Document VpeDQ7E85vKwqoKpz93gEpbo

locomotive drivers Pekka Nokso-Koivisto, Eero Pukkala Abstract Locomotive drivers in the steam engine era were exposed to asbestos during their vocational training for two years while training in workshops. Later in their career they had exposure to coal and diesel combustion products. To assess the level of earlier exposure historical working conditions were reconstructed and hygienic conditions were measured. The average exposure to asbestos (mainly anthophylline) fibres > 5 - was 5.0 fibredcm'. Incidence of cancer in a cohort of 8391 members of the Finnish Locomotive Drivers' Association, 195391, was analysed. The incidence of lung cancer and also total cancer was below the national average, probably due to the low prevalence of smoking among the drivers in the steam engine era. A fourfold risk of mesothelioma was found, most likely caused by exposure to asbestos. Also the observed 1.5-fold incidence of non-melanoma skin cancer and 1*7-foldrisk of cancer of the oral cavity and pharynx may be related to occupa- tion. hygienic conditions in the steam engine era -and b o u g h evaluating routine hygemc 1 %investigations of working conditions be Idiesel engine era. Also, the incidence of + cancer among Finnish locomotive driver 1 was compared w t h that of the total Fmnist: population from 1953 to 1991. Methods -_ ASSESSMENT OF EXPOSURES locomotive drivers of the F b s h starc railways are educated according to defined cwrkula, which before the 1970s mcldcd two years practising in a railway workshop The main part of the training rook place III rbc engine maintenance department of the work- shop. In the 1960s steam engine mamrenancc gradually ended. The drivers who started their m g at the beginning of the 1970s did nor take parr in steam boiler upkeep. The yarn* graduated IO assistant locomotive dnvcn started regular train operations, and were lata promoted to locomotive drivers. TO study the hygienic condmons m steam engine era the working condinons wfl reconstructed in two workshops wth bf assistance of older workers who used to r* ( O m pEmrimn M d 1994;51:330-334) *part in the earlier standard procedm During the procedure the outer coves of boilers of two engines were &srnmdrd Asbestos layers were loosened and tom ayF This study arose fkom a case of mesothelioma and collected in separate contamer. Tb in a locomotive driver, and fkom concern that process looked dusty. The former W O W conservation and recycling of old locomotives confinned that the reconsrmcted procduro and coaches causes exposure to asbestos. resembled old routine dismantling. T h e During the steam engine era all trainee loco- ers pointed out that there were locomow motive drivers in Finland were exposed to under maintenance regularly arid that the dd asbestos in the railway workshops. When hovered all over the workshop. working in traffic operations, the drivers were For the hygienic assessment of d1s exposed to combustion products and to coal eight personal air samples were collec tar. New diesel locomotive drivers are asbestos exposure was m exposed to exhaust fumes from the engine. dardised techniques. The Asbestos exposure is suspected to have fibres > 5 pm was 5.0 (ran caused an increased risk of mesothelioma and indicating medium exposure. lung cancer among locomotive drivers and contained two types of asbest0 other railway w0rkers.l Combustion products anthophyllite and chrysodle fibres O c c u p a t i d Herlth scrvicer ofThe State Trrp.ury, Finland P Noks+Koivisto of coal and diesel have been reported IO be associated with lung cancert and cancer of the bladder.