Document rx7KeGDBKQXvEQpJoaVGOOvBr

environmental research 35, 531-551 (1984) */ Migration of Ingested Asbestos Janet H. Kaczenski* and William H. Hallenbeck* 'Digital Equipment Corporation. 100 Minuteman Road. Andover. Massachusetts 01810-1098. and +School of Public Health. University of Illinois at Chicago. P. 0. Box 6998. Chicago. Illinois 60680 Received July 9. 1983 Tissue samples from one test and one control baboon were analyzed by transmission electron microscopy for the presence of chrysotile and crocidolite asbestos. The lest animal had been gavaged with cumulative doses of 800 mg each of chrysotile and crocidolite asbestos. An earlier evaluation of these tissues led to the conclusion that ingested asbestos fibers do not penetrate the gastrointestinal tract of the baboon and migrate systemically. However, the present study involved more sensitive methodology, and penetration and migration were clearly demonstrated by the recovery of significant levels of asbestos from test stomach, heart, spleen, pancreas, and blood samples, c iwj Ai-aUcm* Pres,, tnc. INTRODUCTION Epidemiological studies have shown that asbestos workers have increased mor tality due to pleural and peritoneal mesothelioma, and cancers of the lung, diges tive system, and related organs (SelikofT, 1979). Exposure to asbestos is not restricted only to those who are occupationally exposed. Asbestos fibers have been found in air (U.S. EPA, 1979) and drinking water supplies (Millette el al., 1980). Approximately 50% of an inhaled asbestos dose is cleared from the respi ratory tract through mucociliary action and subsequently swallowed (Evans et at., 1973). Thus, the gastrointestinal tract becomes the recipient of most inhaled and ingested asbestos. Increased rates of cancers in asbestos workers and others may be related to gut penetration and migration of asbestos. For example, sta tistical associations between asbestos fiber content in drinking water and several types of cancers (digestive system and related organs) have been shown for var ious populations within the United States and Canada (Conforti et al.. 1981; Kanarek et al., 1980; Levy et al., 1976; Mason et al., 1974; Wigle, 1977). We look to controlled animal studies for confirmation of human findings, espe cially with regard to dose. Several chronic animal ingestion studies have produced little evidence that ingested asbestos increases the incidence of gastrointestinal cancer (Bolton et al., 1982; Cunningham et al., 1977; Donham et al., 1980; Gibel et al., 1976; Hilding et al., 1981; McConnell et al., 1984a. b: Smith et al.. 1980; Ward et al., 1980). Research specifically into the area of gut mucosal penetration by asbestos fibers has produced conflicting results. The literature contains ingestion studies in volving rats and baboons. Five rat studies (Cunningham et al., 1977; Donham et al., 1980; Patel-Mandlik and Millette, 1983; Sebastien et al., 1980; Westlake et al., 1965) and three baboon studies (Hallenbeck and Patel-Mandlik, 1979; PatelMandlik et al., 1979; Patel-Mandlik and Millette. 1980) produced some evidence for asbestos gut penetration and migration. These studies were not conclusive 531 0013-9351/84 $3.00 Copyright C 1984 by Academic Pit, Inc. All rights of reproduction in any form reserved. 10002342 532 KACZENSKI AND HALLENBECK because of various shortcomings: (a) the number of fibers observed on a grid opening basis was very small (Hailenbeck and Palel-Mandlik. 1979; Patel-Mandlik etal.. 1979; Patel-Mandlik and Millette, 1980, 1983); (b) a small number of control data were obtained (Sebastien et al., 1980); (c) test and control data were not presented in a quantitative manner (Donham et al.. 1980; Westlake et al., 1965); and (d) fiber count data were reported as fibers/g of tissue (Cunningham et al.. 1977). Unless fiber count data are reported in terms of fibers/grid opening, it is not possible to know whether the authors' conclusions were based on the obser vation of one or hundreds of fibers. Given the problem of background contami nation, positive results are always more credible if they are based on the actual observation of many fibers. Three rat studies (Bolton and Davis. 1976; Bolton et al.. 1982; Gross et al., 1974) and one baboon study (Hailenbeck et al.. 1981) produced no evidence for asbestos penetration and migration. However, there was no discussion of asbestos recovery in these studies. Unless the methodology utilized in sample preparation has been evaluated as to fiber recovery, there is no way of knowing whether the sensitivity was sufficient to detect asbestos, especially if the levels may be very low. Two human studies (Carter and Taylor, 1980; Cook and Olson. 