Document Rag7vXM6Jbdz7d6wxK0ZrNn8

Proposal for the assessment of quantitative dermal exposure limits in occupational environments: part 2. Feasibility study for application in an exposure scenario for MDA by two different dermal exposure sampling methods D H Brouwer, L Hoogendoorn, P M J Bos, P J Boogaard, J J van Hemmen Abstract Objective-To evaluate two different tech- niques for assessing dermal exposure to 4,4`-methylene dianiline (MDA) in a field study. The results were used to test the applicability of a recently proposed quan- titative dermal occupational exposure limit (DOEL) for MDA in a workplace scenario. Methods-For two consecutive weeks six workers were monitored for exposure to MDA in a factory that made glass fibre reinforced resin pipes. Dermal exposure of the hands and forearms was assessed during week 1 by a surrogate skin tech- nique (cotton monitoring gloves) and dur- ing week 2 by a removal technique (hand wash). As well as the dermal exposure sampling, biological monitoring, measurement of h4DA excretion in urine over 24 hours, occurred during week 2. Surface contamination of the workplace and equipment was monitored qualita- tively by colorimetric wipe samples. Results and conclusions-Geometric means of daily exposure ranged h m 81-1762 pg MDA for glove monitoring and from 84-1783 pg M D A for hand washes. No significant differences, except for one TNO Nutrition and Food Research Institute, Division of Toxicology, Zeist, The Netherlands worker, were found between exposure of the hands in weeks 1 and 2. Significant differences between the mean daily exposure of the hands (for both weeks and sampling methods) were found for all D H Brouwer L Hoogendoorn PMJBos J J van Hemmen workers. The results of the colorimetric wipe samples indicated a general contamination of the workplace and equipment. Excretion of MDA in 24 hour urine Shell International Chemicals BV, Shell samples ranged from 8 to 249 pg MDA, whereas cumulative MDA excretion over a Research and week ranged from 82 to 717 pg MDA. Technology Centre, Cumulative hand wash and MDA excre- Department Molecular To~~co~o of ~Y - C1,S Q I ~ tion results correlation over a week showed (R=0.94). The highest a high actual Amsterdam, The daily dermal exposure found seemed to be Netherlands about 4 mg (hand wash worker A on day P J Boogaard 4), about 25% of the external DOEL. Test- ing of compliance by means of a biological Correspondence to: Dr J J van Hemmen, TNO limit value (BLV) led to similar results for Nutridon and Food Research the same worker. It is concluded that both h S o ' N t C , Division of Toxicology, Zeist, The Netherlands. dermal exposure monitoring methods were applicable and showed a compatible performance in the present exposure sce- Accepted 16 July 1998 nario, where the exposure relevant to der- mal absorption is considered mainly restricted to hands. The concept for a DOEL seemed to be relevant and applicable for compliance testing and health surveillance in the situation under investigation. ( O m p Enwiron Med 1998;55:805-811) Keywords: dermal exposure monitoring; h4DA; biological monitoring; compliance; dermal occupationalexposure limit In the fist part of this article a conceptual procedure to derive a dermal occupational exposure limit (DOEL) has been presented.' For testing whether occupational conditions are in compliance with a DOEL, it is essential that validated monitoring techniques are available. In risk assessment the assessment of whether a particular workplace environment or a given combination of tasks ofa worker could give rise to hazardous dermal exposure may be a requirement. As a consequence it may be necessary to monitor the workplace or the workers involved, which requires different approaches. DERMAL EXPOSURE PROCESSES AND PATHWAYS Contamination of the skin can be considered to be the result of dynamic processes that may be characterised as loading and removing.' The loading process is the result of: (a) deposition of and impaction with aerosols present in the work environment; (ZJ) immersion in a liquid or powder; (c) contact with a contaminated surface. The removal processes leading to decontamination are vaporisation of the contaminant, immersion in a clean liquid, and by mechanical forces (vibrations and shocks). These processes are variable and may change