proxy for house dust. In order to allow analysis of whether permethrin would migrate from the wool fiber to the dust substitute, two separate trials were considered. In one trial the dust was allowed to equilibrate with the carpet fibers for 6 weeks, in the other dust was administered to the carpet 1 day prior to testing. As depicted in Fig. 3, a migration or even accumulation of the permethrin contained in wool fibers of carpets to highly surface active kaolin dust could not be ascertained, since the concentration in dust recovered airborne from kaolin treated carpets was almost identical to the starting material, i.e. there was no experimental evidence that any migration of permethrin from wool to dust has occurred. The apparently higher loading of dust following extended brushing (3–
198
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
18 h) appears to be associated with the attrition forces imposed by the high load of dust. Collectively, experimental findings suggest that even under worst-case conditions (continuous mechanical stress for 18 h) only a very small mass fraction of the pyrethroid applied to the carpet was recovered airborne (Fig. 4) in spite of the overall high dust load of atmosphere. The two concentrations shown in Fig. 3 represent the mean concentrations during the first 2 h and remaining 16 h of brushing. More details have been published previously (Pauluhn, 1998). This study demonstrates that under unrealistically highly artificial test conditions only a negligible fraction of the total carpet dust pool appears to be readily available for uptake by inhalation. Experimental data suggest that under such worstcase testing conditions only a small mass fraction (range 0.04–0.2%) of the pyrethroid applied to the carpet, which is considered to represent the carpet-dust pool, is likely to be recovered airborne despite the fact that the test apparatus was particularly designed to use air flow rates which counteract the sedimentation of airborne particulates. The low recovery of airborne contaminant concentrations in relation to that which is potentially available to become airborne suggests that neither the determination of the contaminant nor that of contaminant loading is a suitable means to predict potential inhalation exposure. Taking into consideration that only a 1/500 – 1/2500th of the amount
Fig. 4. Recovery of cyfluthrin from treated carpets in relation to the total time weighted average concentration (TWA) of total suspended particulate matter (TSP).
of pyrethroid administered onto the carpet could be recovered airborne during the 18-h brushing period, it is concluded that a reliable extrapolation of potential airborne concentrations solely on the basis of contamination of sedimented house dust cannot reliably be made.
3. Discussion and conclusion The considerations made appear to suggest that for pyrethroids both concentration and concentration× time, i.e. local and dose-dependent systemic effects, have to be judged separately. For the a-cyano-pyrethroid examined, following inhalation exposure local effects appear to precede systemic ones. Therefore, route-to-route extrapolation of toxicity data may be practical if no portal-of-entry effects within the respiratory tract occur. However, relative toxic potencies may differ markedly following uptake by inhalation and other routes. Default values for the conversion from oral to inhalation are dependent on many mechanistic and substance-related factors and warn against the use of default assumptions. Therefore, default values for route-to-route extrapolation of pyrethroids are not recommended and caution is advised when so doing. In the carpet studies emphasis has been directed to address the more qualitative question, i.e. as to whether a moth-resistant wool carpet subjected to mechanical stress releases toxicologically significant concentrations of permethrin-laden dust particles in the absence or presence of increased load of inorganic dust particles (kaolin). The results summarised in Fig. 4 illustrate that under test conditions that involved continuous brushing and that counteract the sedimentation of airborne particulates, the mass fraction likely to become airborne is too small to allow for a scientifically sound extrapolation from sedimented to airborne dust. Thus, any sampling program should also address actual airborne concentrations of contaminants rather than relying on samples representing the carpet-dust pool, e.g. vacuum cleaner bag analyses. Moreover, tremendous errors and misjudgments are likely to occur relying solely upon analyses of this non-inhalable matrix. For exam-
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
ple, the geometric mean concentration of the readily inhalable and bioavailable volatile organic substances (VOCs) in the personal cloud of people in Germany is equal to 584.1 mg/m3 air (UmweltSurvey, 1996). Especially confusing is that concentrations of pyrethroids in house dust exceeding 1 ppm (mg/g dust) are viewed critically whilst the contamination of metals of toxicological concern are largely disregarded. The geometric mean concentration of some metals in vacuum cleaner dust in German households exceeds that of pyrethroids markedly, e.g. arsenic: 2.1 ppm; lead: 6.1 ppm; chromium: 64.2 ppm; nickel: 16.5 ppm; whereas for permethrin (pyrethroid present in the highest concentration) the respective concentration is 0.194 ppm (Umwelt-Survey, 1995). This comparison suggests that it is especially difficult to causally relate clinical observations or epidemiological findings to arbitrarily selected indicator substances, such as pyrethroids, contained in a matrix not readily available to inhalation exposure without taking into account other factors. Thus, in order to reduce health risks posed by dust-bound or airborne contaminants the focus should be seeking the most relevant medium and indicator contaminant. In the light of the background exposure data (Umwelt-Survey, 1995, 1996), the focus solely on house dust as the only relevant exposure vehicle remains somewhat conjectural (Paustenbach et al., 1997).