1979) produced evidence for the penetration and migration of asbestos. Cook and Olson (1979) detected amphibole asbestos in the urine of Minnesota residents who ingested drinking water contaminated with 5 x 107 fibers/liter. In contrast to Cook and Olson's findings. Boatman et al. (1984) did not find significant levels of asbestos in the urine of humans exposed to 2 x I08 chrysotile fibers/liter. Cook and Olson and Boatman reported their data in fibers/liter instead of fibers/grid opening. Carter and Taylor (1980) identified amphibole asbestos in postmortem tissues of persons exposed to a high oral intake of the mineral. However, they were not consistent in selecting equal masses of tissue, equal numbers of grid openings, and equal ash resuspension volumes for analysis. Hailenbeck et al. (1981) analyzed tissue, blood, and urine samples from a ba boon gavaged with 800 mg each of chrysotile and crocidolite asbestos by trans mission electron microscopy. The. results indicated that asbestos fibers did not penetrate the gastrointestinal tract of the baboon and migrate systemically. Since the test and control baboon tissues were frozen and stored at the end of the study, the purpose of the present research was to reevaluate the test and control baboon tissues using a different asbestos analytical methodology. This reevaluation had several advantages over the original study and other tissue studies in general: 1. Fiber recovery was defined. 2. Instead of the droplet transfer technique used in the original study, the Jaffe washer, considered the current state-of-the-art technique, was used to prepare electron microscope grids. 3. Five grams of tissue were used instead of I g (wet wt), greatly increasing the sensitivity of the technique. The minimum detection limit was reduced from 220 fibers/mg in the original study to an average of 24 fibers/mg dry tissue in the present study. 10002343 MIGRATION OF INGESTED ASHESTOS 533 MATERIALS AND METHODS Preparation of Tissue Samples for Electron Microscopy All tissue samples were rinsed with filtered (0.1 pun polycarbonate) deionized water before use. Five grams (wet wt) of tissue was obtained by cutting several small pieces from each tissue with a surgical blade. A new blade was used for each sample. A blank filter was also processed with the lest and control samples for each tissue type analyzed. All tissue samples were prepared for analysis by transmission electron microscopy in the following manner: 1. Comparable wet weights (5 g) of test and control tissues were placed in preweighed beakers and dried in a vacuum drying oven at 90C for 24 hr. 2. Beakers were removed from the oven, placed in a desiccator until cool, and then weighed on an analytical balance to determine the final dry weights of the tissues. 3. One hundred milliliters of filtered (0.2-fim fluoropore) 59c potassium hy droxide was added to each beaker. Beakers were covered with a glass petri dish and heated at 60-70C for 24 hr on a combination stirrer hot plate. The digestion process was facilitated with a magnetic stirring bar. 4. The tissue digest was cooled to approximately 30C. (Digested tissue at 6070C damaged polycarbonate filters.) The digest was filtered through 47-mmdiameter 0.1-pun polycarbonate filters. The filters were covered during filtration and taken to dryness. Filtration units from Gelman Instrument Company were used along with 47-mm-diameter 0.4-jxm polycarbonate filters as backing fillers. Filtration time varied from 6 to 48 hr. Backing filters were discarded after use. 5. Filters were placed in covered petri dishes and kept inside a desiccator until dry. Once dry, the filters were placed inside 50-ml bottles and ashed in a low temperature plasma asher (13.56 MHz, 80 watts. I torr) for approximately Ihr. For a given tissue type, the test, the control, and blank were ashed together. 6. Ashes were suspended with approximately 30 ml of filtered water by ultrasonication in a water bath (55 KHz) for 15 min. Approximately I ml of filtered (0.2-pun fluoropore) 6 n HCI was also added to each suspension before ultrasonication to aid in the solubilization of the ash. 7. The suspended ashes were filtered through 0.I-pun polycarbonate filters with 5-pun cellulose ester filters as backing filters. The fillers were covered during filtration and taken to dryness. 8. Steps 5, 6. and 7 were repeated to help reduce background debris and or ganic matter to a minimum. 9. The third and final polycarbonate filter with the sample deposit was dried in a desiccator, secured to the bottom of a petri dish with double stick tape, and then coated with a thin layer of carbon in a vacuum evaporator. 10. Random sections of the carbon-coated polycarbonate fillers were removed with a surgical blade and transferred to 200-mesh electron microscope finder grids and subsequently placed in a Jaffe washer until dissolution of the polycarbonate filters was complete. 11. One grid opening from each of four grids was scanned at a magnification of 28,000 x for test, control, and blank samples. 10002344 534 KACZENSKI and hallenbeck Recovery Study A reference suspension of chrysolite (7 x |0~J p.g ml) was made from the same asbestos as administered to the test baboon (Hallenbeck el a!., 1981). Five mil liliters of this suspension yielded about 25 fibers/grid opening. A "prespike" was prepared by adding 5 ml of reference suspension to I g (wet wt) of pig kidney in Step I above. The entire tissue preparation procedure then followed. A "postspike" was prepared by adding 5 ml of reference suspension to I g (wet wt) of pig kidney just before the second ultrasonication of the suspended ash. Chrysolite fiber counts from 10 grid openings per preparation were compared. Fiber Classification and Counting Rules Fibers were identified and classified as chrysotile or amphibole based on their morphology and electron diffraction patterns. A fiber was defined as any particle that had parallel sides and an aspect ratio 2*3:1. Tightly bound bundles of fibrils were counted as a single fiber and an estimate made of the number of fibrils in the bundle, the average width of the bundle, and the maximum length of the bundle. Clusters of fibrils were counted as a single fiber and an estimate made of the number of fibrils in the cluster. STATISTICAL ANALYSIS OF DATA Recovery Study The Kruskal-Wallis one-way analysis of variance by ranks (Daniel. 