`-'l The incidence of non-melanoma type of asbestos exposure was 3n31YS the samples of insulation matenal of demolished engmes. The analvsls Of Fkrnth Cancer skin cancer is increased among people samples mdicated that anth~ph!'ll*~~ R=*w,I.iiuntaN 21 B,Fin-00170 HebinLi,Fialand EPukliPlP exposed to combustion products of coal6 and creosote.' Cancers of the buccal cavity and pharynx and lymphatic tissue have been main type of asbestos used m rnsulaaoa 0 ven5nal of boilers. Chrysotile was der`' woven strings around pipes and i- Request for reprints IO: found in excess among railway workers.48 Asbestos concentrations in cabms Of 4 i i Euro pukL.la, Fwsh Cauca Registry, Lisankaru 21 B, FIN40170HcIsinh, Our study has two components. The life- locomotives with asbestos ~ s d a a o e long work exposures oflocomotive dnvers, as a exhaust pipes has been J I I ~ 4 ~0' ~ ~ ~ ' Finland. constellation of occupational causes of cancer, number of fibres was undetectable ( I Accepted 18 January 1994 were investigated through measurements of fibreskm'). I n e concentration of benz(a)pyrene and dust exposure in rolling stock and during a journey was measured in the cabin of a renovated steam locomotive with standard occupational hygienic methods. While cleaning and declinkering a fire box, whlch took less than half an hour a day, the concentrations of benz(a)pyrene varied from 0-006 to 0.025 pg/m3and dust concentrations from 1.0 to 1-6 mg/m3.During the journey the concentrations of benz(a)pyrene were lower, from 0.003 to 0.010 pgim'. Dust concentrations varied from 0.7 to 2.8 rngim'. Routine hygienic measurernents in cabins of diesel engines and in rolling workshops in Finnish railways indicated that the concentrations of panicles and benz(a)pyrene in diesel engines were similar to those in the steam engine. No nitrogen oxides were detected during steam engine operations, but according to routine hygienic measurements in Finnish railways the diesel engine drivers are exposed to nitrogen compounds emitted by the engines in the cabins and in rolling stock.P As well as the exposure directly linked to train operations, locomotive dnvers were exposed to organic substances such as lubricating oils, coal tar (creosote), and coal. COHORT STUDY OF INCDENCE OF CANCER The cancer risk of locomoave drivers was studied in a retrospective cohort study. The cohort consisted of all members of the Finnish Locomotive Drivers' Associadon between the years 1953 and 1991-that is, all Finnish drivers since 1953. Retired locomotive drivers remain members of the association up to their death. All drivers were men. The cohort consisted of 8693 members; 302 members (33%) were excluded because of inadequate personal identification data. Follow up for Table I Penmycan undrrf o h up among the 8391 Finnuh k m o t i v e drivm, 1953-91, by period and f o h up time sincef i n r emplqymrnr Fdlorr, up dmr 01) Pmod 0-14 15-29 30+ To& 195-5 1966-78 1979-91 1953-91 35 333 14957 25111 75401 24 101 35756 14655 71512 7 506 18516 36819 62841 66940 69229 76585 212754 death and emigration was performed in a record linkage with the Central Population Register of Finland. The calculation of person-years started at the beginning of the membership or on 1 January 1953, whichever was later, and ended at death or emigration or on 31 December 1991, whichever occurred first. At the Finnish Cancer Registry cancer cases of the cohort were identified by means of an automatic record linkage with a unique personal identification number assigned to everyone resident in Finland since the beginning of 1967. For those people who had died between 1953 and 1966 the follow up was performed manually with names, dates of birth, and places of residence as keys. The observed number of cases and person-years at risk were counted for five year age groups for three 15 year follow up periods since first emolovment, senarately for 1953-65, 1966-78, a d 1979-91. -It should be noted that all the Finnish locomotive drivers start their training at about the same age, from 16 to 20. This means that there is no need to report