Document 1gqKbgbzwG8yE4aKX3oeBy8o5

APPLICATION OF TOXICOKINETIC MODELS TO ESTABLISH BIOLOGICAL EXPOSURE INDICATORS* VERA FISEROVA-BERGEROVA Department of Anesthesiology, University of Miami School of Medicine, Miami, FL 33101, U.S.A. (Received 3 April 1990 and infinolform 18 September 1990) Abstmct--This article is a critical review of the application of toxicokinetic models to the biological monitoring of occupational exposure to industrial chemicals.Theexperimentallybased toxicokinetic models are used to determine the elimination half-lives, the metabolic clearance, the elimination rate constants and the volume of distribution. The physiologicallybased multicompartmental simulation models, which describe. the uptake, distribution and elimination of inhaled or percutaneously absorbed organicsolvents,contributed to the understanding of the transport of the xenobiotics in the body. They are used for describing and predicting the dependence of concentrations of indicators of exposure in biological specimens on the extent of exposure and time (duration of exposure and sampling time), and for depicting the contribution of various biological and exposure factors to differencesin biological response to the exposure. In biological monitoring, toxicokinetic models are used for matching biologicalconcentrationsand body burden of indicators of exposure with extent of inhalation or dermal exposure, and for predicting half-lives. They lay the grounds for the strategy used in collecting biological specimens and controlling external and internal factors which alter the biological concentrations and possibly increase the health risk from the exposure. Elimination halflives are used as guidelines in selecting the appropriate indicators of exposure, in designing the procedure for the collectionof biologicalspecimens,and in interpreting the measured data. Predictive models are needed for heavy metals, particulates and compounds undergoing binding to constituents of tissues. INTRODUCTION .- the source of exposure, and personal sampling is used to determine the exposure of individual workers. Air monitoring, however, helps to control only inhalation exposure, neglecting uptake resulting from dermal exposure, from exposure to non- occupational sources,or from unexpected excessiveexposure resulting from peculiari- ties of certain jobs or from poor working practices (FISEROVA-BERGEROVA, 1987a, 1990a). Since there are no external means of measuring the extent of these exposures, 1! monitoring of parent compounds,their metabolitesor biochemicalchangesinduced by exposure in blood and excreta ('indicators of exposure') is recommended for the evaluation of the integrated exposure of the worker (ACGIH, 1989;COMMISSIFOORN THE INVESTIGATIONOF HEALTH HAZARDS, 1989). Moreover, the measurement of these indicators in biological specimens is the only means of measuring the effectiveness of respirators and protective clothing, and of monitoring exposure in the workplace 1 outdoors, or exposure of workers with non-stationary workplaces. Toxicokinetics, which study and describe uptake, distribution, metabolism and ' 'Paper based on a presentation made at the International Workshop on Pharmacokinetic Modelling in Occupational Health, Leysm, Switzerland, in March 1990. 639 I 640 V. FISEROVA-BERGEROVA elimination of toxic compounds from the body, provide information relationship between external (environmental) and internal (body burden) which contributes to our understanding of the link between external exposu development of adverseeffects.The purpose of this article is to review the appli toxicokinetic models to biological monitoring of occupational exposure. In the past, the application of biological monitoring was hindered by th suitable analytical methods, by the variability of biological concen indicators of exposure which resulted from the same extent of expos complexity of the kinetics of uptake, distribution and elimination of co the body. This has changed in the last two decades: modern technolo instrumentsfor suitable analyticalmethods. The development of physi multicompartmental simulation models for organic solvents has pro understanding the fate of these compounds in the body, and for contribution of various exposure and biological factors to the variabili response to the exposure. Simulation models have also provided the strategies used in biological monitoring, for collecting biological spe matching biological concentrations and body burden with extent of in dermal exposure, and for controllingfactors which alter the biological concentra and possibly increase the health risk from the exposure. describing uptakc, dis ,.an be found in the BERGEROVA, 1983;FIS 1986). In order to ir I parameters f bolic clearance,elimir Ii pathway in a moc toxicok (DROZ,1978; FISERO constants (ANDE ANDERSEN, 1984),ani (DRoZ, 1978; DROZ ANDERSEN1, 984; S A percutaneouspenetr. surface must be give Simulation mod1 body, and are a poc biological concentr occupational expos simulation models binding to constituc TOXICOKINETIC MODELS There are three types of toxicokinetic models used in the biological monitori exposure to industrial chemicals: (1) empirical equations; (2) experimentally b toxicokinetic models; and (3) physiologically based simulation models. Empirical equations, such as correlation equations frequently used to