1978) was used to test the null hypothesis that the suspension, prespike, and postspike fiber counts were equal (a = .05). Chrysotile fiber counts were converted to fiber/cm2 values to adjust for the individual variations in grid opening areas and were also adjusted for blank levels of chrysotile contamination before being analyzed by the Kruskal-Wallis test. Significant differences were found among the three groups. A multiple-comparison procedure (Daniel. 1978) was used to determine the source(s) of the significant differences. Baboon Test, Control, and Blank Samples (/) Fiber count analyses. All test, control, and blank fiber counts were con verted to fiber/cnr values to adjust for the individual variations in grid opening areas. A Kruskal-Wallis one-way analysis of variance by ranks was performed on the test, control, and blank fiber/cnr values for each tissue to test the null hypothesis: test, control, and blank fiber/cnr values are equal (a = .05). If sta tistically significant differences were found among test, control, and blank values, a multiple-comparison procedure was used to determine the source(s) of the sig nificant differences. Test and control fiber/cm2 values were then adjusted for blank levels of asbestos contamination and concentrations in fibers/milligram tissue were calculated. A two-tailed Wilcoxon's rank sum test (Snedecorand Cochran. 1976) was performed on the test and control fiber/mg tissue values to test the null hypothesis: test and control fiber/mg values are equal (o = .05). P values were obtained from Dixon and Massey (1969). Detection limits were also calculated for test and control tissues. These calculations were performed as follows. 1000234S MIGRATION OF INGESTED ASBESTOS 535 Calculation of fibers/milligram tissue: No. fibers' No. blank fibers' .V 2) Grid opening area (cm2)' /= I i= i N ^ Grid opening area (cm2)' i~ I A| x -- = fibers/mg tissue N = number of grid openings examined for Ihc sample Af - 9.62 cm2 = effective area of a filter M = dry weight of tissue in milligrams Calculation of detection limit in fibcrs/mg tissue: F x A. -----------= fibers/mg tissue A2 x M F = the observation of only one fiber divided by the total number of grid openings examined. For example, if one fiber was observed in four grid openings, F = 1/4 = 0.25 A | = 9.62 cm: = effective area of a filter A2 = average area of grid openings examined (cm2) M = dry weight of tissue in milligrams The spleen samples were statistically analyzed in a different manner because a control spleen tissue sample was not available. Wilcoxon's rank sum test was used to compare test fiber/cm2 values directly with blank fiber/cm2 values. (//) Mass analyses. The test, control, and blank data were also statistically analyzed using chrysolite and amphibole asbestos mass values. The mass (pg) of each fiber was calculated as follows (Anderson and Long, 1980): where M = L x W'- x D x /0-h M = mass (pg) L = length (pm) IV = width (pm) D = density of fiber (g/cm') (chrysolile = 2.5 g/cm\ amphibole = 3.25 g/cmJ) Calculation of p.g asbestos/mg tissue: pg asbestos' pg blank asbestos' iv X Grid opening area (cm2y /=! /= i ,v X Grid opening area (cm2/ i=i = p.g asbestos/ ,W mg tissue N = number of grid openings examined for the sample A| = 9.62 cm2 = effective area of a filter M = dry weight of tissue in milligrams 10002346 TABLE I Fiber Count and M ass Data by G rid Opening eor T est Samples 10002347 536 KACZENSKI AND HALLENBECK I- la.% V-- 0 9 0 0 ooc o 0 3 0 3330 003 o 3003 0033 33 *"< 33003 30 -- 03003 io-^i cSd e <6 o x ^ 't 30 X -- 6 ^ n *f O n c *f r--, = 2 ^3 rs n A &A - c .f 23C ..-Ca9 Uou Qov. H? ^ 00 oe ^ ^ -- ao ri n ri - ^flov'i^rr^v'ioae o*C 1Oeft *M 7cu O Ho o "X" - -"T r-ci o -- d o o o n6 36 x6 62 --<n n -<C r* -n 66 --6 3 sxnxxxsc -- --6--666--6 5 S "ft g ^ -g5*-5o r ^ ^ C U., ^ g | S. o 3 d = 5Zu Z" ~ U H Z --jHn n------ oooo 3300 o 003 0500 0330 O uh- u H HZH2* ^ 33rz 2*mcs 00 tT" c, o 5- x w% %Ti s r* J6O P6; X- ri AAAA a30e ^^ --r~\ r9*>"i 3 o-- r qo n.s5*C. is -- rs n*. r-i r"> r-. ^ n ri -- <n -- -- <n n pm A AA -s C .5 -3C O</> o0 VO. t. O 611 S 2--_ *, pr-n -r ^a t30 M <6 r4 f**. ^,r m m i MIGRATION OF INGESTED ASBESTOS ccco oooo -ooo OC 30 oooo oooo oooo oooo r- -- r*"i 30 n w, n c Of^ioV 30 O -- -- 9 r- r! o ^ - ^-- n m rMs --O O~ -- nv% --o 9r. ^" -- d d-- nmn ri v n n t ri r- vj tri 9 r*- -- oO-- nC 3C "T T -TO^n oocid dod-- TO*"-" OOO^ -- odd -- -- O -- ^ ^ 3C O odd v V UJ3 O o Vc/t a v o u3 oooo -- o -- -- oooo u3 c rs ^ r n x -- 9 30 -- fN - v. jk r* Ow^ --OO A O A od n A 3^ ri ^ r- -- 30 r n v w E o--o 9 - ri m - S 3 3 Sos^ cOc%tt iJho*s. * 5* JE CQ ___;o 11 >%%> "c3 00 -3cJ v > 3 5a/i h h CD SQ 3 * - 537 10002348 i0002349 j a Baboon (issue Stomach Chrysolile fiber count/ grid opening N O t >C r-i r-4 -- -- Heart >2.12" 3.94 2.03 1.87 9J 5o ^u .y5 f* * peo No. chrysolite cluslers/grid opening (No. fibrils/ cluster) mn --ir. C* ri r-' rj C> L 01 6 t Pancreas ^N- 0.16 0(0) 1.22 0 (0 ) 0.70 0(0) 0.33 0 (0) 0.14 0 (0 ) 0.09 0 (0) 1.35 0 (0) 0.80 0 (0) tl 5 - x X 0 X z 00 e.5cua J''* O oJt * e X 'COSO -- OSS = = = = uzo rs* UJ ^ 2-J3 OX IS c< 10.41 M.48 10.83 10.41 10.41 10.60 10.00 10.61 11.04 10.83 10.41 10.00 10.41 11.70 11.70 10.41 Amphibole fiber count/ grid opening Amphibole (jig/grid 'opening x 10 ) 4.58 0.82 3.55 2.16 n n ITt IT o rC~ tX/", O -- P'l <* o r--i r~- r*- 0.20 2.61 2.59 0.27 III 0.46 0.26 0 rs f-i rj - CSSS csss 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) zu ~ := | "cr 5 ..Si Is tci 9 KACZENSK! AND HALLENBECk 538 MIGRATION OF INGESTED ASBESTOS o * s e s >c ri O S CO O -- --* r* S n -- OS O O S v* AA v*. r-i r-i *t -- r*% m -- r**. r-i r -- >C *'* S 3 t S z -r ^ 3 3 539 AAA r- r-* -- tt rr n r - c * 5 c ^ x ^ n ^ \ ir, ri -- r* w c >c = !?.; 3*g as a* ' ut _?