with time and be described as intermittent and continuous. Fenske defined dermal exposure as the product of skin loading rate (masdskin surface aredunit time) and area exposure (cm').' Dermal exposure is expressed in units of mass per unit time (pglh). Cherrie and Robertson proposed an alternative definition which recognises the biological process involved in skin absorption-that is, the concentration of the substance at the skin surface.4They argue that measurement of mass of material on the skin surface may be misleading in cases in which the concentration of the material is variable. As measurement of exposure with the suggested I ! definition of Cheme and Robertson is pres- only through the skin; 8 hour time weighted ently not possible, the definition of Fenske has average (TWA) concentrations (by personal air been adopted for the aim of the present study: sampling) were < 3 ppb. Also a suitable biologi- DE=Z (SLR,xA,) + .. + (SLR,XA,), cal monitoring method is available for MDA.b These findings suggest that exposure to MDA where DE=dermal exposure (PO)m;ay only have occurred by skin contact. SLR=skin loading rate for each region @g/cm2/ Hence, the exposure scenario for MDA offers h); A=exposed skin area for each region (cm'); the possible study of the applicability of a and n=nh skin region. DOEL in an occupational work environment by comparing external exposure measurements DERMAL EXPOSURE SAMPLING In the present report, it had to be ascertained that the sampling techniques met the accuracy requirements of the occupational exposure limits for precision, specificity, and sensitivity. The following requirements have been specified for an ideal assessment method for dermal with an external DOEL and internal exposure measurements (biological monitoring) with a biological limit value (internal DOEL)). Two different techniques were used to assess external exposure -that is, a surrogate skin and a removal technique. In part I a DOEL for M D A has been derived.' exposure' 5: (a) measure the amount that is available for penetration through the skin; (b) estimate skin loading unbiased by duration and time of sampling; (c) enable repeated sampling in time; (d) is applicable for all relevant anatomical regions; (e) mimics the various processes of loading and removing appropriately; and (r> have high resolution, low limits of detection, and is sufficiently validated. Material and methods PROCESS AND EXPOSURE In the production of reinforced plastic pipes glass fibre sheets are lined and then led, on a mould, through a bath of resin and hardener. The bath contains a (106:ZO w/w) mixture of an Epikote 826 solution in xylene (100:6 W/W) and MDA diluted with methyleneamine (27:13 w/w), so the M D A content is about 12%. The air is removed from the resin by DERMAL EXPOSURE ASSESSMENT TECHNIQUES Skin exposure sampling techniques fall into three categories: surrogate skin-for example, monitoring gloves, patches; chemical removal-for example, washing and rinsing, skin wipes; and tracer techniques-for example, fluorescent tracer in combination with image processing techniques. These categories have been described extensivelyby Fenske' and discussed with respect to the requirements given before by Van Hemmen and Brouwer.* It was concluded that none of the sampling techniques considered meet all requirements for an ideal sampling technique. For the application of such techniques for testing compli- ance of occupational dermal exposure with a DOEL, both the surrogate skin and the removal sampling techniques are limited by their poor resolution properties. The dermal dose/unit area (D4),the mass loading of the skin/unit of surface area, can only be expressed as the average D,over the (skin) area which has been sampled. However, we worked from a maximal (cumulative) skin loading. In the most conservative approach gross skin loading resulting from a dermal exposure process or cumulative mass estimation of all exposed body regions, should be added to ascertain compli- ance with a DOEL. Because of the ease of use for surrogate skin and removal techniques, dermal exposure was assessed by both these techniques in the present study. pressing the sheet with a brush or a roller on to the mould. The temperature of the mould is about 60C. After a period of about 20 minutes in a furnace at 70"C, subsequent