References Alarie, Y., 1973. Sensory irritation by airborne chemicals. Crit. Rev. Toxicol. 2, 299–363.
.
199
Aldridge, W.N., 1990. An assessment of the toxicological properties of pyrethroids and their neurotoxicity. Crit. Rev. Toxicol. 21, 89 – 104. ASTM, 1984. Standard Test Method for Estimating Sensory Irritancy of Airborne Chemicals. Test Method E 981-84. American Society For Testing of Materials, Philadelphia, PA. Bos, P.M.J., Zwart, A., Reuzel, J., Bragt, P.C., 1992. Evaluation of sensory irritation test for the assessment of occupational health risk. Crit. Rev. Toxicol. 21, 423 – 450. Ecobichon, D.J., 1991. Toxic effects of pesticides. In: Klassen, C.D. (Ed.), Casarett and Doull’s Toxicology. The Basic Science of Poisons. 5th ed. Mc Graw-Hill, Health Professionals Division, New York, Chapter 22, p. 643. Moretto, A., 1991. Indoor spraying with the pyrethroid insecticide lambda-cyhalothrin: effects on spraymen and inhabitants of sprayed houses. Bull. WHO 69, 591 – 594. Pauluhn, J., 1998. Hazard identification and risk assessment of pyrethroids in the indoor environment. Appl. Occup. Environ. Hyg. 13 (6), 469 – 478. Paustenbach, D.J., Finley, B.L., Long, T.F., 1997. The critical role of house dust in understanding the hazards posed by contaminated soils. Int. J. Toxicol. 16, 339 – 362. Tucker, S.B., Flannigan, S.A., 1983. Cutaneous effects from occupational exposure to fenvalerate. Arch. Toxicol. 54, 195 – 202. Umwelt-Survey, 1995. O8 kologische Wirkungen und Qualita¨tsziele-Forschungsbericht 11606088/02, Umwelt-Survey in den fu¨nf neuen La¨ndern der Bundesrepublik Deutschland 1991/92 (unter Beru¨cksichtigung der Erhebungen in den alten La¨ndern 1990/91 und 1985/86). Institut fu¨r Wasser-, Boden- und Lufthygiene des Umweltbundesamtes, Dezember 1995. Umwelt-Survey, 1996. WaBoLu Umwelt-Survey 1990/91, Band IV: Personengebundene Exposition gegenu¨ber flu¨chtigen organischen Verbindungen in den alten Bundesla¨ndern. WaBaLu-Hefte, Umweltbundesamt-Institut fu¨r Wasser-, Boden- und Lufthygiene, 4/96, pp.18. Vijverberg, H.P.M., van den Bercken, J., 1990. Neurotoxicological effects and the mode of action of pyrethroid insecticides. Crit. Rev. Toxicol. 21, 105 – 126.