_5|5 2S8W -5 (JE=--a11 1s-i X 10002350 540 KACZENSK1 AND HALLENBECK 0 (ID 0 (0 ) 0. (0 ) 0 . (0 ) oc OS COr C I -- r-r r-x- r W **i 9V0. Vi -- r- v, *r *r rf,i s--C ro* o -- ri V </> Za. t</i * z < _4> XO Q. E < <3Oc-> X oo 'c &o '*C020 - ri ^ f = =------ *> v, X ri v. -- ri --V r+\ V --^ 2 5n""x> rsoi n rooi vr-i n r^* so -n r** ^ *' SO -- -- -- SO (A < 2 2 Z < W 2Z iaie a ~ ,5 e g- >5 JoL"x r- r*\ rM rr"\ wn 5 rv-- r &> m(O --^ V^i ^ T W -- N S f**l o o X 1* -o vg n rs - f\ -- <N -- rt r* n &> r* r>\ GO ~ Stomach Heart Pancreas Blood" 10002351 MIGRATION OF INGESTED ASBESTOS ac o o o o Xoco ooocoeoo o -- -- *gga 2 s -- ` -- --` ^ -- *rv A -- v~ ' ---- OC C 96 -- --X ^ t ^ *T ^ 5 3-- 3~ ?-j 3-i 54! e -- o o o o o o o o so -- oo -- rc~** ^ ^ s--O < ' -- r- ^ -- -- %t 73 t S > -3 s a 2^ r-j -- -- -- N - >0 ri *51 co O asp 2 ax. 10002352 542 KACZENSKI AND HALLENBECK TABLE 4 Dry Weights oe Baboon Tissihs Tissue Stomach" Heart" Pancreas" Blood'1 Blood* Kidney"*' Lung*" Liver" Spleen ! Test (g) 0.9151 1.1200 1.7192 0.59X3 0.2340 1.1260 1.1138 1.2910 0.7681 Control Igl 0.82X8 1.3262 1.4030 0.6500 0.2420 1.2694 1.2334 1.7315 -- Five grams wet weight of test and control tissues were analyzed. 6 Blood withdrawn from the heart at sacrifice. Five milliliters was used for each asbestos analysis. r Blood withdrawn during the last day of gavage. Approximately 2 ml was used for each asbestos analysis. d Right kidney only. ' The lung tissue represented a combination of apical, cardiac, and diaphragmatic lobes. 1 Only 2.5 g wet wt of test spleen was available. No control spleen was available. Calculation of detection limit in Hg asbestos/mg tissue: F X A, A1 x M = M-g asbestos/mg tissue F = the mass of only one fiber divided by the total number of grid openings examined. For example if one chrysotile fiber with a mass of 3.47 x 10*10 p.g (mass of smallest fiber visible at 28.000 x with a 3/1 aspect ratio) was observed in four grid openings. F = 3.47 x 10*10 pg/4 = 8.67 x 10 "M Fg Baboon tissue Fibers/ mg (issue Stomach Heart Pancreas Blood" Blood* Kidney Lung Liver Spleen 26 19 14 34 46 19 19 15 30 " Blood (aken at sacrifice. * Blood taken at last gavage. TABLE 5 DEfECTioN Limits pg Chrysolite' mg (issue x I0-* 9 6 5 1 16 6 6 5 10 M-g Amphibole/ mg tissue x I0-* 10 8 6 15 20 9 9 7 10 i 10002333 MIGRATION OF INGESTED ASBESTOS 543 TABLE 6 Statistically Significant Relationships among Test, Control, and Blank Baboon Samples for Chkysotile Asbestos Tissue Fibers/ mg tissue" Fibers/cm 4g/mg tissue" M.g/cm-'* Stomach Heart Pancreas Bloodr Blood"7 Kidney Lung Liver Spleen* T>C T>C T>C T>C T>C T=C T=C T=C -- T>C T>C T>C T>C T>C T=C T>C T=C -- C>B C> B C=B C=B OB C>B C=B OB -- T>B T> B T>B T> B T>B T>B T>B T> B T> B T>C T>C T>C T>C T=C T = C' T=C NA -- u II T>C T>C T>C T>C T=C NA' NA -- OB C=B C=B C=B OB NA OB NA -- T> B T> B T>B T>B T>B NA T= B NA T>B * Test and control fiber counts/cm: or pg/cm2 were adjusted for blank levels of chrysolite and then converted to concentration values of fibers/mg tissue or p.g/mg tissue. A two-tailed Wilcoxon's ranksum test of the null hypothesis (test concentrations are equal to control concentrations) was performed for fibers/mg tissue and ng/mg tissue for each tissue (a = 0.05). Note that T = test, C - control, and B = blank. * A Kruskal-Wallis one-way analysis of variance by ranks of the null hypothesis (test, control, and blank fiber levels are equal) was performed on fibcrs/cm* and |ig/cm: of test, control, and blank samples (a = 0.05). If there were significant differences between the three groups, a multiple-com parison procedure was used to find the source(s) of the differences. r Blood taken at sacrifice. d Blood taken at last gavage. ' Due lo the occurrence of a cluster in the blank sample, test values were statistically compared directly to control values without adjustments for blank mass levels. ' NA = not applicable. Statistical mass analysis of data not possible due to a cluslerts) in one or more of blank, control, or test samples. ' Control spleen was not available. Wilcoxon's rank sum test was used to compare lest fibcr/cm: and |xg/cm: concentrations directly with those of the blank sample (a = 0.05). A| = 9.62 cm2 = effective area of a filter A: = average area of grid openings examined (cm2) M = dry weight of tissue in milligrams RESULTS AND DISCUSSION Recovery Study The results of the Kruskal-Wallis test demonstrated that there were significant differences among suspension, prespike, and postspike counts in terms of fibers per square centimeter (P < 0.005). Differences were isolated via multiple com parison statistical tests which yielded the following results: (a) suspension and prespike counts were not significantly different (recovery = 95%), (b) suspension and postspike counts were significantly different (recovery >100%), (c) prespike and postspike counts were significantly different (recovery = 61%). Thus the true recovery probably lies in the range 61-100%. 