com under specified conditions, the extent of exposure with biological concentratio indicators of exposure, are based on field or laboratory observations. Empi equations are also used to describe kinetic processes by simple mathe expressions. For example, the pulmonary uptake of vapours is a function blood/gas partition coefficient and the square root of time (LOW, 1972), or elimination of the parent compound or merabolites from the body is an exponen functionfitting the experimentaldata. Empirical equations provide no insight into the movements of the compound in the body, and their use is very restricted. Experimentally based toxicokinetic models are based on experimental data, which show the concentration of indicators of exposureduring and followingexposure.These models determine kinetic parameterssuch as elimination rate constants and the area under the curve (A UC)--which are used to calculate elimination half-life, volume of distribution and metabolic clearance. Metabolic clearance is defined by a sin& constant or by two constants when a saturable process occurs. These models provide, however, very little insight into the effect of biological and circumstantial factors on biological concentrations of the parent compound or its metabolite. Physiologically based multicompartmental simulation models are based on physiological parameters of the exposed subject (pulmonary ventilation, and volume and perfusionof tissues), and on solubility of the compound in blood and tissues (described by the appropriate partition coefficients). These parameters needed for Solutions of simulat Simulation moq number of which is kinetic processes (1 ANDERSEN, 1984). ' At high exposures kinetics deviate fri enzyme systems a compounds can al: integration (DROZ et al., 1986; RAMS] Microsoft Excel S mathematically SC At the low e ) governed by firstelimination of th biotransformati01 BERGEROVA et a] hepatic blood flo' Solutionofsuch 1 transform-whic functions expres functions, expon 1976). Even the used to describe 1 Y.FISEROVA-BERGEROVA usefulness in solving linear models is nevertheless quite high, since many biological monitoring are based on biological half-lives (DROZ1,989). APPLICATION OF SIMULATION MODELS IN BIOLOGICAL MONITOW Selection of indicators of exposure Selection of indicators of exposure is influenced by several factors: (1) sensitivity and specificity to exposure to the particular compound. This means monitored indicator of exposuremust be measurableat the exposurepermi workplace (insufficient biological response or variability in background tions of the same compounds of endogenic or environmental origin can confou&& -result);and, preferably, the monitored indicator should not be observed afterex- to other compounds, though a non-specificindicator of exposurecan be preferred a specific indicator if its biological concentration correlates better with the ex- d exposure; (2) by technical prerequisites such as availability of biological spe-* stability of the sample and the availability of simple analytical methods with st&& sensitivity and accuracy; and (3)by access to laboratories with a good quality programme. The preconditions for the implementation of biological monito* d occupational exposure to toxic compounds are: (1) information on kinetics of& indicator of exposure and information on uptake, distribution and elimination of* parent compound;(2) understanding of the relationshipbetween biological concenw tions of indicators of exposure and the extent of exposure and biological effect; 0) understanding of external (exposureduration and fluctuation,working conditionstad workload, route of entry, coexposure to chemicals) and internal (ethnic, genetic, disease)factors affecting these relationships (DROZand SAVOLAINEN, 1990; FJSEROVA. BERGEROVA, 1987a, 1990a). Toxicokinetic simulation models can provide the following information on whifh biological monitoring is designed and data are interpreted: (1)concentrationeffect; (2) time effect;(3)matching exposure in the workplace with worker's integrated exposure, which includes absorption through all entries (pulmonary, dermal and gasintestinal), accumulated residues and possible non-occupational exposure; (4) depicting effects of external and internal factors which alter the relationship between intensity of exposureand biological concentration and body burden of the indicator of exposure; ( 5 )extrapolation and prediction of biological concentrationsresulting from exposure to new compounds or new exposure conditions; and (6) verification of data. DMAC metabo limit which is c excreted in urin, cadmium conce Evaluation of tir Since the ui metabolites are timing and dura the biological rr life,which deper and the activit! FISEROVA-BER Samplingtir phase (withhall sample for mea polluted area, t the samplewas (a single breatt Evaluation of concentration eflect At low exposures, the relationship between extent of exposure and biological concentrations is usually linear. At high concentrations, saturable processes disturb the linear relationshipand can induce undesirablebiochemical and functionai change& Although saturability and its consequences are usually studied in laboratory anim&,a limited amount of human data is available from field studies. The interference of saturable processes in biological monitoring of exposure and health