Hazard identification and risk assessment of pyrethroids in the indoor environment Ju¨rgen Pauluhn * BAYER AG, Institute of Toxicology, 42096 Wuppertal, Germany Accepted 31 January 1999
Abstract Household insecticide products raise several important considerations concerning safety. These are related to the use of insecticides by untrained individuals, the difficulty of controlling the use of these products once purchased by the consumer and the potential exposure of the very young and very old, possibly with or without pre-existing pulmonary disease. Exposure to pyrethroids contained in mats or vaporizers, being slow release systems, have particular potential for long-term low-level exposure whilst for foggers, spray-cans or sprayed formulations the short-term high-level exposures may be of more concern. According to the volatility of the active ingredient contained in the household insecticide, its persistence in a non-inhalable matrix, i.e. sedimented house dust, may be short or long for highly volatile or low volatile active ingredients, respectively. On the other hand, the potential of exposure is apparently just reciprocal. This demonstrates that the extent and duration of exposure may be highly product-specific. Accordingly, the extent of exposure has to be accounted for and for risk assessment both concentration-dependent (e.g. sensory irritation) as well as concentration×time (= dose) related effects have to be considered and addressed in adequate bioassays. The issue as to whether pyrethroids adhering to house dust is of concern has been addressed in a model study using carpets treated with pyrethroids. This study has demonstrated that the total mass of pyrethroid applied to the carpet and that brushed off within an 18-h period is too small to be of any relevance for risk assessment. Therefore, assessment of health hazards in the indoor environment based simply on methodologies of emptying the household vacuum cleaner and analysing its content, which addresses contamination only, rather than examination of the actual airborne concentration, including other relevant airborne materials, is prone to tremendous errors and misjudgments. Due to the many substances potentially present in house dust and indoor air, e.g. bioaerosols originating from animals, pests and microorganisms, volatile organic substances (VOCs) or metals, prudent expert judgment is needed to assess the relevance of analytical findings. The complex indoor exposure scenario makes it especially difficult to causally relate clinical and epidemiological findings to arbitrarily selected indicator substances contained in a matrix not readily available to inhalation exposure. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Hazard identification; Risk assessment; Pyrethroids; Indoor environment; House dust; Sensory irritation
* Tel.: +49-202-368909; fax: + 49-202-364589. 0378-4274/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 4 2 7 4 ( 9 9 ) 0 0 0 4 7 - 8
194
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
1. Introduction Synthetic pyrethroids are used in agriculture and to a large extent also in the domestic environment for numerous applications. The favorable properties of this class of insecticides have promoted widespread application in virtually all sectors of food protection and pest control. They have also been proven to be useful as space and surface sprays in common houses, airplanes, in active or passive evaporators, in incense products, or to make textiles insect resistant (e.g. uniforms, bednets, carpets). With regard to effectiveness and toxicity, synthetic pyrethroids appear to be the first choice insecticides for this type of use pattern because pyrethroids are much more effective against a wide spectrum of pests than the organochlorine, organophosphate, and carbamate insecticides. Synthetic pyrethroids act directly on the nerve membrane through interference with the sodium channel gating mechanism that underlies the generation and conduction of each nerve impulse. This mode of action is more pronounced in insects than in mammalian species and demonstrates a negative temperature coefficient, i.e. lowering of the ambient temperature causes or promotes the appearance of toxic signs. This means, especially in insects and in the coldblooded vertebrate nervous system the neuroexcitatory mode of action of pyrethroids is more pronounced than in homeothermic laboratory animals or humans. The further prolongation of sodium current by pyrethroids at a lowered temperature accounts for the prolongation of repetitive nerve impulse trains. In conclusion, pyrethroids prolong the open time of voltage-dependent sodium channels, resulting in a prolongation of the inward sodium current during excitation. This effect accounts for the repetitive activity of sense organs at cooler body locations, for example the facial skin (Aldridge, 1990; Vijverberg and van den Bercken, 1990; Ecobichon, 1991). In insects, the major metabolism is related to an oxidative attack, whilst in higher animals and humans detoxification is mediated via hydrolytic (carboxylesterases) and oxidative actions. This ex-
