Document jNgdKGJRe26VN3OBDN1zOwRZy

* Ten Cases of Angiosarcoma of the Liver in Shawinigan, Quebec Fernand Delorme, M.D., and GiHes Theriault, M.D. 3D OG CO cr- Ten cases of angiosarcoma of the liver have been diagnosed in Shawinigan, Quebec since 1955. All have occurred in men who worked in a vinyl chloride polymerizing plant. Cigarettes and alcohol do not seem to be associated with this tumor. The amount of vinyl chloride to which these people were exposed, according to information obtained through questionnaires, ap pears elevated. Angiosarcoma of the liver is accompanied by a fibrosis of the liver which may precede its appearance. In de scribing the Canadian cases of angiosarcoma, this study at tempts to shed more light upon the causal relationship be tween vinyl chloride monomer and angiosarcoma of the liver and thus to provide some dues to understanding the patho genesis of this disease. In January 1974. three cases of angiosarcoma of the liver were discovered among workers of a chemical plant.1 This finding ap peared abnormal and the agent suspected of being responsible for the disease was identified as vinyl chloride monomer (VCM). Since then, many reports on the occurrence of this cancer among vinyl chloride workers have appeared around the world2*1 and the Unit ed States Occupational Safety and Health Administration has enacted a new threshold limit value for concentration of this chemical in the air of the working environment* Ten cases of angiosarcoma of the liver were diagnosed among workers of a vinyl chloride polymerizing plant in Shawinigan, Quebec. This is the largest number of cases ever to take place in a single plant. The present study describes occupational history drinking and smoking habits and some histological features of these Canadian cases. Data on some of these Canadian cases jointly with cases of angiosarcoma from the United States have al ready been described in another article by one of the authors 7 Since more accurate and complete information was retrieved cm the Canadian cases, it is being presented separately in the present article.__________ _________________________________________________ from the Depirtmera ol Pathology, Centre HmpiUlier Regno*! de It Mtunre Shtwinigaft. Quebec (Dr Oelormel *nd the Department c4 Socul *r*d Preventive Med icine. Uvil University. Ste-foy. Quebec (Dr Theriault) 338 Methods Case Finding. -- A search for cases of angiosarcoma of the fiver beginning as far back as 1955 and extending over two decades was carried out bv scrutinizing files of the regional hospitals in the Shawinigan vicinity. All cases of primary tumors of the liver were collected and the histological material (slides, biopsy reports, autopsy reports) were reviewed for the presence of angiosarcoma To confirm the diagnosis of angiosarcoma of the liver, the cases were submitted to consultants from the United States familiar with this disease. Occupational History. -- The questionnaires used to obtain a detailed occupational history of each case culled information con cerning past and present residence, medical history, smoking and drinking habits along with past occupations and employers in chronological order. The answers to most of questionnaires were given by spouses of the men who died from the disease. In one in stance. however, the answers were given by the sister and brother-in-law. and in another, by the worker himself. Two former co-workers of the men who died of an angiosarcoma of the liver confirmed and completed the information obtained through the families. Plant Description. -- The Shawinigan vinyl chloride plant start ed operating in 1943 At that time, it produced and polymerized vinyl chloride monomer Later in the 1950s it stopped producing VCM and kept operating the polymerization unit. The workforce varied from 100 to 350 workers At the moment, it employs 250 men Production of PVC has fluctuated with time. Present produc tion is estimated at 50 million pounds annually Exposure to VCM in the working atmosphere, though not documented, is believed to have been very high in the past. Sharp improvement has taken place in recent years and it is estimated at less than 3 ppm at the moment. ReMitU Table 1 shows the age at death, the smoking habits and alcohol consumption