effects is shownby the followingtwo examples. (1) Saturated metabolism of dimethyl acetamide, DMAC, was observed in workers exposed to approximately 10ppm of DMAC, which is the current limit for occupational inhalation exposure (KENNEDY,1990). Since, at this exposure, the relationship between intensity of exposure and urinary excretion d , measurements t or a couple of , shorter than 5 time and pefic ,, Since it is elimination ha Application of toxicokinetic models metabolites approaches the plateau, biological monitoring of DMAC metabolites in urine provides no quantitative information on the DMAC exposure permissible in the workplace. Moreover, the overloading of the microsomal mixed-function oxidase ~ system may alter metabolism of other concurrently- present industrial chemicals and ' medications, the metabolism of which is mediated by the same enzyme system as ! DMAC metabolism. This raises the question of whether an occupational exposure limit which is outside the range of linear kinetics sufficiently protects the worker ' (FSEROVA-BERGEROV19A8,1). (2) The second example is the binding of cadmium to renal cortex proteins. When the capacity of the binding sites is exceeded as a result of large or long-lasting exposure, renal function is impaired and excessive cadmium is excreted in urine. Consequently, the relationship between the extent of exposure and cadmium concentration in urine is altered (KJELLSTROeMt al., 1984). Evaluation of time effect Since the uptake, distribution and elimination of industrial chemicals and their metabolites are kinetic processes, the outcome of the measurements depends on the timing and duration of the sampling.The concentration of the indicator of exposurein the biological matrix can change rapidly or slowly, depending on its elimination halflife, which depends on biosolubility of the compound, its susceptibiiityto metabolism, and the activity and body build of the worker (FISEROVA-BERGEROVA et al., 1984; FISEROVA-BERGEROVA, 1985). Samplingtime and the duration of the samplingperiod during the rapid elimination phase (withhalf-lives of a few minutes)are very critical. For example,if the exhaled air sample for measurement of a volatile compound is collected shortly after leaving the polluted area, the outcome of analysis very much depends not only on the exact time the sample was collected, but also on whether the samplewas collected instantaneously (a single breath) or over a period of time (Table 1). Similar considerations apply to TABLE 1 . EFFECT OF DURATION OF SAMPLING ON CONCENTRATION OF INDICATORS OF EXPOSURE WITH A SHORT ELIMINATION HALF-LIFE* *The concentrations are related to the concentration in the instant sample, wh~chfor convenience equals 100. ' measurements in random urine samples which may represent voidance of less than 1 h or a couple of hours. If the indicator of exposure is excreted rapidly (with a half-life shorter than 5 h), then the outcome of the measurements is significantlyaffected by the time and period between voiding. Since it is difficult to control sampling for measurements of indicators with short elimination half-lives, the time factor (the time between leaving the polluted workplace and sampling) may introduce an error in measurements. For this reason, me Values, BAT) pro ments of volatile solvents in exhaled air or in blood, collected `during ex respectively. Botk shortly after the end of exposure, are not suitable for routine quantitativ airborne and biok Measurementsof indicators with longeliminationhalf-lives,or in samp likely result from c during the slow elimination phase, are less likely to be affected and are obtained m However, if the half-life is longer than 10 h, the accumulation o and/or its metabolite causes the biological concentrations to week, month or lifetime. Consequently the concentrations in samples collected (170cm tall,70 kg the other hand, ai exposure to MAK beginning of the working week are lower than those in samples collected at higher than BEIs 1 the working week, or after months or years of exposure (DROZ, 1978,1989; exposure derived BERGEROVA et nl., 1974, 1980; FISEROVA-BERGEROVA, 1987a,b). concentrations H. The elimination half-life determines whether the measurement is an indicator toxicokinetic bas: recent exposure or exposure over the day, week, month or lifetime (DROZ, 1989). Table 2 provides guidelines as to how the eliminationhalf-life affects the sampling are re-examined. To simplify th and information on the type of exposure. for samplescollec `end of the shift' TABLE 2. DEPENDENCE OF EXPOSURE EVALUATION AND SAMPLING TIME ON ELIMINATION HALF-LIR `f Exposure Sampling time after the end of 1 weeks),the samp specimenscollect 1986; COMMISSIO tf<2 h Recent Very critical 2<t4<5 h 5 <tf <48 h ti248 h Daily Weekly Monthly or lifetime Critical End of work week Discretionary (after months of exposure) Factors agecting Information factors affecting by comparing 1 indicatorsof exp Matching extent of exposure and biological concentrations 3 toxicokineticba or unrepresenta c Evaluation of occupational exposure is usually based on comparing measured depicted by an concentrations of the airborne compound with the reference values for occupational additional or f inhalation exposure. Simulation models, because of their ability to match the extent of GULLEMIN1,98 exposures associated