Document a1ErQ4ja6p636eDJ8OnnrGv3b

AL AN D APPLIED TOXICOLOG~ 14,299-308 (1990) at Absorption of Organic Chemical Vapors in Rats and Humar1s' JAMES N. MCDOUGALG,ARY W. JEPSONH, ARVEY J. CLEWEL1L11, MICHAEL L. GARGASA,ND MELVIN E. ANDERSEN' Harrv G ArmsrrongAerospace Medical Research Laborarory, Toxic Hazards Division, Wrighr-PattersonA i r Force Base, Ohio 45433 Received March 27. 1989. accepted September 1I , 1989 Absorption of Organic Chemical Vapors in Rats and Humans. MCDOUGALJ,. N., .W.. CLEWELLH,.J., 111, GARGASM, .L.,AND ANDERSEN, M. E (1990). Fundam col 14, 299-308. Quantitation of chemical vapor penetration through skin is necr assessment of health hazards involved in some occupational environments. Informa- on penetration of vapors through human skin is minimal because human exposuresare not ioned. We have investigated the whole-body dermal penetration of styrene, xylene, tolu- lene, benzene, halothane, hexane, and isoflurane in rats and compared the ts with available human studies on vapor penetration. Rats w t h closelq to organic chemical vapors (3000 to 60,000 ppm) while breathing fresh Blood concentrations taken dunngthe 4-hr exposures were determined dwelling jugular cannulas. A physiologcally based pharmacokmetic experimental blood consquare optimization and from 1.75 cm/hr for styrene to 0.03 cm/hr for isoflurane Rat permeability constants an constants which were the literature. These results indicate that organic vapor permeability constants ative estimate of organic vapor permeability constants in humans and that ences in permeability constants between these two species may be due to nces in the skm. o 1990Society ofToxicoiogy ing and quantifying absorption vapors through skin is necessary rate assessment of health hazards al environments where respiran is provided but the skin sur- posed. Until the early 1900s, sidered an effectivebarrier to tion of liquids and vapors. Now it als used in this study were handled in accor- pared by the Comory Animals of the 1 National Research e of Health Publicanimal Welfare Act of ess: Chemical Industry Institute ofToxTnangle Park, NC 27709 is recognized that chemicals penetrate the skin by passive diffusion and that the rates of penetration are related to lipid solubility of the penetrant (see reviews by Scheuplein and Blank, 1971 and Schaefer et al., 1982).A suitable in vivo animal model for predicting rates ofpenetration of vapors through human skin would provide valuable risk assessment information (Clewell and Andersen, 1985). For example, if a safe level in the environment has been set based on inhalation of a chemical, it would be useful to know how respiratory protection extends the range of safe exposure concentrations where skin vapor penetration is still possible. The literature contains several reports of quantitation of organic chemical vapor penetration through skin of human volunteers. 299 0272-0590/90 $3.00 Copynghr C I990 b) the soclctv ofToxicolog\ All nghlsofreproduction in any form reserved I I 300 MCDOUGAL ET AL. Benzene (Hanke et al.. 1961) and aniline the blood during the exposures using an ap. E (Dutkiewicz and Piotrowski, 1961) vapor propriate physiologically based p h a r m a w penetration was studied in Poland in the netic (PB-PK)model to describe the distribp, 1960s. More recently, Riihimaki and Pfami tion, metabolism, and elimination of (1978) studied dermal penetration of xylene, chemical which has penetrated the skin (&. styrene. toluene, I,],1-trichloroethane and Dougal et al., 1986). In this paper, we cfi. tetrachloroethylene (perchloroethylene) va- scribe the transdermal kinetics of eight &- pors. Wieczorek (1985) measured dermd ganic chemical vapors (styrene, r n - x y l q penetration of styrene