544 KACZENSKI AND HALLENBECK TABLE 7 Statistically Significant Relationships among Test. Control, and Blank Bahoon Samples for Amphihole Asbestos Tissue Fibers/ mg tissue" Fibers/cm xA 4gmg tissue" ng/. cnrxA Stomach Heart Pancreas Blood* Blood-7 Kidney Lung Liver Spleen* T=C T>C T=C T=C T>C T=C T=C T=C -- T=C T>C T=C T=C T>C T=C T=C T=C -- CB C=B C=B C=B C=B C=B C=B C=B -- T= B T> B T= B T=B T>B T= B T= B T= B T>B T=C T=C T=C T=C NA' T=C T = Cf T = C' -- T=C T>C T=C T=C NA T=C NA NA -- C=B C=B C=B C=B NA C= B NA NA -- 03 II T= B T>B T= B T= B NA T= B NA NA * Test and control fiber counts/cm: or ng/cm: were adjusted for blank levels of amphibole and then convened to concentration values of fibers/mg tissue or p.g/mg tissue. A iwo-lailed Wilcoxon's ranksum test of the null hypothesis (lest concentrations are equal to control concentrations) was performed for fibers/mg tissue and ng/mg tissue for each tissue (a = 0.05). Note that T = test. C = control, and B = blank. * A Kruskal-Wallis one-way analysis of variance by ranks of the null hypothesis (test, control, and blank fiber levels are equal) was performed on fibers/cnr and ng/cm- of test, control, and blank samples (a = 0.05). If there were significant differences between the three groups, a multiple-com parison procedure was used to find the source(s) of the differences. * Blood taken at sacrifice. * Blood taken at last gavage. * NA = not applicable. Statistical mass analysis of data not possible due to a clusters) in one or more of blank, control, or test samples. 1 Due to the occurrence of a cluster in the blank sample, test values were statistically compared directly to control values without adjustments for blank mass levels. * Control spleen was not available. Wilcoxon's rank sum test was used to compare lest ftber/cnr and pg/cm- concentrations directly with those of the blank sample (a = 0.05). Baboon Test, Control, and Blank Samples Tables 1-3 contain the fiber count and mass data by grid opening for the test, control, and blank samples. Table 4 contains the dry weights of the baboon tissues analyzed. Table 5 contains detection limits. One of the most significant aspects of this study was the consistency of the statistically significant relationships among test, control, and blank baboon sam ples as presented in Tables 6 and 7. The relationships presented in Tables 6 and 7 clearly indicate that the analytical results were not a product of the random occurrence of background asbestos. In ail cases where statistical analyses could be performed, test sample concentrations were greater than or equal to control and blank sample concentrations: control sample concentrations were greater than or equal to blank sample concentrations. For a few tissues, control concen trations were greater than those for their respective blanks. This may have been attributable to environmental factors such as food. air. water, and intravenous fluids. In those tissues where test and control concentrations were equal, asbestos 10002355 MIGRATION OF INGESTED ASBESTOS 545 TABLE 8 Chrysotu.e Asbestos Recovered in Baboon Tissves Concentration adjusted for blank fiber levels (fibers/mg tissue) Concentration adjusted for blank mass levels t^g mg tissue ' 10-') tissue Test Control Test Control Stomach Heart Pancreas Blood taken at sacrifice Blood taken at last gavage Kidney Lung Liver Spleen 9400"* 3425"'* 1649"* 4421"* 5077*' 1139*/ 493*./ 1045* r 11.623' 1744' 54(f 238`/ I84" 2440* 1069" I07J 749* -- 15.7"* 8.0"* 1.6" * 6.7"* 5.9* ' NA" 0.7"' NA 10.8' 2.5' 1.6'' ' 0.2'' 4.8' NA O.l" NA -- " T > C (P = 0.028). Note that T = test. C = control. and B = blank. * T > B (multiple-comparison procedure. P values unavailable). ' C > B (multiple-comparison procedure. P values unavailable). J C = B (multiple-comparison procedure. P values unavailable). ' T > C (/> = 0.020). 'T = C (a = 0.05). ' NA = not applicable. Calculation of mass concentration was not possible due to the occurrence of a clusters) in one or more of blank, control, or lest samples. * T = B (multiple-comparison procedure. P values unavailable). ' T > B (P = 0.028). penetration and migration may have occurred without being detected because less than the whole organ was analyzed and the distribution of fibers in an organ may not be uniform. Tables 8 and 9 contain the concentrations of asbestos recovered from the ba boon tissues. The concentrations were adjusted for blank levels of asbestos. In a few cases, concentrations of zero resulted after adjustments for blank levels of asbestos were made. Tables 8 and 9 should be referred to during the following discussion of individual tissues. Stomach. Penetration of chrysotile asbestos into the stomach wall of the ba boon was demonstrated by the significant level of chrysotile recovered from the test stomach sample. The test sample had chrysotile fiber and mass concentra tions which were significantly greater than those of the control sample (P = 0.028). There was no indication that amphibolc asbestos penetrated into the stomach wall of