with the predetermined dose or biological concentrations of 1985,1987b;FIS indicators of exposure, are a valuable tool in extrapolation of reference values for 1974; SATO et workers with unusual workshifts (ANDERSEN et al., 1987b;SATO et al., 1990b). If the distribution cai health effect studied is associated with the biological concentration, then the concentration in the target organ should be of concern, as in acute effects such as carboxyhaemoglobinaemia, or depression of CNS function by organic solvents. If the health effect studied is associated with the capacity of the defence mechanism of the target organ, then the doses should be of concern, as in the development of renal injury resulting from exposure to some heavy metals. The best known reference values are Threshold Limit Values (TLV) adopted by the American Conference of Governmental Industrial Hygienists (ACGIH, 1989)and BERGEROVA, 19: JOHANSON and 1984; FISEROV! functions induc coexposureto i ` or ethnically k BERGEROVA et L and GUILLEMI the German Maximum Concentrations at the Workplace (MAK)recommended by the Commission for the Investigation of Health Hazards of Chemical Compounds in the Work Area (COMMISSIONFOR m INVESTIGATION OF HEALTHAZARDS, 1989).It is the policy of both organizations that their reference values for biological concentrations of indicators of exposure (Biological Exposure Indices, BEI, and Biological Tolerance r 1 GUBERAN activities (FISEF 1 airborne and i 1 dermal expos1 excessive work Application of tomcokinetic models values, BAT) provide the same extent of health protection as TLV-TWA or MAK. respectively. Both organizations employ the toxicokinetic approach to compare airborne and biological concentrations. BEIs are usually derived as values which most likely result from occupational exposure to TLV-TWA (FISEROVA-BERGEROVA, 1990a), and are obtained by simulation of the occupational exposure of a reference worker (170cm tall, 70 kg weight, light work-alveolar ventilation 15-20 1. min- '). BATs, on the other hand, are derived as the highest values expected to result from occupational exposure to MAK (HENSCHLER, 1990).Therefore,BATs are usually two or three times higher than BEIs (FISEROVA-BERGEROVA, 1990b).If the reference value for indicators of exposure derived empirically from field studies (that is, by comparing biological concentrations with health effect) deviates from the reference value derived on a toxicokinetic basis (equivalent to the airborne reference value), both reference values are re-examined. To simplify the evaluation of biological monitoring data, BEIs and BATs are given for samples collected at the specified time which, for convenience, is usually defined as `end of the shift' (meaning end of exposure) or `prior to the next shift' (meaning 16 h after the end of exposure). For chemicals with a long elimination half-life (days or weeks), the sampling time can be discretionary,but the BE1 or BAT may apply only to specimenscollected after a certain length of exposure (forexample 6 months) (ACGIH, 1986; COMMISSION FOR THE INVESTIGATION OF HEALTH HAZARDS, 1989). Factors affecting the relationship between exposure and biological concentrations Information on possible additional sources of exposure, and on circumstantial factors affecting biological concentrations of indicators of exposure, can be generated by comparing the ratio of measured airborne and biological concentrations of indicatorsofexposure with the ratio of the referencevalues which were determined on a toxicokinetic basis. An unusual ratio, besides pointing to technical or analytical errors or unrepresentative samples, also points to other factors, the effect of which can be depicted by an appropriate simulation model. Examples of such factors are: (1) additional or fluctuating exposures, or unusual exposure duration (DROZ and GUILLEMI1N9,83; DROZ, 1978; FERNANDEZ et al., 1977; FISEROVA-BERGEROVA, 1981, 1985,1987b;FISEROVA-BERGEROVA et af.,1974,1980,1984;GUBERAaNnd FERNANDEZ, 1974; SATO et al., 1990b); (2) altered pulmonary ventilation or cardiac output distribution caused by workload or disease (DROZand GULLEMIN19, 83; FISEROVABERGEROVA, 1985,1987a,b; FISEROVA-BERGEROVA et al., 1980;JOHANSON, 1986, 1988; JOHANSON and NASLUND, 1988);(3) altered metabolism (FISEROVA-BERGEROVA et al., 1984; FISEROVA-BERGEROVA, 1987b); (4) changes in metabolism or physiological functions induced genetically (DROZ and SAVOLAINEN, 1990), by medication or by ` coexposure to other chemicals (ANDERSEN et al., 1987a;SATOet al., 1990a);( 5 )dietary or ethnically based differences in body build and blood composition (FISEROVABERGEROVA et al., 1980,1984);and (6)accumulation over the week, month, etc. (DROZ ' and GUILLEMI1N98, 3;FERNANDEZ et al., 1977;FISEROVA-BERGEROVA et af., 1974,1980; GUBERAaNnd FERNANDEZ, 1974; PERBELLINI et al., 1986),and effect of post-exposure activities (FISEROVA-BERGEROVA, 1987b).If the ratio of measured concentrations of the airborne and indicator of exposure is much smaller than the TLV-BE1 ratio, then 1 dermal exposure, additional non-occupational exposure, bad working practices, excessive workload or interference by other industrial chemicals or medication are indicated. On the other hand, if the ratio of measured values is much larger than BE1 ratio, then interference by another compound(s) is indicated. There