winding of glass fibre filaments on to the mould occurs. The filaments are also led through a bath of resin and hardener at 65C before winding. The M D A content of this bath is about 21%. The bath is placed on a trolly that r u n s down the length of the mould. Baths are filled manually with buckets that are filled at the so called day station (tempory container). Excessive resin is removed from the mould during wind- ing with a flexible spatula or a bucket while pressing the glass fibre to the mould. The thickness of the pipe is checked by a marking gauge or a measuring tape. A label is put on to the pipe, before the last layer of resin and glass fibre. Then the moulds are removed and the pipes are transported to a furnace at 100125C for further hardening. Most tasks are performed with protective gloves (coating of natural rubber on cotton) and a non-woven disposable coverall (Tjvek). Activities performed with bare hands that may result in potential dermal exposure are removal of excessive resin, incidental leading the glass fibre filaments through parts of the machine, measuring large diameter pipes with a measuring tape, and the use of scissors and labelling the pipes. Also, exposure may occur during the removal of the protective gloves, due to cross contamination of surfaces-for STUDY OBJECTIVE example, machine buttons-and due to The objective of the study was to investigatethe splashes during lining and winding. Extensive feasibility of the applicationof dermal exposure air monitoring programmes for the previous 3 sampling techniques for testing compliance of years showed no detectable inhalation expo- occupational dermal exposure with a concep- sure (data not presented). tual DOEL in a field study. A plant producing glass fibre reinforced resin pipes offered STUDY POPIJUTION AND SAh4PLD;G STRATEGY suitable conditions for this. Workers were Six men-two liners and four winders- exposed to 4,4'- methylene dianiline (MDA) participated in the 2 week study. The first week Quantiran-w dermal exposure limits in occuparional envzronments: part 2 during the afternoon shift (1400-2200), and the second week during the morning shift (0600-1400). During the first week dermal exposure of the hands of three out of six workers was assessed by a hand washing method, at the start ofa break, and immediately after the shift. Urine was collected for 24 hours from the start of a workshift until the start of the next shift for five consecutive work days and during the weekend afterwards.Worker F had a six day working week. Before the first day of monitoring a spot sample of urine was delivered. Dermal exposure of the hands of the other three workers was monitored by cotton glove monitors which were worn beneath the protective gloves. During the second week dermal exposure of the first three workers was assessed by cotton glove monitors, whereas the other three workers entered the hand wash biological monitoring programme. To ensure similar exposure scenarios workers did not change machines or major activities-lining or winding-during the monitoring period. Dermal exposure of other parts of the body was assessed by visual inspection of the coverall and wipe sampling of the inside of the coverall and the skin in case of splashes. Also, surface wipes were performed on workplace surfaces. ASSESSMENT OF HAND EXPOSURE Hand wash method Workers were asked to wash their hands twice with about 1.5 ml of a hypoallergic soap (Sporex, Kimberly-Clark, Veenendaal, The Netherlands) for 15 seconds then to rinse them with tap water. A specially designed hand washing device was used, consisting of a tube attached to the tap of the water supply in the changing room, an adjustable flow control set at a flow rate of about 1.5 Vmin, a tap, a funnel, and a 5 1 polyethylene bottle to collect the rinse water. After the hand washing the funnel was rinsed and the total volume of the water was assessed by weighing. Immediately after weighing, 50 ml rinse water was transferred by a 50 ml jug into a 100 ml polyethylene bottle. Ten ml of hydrochloric acid (37%)was added to the rinsing samples and stored at room temperature. Within 24 hours the samples were transported to the laboratory and stored at 4C until analysis. h4DA recovery was found under the conditions of storage. Field blanks of gloves and hand wash