vapors. Although little toluene, perchloroethylene, benzene, information is available about dermal pene- thane, hexane, and isoflurane) in rats tration of chemical vapors, more is known compare these estimated permeability a about the penetration of these industrially stants with published information on pew important solvents in liquid form, through tration of chemical vapors in human volua. guinea pig skin (Jakobson et al., 1982) and teers. human skin (Stewart and Dodd, 1964; Dut- luewicz and Tyras, 1967, 1968, 1969;Aitio et al., 1984; Berode et al., 1985). Since concentrations are far greater in the pure liquid than .%-" 1 L 1MATERIALS AND METHODS &, 'nL in the vapor form, flux across the skin from Animals Male Fischer-344 rats weighing 205 to the liquid form of the chemical will be greater than that from the vapor form (Hanke et al., 1961;Wieczorek, 1985).Factors which determine the rate of penetration of a chemical used in these studies were obtained from Charles Ii ;I :; across the skin are concentration at the skin surface,the surfacearea exposed, and solubility of chemical in the skin. Solubility of chemical in the skin is affected by chemical characteristics, i.e., molecular weight and charge. The physical form of the chemical, ScientificCo. (Fair Lawn, NJ). Perchloroethyle punty) was obtained from Aldrich Chemical C waukee, WI). Halothane (99.9 from Halocarbon Laboratories, rL\- I i.e., liquid or vapor, will affect the concentration at the skin surface but should not affect the solubility of a chemical once it is in the skin unless the liquid physically or chemically alters the skin barrier. Temperature and lsoflurane (99.5%punty) was ob cal Anesthetics (Madison, WI). Dermal vapor exposures Rats were exposed to organic chemicals in a custom built chamber d to allow controlled vapor exposures over the e surface of rats whose fur was closely clipped. relative humidity may affect blood flow to the of this chamber, analytical procedures, and h skin, but with most organic chemicals which have good solubility in the blood, diffusion 1985). Briefly, rats were acclimated to a 1 by at least three 8-hr training periods. On through the skin is the limiting factor rather exposure, fur of the rats was carefully CIS than removal from the skin by the blood. injury to the skin, with electnc clippers and indwd@ I i I We previously developed a chamber system to investigate whole-body dermal vapor absorption in rats (McDougal et al., 1985). In jugular cannulas were implanted under ketamine/* zine anesthesia. Six rats were placed in the expoollg chamber, provided with clean breathing air throu latex masks, and had blood samples (0.I ml) dra i I this system, masks provide respiratory pro- cannulas routed to the exterior of the chamber du& tection and sampling ports facilitate blood sampling during a whole-body exposure to high vapor concentrations of volatile chemicals. The flux of chemical across the skin and permeability constants can be determined the 4-hr exposures. Concentration of the effluent air- monitored every 5 min by flame ionization detection a;! I a gas chromatograph and was maintained at the d e concentration by manual operation of needle valvesa~& trolling airflow through the chemical bubbler. %? Blood samples (0.1ml) were drawn (after d r a w i n g 4 ` i based on the concentrations of chemical in discarding 0.2 ml which is approximately twice the C& 3, I bc.i'the Cbernical :de- I ne in- DERMAL VAPOR ABSORPTION 30 1 TABLE 1 PARAMETERSUSED FOR THE ANALYSISOFORGANIC CHEMICALS IN BLOOD Blood extraction solvent (blood:solvent ) Oven temperature ('C) GC detector Carbon disulfide ( I :IO) Methylene chlonde ( I:1) n-Hexane ( l :l ) n-Hexane ( I :IO) m-Xylene ( I :1) n-Hexane (1:IO) Benzene (1:1 ) n-Hexane ( I :10) I50 Flame ionization 100 Flame ionization 75 Flame ionization 85 Electron capture 70 Flame ionization 80 Electron capture 75 Flame ionization 80 Electron capture ) and total volume drawn was replaced stants (Table 3) were determined for most chemicals by n-saline, via the indwelling jugular can- gas uptake techniques. For modeling purposes, perme- 1 to exposure and at 0.5, 1.0,2.0, 3.0, and 4.0 ability constants were fit both visually and by statistical start of exposure. These samples were ex- parameter estimation techniques with similar results. th either 0.1 or 1.0 ml of solvent (Table I). Visually determined permeability constants were used as I a starting point for a statistical software program (SIMU- I after injection on a IO-ft 10% SOLV, Dow Chemical Co., Midland, MI) which proP(80-100)in astainless-steel vided a maximum likelihood estimate ofthe skin perme- shown in Table 1. With per- ability constant in the model by fitting model predictions ion efficiency from blood was of blood concentrations to experimental observations. ncentrations were corrected for Human permeability calculations. The permeability ion efficienciesfor the other chemicals constant (P)was calculated for each published human study according to r calculations. Organic vapor permewith a physiologically P = A, ABS x C,x t ' the pharmacokinetics. Partition coefficients ) were measured by the vial equilibration of Gargas (Gargas et al., 1989). Metabolic con- where ABS is the total absorbed in milligrams, A, is the surface area exposed in square centimeters, t is time in hours. and C,is the exposure concentration in milligrams per cubic centimeter. The resulting permeability constant (cm/hr) is concentration independent, can easilybe scaled for the exposed surface area, and can be compared TABLE 2 PARTITION COEFFICIENTSUSED IN THE PHYSIOLOGICAL MODEL" L. f { & Chemical I Styrene m-Xylene 1 Toluene Blood/air 40.2 46.0 18.0 Fat/ar 3476 1859 1021 Muscle/air 46.7 41.9 27.7 2 Perchloroethylene 18.9 1638 20.0 Benzene 17.8 499 10.3 Halothane 5 3 I82 4.5 Hexane 2.3 159 2 9 lsoflurane I .8 98 1.6 *%tition coefficients from Gargas ef a1 (1989) Liver/air 140.7 92.0 82.8 69.9 17.8 7.6 12.0 41 I DERMAL VAPOR ABSORPTION 303 tions from individual rats, for most only one exposure concentration was exam- *gpnlcals, tracked very consistentlythrough- ined in these studies. Our experience suggests the entire exposure. This indicates that that permeability constants are not dose de- W t y was probably due to individual rat pendent with vapor studies. Dose-dependent b n c e s rather than analytical variability. permeability constants may result when liq- I a1Penetration Measurements uid is in contact with the skin at high concentrations. Quantitative information on organicvapor ed flux (Table 4) was greatest for roethylene exposure and lowest exane exposure. Estimated permenstants ranged from 1.75 cm/hr for to 0.026 cm/hr for isoflurane. Simu- concentrations based on estieability constants and the mealood concentrations are shown in Figs. e percentage of chemical which absorbed due to dermal absorption dermal and inhalation exposure in etermined to be less than 10%in Is studied (Table 4). exposures in volunteers is minimal (Table 5). For comparison with our rat data, permeabil- ities (P)have been calculated for these chemi- cals by using Eq. (l), where the estimate of the total amount of chemical absorbed was that reported in the human study and `4, was estimated to be 19,000 cm2. The calculated human permeability values in this table are based on a wide variety of experimental methods used to determine the amount of chemical absorbed and the individual studies contain very few subjects. Wieczorek (1985) studied styrene absorption