results by both follow up time since first employment and age. The expected numbers of cases for total cancer and for specific cancer types were cal- culated by multiplying the number of personyears in each age group by the corresponding average incidence of cancer in Finland during the period of observation. The specific cancer types selected for study were cancers earlier reported to be linked with asbestos or derivaties of coal and diesel combustion products including creosote. These are mouth and pharynx, stomach, colon, lung and trachea, mesothelioma, prostate, kidney, bladder, ureter and urethra, skin melanoma, skin (non-melanoma), lymphatic tissue (non-Hodgkm's), Hodglun's disease, and leukaemia. To calculate the standardised incidence ratio (SIR)for each group, the observed number of cases was divided by the expected number. Sigdicance was tested by the Mantel-Haenszel xL test, on the presumption that the number of observed cases followed a Poisson distribution. 'At2 Expaed (Exp) and obserued (Oh) numben of cnncer cases, 1953-91, among - ~ w , n cdrnrm and the standardiredtnca&ncerams (SIR) vnrh char 95% CIS,by SIU Obs ErP SIR (95% CI) k i ut- (I4O-204)'t 915 -a, ~ m c aynd pharynx (143-148) b&m(wIhn()151) (154) rl lad rnchea (162.0-1) -hna (162.2,158) +-talc (177) ra*y (180) *bdcr,ureter, urettm (181.0)' d M O m 9 (190) Iwn-melmomn) (191)t hHodglun's lymphoma (200,202) + disew (201) a-. (204) 17 86 36 29 236 8 122 38 48 17 32 19 7 21 .b h d c s papillomnr of the bladder. 'khrd? basd cell cntcinornaa of the skin. .- 962 10 112 37 34 274 2 109 31 44 17 21 24 8 24 0.95 1.75 0.77 0.98 0.85 0.86 445 1.12 1.25 1.08 0.99 1.53 0.78 0.83 0.87 (0.89-141) (1.02-2.80) (0.61-0.94) (0.69-1.36) (0.57-1.22) (0.75497) (1.75-7.97) (0.92-1'32) (0.88-1.70) (0.80-1.43) (0.57-1.57) (1434-2.15) (047-1.21) (0.34-1.72) (0.54-1.32) Results The number of person-years under follow up for 8391 men in the cohort was 212 800 (table 1). The mean duration of follow up was 25.4 years. During the 38 year observation period, 9 15 cases of cancer were found; the expected number was 962 (table 2). The SIR for total cancer was sigmficantly lower than expected only in the first period, 1953-65 (131 observed cases (Obs) v 163 expected (Exp), SIR 0.81, 95% confidence interval (95% CI) 0 . 6 7 4 9 4 ) . In later periods the SIR for total cancer was close to unity. The reduction in the risk in the first period was totally attributable to cancers of the stomach (SIR 0.60, 21 Obs w 35 ET, 95% CI 0-37-0.92) and l u g (SIR 0.64, 29 Obs v 46 Exp, 95% CI - ---- - _ _ _ _ _ _ _ __-__ _...... .I I-.... . 95% CIsjor sektod stm among~~nnub'loc~moaduncvm, 1953-91 bjiii- up nmc SulcejiKt m m n r F&w-6!! 0-1 4 15-29 2 30 -,,(., - rnwv m ~ 7 ~ ~ 7 1 Obs SIR PS%CI) Obs SIR @S%ClI Obs SIR (SS%c0 w Utes Wcantyandphnrynr Lung and uachea Mesorhel~ona Kidney Bladder, ureter, urethra Shn (non-melmoma) 26 0.64 -- -6 - 1.02 0.-05 -- 1 0.99 (0.42-0.93) (04.15) (0.37-2.22) (0-72.6) (0-4 35) (04.94) (0.02-5.49) 190 0 9 6 (04S1.10) 699 0 9 7 (0.90-1 03, 4 1 5 6 (0.43-4.00) 13 1 9 6 (1 05-3s - -43 0.73 (0.53497) 187 0.89 (0.77-1 02 (0.00-10.9) 8 5 4 4 (2 18- 99] 9 1 31 (0.60-2.49) 29 1.27 (0 85-1 82 9 135 (062457) 39 1 0 5 (075-1 43 6 1.67 (0 61-3.64) 25 1 53 (0 99-22) 0.43-0.91). In the more recent observation periods the relative risk of lung cancer approached unity: The SIR was 0.87 (95% CI 0.70-1.07) in 1966-78 and 0.93 (95% CI 0.77-1.10) in 1979-91. No cases of mesothelioma were diagnosed dunng the first 30 years of follow up since first employment, but after that the risk was sigxllficantly increased (SIR 5.04, 95% CI 2.18-9.93, table 3). The sigxllficantly increased risks for cancer of the oral cavity and pharynx (SIR 1.75) and for non-melanoma skin cancer (SIR 1.53) were due to the excess of cases after 15 years of follow up. The SIR for other analysed