the baboon. Test amphibole fiber and mass concentrations were not significantly different from those of the control sample. Heart. Penetration and migration of chrysotile asbestos were demonstrated by the significant level of chrysotile recovered from the test heart sample. The test i j j | 10002356 546 KACZENSKl AND HALLENBECK TABLE 9 Amphibole Asbestos Recovered in Baboon Tissues Concentration adjusted for blank fiber levels (fibers/mg tissue) Concentration adjusted for blank mass levels (ligmg tissue x 10"') tissue Test Control Test Control Stomach Heart Pancreas Blood taken at sacrifice Blood taken ai last gavage Kidney Lung Liver Spleen 269"'' 6811'' 4"* 736" * 213 y-' 102"* 0"* 0"* 531* 72' 87' 61' O' 553' 57' 0 O -- 10.8-* 10.2" 0.1"* 6.1" NA' 1.5" NA NA 2.6 0.6` 0.8' 0.02' O NA 0.7' NA NA -- "T = C (o = 0.05). Note that T = lest. C = control, and B = blank. *T = B (multiple-comparison procedure. P values unavailable).' r C = B (multiple-comparison procedure. P values unavailable). JT > C(P = 0.028). ' T > B (multiple-comparison procedure. P values unavailable). rT > C (P < 0.001). * NA = not applicable. Calculation of mass concentration was not possible due to the occurrence of a clusteris) in one or more of blank, control, or test samples. * T > B (P = 0.028). 1 T = B (q = 0.05). TABLE 10 Characterization of the Lengths and Diameters of Chrysotile Fibers" Recovered in Test Stomach Tissue Length (M-m) 0.018-0.036 Chrysotile diameter (|im) 0.037-0.071 0.072-0.125 >0.125 Total |<2)* <0.2 0.2-0.5 0.51-1.0 1.1-2.0 2.1-5.0 >5.0 Total [%] 0 (0)' 1 (0) 1 (0) 0(0) 0(0) 0(0) 2 10] 2 (0) 61 (1) 114 (1) 70(21 9(1) 5(3) 261 |70] 0(0) 21 (9) 47(21) 27(12) 6(3) 1 (1) 102 [27| 0(0) 0(0) 7(7) 3 (3) 0(0) 0(0) 10 [31 2 [<l| 83[22| 169 [45| 100 [27| 15 |4| 6[2| N" = 375 " Fibers = fibrils and bundles of fibrils. All entries were adjusted for blank chrysotile levels. * The proportion of the total number of fibers in a specified length or diameter interval is shown in brackets. ' Numbers in parentheses refer to the number of bundles included in the cell total. d N = Total number of fibers. MIGRATION OF INGESTED ASBESTOS 547 TABLE II Characterization oi the Lengths and Diameters oi Chkysodi.e Fibers'' Ri covered in Test Heart Tissue Lengcn (p-m) 0.018-0.036 Chrysolile diameter (pm) 0.037-0.071 0,072-0.125 >0.125 Total <0.2 0.2-0.5 0.51-1.0 LI-2.0 2.1-5.0 >5.0 0 tor 2 (0) 0 (0) 0(0) 0 (0) 0 (01 0 (01 23(1) 37 (0) 14 (1) 4(1) 0 (0) 0 (0) 8 (5) 3V (11) 26 (5) 6(0) 2 (0) 0 (0) 0 (0) 3 (3) 0(0) III) Ml) 0 |0) 33 |20| 79|47| 40 |24| II 17) 3 121 Total m 2IH 78 |47| 81 |49| 5 13] N4 = 166 " Fibers = fibrils and bundles of fibrils. All entries were adjusted for blank chrysotile levels. * The proportion of the total number of fibers in a specified length or diameter interval is shown in brackets. e Numbers in parentheses refer to the number of bundles included in the cell total. 4 N = Total number of fibers. sample had chrysotile fiber and mass concentrations which were statistically greater than those of the control sample (P = 0.028). Penetration and migration of amphibole asbestos was indicated by the signifi cant level of amphibole fiber recovered from the test heart sample. The test sample had an amphibole fiber concentration which was statistically greater than that of the control sample (P = 0.028). However, when the amphibole data were analyzed on a mass concentration basis, the test concentration was not statisti cally different from that of the control sample. Pancreas and blood taken at sacrifice. Penetration and migration of chrysotile TABLE 12 Characterization of the Lengths and Diameters of Chrysotile Fibers" Recovered in Test Pancreas Tissue Chrysolile diameter (pm) (pm) 0.018-0.036 0.037-0.071 0.072-0.125 >0.115 Total [%)* <0.2 0.2-0.5 0.51-1.0 1.1-2.0 2.1-5.0 >5.0 Total \%\ 4 (0)' 3 (0) 0(0) 0(0) 0(0) 0(0) 7 [6] 0 (0) 35 (6) 48 (2) 9(1) 2 (0) 0(0) 94 [73) 0(0) 5 (2) 14 (5) 6(1) 2 (0) 0(0) 27 |2I| 0(0) 0 (0) 0 (0) 0(01 0(0) 0 (0) 0|0| 4|3| 43 |34| 62 |48| 1-5(12] 4(3) 0|0| N4 = 128 " Fibers = fibrils and bundles of fibrils. All entries were adjusted for blank chrysotile levels. * The proportion of the total number of fibers in a specified length or diameter interval is shown in brackets. f Numbers in parentheses refer to the number of bundles included in the cell total. 4 N = Total number of fibers. 10002358 548 KACZENSKI AND HALLENBECK TABLE 13 Characterization of thf: Lengths and Diameters of Chrvsotii.e Fibers" Recovered in Test Blood'' Length <M.m) 0.018-0.036 Chrysolile diameter Ijiml 0.037-0.071 0.072-0.125 >0.125 Total \9t ] ri II V 0.2-0.5 0.51-1.0 2.1-5.0 >5.0 Total [9c] IJ1 3 (0)1' 4 (0) 0(0) 0(0) 0(0) 0(0) 7 [6] 0(0) 24 (5) 42 (2) 10 (0) 2 (0) 0(0) 78 |64| 0 (01 8(6) 16 (12) 5 (5) 1 (0) 0 (0) 30 |25| 0 (0) 0 10) 2 (2) 4 (2) 0 (0) 0(0) 6|5| 3[2| 36 |3()| 60 1501 I9|I6| 3 |2| 0 [0] " Fibers = fibrils and bundles of fibrils. All entries were adjusted for blank chrysolile levels. * Blood taken al sacrifice. ' The proportion of the total number of fibers in a specified length or diameter interval is shown in brackets. d Numbers in parentheses refer to the number of bundles included in the cell total. ' N = Total number of fibers. asbestos was demonstrated by the significant levels of chrysotile recovered from the test pancreas and sacrifice blood samples. The test samples had chrysotile fiber