are four causes of interference by another compound: (1) the compound is either the studied indicator of exposure or it is biodegra studied indicator. Under such circumstances,the ratio of the measured c is smaller than the TLV-BE1 ratio; (2) the other compound inhibits the meta the compound under study, the inhibition being manifested by the increa level of the parent compound (for which the measured values ratio 6 T L and by the decrease of biological concentrations of the metabolite (for measured values ratio 9TLV-BE1 ratio); (3) the other compound acts stimulator(inducer)or releasesthe indicatorof exposurefrom the bond to tissues. This interference has the opposite effect of inhibition; (4) the other c o r n p o w altersthe distributionof cardiac output or it alters the functionof an organ (lung,kidnq, liver, skin permeability)which plays a role in uptake and elimination of the cornpow. Dermal exposure A simulation model for dermal exposure of volatile solvents is shown in Fig. 1, Dermis under the exposed area is treated as a separate compartment for which & inflow of the compound is defined by flux. Flux can be determined experimentally, predicted from aqueous solubility, octanol-water distribution coefficient and m o b cular weight of the compound (FISEROVA-BERGEROVAet al., 1990).Figure 2 showsb profound increase of alveolar concentration of methanol and toluene caused by short intermittent dermal exposures of 2.5% or 5% of body surface to the liquid solvent during inhalation exposure to TLV-TWA. The simulation study shows that the larger the exposed area, or the better the perfusion of dermis under the exposed area, the larger the increase in alveolar concentration and the smaller the pulmonary uptake. If the dermal exposure is extensive, the pulmonary uptake can be suppressed and the percutaneously absorbed solvent exhaled. Under such circumstances, the vapour concentration in exhaled air is larger than in the ambient air. Thus, concentrations of highly volatile solvents in exhaled air samples can be the most sensitive indicator of dermal exposures. Modelling of dermal penetration is used for identification of compounds with the potential to affect biological concentration of the compound or its metabolite and toxicity resulting from occupational exposure (FISEROVA-BERGEROVAet al., 1990). Prediction, extrapolation and verification Simulationmodelscan be used for description and verificationofexperimental data (CLEWELeLt al., 1988; DROZ and GUILLEMIN19, 83; DROZ,1978; FERNANDEZ et al., 1977; FISEROVA-BERGEROVA et al., 1974; GUBERAN and FERNANDEZ, 1974;JOHANSON and NASLUND19,88,JOHANSO1N9,86;LIRAet af.,1990; PAITRSONand MACKAY, 1986; PERBELLMet Ial., 1986; RAMSEY and ANDERSEN, 1984). The following example shows how simulation models can be used to explain the inconsistencyof information. FERNANDEZ et al. (1975)determined the elimination half- lives for triphasic pulmonary elimination of trichloroethylene to be 5-20 min, 1-3 h and 10-30 h. SATO et al. (1977), in similar experiments in volunteers, measured elimination half-lives of 2.7 min, 0.4 h and 4 h, respectively. The five-compartmental simulation model, solved by Laplace transform, showed the observable elimination FIG. 1. Toxicoki (depicted by the exposed area); h VRG (well-perk given in the UPF shown to the rig volume valuesfc lower cornen of `c,; denotes ir half-lives of t elimination differences b ! concentratic provide dat; 1 concluded t II concentratic 1 determinatil I Simulati Application of toxicokinctic models 647 31.3 4.O MG *cn NO - d e m l s ) 4 *5' 3a P w> II nces, the v ed to explain I 2 eliminationbr 15-20min, 1 1 nteers, mas03 e-compartmcn ,able eliminati CI (I/min) "d FRC+"Vtld+Vad'bl/gas+ vlunghng /gar FIG. 1. Toxicokinetic model for simulation of dermal absorption of organic solvents. The compartments (depicted by the rectangles) are: LUNG (connective iiSSueS and airspace); DERMIS (dermis under the exposed area); MG (dermis under unexposed area and muscles); FG (adipose tissue and white marrow); VRG (well-perfused organs,except the liver); and LIVER. Volumes of the compartments `V`(in litres) are given in the upper left Corner of each rectangle. The perfusion for each compartment, `F(in I. min-I), is shown to the right, above the arterial flow. For the DERMIS and MG compartments, the perfusion and volume valuesfor 5% of body surfaceexposure to liquid are given underneath the arterial flow and in the left lower comers of the rectangles, respectively (indicated by t).Alveolar ventilation, `v,,v',equals 10 1. min- I ; `cair'denotes inspired concentration, FRC denotes functional residual capacity and `1'denotes the appropriate tissue-gas partition coefficient. half-lives of 3 min, 0.8 h and 26 h. Further simulation revealed that the short elimination half-lives reported by SATOet ai. (1977) cannot be attributed to ethnic differences but to the experimentaldesign. These authors measured trichloroethylene concentrations in exhaled air for 10 post-exposure hours. This period is too short to provide data for the determination of the half-lifeof the slow elimination phase. We concluded that the study of FERNANDEZ et al. (1973, in which the trichloroethylene concentration in exhaled air was measured for 3 days, provides better data for the determination of elimination half-lives than the study by SATOet al. (1977). Simulation models can also be used for prediction of uptake, distribution and 648 V. FISEROVA-BERGEROVA EFFECT OF IMMERSING HANDS IN SOLVENTS -1 TOLUENE 5min \ h 2 10 min 5% A 5 min I\i-l8nin . 