solutions were taken to determine possible contamination during the sampling procedure. ASSESSh4EhT OF BODY EXPOSURE AND SURFACE CONTAMINATION Contamination of the coveralls was found during and at the end of the shift and the positions of the contamination were registered. After the shift the coverall was removed and turned inside out. At positions on the inside of the coverall where contamination was found surface wipe samples were taken with colorometric surface swypes, according to the manufacturers' procedure (CLI Laboratories, Des Plaines, IL,USA). This is a semiquantitative indicator; the colour indicationis sensitivewith a dection limit of about 3-5 pg of MDA.' Similar procedures were followed during the sampling of contamined surfaces, operation buttons, equipment, and latches of doors. Surface swypes were contaminated with MDA and MDA/Epikote 828 mixture (100:28 w/w) to check their specificity. A colour indication was found, whereas contamination with Epikote 828, mehtyleneuiamine, and xylene did not show a colour change. Positions on the skin of the worker similar to contaminated positions of the coverall were wiped with skin swypes (CLI Laboratories, Des Plaines, IL,USA). All work- ers wore jeans underneath a non-woven disposable coverall (Tjvek). Also, workers A, G, and I wore long sleeved shirts, whereas workers F and H wore T-shirts. The torso of worker E was not covered by work clothing. BIOLOGICAL MONITORING Workers collected urine samples for 7 days in portions of 24 hours in 2.5 1 jerry cans. Spot samples were taken before the start of the first shift and before and after the 7 day period of biological monitoring: After delivery of the samples the volume of the urine samples was measured. From each 24 hour urine sample 20 ml was transferred into a 100 ml polyethylene bottle and 10 ml of hydrochloric acid (37%) was added and stored at room temperature. Within 24 hours the samples were tranported to the laboratory and stored at 4C until analysis. Cotton glove monitoring At the start of the shift and after each break prewashed long sleeved cotton gloves (cotton stretch 200 g/mz,Van der Wee, Riel, The Netherlands) covering about 1300 cm' were provided to the workers. At the start of the break and after the shift the gloves were removed by the investigators and each pair was transferred into a 11 bottle containing 500 ml of a 0.1 N potassium hydroxide solution in methanol and stored at room temperature. Within 24 hours the samples were transported to the laboratory and stored at 4C until analysis. Validation experiments in the laboratory showed a decrease of MDA recovery in MDNresin mixture on the glove of about 30%within 2 hours when the glove was stored at 30C.' No effect on EXTRACTION AND CHEMICAL ANALYSES By addition of potassium hydroxide the pH of the rinse samples was adjusted to 12. After the addition of 4,4`-diaminodiphenylethane (EDA) as an internal standard hexane was used to for extraction. After evaporation of the hexane at 50C the residue was added to the mobile phase (methanol and Pic A (100:95) and injected to an high performance liquid chromatography (HPLC) system (SpH 125 FIX) and detected by ECD (LDC, Milton Roy, USA). A part of the extraction solution of the cotton glove samples was transferred into a tube and inserted into the mobile phase, including the internal standard EDA. The detection method of MDA was similar to that for the hand rinse samples. The limit of 808 Brouwer, Hoogendim-, BOSe, t d Table I Results of &dy exposure aisessment of the hands mrh %ker yobride rrarwn Hand w h rerultr (ug MDA) Samples Samples (n) d O Q in) Ranget$ AM$ SD GM$ 95%CI Glow rrrvlrr (pg MDA) Samples Samples (n) <LOQ(n) Ranget$ Ah4 SD GM$ M%CI A Winder Machine7 19 F Winder Machine7 23 E Winder Machine3 21 H Winder Machine3 20 G Liner Machine3 20 I Liner Machine7 20 0.00 0.00 0.00 0.00 11 0.00 213-4046 995-1875 777-1540 240-927 <39.5-203 288-950 2040 1390 1762' 1717 to4332 1582 261 1562 1310to 1862 1119 290 1089 789 to 1503 397 297 339 167to688 92 64 81 41 to 160 542 260 496 278to886 20 19 19 19 19 18 0.00 0.00 1 2 14 1 235-1 108 656 362 568'262 to 1233 869-3328 1986 1088 1783 842to3772 865-1240 1323 375 1324 924to1899 188-438 313 104 316 226to443 <20-925 285 396 84 8 t o 8 6 2 70-559 264 185 237 98to574 * Significant differences (ftest, ps0.05). t Range of overall day exposure. $ < LOQ samples included as I/z LOQ. AM=arithmetic mean; GM=geometric mean. measurement was 10 pgA for hand rinse samples and 10 &pair of cotton gloves . The stability of MDA on cotton gloves, on gloves in the presence of Epikote 828, and the recovery of MDA from a soap and water solution were acceptable and are described elsewhere.' Analyses of urine samples have been described extensively elsewhere.