plains why inhibitors of the oxidative metabolism increase the potency of pyrethroids in insects synergistically, whereas in mammalians appreciable effects had not been observed. The rapid metabolic degradation of pyrethroids in mammalian systems ensures against bioaccumulation or biomagnification. Also toxicokinetic mechanisms are described to enhance the selective toxicity of pyrethroids to insects. In laboratory animals and humans, due to the high lipophilicity of this class of chemicals, the toxicity is highly dependent on route of administration or exposure, which is highest by injection and lowest by the dermal route. Following inhalation exposure, major barriers which might prevent or protract absorption are absent thus making the inhalation route of exposure particularly important. One notable form of toxicity associated with synthetic pyrethroids has been a cutaneous paraesthesia and respiratory sensations mainly observed in workers spraying pyrethroids in crops or in occupational settings (Tucker and Flannigan, 1983; Moretto, 1991). This means that exposure to higher concentrations or doses of pyrethroids often causes hypersensitivity to sensory stimuli, and a number of compounds have been demonstrated to induce tingling sensation in the directly exposed skin. The time course of these sensations is usually immediate or within a few minutes after contact. The respiratory sensations are known to occur more frequently in context with the use of the more potent a-cyano pyrethroids, also known as type II pyrethroids. Thus, these respiratory sensations appear to occur particularly following brief, high-level peak exposures rather than longterm low-level exposure of humans. An approach to assess the acute respiratory tract sensory irritation potential of pyrethroids in a highly controlled and standardised animal model is described below. This bioassay provides a readily available means to assess whether irritation of the upper and/or lower respiratory tract is likely to occur (ASTM, 1984). Public perception and concern may trigger complex and demanding toxicological examinations and, as far they are considered to be part of the registration process of the product, standardisation of such complex studies with regard to the
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
generation of test atmospheres, mode and duration of exposure, and selection of adequate toxicological endpoints is challenging. However, in order to satisfy the concerns of the public as well as of the regulators who strive to achieve to protect the public from hazardous exposures from household insecticides, the described approaches can serve as a prerequisite for an up-to-date and scientifically sound risk assessment. In this instance, the toxicological assessment of pyrethroids adhering to house dust appears to be most challenging since the shuttle function of such particulates may facilitate exposure to non-volatile agents, at least on a hypothetical basis. In this context, the analysis of exposure levels causing sensory irritation, sensations causing discomfort or unpleasant odour perception are equally important. Thus, the objective of the studies presented is to analyse the suitability of various regimens to evaluate for what type of household insecticide which bioassay provides data most relevant for risk assessment. Particular emphasis is made to assess the extent of exposure originating from pyrethroids bound to house dust.
2. Test models
2.1. Upper respiratory tract sensory irritation potential of pyrethroids Stimulation of the trigeminal nerve endings located in the respiratory epithelium of the nasal mucosa by irritant airborne substances causes a variety of reflex reactions. Sensory irritation includes both nasal pungency and eye irritation. Sensory nerve endings respond to chemical as well as physical stimuli. These nerve endings respond to a variety of chemicals and stimulation can be dissociated from smell or taste. Reflex reaction originating from this stimulation are considered to be of a protective nature, i.e. to limit contact to a variety of potentially hazardous chemicals. Although other nerves are involved with the effects of sensory irritants, e.g. vagal nerves, the trigeminal nerve appears to be the only nerve directly involved with the decrease in respiratory rate in laboratory rodents. The decrease in respiratory
195
rate appears commonly within a few minutes, depending on the chemical. Thus, the sensory irritation potential of airborne substances is of crucial importance in regard to air quality, especially indoor air quality. These reactions to airborne materials are of great interest to toxicologists because they are indicative of a chemical’s likelihood to irritate and potentially damage the respiratory tract and are easy to measure in experimental animals. Using mice or rats, animal models have been developed based on their characteristic effect on breathing patterns. The concentration causing a decrease of respiratory rate of fifty percent (RD50) provides a means to compare the relative irritating potency of exposure atmospheres (Alarie, 1973; Bos et al., 1992). Utilising a 45-min single exposure regimen for rats, for cyfluthrin an RD50 value of approximately 50 mg/m3 was determined. The measurements made with aerosolised cyfluthrin used spontaneously breathing, conscious rats in modified nose-only exposure restrainers modified as plethysmographs. Prior to measurements of respiratory rate, tidal volume and apnea time during a 45-min to 1-h exposure period to the pyrethroid, the animals were allowed to acclimatise to the exposure conditions for 15 min, followed by an exposure and recovery period of short duration (Fig. 1). This experimental finding demonstrates that the a-cyano-pyrethroids especially appear to evoke functional changes in the breathing pattern (reflexively induced bradypnoeic response) which is a characteristic finding of agents acting as upper respiratory tract sensory irritants. Furthermore, from Fig. 1 it is apparent that the onset of changes is immediate and that no exacerbation of effects occurs following extended duration of exposure. This means that the change in breathing patterns is a concentrationrather that dose-dependent phenomenon. Based on analysis of the reflexively increased duration of the pause between inspiration and expiration (which leads to an increased apnea time), for this pyrethroid, the sensory irritant threshold concentration was estimated to be approximately 0.1 mg/m3 air (single 1-h exposure). Respiratory measurements were also made repeatedly during a subacute 4-week inhalation study.