7 4.9 t 0' I I -I, elimination of new (FISWOVA-BERGER indicators of exposur et al., 1977; FISWOV 1974; PERBELLINI et Simulation modc and biological conce occupational exposi resting volunteers. R of ethnic difference: enzymeactivity and SAVOLAI", 1990). The simulation I elimination of vapo animal to man (FIS RAMSEYand ANDER and quantitative sp The simulation moa by the suckling of I Toxicokinetic r of inhalation andl indicators ofexpo! contribute to the 5=0O I compounds. They ? 0 t -ci 2 8 30 ~ Smin 5 min verification of da! exposureare deter for the evaluatior relationship betw -2 a 20 concentrations of selection of the 10- 5min 2.5% procedure, and ir develop similar r -_ _ _ _ _ _ _ _ _ - . - . - - - - - - _ 0 _. 00035 _ _ _ _ . . _and compounds I ,I 100 -- - TLV only 2 00 300 -TLV +skin 400 min 2 4 hrs increased perfusion FIG. 2. Effect ofshort-term dermal exposure to liquid toluene and methanol on alveolar concentrations. The simulation is done using the model and values shown in Fig. 1. Broken lines depict alveolar concentratiom during 8-h inhalation exposure to TLV-TWA. Solid lines depict alveolar concentration when dermal absorption is added. The peaks result from dermal exposures of given area (% of body surface) for a short period b i n ) , both given at the top of the peaks. Flux values are taken from FISEROVA-BERGEROVAet at. (1990),and the partition coefficientsare taken from FISEROVA-BERGEROVAand D1az (1986). Note, that after two dermal exposures the alveolar concentrations exceed the inspired concentration, thus reversing pulmonary uptake in pulmonary wash-out. The differences in the patterns of alveolar concentrations of toluene (relatively rapid decline to the concentration resultingfrom inhalation exposure only) and methanol (relatively slow decline resulting in a rising alveolar concentration to a level about 200 times higher than a concentration resulting from inhalation exposure only) reflect the differences in their solubility in blood. I ACGIH (1986) Doc1 supplements to I' USA. ACGIH (1989) Thre * of Governmental ANDERSEN,M.E.,G ofthemetabolic 1 1 'methods. Toxic0 ANDERSENM, .E., systemicextract1 in rats based on --1) Application oi toxicokinetic models elimination of new compounds if their solubility in water and lipids is known (FISEROVA-BERGEROVAet al., 1984),and for prediction of biological concentration of indicators of exposure in organs and tissues (DROZand GUILLEMI1N98, 3;FERNANDEZ et al., 1977;FISEROVA-BERGEROVAet al., 1974,1980,1984;GUBERAaNnd FERNANDEZ, 1974; ~ERBELLINetIal., 1986). Simulation models are also used for extrapolation of data on pulmonary uptake and biological concentrations of the compound or its metabolitesduring and following occupational exposure (with workload) from data obtained in controlled studies in resting volunteers. Recently,the simulation model was employed to evaluate the effect of ethnic differences in physiological parameters (mainly body build, metabolizing enzyme activity and life style)on uptake and elimination oforganic solvents(DROZand SAVOLAINEN, 1990). The simulation models are also used for extrapolation of uptake, distribution and elimination of vapours inhaled by differentanimal species and for extrapolation from animal to man (FISEROVA-BERGEROVAand HUGHES,1983; PAUSTENBAeCtHal., 1988; RAMSEYand ANDERSEN, 1984; REITZ et al., 1988). However, unpredictable qualitative and quantitative species differences in metabolism make the extrapolation uncertain. The simulation model was also modified to study the uptake of industrial compounds by the suckling of a breast-feeding worker (SHELLEY et al., 1988). -- 8 , - oOo03s 24 hrr tn CONCLUSIONS Toxicokineticmodels are used for determiningthe relationships between the extent of inhalation and/or dermal exposure and the biological concentrations or dose of indicators of exposure. Physiologically based multicompartmental simulation models contribute to the understanding of the fate of inhaled or percutaneously absorbed compounds. They are used in biological monitoring for prediction, extrapolation and r verification of data and principles on which the reference values for indicators of exposure are determined and the monitoring data are interpreted. They have been used for the evaluation of the effects of a variety of external and internal factors on the relationship between extent of exposure to organic solvents and dose or biological concentrations of indicators of exposure. The elimination half-life is considered in the selection of the appropriate exposure indicator, in the design of the sampling procedure, and in the interpretation of the measured data. There is a great need to develop similar predictive toxicokinetic models for exposure to metals, particulates and compounds undergoing binding to constituents of tissues. REFERENCES ACGIH (1986) Documentation of the threshold limit values and biological exposure indices (1986 with ACGIH (1989)Threshold limit va!ues and biological exposure indices for 1989-1990. Amencan Conference of Governmental Industrial Hygienists, Cincinnati, Ohio,U S A . t, ANDERSEN, M. E., GARGAMS,. L., CLEWELHL., J., 111and SEVERYKN., M. (1987a)Quantitativeevaluation of the metabolic interactions between trichloroethyleneand 1,ldichloroethylene in oioo using gas uptake methods. Toxicol. Appl. Pharmacol. 