%"Briefly, after addition of the internal standard (EDA) to the urine samples, N-acetylated metabolites were hydrolysed by heating and the addition of potassium hydroxide and extracted by diethyl ether. With hydrochloric acid, free MDA is extracted again from the organic phase, and by diethyl ether, after adjustingto pH 13,from the water phase. Evaporation of the organic phase was performed by a stream of nitrogen at 30C. The residue was inserted in the mobile phase and MDA was detected as described previously. The limit of measurement was 2 pgA urine and the coefficient of variation was <10% for a range of 20-50 pgA urine. Completeness of 24 hour urine collectionwas checked by measurement of creatinine concentration in all urine samples. STATISTICAL ANALYSES AND CALCULATIONS The data were statisticallyanalysed with SOLO statistical system (BMDP Statistical Software, Los Angeles, USA) for personal computers. Differences between average MDA excretion between liners and winders were studied with the Mann-Whitney U test. Hand wash and glove monitor data were log transformed, and geometric means (GMs) and 95% confidence intervals (95% CIS)were calculated. Analyses of variance (ANOVAs) were performed on the mean exposure data of the hands for all workers to study differences between exposure assessment methods, whereas t tests were used to study differences in mean daily exposure between the two weeks studied for each worker. For all analyses a probability of p<0.05 was considered significant. For data below the limit of detection the half of this limit was used in analyses. This is an acceptable method for handling non-detectable values when the distribution is log normal and the number of detectable values is low." The relation between the cumulative amount of MDA excreted and the cumulative hand exposure was studied by means of regression analysis. Results ACTUAL EXPOSURE OF THE HANDS The results of the exposure assessment during the 2 weeks of steady work are given in table 1. Worker F worked 6 days during the week in which the hand wash method was used. Average daily hand wash and glove monitoring results (GM) ranged from 81 to 1762 pg MDA, and from 84 to 1783 pg &ADA, respectively. For risk assessment purposes the average daily exposure is based on the arithmetic means and can be expressed as dermal dose/unit area (0,)A.ssuming a surface (skin) area of the skin that was monitored with gloves or by hand washing of about 1300 cm', the ranges of contamination were 0.07 to 1.56 pg/cm' and 0.20 to 1.53 pg/cmz, respectively. The winders (workers A, F, and E) showed the highest exposure during both weeks, whereas the lowest exposure was found for liner G. As well as his work as a liner, this worker performed other activities not related to MDA during a substantial part of his shift. For this worker about 55% of both hand wash and glove monitor samples were below the limit of measurement. t Tests showed significantdifferences of hand exposure in week 1 (hand wash method) and week 2 (glove method) for worker A only. Analyses of variance showed significant differences between the mean daily exposure of the hands (for both weeks and sampling methods) for each worker (peO.05). Cumulative MDA exposure over the week ranged from 385 pg MDA (worker G) to 10 202 pg MDA (worker A) and from 430 (worker G) to 9932 pg MDA (worker F) assessed by hand wash and glove monitoring, respectively. During the week in which hand washing was performed worker F worked for 6 days instead of 5 . BODY EXPOSURE A") SURFACE CONTAMINATION Au workers were inspected once or twice before each break. Visual contamination of the outside of the coverall was found 15 times; most spots were found on the lower legs (front and back) and forearms of the coverall. In nine cases surface =?pes showed a colour change. The remaining six spots were due to small splashes. Wiping the skin correspondingto