196
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
Fig. 1. Average respiratory rate, minute volume and apnoea time measured in flow nose-only plethysmographs on male rats (n=8) during a 1-h exposure period. Prior exposure to 80 mg cyfluthrin/m3; base-line data were collected during an air-exposure period of 15 min ( = 100%).
From this study it is apparent that repeated exposure of rats was tolerated without time-related alterations in responsiveness (Fig. 2) even when exposed to highly effective RD50 concentration. This means that from the data shown in Fig. 2 it emerges that the single 45-min to 1-h and 5 ×6-h/
Fig. 2. Repeated nose-only exposure of rats to actual concentrations of 0 (air), 0 (vehicle), 0.44, 6 or 47 mg pyrethroid/m3 air (exposure 6 h/day, 5 days/week for 4 weeks). In each group measurements of respiratory rate were made on four rats/ group on 1 day per week during weeks 2, 3 and 4. WP10%: single 45-min exposure of rats to a dust containing 10% cyfluthrin. The respective airborne concentrations were 5.9 mg cyfluthrin/m3 and 76 mg TSP/m3 (TSP: total suspended particulate matter). The decrease in respiratory rate as a result of exposure to the vehicle was taken into account for the calculation of the sensory irritant threshold concentration.
day 4-week studies yielded virtually the same NO(A)EL (no observable adverse-effect level) and susceptibility did not change during the course of study. In an ancillary study rats were exposed for 45 min to dust containing 10% cyfluthrin (WP 10%); the resultant concentration of the pyrethroid in the vicinity of the rats’ breathing zone was approximately 6 mg/m3 air. As illustrated in Fig. 2, the magnitude of response following this single exposure was similar to that observed following 5× 6-h/day 4-week exposure. This finding demonstrates that the carrier-matrix of the pyrethroid does not appear to play any significant role. Taking into account the accumulated experimental evidence from single 45-min to 1-h, repeated 4-week and 13-week (6-h/day for 5 consecutive days/week) inhalation studies it can be demonstrated that the respective NO(A)ELs are almost identical (range: 0.09–0.44 mg/m3 air) and, accordingly, appear to be determined by local rather than systemic effects.
2.2. Analysis of exposure potential to pyrethroid-contaminated house-dust More recently, in the western world public concern has been raised that pesticides bound to house dust are likely to accumulate in this matrix. In this context, carpets are considered to be the most important sink for dust bound pyrethroids. Such dust can readily be sampled by vacuum cleaning and appreciable amounts of pyrethroids in that matrix are collected by analytical chemists, despite the absence of standardised sampling, sighting, and balancing procedures. Furthermore, house dust is considered to be a poor passive sampler, because neither the most effective subfraction of dust nor its true duration of contact with the incriminated agent is known. Moreover, house dust is a heterogeneous mixture and a number of sources contribute to this mix, including tracked-in or re-suspended soil particles, clothing, atmospheric deposition of particulates, hair, fibers (artificial and natural), molds, pollen, allergens, bacteria, viruses, arthropods, ash, soot, animal fur and dander, smoke, skin particles, cooking and heating residues, and building components among others. In Germany, for example,
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
197
Fig. 3. Dislodgment of permethrin contained in carpet wool from carpets. The carpet was dusted with kaolin as house dust substitute (3 g/m2). The contact time of kaolin was 1 day or 6 weeks. 0–2 h: time weighted average concentration following onset of brushing; 3–18 h: time weighted average concentration during the remaining brushing period.
