Document NeGKELvmj2OKDGJDBw5pgYEBw

_______________ M MOV .. W' In: Pneumoconiosis. Proceedings of the International Conferenp^jp yr Johannesburg. 1969. / DAR LOW THE INHALATION OF FIORES V. Tiv.nr tLL Pneumoconiosis Reiccrch Unit of the Medics! Res:arch Council, Pensrih, Glamorgan, U.K. A previous study (Timbrcll, 1965) examined vlc influence of the dixmeter and length of fibres on the respirabili'.y of these panicles. paper extends the diseu.siun to the ii.finei-cc cf the fibre shape, .vith specific refeicncc to the relative rerpirebiiity of chryiotiie r.ttd arnphibole asbestos. 1'Ar.TiCLE Deposition Meoi'wisms The mechanisms operating to deposit particles in the lung and the relevant particle para meters are: 1. Gravitational settlement: free falling speed, ii. Inertial impaction: proportional to free failin' speed. lii. Interception: size, it. Diffusion: size. Diffusion is efficient only for particles smaller than 0.5 microns in small air spaces and is of little importance for the long fibres that can penetrate deeply into the lung. Fibre Diameter The observation that fibres of a range of different materials recovered from lungs are very slender is explained by the fact that the free failing speed- of a fibre is approximately proportional to the diameter squared and almost independent cf the length. Only if a fibre is of small diameter can it avoid deposition from gravitational settlement ar.d inertial impaction high in the respiratory tract and succeed in penetrating to the pul monary air spaces. Fibres of atnphibolc asbestos (arm site, aathophyllite and crocidolite) found in lungs are straight and have a maximum diameter of about 3 microns (Timbrcll, Pooley and Wagner, this Conference, p. 120). Chrysotiie fibres, on the other hand, arc often partially opened bundles cf very fine fibrils, whose length and diameter arc difficult to define. When, however, chrysotiie fibres in lung sections are tight bundles, the maximum diameter of these is approximately the same as for the amphibolcs. Aerosol spcc:;om:tc: examinations of partially opened bundles and aggregates of chrysotiie fibres ha/c .shown that some of these can have low falling srccds for their size. This explains why `thick' chrysotiie fibres are sometimes found in lungs. Fibre Length ?s R--GVs1'^^ For particles of compact shape, such as tho'erv of coal and silica, the important depo a 1* mechanism; operating in the tipper retpit;.- \ jjAv Tory tree: arc gravitational settlement and VH inertial impaction. Only if these ccmp.vt l< ' particles ate less than about 10 re:::: diameter can they penetrate deeply hue tne lung. Even tllC largest of these par ticles is small compared with the diameter; of the narrow airways so that inrsrrcpticn, which depends on the relative sizes of p.rdcit and airvrav diameter, is of little consequence. For fibres, however, interception cm be very important. A long fibre, if it :s slender, may avoid deposition in the upper respirr.t'.-ry tract from gravitational settlement and inert;,d action and penetrate deeply to the ionary air spaces. In these regions fee _ :h cf the fibre may be ccinparaJic to the diameters of the airways, and intercept lam becomes a major deposition mechanism. The previous study showed that the efficiency of deposition cf fibres by intercep tion increases with increase in fibre length and with decrease, in airway diameter. Inter ception must be expected to concentrate i.:m; fibres in the narrow air ducts and p.trticj.'.riv at bifurcations. Elsewhere at this Coitferenea (Timbrcll, Pooley and Wagner, p. 1 .to) tee length distributions cf amesite fibres in rai lung sections arc compared with the length, distribution of the fibres to which the anitiiaia were exposed. These results show that fibres 50 microns and longer were present in the lungs: the fibres in the lung were on average shorter than those in the cloud and the longer fibres were mainly iu the narrow air ducts, particularly at bilincations. Although other factors were probably involved the results agree with the conclusions reached from analysis of the interception effect in the previous study. Fibre SiiArE The most striking physical difference between the 2 main types of asbestos fibres is their shape: the amohiboles are straight and the chrvsotiies curly, roughly rerembling a stretched cod. This suggests the possibility cf // ` r\ ASARCO ALV 0002159 4 V. TimbrcU important differences between the ir.3!Titudcs of the size-sensitive interception effects in Fif. I. Measurement of coil diameter and coiJ length of a curly fibre. the lung for the 2 types of fibre. Data have therefore