for the ten cases of angiosarcoma of the liver diag nosed in the Shawinigan region over the last two decades All Original Articles URL 18861 Effect of Occupational and Journal of Occupational Medicine October 1976 Vol. 18 No. 10 Nonoccupational Factors on the Respiratory System of Vinyl Chloride and Other Workers )ohn Gamble, Ph.D.; Shuguey Liu, M.S.; A. |. McMichael, M.D., Ph.D.; and Richard |. Waxweiier, M.S.I.E. There are suggestions in the literature that vinyl chloride (VC) acts as a lung irritant. Respiratory questionnaires and lung function tests were administered to 174 chemical (VC) workers, 81 polyvinyl chloride (PVC) workers, 72 former VC workers, and 136 rubber workers, and 68 maintenance workers with exposure to VC, PVC, and rubber. Except for small air ways obstruction associated with rubber, increased respiratory symptoms and decreased pulmonary function were not associated with working in chemicals, plastics, or rubber. Some increases in baseline pulmonary function were associated with VC exposure. Acute reductions in pulmonary function were observed in smokers working in chemicals, plastics, and rubber. Heavier cigarette smokers over 40 years of age had the most adversely affected respiratory system. Work was not associated with chronic respiratory effects, but all exposure groups experienced some acute respiratory insult. At high exposures, vinyl chloride (VC) is a nervous system depressant, and in fact was investigated as a possible anesthetic, but was not used because of circulatory and cardiac effects at concentrations of 10-20 vol. %.' In the late 1960's, acroosteolysis was associated with the polymerization of VC to form polyvinyl chloride (PVC).2*4 Early in 1974, four fatal cases of angiosarcoma of the liver were reported in reactor operators working in a VCpolymerization plant.5 Subsequently the disease has been discovered among VC workers all over the world. Experimental and epidemiologic studies suggest that exposure to VC may carry an increased risk of respiratory impairment.6'14 As a result, the Occupational Health Studies Group (OHSG) at the University of from Occi^Htional Health Studies Group. School ot Public Health. University o! North Carolina. Chapel Hill. NC 27S14 Reprint requests to NCNB Plaza. Suite 32. Chapel Hill. NC 27514 IDr Gamble) North Carolina and the National Institute of Occupational Safety and Health INIOSH) and CDC jointly carried out an epidemiologic study of VC-PVC workers. A part of this study was the investiga tion of respiratory function and symptoms in currently employed VC and PVC workers, former VC workers, and a sample of control rubber workers employed in the same plant. Methods and Materials The plant under study fas three major divisions. The chemical plant produces primarily polyvinyl chloride (PVC) resin, although some copolymers such as vinylidene chloride and styrenebutadiene are also produced. The polymers are either shipped or used in the Film and Sheeting division of this same plant in the production of plastic products. Film and Sheeting division (plastics) is in a building adjacent to the Chemical Plant, which also contains the Rubber Products division manufacturing primarily tires and tubes. All workers currently employed in the VC-PVC plants were invited for examination. A sample of 136 rubber workers from cleaner areas of the rubber division (i.e.. tire building, final inspection, shipping and receiving) and without prior exposure to either VC or PVC, were selected and groupmatched with the VC workers on age and sex. A total of 71 former VC workers and 68 maintenance workers were also examined. The response rate varied from a low of 62% in rubber workers to 77% in chemical workers. Clinical, smoking and occupational histories, complete physical examinations, chest roentgenograms, sputum cytology, symptoms, blood and urine chemistries were obtained on all 531 workers, and will be reported elsewhere. The British Medical Research Council respiratory questionnaire was administered by trained medical interviewers. Each worker performed a minimum of five forced exhalations, seated and with a noseclip. The three best performances were selected on the basis of forced expiratory volume in 1 second (FEV) and forced vital capacity (FVC) within 10% of each other, and acceptable re producibility of the flow-volume curve. In addition to FEV and BoprUrtod from Jownal oI Oom--on* HUklM October. 