cancers remained close to umty over the whole observation period and there were no systematic changes by follow up time. Discussion The 8391 locomotive drivers in our cohort form a homogeneous group starting their career at the same age and having similar training. The loss of 3.5% of the members of the original cohort due to inadequate identification data is likely to be non-systematic and therefore should not have caused any bias in the results. In the steam engine era they were exposed to asbestos during their two years of training. h aaffic operations they were exposed to combustion products of the fuels of engines. The reconstruction showed that the insulation of the boilers caused at least a medium exposure to asbestos. The dust hovered in the air all the working day. AmericanIO and Swedish1studies based on enquiries of workh g conditions obtained similar estimations of exposure. The mean fibre counts m asbestos t e d e factories rn the late 1940s and early 1950s have been estimated to be 5-10 fibredan'. The levels of asbestos exposure in the railway workshops were typical of heavy industry from the 1920s to the 1970s. In Finland the main component was anthophylline produced in domesac mines. The exposure of locomotive drivers was not exclusively to anthophylline fibres as chrysotile was detected in samples taken around the pipes of locomo- fives. In minerological invesugauons h the Finnish asbestos mining area, the source of asbestos used in the railways, traces of tremo- lite have also been found.I* Amphibole varieties of asbestos have beer. ! considered to be the main causal agent of i mesothelioma." Chrysotile has been shown increase risk of lung cancer-for example. L' asbestos textile operations." Also, mcreass: risk of bronchial cancer occurs among anhv phylline miners even when cigarette s m h t is taken into account.15The Finnish locome tive drivers have been exposed to anrhophylline asbestos contaminated with chrysoolc and tremolite. Results of earlier studies that show an increased risk of mesothelioma among railwit workers vary considerably due to difference in type of asbestos and duration and level of the exposure. M a n c u ~ ofo~u~nd that one m I! machinists exposed to chrysotile develod mesothelioma, and mesothelioma consututd 34% of all cancer deaths. Railway occupaoom in the United States with regular asbesta exposure (skilled trades, steam locornon~~ repair) had an odds ratio of 21.4, but the occupations with potential intermittent exPo s u r e (engineers) had a non-sipficant odb ratio of only 2.3.'' In a Swedish cancer nB' istry study an increased risk of 9-5 (four Cam observed) was found among workers worm in the railway equipment m a n u f a c m industry.ls Pet0 has stated that brief early expos- IC asbestos may produce a substanaal risk mesothelioma but only a small increased n3 of bronchial carcinoma, whereas 20 years ''asbestos exposure from the age of 45 to f4might increase the risk much more bronchial carcinoma than for mesothelioma The results of our study are consistent *I' this model. The short term medium exPosur' to asbestos seems to be associated wth a tive risk of 4 for mesothelioma but the risk lung cancer is not increased. The mcldenCe lung cancer was below the national aveM and there was no mcrease ~ t mhcreasmg low up smce first employment. In our cohort all the cases of mesotheliorP' **occurred more than 30 years SmCe employment, and only one of eight caseS younger than 60. This finding IS SunJar 'those reponed earlier. The malignant cO* quences of asbestos exposure appear 3O to years after exposure.I4 *O Smoking is the most important nond pational risk factor for mesothelloma especially for lung cancer. The upward in the SIR of lung cancer from Oe6 yl t,t 1950s to 0.9 in the 1980s in our stud)' cfl . explained by changes in exposures (asbestos, be work related. Because of the low prrrr- combustion products) and in smoking habits. Ience of smoking in the steam engine e n dac Smoking, diesel exhaust fumes, and asbestos risk of lung cancer has until now been clarly have been suggested to have interactive effects below the national average. The exposure to on risk of lung cancer.22Diesel locomotives, diesel engine exhaust fumes seems to be SmJl causing exposure to diesel exhaust, were and its role in causing cancer neghble. 