been collected on the shape as v/ell as the size of fibres in industrial clouds of amphibolcs and chrvsotile and in clouds generated under laboiatory conditions. A complete mathematical description of the shape of a curly fibre is difficult but a simple definition is adequate for present pur poses. The coil is enclosed in an imaginary cylinder. Under the microscope this cylinder normally has its axis parallel to the surface of the slide or membrane filter on which it is lying. A rectangle is visualized around the microscope image of the fibre as illustrated in Fig. 1. and the width and length arc measured. These 2 dimensions represent approximately the diameter and lcnttth cf the cylinder enclosing the coil. The ratio of roll diameter and mil lrngrh ic referred to as the `coil aspect ratio'. The ratio of the diameter and length of the fibre remains the `aspect ratio'. Ic may be noted that if the fibre is S'<9 o. Oi RF ER LE AQ TU IE VN EC Y <> ^ O'* 70 <> 9i to' 3 ^ 2` ..a0 n Coil Length o^ 3 4-5 6 0 (Microns) 9) 5 6' o-8 0 to a ao o<c> V o <x 0., 3.. '4.,, '* Coil Aspect Ratio Fig. 2. Distribution of coll length and coil aspect ratio for UICC Rhodesian chrysoule. ASARCO ALV 0002160 The Inhibition of Fibres 5 straight '.lie coil aspect ratio is equal to the aspect ratio. Fig. 2 shows the joint distribution of fibre cc-il length and coil aspect ratio obtained flora a membrane filter sample of the UICC Rhodesian clirysotile reference sample; Fig. 3 shows the distribution for an industrial cloud of chrysotile. Although these distribu tions diller in detail, they both exhibit the characteristic of a relatively small proportion of fibres with coil aspect ratio < 6.1: in both clouds the proportion of fibres that could be described as straight was even smaller. Fig. 4 shows examples of distributions of coil length and coil aspect ratio of laboratory and industrial clouds of amphibolc fibres, because these fibres were straight the majority of the coil aspect ratios were low (<0.05), compared with the much higher ratios for the chrysotile fibres. Fidre Shape and the Interception Effect In the study mentioned earlier, in order that the calculations should produce conclusions applicable to asbestos fibres in general, it was supposed that the fibres were straight and had random orientation in airways. This was justifiable on the grounds that asbestos fibres arc often dirty and that orientation is markedly influenced by attached particles. It was pointed out however that in practice, penetration of straight fibres into the respiratory tract would be assisted by a marked tendency for such fibres to digit themselves parallel to the axis of an airway: RF ER LE AQ _ TU to IE VN to>EC 9. to to to oto to to 0.i0 Q Coil Length to ,o-2 3o (Microns) m to 2'3o-45o Ri 0^ 4>% *0' o-*\o U.0) 0' o . to.to to 2 . 0. A./. 5-r '3. -4 Coil Aspect Ratio Fig. 3. Distribution of coil length and coil aspect ratio for an industrial chrysotile. t ASARCO ALV 0002161 6 V, Tiinbrell this tendency is the greater the longer the fibre and the smaller the airway diameter. If a straight fibre has such an orientation in an air duct, the effective interception diameter can be as small as the actual diameter of the fibre, and if the air duct is one of the narrower airways this diameter must be less than about 3 microns for the fibre to succeed in reaching the region. This inches it possible for a fibre ioo-microns long to have an interception diameter of 3 microns or less. Fig. 4 shows that virtually all the fibres in the amphibcle clouds could have small interception diameters in the narrow lung airways. fibre has to settle to make contact with the wall of an airway. The influence of curvature on the orientation of a fibre could result in this distance being shortened by as much as half the length of the fibre": for long fibres in narrow airways the reduction could be substantial. Curvature in a fibre therefore works to decrease the efficiency of penetration, especially in narrow airways. Looked at in another way, introduction of curvature into a fibre causes deposition to be caiiitr and the site to be higher in the respiratory tract. The fibres most affected are the long fibres, which ace also in general the most massive. UICC. Anlhophyllite F R E Q U E N C Y of10 U.I.C.C. Amosite U.l.C.C. Crocidolite ft* Industrial Amosite Industrial Crocidolite 010 jo 4OJ Q) Q ^5 <0<*0 ^ 3O <0"5j ,1 Coil Length (3)~>>< (0 30 p'0 0J ' <*S <>5 Coil Aspect Ratio Fig- 4. Distributions of coil length and coil aspect ratio for amphiboles. In contrast, a curly fibre in an airway can never exhibit an