through the skin at two exposure con- centrations 3250 mg/m3 (one volunteer) and DISCUSSION 1370 mg/m3 (three volunteers). Volunteers, ncrn- clothed only in shorts, were placed in a cham- permeability constants for this set of ber maintained at 23-25Cand 33-3670 rela- inged ir. chemicals generally decrease as solun fat decreases (Table 2) showing the relationship to lipid solubility (r2 . Styrene, with fat solubility more tive humidity wearing respirators connected to fresh breathing air outside the chamber during a 2-hr exposure. Urine was collected half-way through the exposure and at inter- imes greater than that of hexane, has vals of several hours for 24 hr after exposure. ability constant approximately 67 Wieczorek estimated the amount of styrene . Solubility in muscle tissue also absorbed to be 175 mg in one subject at the 0.om the permeability constant de- high concentration and an average of 45 mg 1% m`: = 0.827). Skin partition coeffi- in the subjects exposed to the low concentra- b t s would be expected to be the best pre- tion using Qctorofthe permeability constant; however, k t 0 the difficultyof homogenizingskin, we +Y = 0.64X 4.8, (3) rn] In %e been unable to measure reliable skin where Y is the excretion of mandelic and phe- m t i o n coefficients. nylglyoxylic acids in 24 hr and X is the dose Determination of these permeability con- absorbed as determined from inhalation *& extends the range of values we pre- studies. Riihimaki and Pfaffli ( 1 978) exposed *UlY reported for dibromomethane (1.32 two volunteers wearing respirators for respi- m h r ) , bromochloromethane (0.79 cm/hr), ratory protection and thin pajamas and socks ~flur- dichloromethane (0.28 cm/hr) in rats to 600 ppm styrene vapor for 3.5 hr. Once an - and t WChugal et al., 1986). The permeability hour the subjects cycled on a bicycle ergome- %es for the dihalomethanes were shown to ter at 100 W for 10 min and were sedentary ation 1 bcmnsistent over a three order of magnitude for the remainder of the hour. Venous blood t con- 1 `i$ge of exposure concentrations; therefore, and exhaled air samples were taken four a 2 0 ] STYRENE 3.000 ppm 16 50 ]M-XYLENE 5.000 ppm 1-- 40 E \OI 5 30Vz 8 0 20- 10" l 10 le 012 34 TIME (hours) 0 0 1 TIME 2(hours) 3 2 12.5 C TOLUENE 3.000 ppm PERCHLOROETHYLENE 1 2 5 0 0 ppm , 1 100 -- E \0) It 7.5 40- -- -E0 0 \ OI t 30- 2 03 4 vz 8 5.0 zV 0 0 3 m 2.5 0.0 0 12 3 TIME (hours) 4 1V, 0 01 , 2 TIME (hours) . 3 FIG. I . Fits of simulated blood concentrations using optimized permeability constants to experiment blood concentrations (various symbols) during 4-hr exposures to styrene (n = 5 ) (a), m-xylene (n = 6)0 toluene (n= 6) (c), perchloroethylene (n= 4) (d), benzene (n = 5 ) (e), halothane (n = 4) (f),hexane (n= 4 (g). and isoflurane ( n = 6)(h). Exposureconcentrations and blood collections are described under Materia and Methods and Results. In each graph the solid line is the best-fit curve for the entire data set. times during the exposure and four times postexposure. Postexposure urine samples were collected for 24 hr. They estimated that 60.1 mg of styrene was absorbed during the experiment. Giving equal weight to the three studieswith styrene indicate a human permeability constant of approximately 0.9 cm/hr. Riihimaki and Pfaffli (1978) exposed two volunteersto 300 and three volunteers to 600 ppm rn-xylene using the same experimental design as with styrene. On the basis of u1 metabolite and expired air concentn they calculated that 20.8 and 44.5g, re tively, were absorbed during the 3.5-hrr sures. Permeability constants calcu from these amounts absorbed averaged cm/hr. Riihimakiand Haffli (1 978)alsoexpose volunteersto toluene usingthe sameexperi tal design as with styrene. They