concentrations exceeding 1 mg pyrethroid/kg house dust have often been misconstrued as ‘heavily contaminated’ and decontamination measures are communicated or mandated. The fear of the public is that a redispersion of such sedimented dust containing pyrethroids may cause short-term peak exposures in dwellings. This allegation prompted complex and costly inhalation studies with carpets and were designed to arrive at conclusions relevant for real-life human exposure, i.e. whether sedimented dust (vacuum cleaner sampling) can indeed be used to predict airborne concentrations likely to occur following mechanical stress to surfaces. For this purpose wool carpets made moth-resistant with permethrin were subjected to continuous brushing for approximately 18 h. Following treatment of carpet with an aqueous (EW) or dry powder formulation (WP) containing a type II pyrethroid, a series of well controlled measurements was made. Analytical characterisation of exposure atmospheres revealed that airborne pyrethroids were solely associated with the dust particles which is consistent with the physicochemical properties of this class of chemicals. The relative
percentage of pyrethroid applied to the carpet in relation to the average mass brushed off the carpet was used to calculate the dislodgment factor for each trial and was compared with the timeweighted average (TWA) of airborne total suspended particulate matter. Furthermore, a carpet made moth-resistant with permethrin (the wool contained approximately 150 ppm permethrin), was dusted with 3 g kaolin/m2 as proxy for house dust. In order to allow analysis of whether permethrin would migrate from the wool fiber to the dust substitute, two separate trials were considered. In one trial the dust was allowed to equilibrate with the carpet fibers for 6 weeks, in the other dust was administered to the carpet 1 day prior to testing. As depicted in Fig. 3, a migration or even accumulation of the permethrin contained in wool fibers of carpets to highly surface active kaolin dust could not be ascertained, since the concentration in dust recovered airborne from kaolin treated carpets was almost identical to the starting material, i.e. there was no experimental evidence that any migration of permethrin from wool to dust has occurred. The apparently higher loading of dust following extended brushing (3–
198
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
18 h) appears to be associated with the attrition forces imposed by the high load of dust. Collectively, experimental findings suggest that even under worst-case conditions (continuous mechanical stress for 18 h) only a very small mass fraction of the pyrethroid applied to the carpet was recovered airborne (Fig. 4) in spite of the overall high dust load of atmosphere. The two concentrations shown in Fig. 3 represent the mean concentrations during the first 2 h and remaining 16 h of brushing. More details have been published previously (Pauluhn, 1998). This study demonstrates that under unrealistically highly artificial test conditions only a negligible fraction of the total carpet dust pool appears to be readily available for uptake by inhalation. Experimental data suggest that under such worstcase testing conditions only a small mass fraction (range 0.04–0.2%) of the pyrethroid applied to the carpet, which is considered to represent the carpet-dust pool, is likely to be recovered airborne despite the fact that the test apparatus was particularly designed to use air flow rates which counteract the sedimentation of airborne particulates. The low recovery of airborne contaminant concentrations in relation to that which is potentially available to become airborne suggests that neither the determination of the contaminant nor that of contaminant loading is a suitable means to predict potential inhalation exposure. Taking into consideration that only a 1/500 – 1/2500th of the amount
Fig. 4. Recovery of cyfluthrin from treated carpets in relation to the total time weighted average concentration (TWA) of total suspended particulate matter (TSP).
of pyrethroid administered onto the carpet could be recovered airborne during the 18-h brushing period, it is concluded that a reliable extrapolation of potential airborne concentrations solely on the basis of contamination of sedimented house dust cannot reliably be made.