1976. Volume II. No. 10 p. 6M-S70 FVC, expired flow at 25% (FEFjs), 50% (FEFso), and 75% (FEF?s) of exhaled FVC were obtained. At least three single-breath nitrogen tests were also performed. Recommended procedures and selec tion criteria were followed for obtaining closing volume (ex pressed as ratio of closing volume to vital capacity. CV/VC." In addition, a random sample of fifty-six VC, PVC and rubber workers were administered lung function tests before work and again after work. The difference in flow rates is calculated as after - before/mean with afterwork flow rates measured using the maxi mum pre-exposure FVC. The following questions were of interest in this study 1. What are the effects of work on respiratory symptoms and function? Symptoms and baseline lung function were analyzed ac cording to three work exposure categories: (a) current job (hourly VC salary VC. PVC, rubber, maintenance, former chemical): (b) all lobs, past and present (only in VC. only in PVC, only tn rubber, or in a combination of the three). In this mixed exposure group an analysis of covariance was used to sort out the contributions of the VC. PVC, and rubber work experience to baseline pulmonary function: lc) man-months of exposure to VC: All VC jobs were given a high (10), medium (3) or low (1] exposure rating. Time (in months) spent in each job multiplied by the exposure index provided a cumulative exposure to VC. This cumulative exposure score ICESi was then compared to baseline pulmonary function. The association of dose (years worked in the exposure categories or CES) and baseline lung function were assessed by calculating lung function regression equations for each smoking category of the total population. Then, within each exposure group, actual values of each smoking category were expressed as a percentage of expected values for all workers in that same smoking category. In this way, the effect of work exposure on lung function was adjusted for differences in age, height and smoking between exposure categories. 2. What are the potential hazards of work exposure as measured by changes in pulmonary function over a work shift IAPFH A reduction in lung function suggests the likelihood of some harmful exposure in the working environment. 3. Are symptoms rates and lung function parameters com parable to other occupation groups? Because the PVC and rubber control groups were themselves exposed to varied levels of dust and fumes, comparisons with other studies are made. Results The age-smoking distribution of the population is shown in Table 1. Slightly over half of the population were smokers, 20% nonsmokers, and 16% exsmokers. The exsmokers were older than the other smoking categories, with proportionally fewer ex smokers in the .1.39 year age group. Smokers and nonsmokers had approximately the same distribution in all age groups, although a higher proportion of nonsmokers were older, and a higher propor tion of smokers were in the middle age groups. None of the light smokers ! < 1 pack/day) smoked more than 25 pack-years, and they were equally distributed in the < 25 pack-years category. Most (61%) of the heavy smokers ( >.1 pack/day) were in the > 25 pack-year category: 25% smoked 10-25 pack-years, and 14% < 10 pack-years. Table 2 shows the distribution of symptoms related to smoking and age. Prevalence of cough, phlegm, and persistent cough and phlegm did not increase with age. Breathlessness increased slightly but steadily with age, with the oldest age group reporting a rate of breathlessness 2.5 times greater than the youngest age group. Smoking had a strong effect that was dose-related for all respiratory symptoms. Light smokers ( < 1 pack/day) had the same symptom rate for cough, phlegm, and persistent cough and phlegm as exsmokers and pipe or cigar smokers. Exsmokers, pipe, cigar, and < 1 pack/day smokers had similar rates of breathlessness. Because of the potential confounding effects of age and smoking, comparison of symptoms in each exposure category were adjusted for smoking and age differences. Table 3 shows the prevalence of respiratory symptoms according to current job. The