the positions of visual contamination at the coverall by skin sa.lpes did not show exposure. All surface su-ypes of the buttons at machine 7 (workers A and F,n=7) and all but two but- tons at machine 3 (workers E and H, n=7) showed a colour reaction. Six out of 20 swype 1 151 $ 1 14 9 2 22 15 3 69 39 4 178 102 5 249 128 6 NA NA + I 125 54 +2 60 38 +3 411 24 54 65 68 115 97 134 99 69 56 133 57 NA XA 9 20 15 11 75 63 103 52 122 65 104 53 73 36 66 35 66 45 63 44 27 16 -10 9 63 10 5 12 5 83 12 4 11 6 NA NA 21 9 87 -5 2 *p<0.05 Mann-Whimey, u group of winders (A,F,E) and u liners (GJ). t M D A excretionexpressed in pgil. $Biologicallynon-representative sample. NA = not appropiate 14 4 17 10 29 16 45 28 61 36 50 27 N.4 N A 61 45 76 -9 -2 42 18 11 28 18 40 25 34 18 55 29 NA NA 46 21 26 13 54 Spot sample after 2 days OW 24 h urine from start day 1 24 h urine from start day 2 24 h urine from start day 3 24 h urine from stan day 4 24 h urine from start day 5 24 h urine from stan day 6 24 h urine from start first day off 24 h urine fromstart second day off spot sample after 2 days OW r250 samples indicated a contamination of the buttons of liner unit 3 (worker G), whereas five out of 1 1 buttons ofliner unit 7 (worker I) were contaminated. The handles of all scissors present in the workplace were contaminated. The results of swype tests from other surfaces indicated a general contamination of the workplace,-for example, door handles and doors of cupboards (including the cupboard with protective clothing). BIOLOGICAL MONITORING Table 2 summarisesthe results of the biological monitoring. In all urine spot samples taken before both working weeks MDA was detected, indicating a non- completed excretion of MDA during the period of non-exposure (minimal 56 hours for each worker). For worker A, who was the worker with the highest amount of MDA excreted, the MDA concentration in urine differed by more then a factor of two between the two samples. Figure 1 depicts the excretion pattern of MDA for worker A. Mean amount of MDA excreted in 24 hour urine samples ranged 6om 8 to 249 pg MDA and differed significantly between the liners (n=2) and the winders (n=4). RELATION BE'IWEEN URINARY M D A EXCETION A!!D HAND WASH RESULTS Cumulative MDA excretion over the week ranged &om 81 pg MDA (worker G ) to 717 pg MDA (worker A). A sigmficant correlation between cumulative MDA excretion and hand wash results over the week was found (fig 2). Linear regression analyses showed a rela- tion of MDA,,,=115.7+0.058 M D L wmH (I?=0.94). Discussion and conclusions The present study aimed to investigate the feasibility of a quantitative DOEL in an occupational setting.' The feasibility of testing that dermal exposure conditions are in compliance with a DOEL is strongly related to the performance of the dermal exposure sampling technique. Both a hand wash (removal) and a glove monitoring technique (surrogate skin) were used; however, validation of both methods is limited. Generally, surrogate skin monitoring tech- niques are considered to overestimate actual skin exposure, as the retention of the cotton fabrics compared with retention of the skin is high.' Average sampling time in the field study was about 80 minutes but never exceeded 120 minutes. Laboratory tests at the temperature conditions underneath the protective glove (about 3OOC) showed a decrease of recovery of MDA in the resin mixture from the monitoring glove of about 30% within two hours.' Therefore, in the present study a slight underestimation of h4DA exposure may have occurred. The removal efficiency of MDA from the skin by hand washing has not been investigated. Results of tests of efficiency of removal of pesticides by hand washingperformed by volunteersin our institute n 15 Frqure 1 I 0: I I ' 0 02 4 68 Days Amount of M D A washedfrom and excreted by worker A during week 2. 'c loob :Ob 2d00 4doo 61300 8i00 1 0 h 12\00 Amount of MDA washed from hands (Wveek) Figure 2 Relarion between the cumulatim amount of M D A excreted and the cumulative amount of M D A washed offfrom the hands in 1 week of monitoring. . 