89, 149-157. ANDERSENM,. E., GARGASM, . L. and RAMSEY,J. C. (1984) Inhalation pharmacokinetics: evaluating 650 V. FISEUOVA-BERGEROVA ANDERSEN,M.E.,MACNAUGHTOMN.,G.,CLEWELLH, . J., 111and exposure limits for long and short exposure periods using a phy ind. Hyg. Ass. J . 48,335-343. CLEWELLH..J., 111,ANDERSENM, .E., MACNAUGHTOMN.,G. and STUART,B. 0.(19 analytical tool for assessing chemical hazards to man. Auiut. Space Enuiron. (November). COMMLSIOPONR THE INVESTIGATION OF HEALTH HAZARDS OP CHEWCUC. OMPOUNDINSTW (1989)Maximum concentrationsat the workplace and biological tolerance valuesfor worki Report No.XXV. VCH Publishers, Weinheim, F.R.G. DROZ, P.0.(1978)Contributionto the research ofbiologicalexposureindicesforsolvents. their partition codlicients and a study of their compartmentsin the organism using simu Institut de Chimie de I'Universite de Neuchatel, Switzerland (in French). DROZ.P.O. (1989) Biologicalmonitoring I: Sourasof variability in human rcsponscto chemicalsx App. i d . Hyg. 4, F-20-F-24. DROZ, P.0.and GULLEMIN,M.P.(1983) Human styrene exposure V. Development of a m biological monitoring. Int. Archs occup. Enuiron. Hlth 53, 19-36. DROZ,P. 0.and SAVOLAINEN, H.(1990) Effect of physiological and metabolic differenas bet Japanese and American populations on uptake, distribution and elimination of toxic c Biological Monitoring of Exposure to Industrial Chemicals (Edited by FISEROVA-BERGER OGATAM,.),pp. 185-188. Proceedings of the United States-Japan Co-operatiw Seminar on Blo Monitoring, Honolulu, 1989. American Conference of Governmental Industrial Hygienists, C Ohio, U.S.A. DROZ, P.O., Wu, M.M.and CUMBERLANWD. G, . (1989a) Variability in biological monitorin solvent exposure. 11. Application of a population physiological model. Br. J. ind. Med. DROZ, P.O., Wu, M.M., CUMBERLAWND. G, . and BERODEM,.(1989b of solvent expoaure. I. Development of a population physiological FERNANDEZ, 1. G., DROZ, P.O., HUMBERBT., E. and CAPER-, J. R Simulation of uptake, excretion,and metabolism using a mathemat FERNANDEJZ. G, ., HUMBERBT., E., DROZ, P. 0.and CAPEROSJ., R. Balance of absorption,excretion and metabolism in humans. A r c h French). FISEROVA-BERGEROVA,V. (1981)Modeling of uptake and clearance of inhaled vapors and gases.In and Enuironmental Xenobiotics (Edited by GUT,I., CIKRTM, .and PLM, G . L.),pp. 211-220. Springpr, Berlin. FISEROVA-BERGERVO. V(EAd,itor)(1983)Modeling of Inhalation Exposure to Vapors: Uptake, Distributhrt, and Elimination, Vola I and 11. CRC Press,Boca Raton, Florida, U.S.A. FISEROVA-BERGEROVA, V. (1985)Toxicokinetics of organicsolvents. Scond. J. W k Environ. HIth 11, SUM1., 7-21. FISEROVA-BERCERVO.V(1A9,87a)Simulationmodel asa tool for adjustment of biological exp0sure-b exposure conditions.In Biological Monitoring of Exposure to Chemicals: Organic Compounds (Edited Ho,M. H. and DILMN,H. K.),pp. 29-57. John Wiley & Sons, New York. FISEROVA-BERGEROVA(THOMAS),V. (1987b) Development of biological exposure indices (BEIs) and implementation. Appl. ind. Hyg.2, 87-92. FISEROVA-BERGEROVA (THOMAS), V. (1990a) History and concept of BEIs. In Biological Monitoling 4 Exposure to Industrial Chemicals (Edited by FISEROVA-BERGEROVA, V. and OGATAM, .), pp. 19-23. Proceedings of the United States-Japan Co-operatiw Seminar on Biological Monitoring, Honolulu, 1989. American Conference of Governmental Industrial Hygienists, Cincinnati, Ohio, U.S.A. FISEROVA-BERGEROVA (THOMAS), V. (1990b) Comparison of biological exposure indices (BEIs)and biological tolerance values (BATS).in Biological Monitoring of Exposwe to Industrial Chemicals (Edited by FISEROVA-BERGERVOVanAd,OGATAM,.),pp. 31-33. Proceedings of the United States-Jam Co. operative Seminar on Biological Monitoring, Honolulu, 1989. American Conference of Governmaat Industrial Hygienists, Cincinnati, Ohio, U.S.A. FISEROVA-BERGEROVA, V. and DLAZ,M.L. (1986) Determination and prediction of tissue-gas partition coefficients. Int. Archs occup. Enuiron. Hlth 58,75-87. FISEROVA-BERGEROVA, V. and HUGHES, H. C. (1983)Species differences in bioavailabilityof inhaled vapon and gases. In Modeling of Inhalation Exposure to Vapors: Uptake, Distribution and Elimination, Vol. U (Edited by FISEROVA-BERGEROVA, V ), pp. 97-106. CRC Press, Boca Raton, Florida, U S A . FISEROVA-BERGERVO.,VPAIE,RCE,J. T and DROZ, P.0.(1990) Dermal absorption potential of indwtrirl chemica!s: criteria for skin notation. Am. J. ind. Med. 17, 617435. FISEROVA-BERGEROVA, V., TICHY, M.and DI CARLO, F. J. (1984) Effects of biosolubdity on pulmoauy uptake and disposition of gases and vapors of lipophilic chemicals. Drug Metab. Rev. 15, 1033-1070. FI~OVA-BERGEROVA and excretion of g; FI~OVA-BERGEROVA. distribution. and e G-ERAN, E. and FER! alveolar mncentra HENSCHLER,D. (1990 Monitoring of Exp pp.25-30. Procee Honolulu, 1989.A JOHANSON,G . (1986) ' Toxicol. Lett. M, JoHANSON. G. (1988) JOHANSON, G. and NI based modeling 0 KENNEDY,GL..,JR ( 1 Exposure to lndu proceedings of the Amencan Confer1 K E L ~ O M T, ., ELI! concentration of 1 LURA,J., JOHANSON, man. Toxicol. Le LOWE, H.J. (1972) D Illinois, U.S.A. PATERSON, S.and M approach. Toxic( PAUSTENBACKD. J. based pharmacok pERBELLINI, L., Mod