and mass concentrations which were statistically greater than those of their respective control samples (P = 0.028). There was no indication of amphibole asbestos penetration and migration due to the insignificant levels of amphibole recovered from the test pancreas and blood samples. The test samples had amphibole fiber and mass concentrations which were not statistically different from controls. Blood taken at last gavage. Penetration and migration of chrysolile asbestos was indicated by the significant level of chrysotile fiber recovered from the test blood sample. The test sample had a chrysotile fiber concentration which was statistically greater than that of the control sample (P = 0.020). The test sample mass concentration was not significantly different from that of its control. Penetration and migration of amphibole asbestos was indicated by the signifi cant level of amphibole fiber recovered from the test blood sample. The lest sample had an amphibole fiber concentration which was statistically greater than that of the control sample (P < 0.001). A statistical analysis based on amphibole mass was not performed due to the occurrence of amphibole clusters in the test and control sample data. Kidney, lung, and liver. Test sample chrysotile and amphibole fiber and mass concentrations were not significantly different from those of the control. Spleen. Penetration and migration of chrysolile was indicated by the significant level of chrysotile recovered from the test spleen. The test sample had chrysotile fiber and mass concentrations which were significantly greater, than those of its blank (P = 0.028). 1000235? MIGRATION OF INGESTED ASBESTOS 549 Penetration and migration of amphibolc was indicated by the significant level of amphibole fiber recovered from the test spleen. The test sample had an amphibole fiber concentration that was significantly greater than its blank (P = 0.028). However, when the amphibolc data were analyzed on a mass concentra tion basis, the test concentration was not significantly different from that of its blank. Penetration and migration of chrysotile asbestos was demonstrated by the sig nificant fiber and mass levels of chrysotile recovered in the test stomach, heart, pancreas, sacrifice blood, and spleen samples. Characterizations of the lengths and diameters of the recovered chrysotile fibers are presented in Tables 10-14. Approximately 73% of the chrysotile fibers recovered from (he test stomach, heart, pancreas, sacrifice blood, and spleen samples were 0.2-1.0 jj.m in length. Approximately 62% of the chrysotile fibers administered to the test baboon by gavage were 0.2-1.0 p.m in length (Hallenbeck et at., 1981). CONCLUSIONS 1. The methodology used to prepare samples was sensitive and nondestructive to asbestos fibers. 2. Penetration and migration of chrysotile asbestos in the baboon due to gavage was demonstrated by both fiber count and mass data obtained from the test stomach, heart, pancreas, sacrifice blood, and spleen samples. Penetration and migration of chrysotile asbestos in the baboon due to gavage was indicated only by the fiber count data obtained from the test blood taken at last gavage. 3. Penetration and migration of amphibole asbestos in the baboon due to gavage was indicated only by fiber count data obtained from the test heart, blood taken at last gavage, and spleen samples. Amphibole penetration and migration TABLE 14 Characterization of the Lengths and Diameters of Chrysotile Fibers" Recovered in Test Spleen Tissue Length (M-m) 0.018-0.036 . Chrysotile diameter (|im) 0.037-0.071 0.072-0.125 >0.125 Total |T I" <0.2 0.2-0.5 0.51-1.0 1.1-2.0 2.1-5.0 >5.0 Total f%) 15 (Of 1 (0) 2 (0) 0(0) 0(0) 0(0) 18(51 0(0) 90 (4) 162 (2) 83 (0) II (0) 4 (1) 350 |9I) 0 (0) 6 (5) 7 (7) 0 (0) 2(1) 0 (0) 15 |4| 010) 0 (0) (I (0) 0 (0) 1 (1) 0(01 1 [<H 15 1-41 97 |25| 171 |44| 83 1221 I4|41 4 III NJ = 384 * Fibers = fibrils and bundles of fibrils. All entries were adjusted for blank chrysotile levels. 4 The proportion of the total number of fibers in a specified length or diameter interval is shown in brackets. r Numbers in parentheses refer to the number of bundles included in the cell total. 4 N = Total number of fibers. 550 KACZENSKI and hallenbeck may not have occurred to a great extent due to the larger size of amphibolc asbestos fibers as compared to chrysotile asbestos fibers. 4. It appears that the length distribution of gavagc fibers was not greatly changed in migrating through the body or during sample preparation. 5. Chrysotile concentrations in those test tissues (stomach, heart, pancreas, sacrifice blood, and spleen) where chrysotile penetration and migration due to gavage was demonstrated by both fiber count and mass data averaged 6100 fibers /mg tissue and 8.6 x 10"5 p.g/mg tissue. REFERENCES Amlerson. C. H.. and Long. J. M. <19801. ''Interim Method for Determining Asbestos in Water." U.S. EPA 600/4-80-005. Boatman. E. S.. Merrill. T.. O'Neill. A.. Polissar. L.. and Milletle. J. R. <19841. The use of quanti tative analysis of urine to assess exposure to asbestos fibers in drinking water in the Pudget Sound region. Environ. Health Perspect. S3. 131-141. Bolton. R. E-. and Davis. J. M. G. (1976). The short-term effects of chronic asbestos ingestion in rats. Ann. Occup. Hvg. 19. 