interception diameter as small as the actual fibre diameter. A minimum value is obtained if the axis of the fibre coil is parallel to the axis of the airway. Figs. 2 and 3 show that for a substantial proportion of long chrysotile fibres the coil diameters, and consequently the minimum interception diameters, are cf rhe order of 20 microns or greater. Also, since curly fibres do not show a marked tendency to align themselves with respect to the axis of an airway even these minimum interception diameters would only be exhibited occasionally. The gravitational settlement and inertial precipitation mechanisms are not entirely independent of the shape of a fibre. Curvature in a fibre increases the efficiency of these 2 mechanisms, for instance, introduction of curvature into a fibre reduces the distance the The interception effect is important only in the narrow lung airways. These arc short compared with the length of the respiratory tract. But they arc of disproportionate impor tance in the present context since it is in these air ducts that normally the ciliary epithelium and mucus production terminate. Although curvature in a fibre might cause only a snort reduction in the distance penetrated along one of these narrow airways it could mean a difference between landing on this escalator and travelling beyond it to reach a region whose geometrical complexity would make subsequent clearance of the fibre up the respiratory tract extremely difficult. Animal Experiments Recently the first results became available from an experiment in which groups of rats I ASARCO ALV 0002162  L. The Inhab'.ion of Fibres 7 v.-crc exposed to dust clouds of the UICC refcrcr.ce fibres in cqurd mass concentrations of respirable dust as measured with the Ccs-'iin Type 113A horizontal clutriruor gravi metric sampler. The animals were killed 24 hours after the end of a 10 weeks' exposure. Approximately equal weights of dust were estimated in the lungs of the rats exposed Physical Experiments The measured distributions of fibre length and of aerodynamic equivalent diameter are slightly coarser for the UICC chrysotilc than for the UICC amphiboles (Timbrcll, this Conference, p. 28). The percentage `respir able' fractions determined using the Cas'clla RP (Oi EE LN AE T T Ci i? Oi (Oi fO, Ol Ot Cl M 8}ki. 0 .2 I0'*,,.3 Coil Length 3',d-5fio (Microns) <> a 5 70 6 8 7 9 ,, 8 1O0 - 0 o. e.<9> e. o. 0. O 0.,, 0., 4., 5 5.. 3 '-4 Coil Aspect Ratio Fig. 5. Tube bundle experiment Relative penetration of fibres in terms of coil length and coil aspect ratio. lo the 3 amphiboles. This was about 6 times the weights (approximately equal) estimated to be in the lungs exposed to the 2 chrysotiles. This ratio is about the same as that obtained between amosite and chrysotilc in an earlier experiment when the materials used were not the UICC reference samples. (Wagner and Skidmore, 1965.) Type 113A sampler are slightly greater for the amphiboles than the chrysotiles, although a reverse result has been obtained using the Unico 18 cyclone device. Overall, these measurements, which take little cr no account of fibre shape, suggested that in animal experiments with the UICC reference samples more amosite than ebrysotile would be 1 ASARCO ALV 0002163 ... L_ 8 V. Timbrell deposited in rats' lungs in our experiment, but the difference should only be a few per cent. There are a few possible explanations for the ratio of 6 actually obtained, including a suggested difference in the solubility rates of chrysotiie and araphiboles in lung fluids. Clearly, the rates would have to be substantially different to effect this ratio of retained fibre after an exposure spread over only to weeks. The different shape of the 2 types of fibre presents another possibility, and to investigate whether this is capable of producing a ratio large enough to be an explanation, experi ments have been performed on the relative efficiency of penetration of curly and straight fibres through narrow tubes and hollow lung casts. Fig. 5 shows the result obtained from an experiment in which a cloud of UICC Rhodesian chrysotiie fibres was sucked through a bundle of loo tubes I cm. long. The cloud contained enough straight fibres to give a result comparable with that which would be obtained from an amphibole. The tubes were 400 microns internal diameter and air velocities were arranged to simulate those in the narrow lung airways. Distributions of coil length and coil aspect ratio were con structed for the fibres presented to the tube bundle and for the fibres which penetrated. These were compared to construct the pene tration diagram in Fig. 