estimated$ DERMAL VAPOR ABSORPTION BENZENE 40.000 PPm 305 -- E \ 3 cj Z 8 04 0 3 m 2 0 0 0 12 3 TIME (hours) 4 0 0 1 2 3 4 TIME (hours) HEXANE 60.000 ppm h 5 ISOFLUAANE 50.000 PPm 4 0 -- E -0 0 \01 a3 0 0 Vz 8 02 0:0 9 m 1 1 2 3 4I TIME (hours) C 1 2 34 TIME (hours) FIG. 1-Coniinued l was absorbed during the exposure on amount of phenol recovered from the urine, t of exhaled breath concentrationsof tol- they estimated that approximately 10 mg of %assuming that 16% of the absorbed tolu- benzene was absorbed because 30% of the *Was exhaled. From their data we calculatea benzene absorbed is excreted as phenol in addition to the physiological amount of phenol tion studies with perchlo- which is normally in urine. The calculated ki and Pfaffli, 1978) 47.1 human permeability constant in these expoW, absorbed during the exposure, which sures is 0.08 cm/hr. h a permeability constant of 0.17 cm/hr. Absorption of aniline vapor through the ' b k e and co-workers( 1961) exposed two skin was studied in an unspecified number of volunteers (provided with fresh breath- volunteers (Dutkiewin and Piotrowski, 1961) @@airthrough a mask) to 1 mg/liter of ben- who were placed in a chamber maintained at *vapor for 7 hr and collected urine for a minimum 25C and at least 35% relative 9 followingexposure. On the basis of the humidity by determining the amount of p exposed to 600 ppm, on the basis of the assumption that 99% of the absorbed methylchloroform is exhaled. The calculated permeability constant was 0.01 cm/hr. There is good agreement of the relative ranking with chemicals, which were done in both rats and humans. Generally, our experiments show that the permeability constant for each chemical in rats is from two to four times greater than the human permeability constants calculated from the literature data (Table 5). The apparent greater permeability in rats could have at least two explanations. ences may also account for differe etration, e.g., Riihimaki and Pfa their volunteers with thin pajam could have afforded some protecti styrene, xyIene, toluene, perchlor and methyl chloroform vapo This may be the reason that calculated from Riihimaki and with styrene are lower than from the studiesof Wieczore Our study demonstrates that de tion of rates of penetration of orga in rats usingthese methods correlat DERMAL VAPOR ABSORVTION TABLE 5 CALCULATED HUMAN PERMEABILITY CONSTANTS" ABS Flux (mg) (mg/cm/hr) 2.55 x 10-3 3.5 60.1 3.25 x 10-3 2.0 175.0 1.37x 10-3 2.0 45.0 1.30 x 10-3 3.5 20.8 2.61 x 10-3 3.5 44.5 2.26 x 10-3 3.5 26.4 4.07 x 10-3 3.5 47. I 1 . 0 01~0-3 5 . 0 0 ~10-3 7 .O 6.0 10.0 I .90 x 1o - ~ 1.00 x 10-2 6.0 2.50 x IO-' 2.00 x lo-* 6.0 4.00 x 3.27 x 10-3 3.5 2.1 307 P (cm/hr) 0.35 I .42 0.87 0.24 0.26 0.18 0.17 0.08 0.04 0.03 0.02 0.01 ewin and Piotrowski, I96 I. chemical absorption which is due to dermal s in permeability may be due uptake in whole-body exposures. differences in skin structure. ACKNOWLEDGMENTS able to examine physiological n skin from different species (Le., partition coefficients, capillary nd hair follicle density and depth in The authors acknowledge the expert technical assis- tance of Kenneth R.Thimling and Gregory L. Sudberry. We also thank Dr.Carl C. Peck for stimulating discussions of pharmacokinetics of skin absorption. REFERENCES ate permeability constants from one to another on the basis of those ces. Styrene inhalation exposures in AITIO, A., PEKARI,K., AND JARVISALO, J. (1984). Skin absorption as a source of error in biological monitoring. Scand. J. Work Environ. Health 10,317-320. BERODEM, ., DROZ, P.-0.. AND GUILLEMIN, M.