salaried chemical workers consistently had the lowest prevalence for all symptoms. Plastics, former chemical, and maintenance workers reported the highest rates for cough, phlegm, and per sistent cough and phlegm: rubber and hourly chemical workers were intermediate for these symptoms. All workers except the salary chemical workers observed similar rates of breathlessness. Table 4 reports the age and smoking adjusted prevalence of respiratory symptoms in workers working only in the rubber division, only in plastics, and only in the chemical plant. Main tenance workers move back and forth among chemical, plastics, and rubber, with no epidemiologically appropriate records kept of time spent in any of these areas. The mixed group comprised those who had worked in two or more exposure categories. The plastics only and chemical plant only workers had marked reduc tions in reported symptoms when compared to all currently em ployed workers in these same categories. Symptom rates for the mixed group were slightly higher than observed in the total population, and similar to the former chemical group- Those working only in chemical, plastics, or rubber appear to be a select group of healthier workers. A high proportion of those with symp toms in current job categories had a mixed exposure of rubber, chemicals and/or plastics URL 18862 'Monsmofcar = has sntoud Im Han IX cigarettes m > Irfetna. fEcsmoka = las stops*) smdnng to mors than 1 yea. and doss not no* smoke e pips or ciga tSmokv = las smatod more than IX agntts m a Irhrtime. end has smdud within the test yew Ripe or ciga smokers who also smoke Cigarettes are included m this category IPs* and Cigar = all those who smoke i pipe v ciga now. tut do not smoke cigarettes or ha< not smoked cigarettes to a yea or more Percentages n the margsals are % of totals, pacantagee m calls re axumn parentages f NM..V--M.Mf Emd* All Mhm *1 1 eed^day >1 ed/dy Pi* and Cipr Ml in eojw 94 (ISO*) 292 65.0%) (1 U15% 143 (260%) >8 (16.6%) 47 ( |J%J 531 5.6 7.1 210 M 119 34.1 (5 149 CMtwy 29 30-39 4049 SOU 4(0 107 C02W II UftJW 1(2 <M5W at now 22 I4.1W 144 157 15.4 143 133 * - ScorW by criteria from British Medal Retarch Ccuncil Rnpaitwy ouestionmire t -- 1 m>ume t = 3 mining U 13.1 26.4 13.1 23J 390 19.6 197 211 157 200 103 227 23 (0 137 U iti 26.1 2.1 92 93 90 103 70 Oi 11.1 193t 2(.5t 233t 119 400 15.2t 2122 1252 19.1 21.6 215 310 Figures 1-6 display the regressions of pulmonary function tests for the different smoking categories. Age and height significantly affected FEV and FVC values; age. but not height, was a statistically significant predictor of flow rates and closing volume (CV/VC). Heavy smokers (il pack/day) had markedly reduced ventilatory capacity compared to other smoking groups (except for FVC where smoking did not significantly affect lung volume). Lung function values of exsmokers decreased at a rate in termediate between the heavy smoker and nonsmoker groups. Pipe or cigar smokers resembled nonsmokers for FEV. FEFso, and FEF75. but resembled exsmokers for FEF2S-75 and closing volume (CV/VC). Except for CV/VC. the least reduction of ventilatory capacity was observed in the light smoker (<1 pack/day) group. All smokers had similar functional reductions of CV/VC with in creasing age. Current smoking behavior showed a better correlation with pulmonary function than did pack-years of smoking, and for this reason regressions based on current smoking categories were used in calculating predicted lung function values. Table 5 summarizes the prevalence of impaired lung function as related to age and smoking. Except for FVC and CV/VC pulmonary function impairment (see Table 5 for definitions! con sistently increased with age; and for all parameters consistently in creased with increased smoking. Impairment rates for FEFas-7-s. FEFso/FVC and CV/VC were significantly higher for smokers than nonsmoking categories; the smaller number of pipe or cigar smokers precluded them from achieving statistically significant dif ferences. Age and smoking were then analyzed simultaneously. Regressions of pulmonary function with age (Figs 1-6) suggested that after about age 40, smokers of pack/day clearly had lower lung function values than the other categories of smokers. As there was little