5. The results show decrease in penetration of the fibres with increase in coil length or coil aspect ratio and agree with the conclusions reached from con sideration of the interception mechanism of particle deposition. Using Fig. 5 and the distributions for UICC chrysotiie and amosite in Figs. 2 and 4, computations have been made of the rela tive mass of chrysotiie and amosite that would penetrate the bundle (and which in the analogy with the lung should travel beyond the ciliary epithelium) if equal masses of the two types were presented. This procedure was necessary because of difficulties, such as serious blockage of the tubes, in using a direct gravimetric method, if weighable samples were to be obtained. The computations indicated that 5-6 times as much amosite as chrysotiie should pene trate the tube bundle. This good agreement with the animal experiment depends to some extent on our favourable choice of tube dimensions, etc. in the physical study. The good agreement suggests, however, that fibre shape is an important factor in the efficiency of penetration through narrow ducts and must be considered a probable cause of the difference in the retention rates recorded for chrysotiie and amphibolcs in the animal experiment. Hollow casts have been produced of pig and human lungs. The `airways' can be cut at a required diameter and left open for the fibre cloud to be drawn through. The results have been more variable than in the studies on tube bundles, but even higher values have been obtained for the ratio of weights of amphibole and chrysotiie. Discussion Long curly chrysotiie fibres have been found in lungs, sometimes in numbers exceeding those observed for amphiboles. This however is not in conflict with the possibility that chrysotiie fibres are generally less respirable than amphiboles, since chrysotiie fibres readily divine into numerous flexible fibrils which are small in diameter compared with amphibole fibres. For instance, a chrysotiie fibre of 1 micron diameter could yield 1,000 fine fibrils of the same length. An observation of large numbers of chrysotiie fibres in lungs is thus not a good indication of a large mass being present. The experimental study of respirable asbestos fibres which was men tioned earlier suggested no similar subdividing of amphibole fibres, and comparison of fibre numbers is unlikely to be a good measure of relative mass of chrysotiie and amphiboles in lungs. An experiment reported elsewhere (Tim brell and Skidmore, 1968) using `Icng-fi'orc' and `short-fibre' clouds of amosite of approximately equal mass concentration of respirable fibres, indicated that the long fibres were the more fibrogenic. Other experiments (Vorwald, 1951) showed a similar relative importance of long fibres. The influence of fibre curvature on penetration is greatest for long fibres. These two factors working to gether could considerably reduce the fibrogenicity of inhaled chrysotiie relative to that of an amphibole. Some chrysotiie clouds clearly visible to the naked eye have yielded respirable frac tions as low as 5% when measured with the Casella sampler. These fractions are very much lower than those recorded for the chrysotiie and amphibole UICC reference samples and some industrial dust clouds. There is therefore a need for more data of this type to determine whether there is t ASARCO ALV 0002164 The Inhalation of Fibres 9 another factor, quite apart from shape, which could make inhalation exposures from chrysotile clouds generally less than from the amphiboles. Injection experiments do not simulate the deposition mechanisms which operate on inbaled particles, and such studies with fibres should be considered as examinations of their biological activity. This needs to be borne in mind when comparing the resuits of injection experiments with inhalation studies which are relevant to human exposure. Summary Gravitational settlement and inertial impaction have the effect of limiting the diameters of asbestos fibres that can reach the pulmonary air spaces to less than about 3 microns. Interception is important only in narrow airways and has 2 effects: i. It concentrates long fibres in the narrow airways especially at bifurcations. ii. It can cause a marked difference between the magnitude of retention of amphibole and chrysotile fibres. Timbrel!, V, (:96s): Ann. N.Y. Acad. Sci., 132, 255-273- Wagner, J. C. and Skidmore, J. W. (1965): Ann. N.Y. Acad. Sci., 132. 77-86. Timbrel!, V. and Skidmore, J. W. (1968): Proceedings of Dresden Conference. In the press. Vorwald, A. J. (1951): Arch, industr. Hyg., 3, 1-43- I ASARCO ALV 0002165