studying human 2,5-hexanedione RA-, J. C. and pharmacokinetic ~ I T ZR, . H.,MENDR of in vitro WYm chloride: impha %TO, A., ENWH, K inhaled vapors 0 by FISEROVA-BEE perotive Seminar Industrid Hy@e SATOA, ., KANEKO, doses of oganic FISEROVA-BERG[ operative Semina Industrial H Y P Sam, A., NAKAJIM) excretionoft r d S ~ L L E Y , ML A. N mother and nu* VLACH. J. (1983) 5 Elimination, VO s U.S.A. WAGNER,J. G . pharmacokinet have been fitkc 8 i Application of toxicokinetic models 651 FISEROVA-BERGERVO.V, VAL,ACH,J. and CASSADYJ,. C. (1980)Predictable 'individual differences' in uptake and excretion of gases and lipid soluble vapours simulation study. Er. J. ind. Med. 37,4249. FISEROVA-BERGEROVA, V., VLACH,J. and SINGHALK, . (1974) Simulation and prediction of uptake, distribution, and exhalation of organic solvents. Br. J. ind. Med. 31.45-52. GUBERAEN.,and FERNANDEJZ.,(1974)Control of industrial exposure to tetrachloroethyleneby measuring alveolar concentrations: theoretical approach using a mathematical model. Br. 1.i d . Med. 31,159-167. HENSCIUER, D. (1990) Biological monitoring and exposure limits as perceived in Germany. In Biological Monitoring of Exposure to Industrial Chemicals (Edited by FISEROVA-BEXGEROVVaAn,d OGATAM, .), 1 pp. 25-30. Proceedings of the United States-Japan Co-operative Seminar on Biological Monitoring, -Hon~olul~.~u. 1~989..Ame~ric~an C~onferenceofGovernmentalIndustrial Hygienists, Cincinnati, Ohio, U.S.A. JOHANSON, G:(1986)Physiologicallybased pharmacokineticmodeling ofinhaled 2-butoxyethanolin man. Toxicol.Lett. 34.23-31. JOHANSON, G. (1988)Aspects of biological monitoring of exposure to glycol ethers. Toxicol. Lett. 43.5-21. JOHANSOGN.,and NASLW, P. H. (1988) Spreadsheetprogramming-a new approach in physiologically based modeling of solvent toxicokinetics. Toxicol. Lott. 41,115-127. KENNEDY, G. L., JR (1990) Biologicalmonitoringin American chemicalindustry. In Biological Monitoring of Exposure to Industrial Chemicals (Edited by FISEROVA-BERGEROVV.Aa,nd OGATAM, .), pp. 63-47. Proceedingsof the United States-Japan Co-operatiw Seminar on Biological Monitoring, Honolulu, 1989. American Conference of Governmental Industrial Hygienists, Cincinnati, Ohio, U.S.A. KJELLSTROMT, ., ELINDERC,. G. and FRIBERGL, . (1984)Conceptual problems in establishing the critical concentration of cadmium in human kidney cortex. Environ. Res. 33, 284-295. LIIRAJ,., JOHANSOGN,. and RIWMAKVI,. (1990)Dosedependent kinetics of inhaled methyl ethyl ketone in man. Toxicol. Lett. SO, 195-201. LOWE, H. J. (1972)Dose-regulated Penthrane" methoxyfluraneanesthesia. Abbott Laboratories, Chicago, Illinois, U.S.A. PATERSONS, . and MACKAYD,. (!986) A pharmacokinetic model of styrene inhalation with the fugacity approach. Toxicol. Appl. Pharmacol.82,444-453. PAUSTENFIACDH. J, ., CLEWELLH,. J., 111, GARGASM,.L. and ANDERSENM, . E. (1988) A physiologically based pharmacokinetic model for inhaled carbon tetrachloride. Toxicol. Appl. Pharmacol. %, 191-211. PERBELLINLI,., Mozzo, P., BRUGNONFE., and %DE, A. (1986) Physiologicomathematical model for studying human exposure to organic solvents: kinetics of blood/tissue n-hexane concentrations and of 2,s-hexanedione in urine. Br. J. ind. Med. 43,700-768. RAMSEY, J. C, and ANDERSENM,. E. (1984) A physiologically based description of the inhalation pharmacokinetics of styrece in rats and humans. Toxicol. Appl. Pharmacol. 73, 159-175. REITZ, R.H., MENDRALA, A. L., PARK, C. N., ANDZRSEN,M. E.and GUENGWCHF,. P. (1988)Incorporation of in vitro enzyme data into the physiologically-based pharmacokinetic (PB-PK) model for methylene chloride: implications for risk assessment. Toxicol Lett. 43.97-116. SATO, A., ENDOH,K. and KANEKOT, .(1990a) Effectsof ethanol on uptake, distribution,and elimination of inhaled vapors of organic solvents. In Biological Monitoring ofExpsure to Industrial Chemicals (Edited by FISEROVA-BERGERVO.VaAnd, OGATAM, .), pp. 155-158. Proceedings ojthe United States-Japan Co- peratiue Seminar on Biological Monitoring, Honolulu, 1989. American Conference of Governmental Industrial Hygienists, Cincinnati,Ohio, U.S.A. SATOA, ., KANEKOT,., ENDOH,K. and JOHANSON,G. (1990b) Relationship between external and internal doses of oganic solvent vapors. In Biological Monitoring of Exposure to Industrial Chemicals (Edited by FISEROVA-BERGERVO.VaAnd, OGATAM, .), pp. 137-143. Proceedings of the United States-Japan COoperative Seminar on Bobgical Monitoring, Honolulu, 1989. Amencan Conference ofGovernmental Industrial Hygienists, Cincinnati, Ohio, U.S.A. SATO, A., NAKAIIMTA.,,FUJIWARAY., and MURAYA~N~. A(1,977)A pharmocokinetic model to study the excretion of trichloroethyleneandits metabolites after an inhalation exposure. Br.J . ind. Med. 34.5643. SHELLEMY., L., ANDERSEMN,. E. and FISHERJ., W. (1988)An inhalation distribution modelfor the lactating mother and nursing child. Toxicol. Lett. 43, 23-29. VLACH, J. (1983) Simulation. In Modeling of Inhalarion Exposure to Vapors: Uptake, Distribution, and Elimination,Vol. I (Edited by FISEROVA-BERGERVO.)V, pAp,. 133-153.CRC Press, Boca Raton, Florida, U.S.A. WAGNER, J. G. (1976) Linear pharmacokinetic equations allowing direct calculation of many needed pharmacokinetic parameters from the coefficients and exponents of polyexponential equations which have been fitted to the data. J. Pharmacokinet. Biopharm. 4,443-467.