3. Discussion and conclusion The considerations made appear to suggest that for pyrethroids both concentration and concentration× time, i.e. local and dose-dependent systemic effects, have to be judged separately. For the a-cyano-pyrethroid examined, following inhalation exposure local effects appear to precede systemic ones. Therefore, route-to-route extrapolation of toxicity data may be practical if no portal-of-entry effects within the respiratory tract occur. However, relative toxic potencies may differ markedly following uptake by inhalation and other routes. Default values for the conversion from oral to inhalation are dependent on many mechanistic and substance-related factors and warn against the use of default assumptions. Therefore, default values for route-to-route extrapolation of pyrethroids are not recommended and caution is advised when so doing. In the carpet studies emphasis has been directed to address the more qualitative question, i.e. as to whether a moth-resistant wool carpet subjected to mechanical stress releases toxicologically significant concentrations of permethrin-laden dust particles in the absence or presence of increased load of inorganic dust particles (kaolin). The results summarised in Fig. 4 illustrate that under test conditions that involved continuous brushing and that counteract the sedimentation of airborne particulates, the mass fraction likely to become airborne is too small to allow for a scientifically sound extrapolation from sedimented to airborne dust. Thus, any sampling program should also address actual airborne concentrations of contaminants rather than relying on samples representing the carpet-dust pool, e.g. vacuum cleaner bag analyses. Moreover, tremendous errors and misjudgments are likely to occur relying solely upon analyses of this non-inhalable matrix. For exam-
J. Pauluhn / Toxicology Letters 107 (1999) 193–199
ple, the geometric mean concentration of the readily inhalable and bioavailable volatile organic substances (VOCs) in the personal cloud of people in Germany is equal to 584.1 mg/m3 air (UmweltSurvey, 1996). Especially confusing is that concentrations of pyrethroids in house dust exceeding 1 ppm (mg/g dust) are viewed critically whilst the contamination of metals of toxicological concern are largely disregarded. The geometric mean concentration of some metals in vacuum cleaner dust in German households exceeds that of pyrethroids markedly, e.g. arsenic: 2.1 ppm; lead: 6.1 ppm; chromium: 64.2 ppm; nickel: 16.5 ppm; whereas for permethrin (pyrethroid present in the highest concentration) the respective concentration is 0.194 ppm (Umwelt-Survey, 1995). This comparison suggests that it is especially difficult to causally relate clinical observations or epidemiological findings to arbitrarily selected indicator substances, such as pyrethroids, contained in a matrix not readily available to inhalation exposure without taking into account other factors. Thus, in order to reduce health risks posed by dust-bound or airborne contaminants the focus should be seeking the most relevant medium and indicator contaminant. In the light of the background exposure data (Umwelt-Survey, 1995, 1996), the focus solely on house dust as the only relevant exposure vehicle remains somewhat conjectural (Paustenbach et al., 1997).
References Alarie, Y., 1973. Sensory irritation by airborne chemicals. Crit. Rev. Toxicol. 2, 299–363.
.
199
Aldridge, W.N., 1990. An assessment of the toxicological properties of pyrethroids and their neurotoxicity. Crit. Rev. Toxicol. 21, 89 – 104. ASTM, 1984. Standard Test Method for Estimating Sensory Irritancy of Airborne Chemicals. Test Method E 981-84. American Society For Testing of Materials, Philadelphia, PA. Bos, P.M.J., Zwart, A., Reuzel, J., Bragt, P.C., 1992. Evaluation of sensory irritation test for the assessment of occupational health risk. Crit. Rev. Toxicol. 21, 423 – 450. Ecobichon, D.J., 1991. Toxic effects of pesticides. In: Klassen, C.D. (Ed.), Casarett and Doull’s Toxicology. The Basic Science of Poisons. 5th ed. Mc Graw-Hill, Health Professionals Division, New York, Chapter 22, p. 643. Moretto, A., 1991. Indoor spraying with the pyrethroid insecticide lambda-cyhalothrin: effects on spraymen and inhabitants of sprayed houses. Bull. WHO 69, 591 – 594. Pauluhn, J., 1998. Hazard identification and risk assessment of pyrethroids in the indoor environment. Appl. Occup. Environ. Hyg. 13 (6), 469 – 478. Paustenbach, D.J., Finley, B.L., Long, T.F., 1997. The critical role of house dust in understanding the hazards posed by contaminated soils. Int. J. Toxicol. 16, 339 – 362. Tucker, S.B., Flannigan, S.A., 1983. Cutaneous effects from occupational exposure to fenvalerate. Arch. Toxicol. 54, 195 – 202. Umwelt-Survey, 1995. O8 kologische Wirkungen und Qualita¨tsziele-Forschungsbericht 11606088/02, Umwelt-Survey in den fu¨nf neuen La¨ndern der Bundesrepublik Deutschland 1991/92 (unter Beru¨cksichtigung der Erhebungen in den alten La¨ndern 1990/91 und 1985/86). Institut fu¨r Wasser-, Boden- und Lufthygiene des Umweltbundesamtes, Dezember 1995. Umwelt-Survey, 1996. WaBoLu Umwelt-Survey 1990/91, Band IV: Personengebundene Exposition gegenu¨ber flu¨chtigen organischen Verbindungen in den alten Bundesla¨ndern. WaBaLu-Hefte, Umweltbundesamt-Institut fu¨r Wasser-, Boden- und Lufthygiene, 4/96, pp.18. Vijverberg, H.P.M., van den Bercken, J., 1990. Neurotoxicological effects and the mode of action of pyrethroid insecticides. Crit. Rev. Toxicol. 21, 105 – 126.