'Ihe introduced to the railways in the 1950s. The classic occupational exposures have been exposure to combustion products among waning with new working conditions arid locomotive drivers is, however, only slightly gradual elecmfication of the railways in h e above the national average. The exposure to 1970s and the 1980s but this benefit is not benz(a)pyrene is above that of Finnish town strong enough to compensate for the incrrcls- air.= In a Swedish study of aaftic workers ing risks associated with modem working con- with the same level of diesel exhaust exposure ditions including more enclosed cabins, rmrc as Finnish locomotive dnvers there were no monitoring, and more smokmg. detected signs of urinary mutagenicity or increased rhioether excretion in urine.' In the steam engine era Finnish locomotive drivers were not heavy smokers. Anecdotal evidence indicates that draughts in the cabin prevented drivers from appreciating smoking. Thu work was funded by an Occupaaod Health Gnnt 5um the Treasury of Fdand. We thank Hanau Ripmen .Hssfw exposure assessmmts and Mam Romo .MD and HcJrh Hirvonen MD, Finnuh State Railways, for theu comme0. We express our slncere rhnak to Professor Nick Day d to the MRC Biascansacs Umt, Cambndge. for support m nuon of the ~ C ~ O R . It may be assumed that the increased smoking in the less draughty cabins of diesel locomo- tives with less physical work and more passive monitoring would have increased the risk of lung cancer close to the national average. A cross sectional study of Finnish locomotive drivers in the year 1976 showed that 40% of them were current smokers, 34% had stopped smoking, and 26% had never These percentages are similar to those of the whole Finnish population in 1979.2s Asbestosis insulation workers in the United States and Canada have an increased risk of death from cancers of the oesophagus, colon and rectum, larynx, oropharynx, kidney, and perhaps stomach.z6 Our study showed an increased SIR for oropharyngeal cancer but not for the other sites, possibly due to differ- 1 Ohhon C-G, Klaesson B, Hogtedt C. . M o d y lmaao asbestos-exposed workers in a railway worMop. Scad J Work En- He& 1984;10283-91. 2 W~atnsRR, Stegens NL, Goldsmith JR.AuocLdopr of cancer site and type wrh occupation and indu- h m third ~ t i o n aclancer survey interview. 3Nad CorvsI~ut 1977;59:1147-85. 3 V i c h P, Simonoto L Esdmates of rhe proprodolr of bladder cancers amibutablc to occupation. .Scad J Wvk Emnnm He& 1986;12:5%50. 4 Schenk TU'. A mortality study of locomotive a q b c a S , Int-B. Buffalo, NY:State Univasity of New Yo+ 1989. (Dissemtion.) 5 schifavs E,JvnvrJ, Renard V.Tobacco and occupawn as risk factors in bladder cancer: a casf-conaul rmdy in Southern Belgium. I n t J C a m 1987;39287-92. 6 Whitakcr CJ, k WR, Downes J. Squamous cell skin cancer in the Nodwest of England, 196749, md its relation to occupation. Er3Id.Uuf 1979;36.4>51. 7 Karlehagen S,Andenen A, Ohlson C G. Cancer inddence among crcosote-crposed workers. Scand J&'FI .?3wm ent exposures. On the other hand our study shows an increased risk of non-melanoma skin Health 1992;18:26-9. 8 Viana E, B m s DJ,Houren L Cancer &axe of men exposed to inhalanon of chemicals or to com- cancer, consistent with other experiences, in occupational medicine. Locomotive drivers bustion products. 3 Occup Med 1976;18:787-92. 