(1985). were used to extramlate in inhala- exposures in humans with very good re(Ramsey and Andern% 1984)* Our currently suggests a means to estiman vapor penetration for occupa- Human exposure to styrene. VI. Percutaneous absorp tion in human volunteers. Int Arch Occup Envrron ss, 33,-336, BRONAUGHR, . L., STEWART, R. F.. AND CONGDON, E. R (1982).Methods for in vitro percutaneous ab- sorption studies. 11. Animal models for human skin. emicals which could not be ethically Toxlcol Appl Pharmacol 62,48 1-488. ,gated in human volunteers. ~ ~ of the dermal route of absorption in the CLaEssWesEs~mLeLHn,.t Jex. ItIrIa,p~AoNlaDtioAnNsDaEnR~dSEpNhyMsroElo.g(i1ca9l85m)oRdieslk-~ Ine Toxlcol Health I -I ~ - of respiratory protection also pro- DUTUKIEWICZ, T., AND P~OTROWSKJ.I,(1961). Expeninformation about the proportion of mental investigations on the quantitative estimation 308 MCDOUGAL ET AL. of aniline absorption in man Pure 4ppl Chem 3, 3 19-323. DUTKIEWICZ, T.. AND TYRAS, H. (1967). A study ofthe skin absorption of ethylbenzene in man. Brit J lnd .%fed24,330-332. DUTKIE-ICZ.T ,AND T) RAS, H. (1968). Skin absorption of toluene. styrene. and xylene by man. Brrr J Ind .Wed 25,243-246. DUTKIEWICT.Z. A. ND TYRAS, H. (1969). [Comparative studies on the absorption oftoluene, ethylbenzene, xylene. and styrene through the human skin.] Med Pr. 20,228-234. GARGASM, . L., AND ANDERSEN, M. E. ( 1 982). M e t a b lism of inhaled brominated hydrocarbons: Validation of gas uptake results by determination of a stable metabolite. Toxicol Appl Pharrnacol ,66,55-68. GARGASM, . L., ANDERSEN, M.E., ANDCLEWEHL.LJ.,, 111. (1986). A physiologically based simulation a p proach for determining metabolic constants from gas uptake data. Toxicol .4ppl Pharmacol 86,34 1-352. GARGAS, M. L., BURGESS, R. J., VOISARD, D. E., CASON, G . H., A N D ANDERSEN, M. E. (1989). Partition coefficients of low-molecular-weight volative chemicals in vanous liquids and tissues. TOXICAOp~p.l Pharmacol 98,87-99. HANKE, J.. DUTKIEWICZ, T.. AND PIOTROWSKI, J. (1961). [The absorption of benzene through human skin.] Med Pr. 12,413-426. JAKOBSON, I.. WAHLBERG, J . E., HOLMBERG, B . , AND JOHANSSON, G. (1982). Uptake via the blood and elimination of IO organic solvents following epicutaneous exposure of anesthetized guinea pigs. Toxicol Appl Pharmacol 63, 18I - 187. dermal absorption of lapors in the rat. Toxicol. Pharmacol 85,286-294. MCDOUGAL, J . N.. JEPSON, G . w . , CLEWELL, H,J AND ANDERSEN. M E. (1987). Pha organic vapor absorption In Pharma Skin Vo1 I (B Shroot and H. Schaefer, 25 1 . Karger, Basel. RAMSEY. J. C . ,AND ANDERSEN, M.E. ( 1984). logcally based description of the inhalation cohnetics of styrene in rats and humans. Tax .4ppf Pharmacol 73, 159-175. RIIHIMAKvI,., AND PFAFFLI, P. ( I 978). PercU absorption of solvent vapors in man. Scund. J. fi Envrron Health 4,7345. SATO,A . , AN D NAKAJIMTA.,(1979). A Vial-eqa zyme activity for volatile hydrocarbons. TOXI Pharmacol 47,4146. ~ H A E F E R ,H., ZESH, A . , A N D STUTTGEN, G . Skin Permeability Springer-Verlag,Berlin. SCHEUPLEIN, R. J., ANDBLANK, I. H. ( ity ofthe skin. Physiol Rev 51,702 STEWART, R. D . . AND DODD, H. C. ( 1 of carbon tetrachloride, tnchloroeth ethylene. methylene chloride, and I , ane through human Skin..her Ind HYg 439-446. TREGEAR, R. T. (1964) ecules ofwidelydiffe ological Sciences in bridge University Sc c.,WESTER, R. AND MAIBACH, H. I. (1983). In Vivo A N D ANDERSEN, M. E. (1985). Dermal absorption of dihalomethane vapors. Toxicol.Appl Pharrnacol 79, 150-158. MCDOUGALJ,. N., JEPSON, G.W., CLEWELLH,. J., 111, MACNAUGHTONM, . G., AND ANDERSEN, M. E. ( I 986). A physiological pharmacohnetic model for Marzulli and H. 1. Maibach, Eds.), pp. 13 Hemisphere, Washington, DC. WIECZOREKH, . (1985). Evaluation of low e styrene. 11. Dermal absorption of styrene humans under experimental conditions. Occup Envrron Health 57.71-75. t