810 after dermal application ranged from about 40% to 95%.Therefore, a slight underestimation of MDA exposure may occur. Concurrent MDA exposure monitoring by hand washing and biological monitoring is considered applicable for MDA. From data reported by Hewitt et al," it can be concluded that washing the hands with liquid soap and water 30 minutes after contamination has occurred will have little effect on removing MDA, as a cutaneous reservoir will already have been established. Therefore, it is indicated that the hand wash method to determine MDA exposurewill not interfere with the process of absorption. The design of the study, however, enables comparison within workers of both methods, assuming a limited variance of exposure due to work activities. The results for hand exposure show large variances within all workers be- tween the days. Despite this finding, for all but one worker no significant differences were found between both weeks of exposure, indicating a compatible performance of both methods of exposure assessment. Quantitative assessment of exposure was limited to hand exposure, assuming no substantial exposure of other parts of the skin.The exposure of the hands is likely to result from the performance of tasks without protective gloves and the cross contamination of handles, equipment, and buttons, and seems to be more structural, whereas exposure of other parts of the body-for example, resulting from splashes-seems to be more incidental exposure and not to contribute substantially to the total external exposure. This is supported by the results from the swype tests performed at positions of the body underneath locations where splashes were found visually, although the relatively high limit of detection (about 3-5 pg) does not entirely exclude a minor exposure of other body parts. The surface swypes seem to be useful as part of a workplace contamination control programme to prevent exposure due to contact with contaminated surfaces. As 62% of the total excreted dose of I4C labelled MDA after dermal administration to rhesus monkeys was excreted in the first 24 hour^,^ comparison of MDA from hand washes on a daily basis and the corresponding MDA in 24 hour urine samples was considered inappropriate. Therefore, results of cumulative hand exposure and cumulative MDA excreted in urine over 1 week (including about 2 days of non-exposure) were used for comparison. The strong association between the cumulative amount of MDA excreted in urine and washed off, the absence of detectable contamination at other body parts, and the assumption of negligible inhalation exposure, based on a previous study' suggest that the exposure relevant for dermal exposure is mainly restricted to the hands. As documented in part 1 for MDA the DOEL associated with an excess incidence of mortality of 4/1000, interpreted as DAM,was set at 16 mglday.' In the present study the high- est actual daily dermal exposure found seemed to be about 4 mg (hand wash worker A on day 4), about 25%of the'external DOEL. Testing of complianceby means ofa biological limit Value (BLV) led to similar results. A BLV of 1280 I.% MDA in 24 hour urine samples can be derived from the DOEL of 16 mg/day, assuming a urinary excretion of 8%of the absorbed amount of MDA. The highest excreted amount of MDA for worker 4 (249 pg 'MDA on day 5) was about 20% of the BLV. This agreement confirms the applicability of a DOEL as an instrument for surveillance. However, the fact that MDA excretion for worker A was highest on the day subsequent to the day with the highest exposure illustrates that MDA may still be excreted after 48 to 72 hours.'' Therefore, with dermal exposure to MDA on several subsequent days, the amount of h4DA (conjungates) excreted in 24 hour urine samples cannot be clearly appointed to a specific exposure day, indicating the need for a well designed sampling strategywhen spot samples are taken. Brumark et ul showed a peak of excretion of MDA within 6-1 1 hours after a 1 hour skin exposure and prolonged excretion after 24 hour^.'