9 Heino M, Ketoln R, Mike13 P, W e n R, Niemeli R Snuck J, Partanen T. Work conditions and halrh of are exposed to different organic substances (lubricating oils, creosote) that can cause dermal irritation and t o ~ i d e r m i aA. ~s~light excess of squamous cell skin cancer has been found locomotive engmcen. I. Noise, vibration, rherrml dimate, diesel exhaust constituents, ergonomics. Sumd vmJ Work E n v i m He& 1978;4(suppl 3):3-14. 10 Nicholson W,Pcrkcl G,Selikof IJ. Occypauonal to asbestos: population at risk and pmlecrcd m o d 1980-2030. A m l I n d Mcd 1982;3:259-311. before among engine driversz8and creosote exposed impregnators.29 11 Bcny G, Lewissohn HC. Dose-response relationsbps for asbestos related disease: implications for h y @ a c swdards. Ann .WAcad Sci 1979;330 185-94. Among the members of the Brotherhood of 12 Karjalainen A, Meurman L, Puklrala E. Four cases of Locomotive Engineers in the United States mesothelioma among Finnish anrhophylline m e n . Occup Envinm Med 1994;51:212-5. there were sigmficantly raised proportionate mortality ratios for cancers of the bladder 13 Pelnv W .Further evidence of nonasbestos-di.ted mesothelioma, a review of litenme. S c d J Work Enwimn H c d h 1988;141414. (2-05),kidney (2-65), digestive organs ( 1 ~ 3 0 ) ~ 14 and Ijmphoreticular malignancies (2-05.)4 Dement JM, Harris RL, Symons and monalip among chrysoule MJ,Shv asbcsto; wCoMrk.eErsx.pPo+sunrarS: Monalitv. rlm f Ind ,Med 1983;4:421-33. Mortality ratios for the respiratory system were not increased. The increased risks of 15 M c u H ~ ~ ~ . LKiOvil,uoto R Hakama .M. Combined effect of asbestos exposure and tobacco smoidng on Finrush anrhophylline miners and millers. Ann .VY Acad Sci cancer of the lower urinary tract associated with occupational exposure of locomotive dri- 1979;330:49 1-5. 16 Mancuso TF,Relative risk of mesothelioma among rail- road workers exposed to chrysodlc. Am J Ind Wed vers has been confirmed by later studies,'3o even when smoking was controlled We 1988;13:639-57. 17 Schenker MB,Ganhick, et d.A populaaon-based casecontrol study of mesothelioma deaths among U.S. found only slightly higher risks for cancers of the bladder (SIR 1.08) and kidney (SIR 1-25), which may, t a h g into account the low prevalence of smoking in our cohort,include a railroad workers. A m RN R+ir Dir 1986;134:461-5. 18 Malker HSR Mclaughli JK, ,Malka BK, Stone BJ, Weiner JA, Enkon JLE, Blot WJ.Occupaaod nsk for pleural mesochelioma in Sweden, 1961-1979. 3 N d Cancer ins^ 1985;7461-6. 19 Peto J. Doseresponse and mortality. Dose L"JW component of occupational aetiology. In conclusion, the fourfold risk of mesothe- relationship for asbestos related disease: implicadom for hygienic standards. Ann NY A d Sci 1979;330: 195-203. lioma in locomotive drivers is likely to be wiggered by short term asbestos exposure. Also 20 Jones RD, Smith DM Thomas E.Mesothelioma in Britain in 1968-1983. Scand 7 Work Em'nm Health 1988;14:-45-52. -- the increased incidence of non-melanoma skin cancer and oropharyngeal cancer might 21 Bignon J, Sebastien P, Di Menzp L, Payan H. French mesothelioma r e b r a . Ann NY Acad Sci 1979;330: 45546. ! Correspondence and editorials -- OGUpaWdmrdE- Mrdicine wel- comes comspondQlce relating to any of the m a t d appearing in the journal. Results h m prrlimuLary or smaU scale studies may also be published in the correspondence column if this scam appropriate. Lett- should be not more than 500 words in length and contain a minimumof rrfcrrnces.Tables and figures should be kept to an absolute minimum. Letten are accepted on the understanding that they may be subject to editorial revision and shortening. The journal also publishes editorialswhich are nonnally specially commissioned. The Editor welcomes suggations r e e g suitable topics; those wishing to submit an editorial, however, should do so only after discussion with the Editor.