121-138. Bolton, R. E.. Davis. J. M. G.. and Lamb. D. (1982). Pathological effects of prolonged asbestos ingestion in rats. Environ. Res. 29, 134-150. Carter, R. E.. and Taylor. W. F. (1980). Identification of a particular amphibolc asbestos fiber in tissues of persons exposed to a high oral intake of the mineral. Environ. Res. 21, 85-93. Conforti. P. M.. Kanarek. M. S.. Jackson. L. A.. Cooper. R. C.. and Murchio. J. C. (1981). Asbestos in drinking water and cancer in the San Francisco Bay area: 1969-1974 incidence. J. Chronic Du. 34, 211-224. Cook. P. M.. and Olson. G- F. (1979). Ingested mineral fibers: Elimination in human urine. Science (Washington. D.C.) 204. 195-198. Cunningham. H. M.. Moodie. C. A.. Lawrence. G. A., and Pontefract. R. D. (1977). Chronic effects of ingested asbestos in rats. Arch. Environ. Contain. Toxicol. 6. 507-513. Daniel. W. W. (1978). Applied Honparametric Statistics. Houghton Mifflin. Boston. Dixon. W. J.. and Massey. F. J. (1969). Inirodncrion to Statistical Analysis. McGraw -Hill. New York. Donham. K. J.. Berg. J. W., Will. L. A., and Leininger. J. R. (1980). The effects of long-term ingestion of asbestos on the colon of F344 rats. Cancer 45. 1073-1084. Evans. J. C.. Evans. R. J.. Holmes. A.. Hounam. R. F.. Jones. D. M.. Morgan. A., and Walsh. M. (1973). Studies on the deposition of inhaled fibrous material in the respiratory tract of the rat and its subsequent clearance using radioactive tracer techniques. Environ. Res. 6. 180- 201. Gibel. W.. Lohs. K. H.. Horn. K. H.. Wildner. G. P.. and Hoffmann. F. (1976). Animal experiments concerning cancerogenic effects of asbestos Fiber material following oral administration. Arch. Geschwulstforsch. 46, 437-442. Gross. P.. Harley. R. A.. Swinburne. L. M.. Davis. J. M. G.. and Greene. W. B. (1974). Ingested mineral fibers: Do they penetrate tissue and cause cancer? Arch. Environ. Health 29, 341-347. Hallenbcck. W. H.. Markey. D. R.. and Dolan. D. G. (1981). Analysis of tissue, blood, and urine samples from a baboon gavaged with chrysotile and crocidolile asbestos. Environ. Res. 25. 349-360. Hallenbeck. W. H.. and Patel-Mandlik. K. J. (1979). Presence of fibers in the urine of a baboon gavaged with chrysotile asbestos. Environ. Res. 20. 335-340. Hilding. A. C.. Hilding. D. A.. Larson. D. M.. and Aufderheide. A. C. (1981). Biological effects of ingested amosile asbestos, (aconite tailings, diatomaceous earth and Lake Superior water in rats. Arch. Environ. Health 36, 298 - 303. Kanarek. M. S.. Conforti. P. M.. Jackson. L. A.. Cooper. R. C.. and Murchio. J. C. (198(1). Asbestos in drinking water and cancer incidence in (he San Francisco Hay area. Amcr. J. Epidemiol. 112. 54-72. Levy. B. S.. Sigurdson. E.. Mandel. J.. Laudon. E.. and Pearson. J. (1976). Investigating possible I MIGRATION OF INGESTED ASBESTOS 551 effects of asbestos in city water: Surveillance of gastrointestinal cancer incidence in Duluth. Minnesota. Amer. J. Epidemiol. 103, 362-368. Mason. T. J.. McKay. F. W., and Miller. R. W. (1974). Asbestos-like fibers in Duluth water supply, relation to cancer mortality. J. Amer. Med. Assoc. 228. 1019-1020. McConnell. E. E.. Rutter. H.. Ulland. B.. and Moore. J. A. (1984a). Chronic effects of dietary ex posure to amosite asbestos and tremolile in Fischer 344 rats. Environ. Health Persped. 53. 27-44. McConnell. E. E.. Shefner. A. M., Rust. J.. and Moore. J. A. (1984b). Chronic effects of dietary exposure to amosite and chrysolile asbestos in Syrian Golden hamsters. Environ. Health Persped. 53, 11-25. Millette. J. R.. Clark. P. J.. Pansing. M. F.. and Twyman. J. D. (1980). Concentration and size of asbestos in water supplies. Environ. Health Persped. 34. 13-25. Patel-Mandlik, K. J.. Hallenbeck, W. H.. and Milletee. J. R. (1979). Asbestos fibers: I. A modified preparation of tissue samples for analysis by electron microscopy. 2. Presence of fibers in tissues of baboon fed chrysolite asbestos. J. Environ. Pathol. Toxicol. 2, 1385-1395. Patel-Mandlik. K. J.. and Millette. J. R. (1980). Asbestos in baboon organs following its ingestion. Scanning Electron Micros1, 347-354. Patel-Mandlik. K. J.. and Millette. J. R. (1983). Chrysolile asbestos in kidney cortex of chronically gavaged rats. Arch. Environ. Contain. Toxicol. 12. 247-255. Sebastien. P.. Masse. R.. and Bignon. J. (1980). Recovery of ingested asbestos fibers from the gas trointestinal lymph in rats. Environ. Res. 22. 201-216. Selikoff. I. J. (1979). Mortality experience of insulation workers in the United Slates and Canada. 1943-1976. Ann. N.Y. Acad. Sci. 330. 91-116. Smith. W. E.. Hubert, D. D.. Sobel. H. J.. Peters. E. T.. and Doerfler. T. E. (1980). Health of ex perimental animals drinking water with and without amosite asbestos and other mineral particles. Environ. Pathol. Toxicol. 3, 277-300. Snedecor. G. W.. and Cochran. W. G. (1976). Statistical Methods. Iowa State Univ. Press. Ames. U.S. EPA (1979). "Asbestos Ambient Water Quality Criteria." Criteria and Standards Division. Wash ington. D.C. Ward. J. M.. Frank. A. L.. Wenk. M.. Devor. M.. and Tarone. R. E. (1980). Ingested asbestos and intestinal carcinogenesis in F344 rats. J. Environ. Pathol. Toxicol. 3. 301-312. Westlake. G.' E., Spjut. H. L.. and Smith. M. N. (1965). Penetration of colonic mucosa by asbestos particles. Lah. Invest. 14, 2029-2033. Wigle. D. T. (1977). Cancer mortality in relation to asbestos in municipal water supplies. Arch. En viron. Health 32, 185-190. 10002362