^ Boeniger et al indicated that a biphasic pattern of elimination might be expected in cases where exposure will comprise both the inhalation and the skin routes. They suggested that for screening purposes urine samples at the end of shift or at the first void next morning should be collected, in which the next morning elimination might reflect the absorption by skin contact.' In conclusion, both dermal exposure monitoring methods were applicable and showed a compatible performance to assess exposure to MDA in the present exposure scenario where relevant exposure for dermal absorption is mainly restricted to the hands. The concept for a quantitative DOEL seemed to be relevant for MDA and applicable for compliance testing and health surveillance in the situation under investigation, where occupational exposure is dominated by the dermal route. This study was financially supported by the Dutch M i n i s t r y of Social Affairs and Employment.The important conmbutions of H van der Waal of Shell, the management of the factory that makes glass fibre reinforced resin pipes, and employees who participated in the study are gratefully acknowledged. 1 Bos PMJ, Brouwer DH, Stevenson H, er al. A proposal for the assessment of quantitative dermal exposure h i t s in the occupational environment:pan 1. Development of a concept for the derivation of a quantitative dermal occupational exposurelimit. Occup Envita MLd 1998;55:0oo-O00. 2 Van Hemmen JJ,Brouwer DH. Assessment of dermal expo- sure to chemicals.Sci ToralEm'nm 1995;168:13141. 3 F a k e R Dermal exposure assessment techniques. Ann Ocmp Hyg 1993;31:687-106. 4 Cherrie JW,Robertson A. Biologically relevant assessment of dermal exposure. An>' Occup Hyg 1995;39:387-92. 5 Brouwer DH, van Hemmen JJ.Elementsof a sampling srrazegy fm &,mal exposure assessmeni [abstracij. BNSSC~BSe,lgium: International Occupational Hygiene Association. First International S a e n ~ Cc onference, 7-10 December 1992. 6 SdejnnimorhPylLen, aviainanSiliinneen(NMJ.UAf)icdWursinrwR& of rk ftahbe reicxaproisounreowf 4,4'glass jibn reinfaccd resin p q x ~The Hague, The Netherlands: Shell HSE, 1990. (HSEW n 90.@03.) 7 Van der Wad H. . U ~ ~ ~ ~ k k <miori den agnalyse wan 4,4'-dirnnhy&mdw~11/1n(e.%IDA/ 111 zdep opbssingo? en in kdiwnhydmndr-oplornng in mahanol. fDevebpmenroj merh- odc for chemd Onuh?rc Of J,.l'~mrrm.lniedianiline(MDA) in nravr soup r o h r m ~and poiasmm b d m & in methanol). Pemis, The Nethdandr. Shell Sederland Chemie Bv, Biomedical Labraton-. 1995 con-8 a g e r T. Bm d-rlopmcnrs m surface contaminadon rnonito* for m m ~ amt m~- In- jfoizdhzgs of the / m e ,4lfrWUd C'mfN'Sllir CinimMan, OH: ACGIH, 1995343-6. 9 Boeniper M. .Mason R. Hcrcko J. 5: d.CSCof to . u c ) s mJ i o n u d exposures to samp4l,e4s- dimcCtr).kmdunJmc the acrorpace indusq, I,-,:fioceed- inn of the Cmjerence a Advanced COmpOn'tos. Cincinnati, OH:ACGIH, 1995:87-111. 10 Cocker J, Gristwood W,Wilson HK. 4 biological monitormg assessment of exposure to methylene dianiline in manufacturers and users. Occup Envimn Med 1994;51: 519-22. 11 Boogaard PJ,van der Waal H. Biological monitoring of drrmal exposure w 4,4`-dimetlrylenedianiline (MDA) by determination of MDA in hydrolysed unne. A human volunteer study [internal report]. Rotterdam, The Netherlands: Shell Biomedical Laboratory, 1994. 12 Hornung RW, Reed LD.Estimation of average concentntion in the presence of undetectable values. Appl Occvp Em'ron Hyg 1990;5:46-51. 13 Hewitt PG, Hotchkiss SAM, Caldwell J. Decontamhadon procedures after in vitro topical exposure of human and rat skin to 4,4`-methyIenebis(2-~hloroaniline]and 4,4merhylenedianiline. Appl Toxic02 1995;26:91-8. 14 Brunmark P,Bruze M, Skerfving S, er al. Biomonotoringof 4,4-methylenedianiline by measurement in hydrolysed urine and plasma after epicutaneous exposure in humans. Int Arch Occup Emiron Health 1995;67:95-100. Occupational and Environmental Medicine - http://www.occenvmed.com Visitors to the world wide web can now access Occupational and Environmental Medicine either through the BMJ Publishing Group's home page (http://www.bmjpg.com) or directly by using its individual URL (http://www.occenVmed.com).There they will find the following: Current contents list for the journal Contents lists of previous issues 0 Members of the editorial board Subscribers' information Instructions for authors 0 Details of reprint services. A hotlink gives access to: 0 BMJ Publishing Group home page 0 British Medical Association website 0 Online books catalogue 0 BMJ Publishing Group books. The web site is at a preliminary stage and there are plans to develop it into a more sophisticated site. Suggestions from visitors about features they would like to see are welcomed. They can be left via the opening page of the BMJ Publishing Group site or, alternatively, via the journal page, through "about this site".