difference in prevalence of impairment among ex smokers. nonsmokers and pipe or cigar smoker categories, they were combined into a nonsmoker category, divided into age groups above and below 40 years of age, and compared with cigarette smokers. Smokers and nonsmokers in the younger age group had similar rates of lung function impairment. Smokers over URL 1886 co ' = 1 IDMUng t =3 mttsinc hSMu f = i fURlUJ t = , URL 18864 didM suoqenba paquosap AjsnoiAaid lauuEUJ aqi ui pajsnlpp dJdM seiqpuEA Sui^oujs pus 'jqSjaq 'aSy pazA|EUP uaqi sem uoqaunj 8un| uo ajnsodxa jo uoijpjnp jo joajja aqj_ saujajixa asaqi uaaMjaq sajpj )udujjieduj| ppq A||ejauaS sapoSajpa qol jaqjo aqx -jsaMO| aqj sjaxpo/w iPDiiuaqa Ajejps 'juaLUJiediui jsaqSjq aqj bapij oj papuaj sja^jo.M [Eaiujaqa jaujjo-j -qol juajjno Aq uoqounj Ajpuouijnd pajieduj! jo aaua|EA3Jd aqj sazueuiLuns 9 e|qpj_ sdrtojS qol tuajjna ||p ui auies aqj sbm pajaipaid juaajad ueaui Aq pajrtspauj sp uoqounj 8un| auipreg uoqeindod upjoj) pjepiiEjs aqi jo uoqnquviip Supjouis aq; 9uisn fuipows *q UMWiMod IW! ui joj wo)|Mb3 uo<japJ4 -- j Jy pwnlpe Ap3ap aidM ssjej juaujjipduji paisnlp^-9ui^oius ueaiu aqj 8uiiP|na|P0 pup AjoSaiPD 8ui>)Olu5 ajeudojdde aqj mop paAuap san|PA pajoipajd Aq san|EA uoijaunj Ajpuoujfnd |pnpiAipui SujpjAjp Aq pd)P|n3|C3 seai\ dnojS qol qopa joj uoqounj Ajpuouj|nd juaruad upaiu -ajnsodxa ipuoqpdnaao jo aaueuoduji aqj Suissassp ui 8u!^oujs pup `jqSiaq 'aSp jo sjaajja Suipunojuoa apnpajd oj_ (aSp A up je sjdjjoiusuou upqj jaqSiq iuaiujipdiiii jo sajpj ppq sja^otus pjo puE 9unoA qjoq] 3A/AD puE (Suborns jo 38e Aq pajaajjE jou sea* juaLUJjEdiuj) }AJ joj paAjasqo jou sem 9ui5|Oujs pup a&e jo jaajja anstUiauAs y sjjpdjsjunoo fcuofoujsuou jiaqj 01 pajpdujoa juaujipduj| paseajaui Apup3ij|u8|S peq aip jo sjpaA -Suiqouis Aq tiqiqndod inoj. u| MOiimbj uqjbiJd -- j ly ......................... M . OS 41 _ 0> SC. . W tz , y Avp/^ad 1 4*p/q3*4 j , *p/qjd r AA*Fp/*/iad1t A*p/*7*< l> K> d'HOWM'l ^1 > \V X X . Vx X \ \. \VX\ \\ \ \,_ \ PI>1Iipppp**acchr/t*//4**Cy*ai?gvar w 1 rt " * 1 >} Fig 3. -- Prediction Equations for FEV25-7S In Total Population by Smoking. XV \X\ \\x>x \ \ xX \ \ '\ \\ \ V\ \ w\ \X x ^\ \ \ vx \ \ x\ \ X\ X\ '\ \ \, \ iu yan> 1 mcUM* paek/4p 1 Ik/<r m *t ct|ir >80 .0 i so Fig 4. -- Prwttction Equations for FEFso in Total Population by Smoking. .- LJRL 18865 *NonamdMrs=immoUR. eismcOare and pe or cigar smokars flmpand Pulmonary Funetien= FEV.TVC < 075 FVC/0.148 (height) 0.025 (age) 4 241 < 010 (Irani Manna 31) FEF2S-75/0 040 (heightJ 0090 (a|() + 4 139 < 0.75 (Irani this gopulatnn) FEFS0/FVC < 0.56 FEF7S/FVC < 0.30 CV/VC/0562 + 0 357 (age) >030 s0o predicted (Iran Reference 34) > --1 missing. 2=2 missing: 3 = 3 missing. 4 = 4 missing. 5 = 5 missing. 6=6 missmg /l !/ Nn <\ X'N' W V. "TT--- 1 1 W 1 \X V ^ Si \ i ?</? // y' *, y ./ // y/ // '/ // // ///./- '/*// >/V/ "* // X V/ '/ // //// //** 4///* y insertaye 'ZQ.fl (<M m yeete) (K<nm |fi iii<>mh >1F1tp* Loflaclf/t 20 n 30 O.i* l.H Fij s. -- Prediction Eouattom for FEF?s in Total Population by Smoking. Fig 6. -- Prodiction Equations for Cloting Volumo/Vltal Capacity (CV/VC) in Total Population by Smoking. calculated using length of work in years as the independent vari able, i.e.. predicted pulmonary function * % predicted + % change m pulmonary function/year spent in exposure category. Table 7 summarizes regression for which a significant change oc curred. For exposure category 1 (rubber only, chemical only, plastics only), rubber workers observed a significant decrease of -0.8%/year in FEF's. plastics workers a decrease of -1%/year. All other significant changes were increases in lung function. Health-related selective migration of workers is a potential source of bias in a cross-sectional study In an attempt to take ac count of this, the effects of working in rubber and chemical, rub ber and plastics, chemical and plastics, rubber, chemical, and plastics, or any one alone, were examined by analysis of covariance (Table 7). By backwards elimination method, exposure variables not statistically significant (p >0.05! were sequentially re moved until only the most significant exposure variables remained in the model. The analysis was done first on the total population (No. 2), then on the mixed exposure group (No. 3) and finally on the combination of categories within the mixed exposure group INos. 4-61. For the total population, working in chemicals was as sociated with increased FEV, FEF25-?s. FEFjs, and FEFso. The only re ductions in ventilatory capacity were observed in FEF75, a highly significant reduction associated with rubber, and increased CV/VC, a functional decrease associated with plastics The analysis of the mixed exposure groups (exposure categories 3-6) tended to support the results observed in the total population. URL 18866 Qafinilion of Impaired Pf in TMHt 5 1 = 1 missing 2--2 missing 6= 6 missing 99 mtssrtf URL 18867 f 1 I ] FEY m*% +4.47* 2 mi 3 mi 125 I FVC 931 +&M 6 M3 +452 1 FEFS-75 931 +044 2 M4 433 4 UR3 1M 9M I Wd Mi) 1 W +1M 3 MJ 0J9 4 MU m Mi MB 1 mm MS +440 H2 *440 1 MU 127 1 IIIJ IS MU 4J2 3 *U W 4 MM im 5 MS 44 jxm M3 971 MS S MM 125 1 >V US `EiMturt Category 1 = Rubber Only. PVC Only Chemical (H & Si Only 2 =Tai Potulation (Ruttw Chemical. PVC) 3 =Mned Expoture (Rubber & Chemol; Plastics & Chemical: Rutber i Plastics. RutM, Chemical 4 Plastics) 4 = Rubber & Plastics (N=S2> 5 --Rubber & Chemical fli=86) 6= Rubber. Chemical 4 Plastics (H~61) Satey CM My QaM MMr rvc OMy FVC Mow* CtemcM My Chwniid MM omm Has* CMM (My CM Ctatac* MM MM' ' MM MM Mr nc+ MM PM MM ' MM WC CMM fW ' * 042 046 0.10 004 04s 0402 0.009 0.04 003 oon 0405 040 047 012 S3 049 Ul OS7 04S91 ELM 0404 04* 044 440 B4S 045 Work in the chemical plant was associated primarily with in creased flow measured over mid vital capacity. Work in plastics was associated with functional reductions of from 0.4 to 1.2%/year in CV/VC. The most consistent and greatest changes were reductions of from -0.7 to 2.3%/year in expired flow rates at low lung volumes (FEF75), associated with work in rubber. Table 8 summarizes the relation of pulmonary function and cumulative exposure to vinyl chloride (see Table 8 for definition of CES-cumulat*ve exposure score). Increasing exposure to VC in all workers ever exposed to VC was not associated with any signifi cant reduction in pulmonary function. The general tendency was for pulmonary function to rmprove with increasing exposure. Only reductions in FVC approached statistical significance. Dividing the group into those with current VC exposure, and those that worked in chemical or plastics in the past (former chemical) and repeating the analysis produced the same results. The relation between respiratory symptoms and baseline pulmonary function is summarized in Fig 7. All respiratory symp toms were significantly associated with reduction in expiratory flow (FEF?5-7s. FEF:s) Cough, phlegm, and persistent cough and phlegm were significantly associated with reductions in FEF50. Those with cough 2. phlegm 2 and persistent cough and phlegm had reduced percent predicted CV/VC. Reduced FEF75 was asso ciated with symptoms of cough and persistent cough and phlegm. In addition to baseline pulmonary function, 56 workers were tested before work and again after work. Since the workers were also taking physical exams, actual working time averaged about 4hours. As there were about twice as many smokers 182%) among chemical workers as plastics (36%) and rubber (41%) workers, acute effects were compared in cigarette smokers and all others combined (nonsmokers). Current smokers observed significant `CESsamitotn* iipbwrt tan. All iota in th ctamnal pbnt wr* rmd ter VC upon** m high (TWA =iffroimtt>r 250 ptm). mnbun (TWA mnnmnWi 75 [ten). M tew flYA wpraunawy 25 ppnJ CES * aleuiMM by muHbb** mentta wetted limn * rekitn* axpown rating (high = 10 oMdum =3. low * 1) tor ttet jcb. nb winning <jm ill iota. Rwgt = 0 2877 npMim+wtta 0 1 2 0 12 012 012 (T I-2 ^25-73 ^29 PEFjq FEF,, CV/VC Fi( 7. -- Ptrcont Prmltetud Pulmonwy Function in Woitun With am3 Without Reipiritory Symptom Adjustud by Af*. Hoifht and Smoking. reductions in both flows and volumes. Nonsmokers observed in creases in FVC and peak flow, but other lung function parameters decreased, significant only for FEFso iTable 9). Smokers in all job groups had reductions in ventilatory capacity, the greatest decre ment occurring in flows at low lung volumes. The reductions were of the same order of magnitude for all three exposure groups, but were significant only for vinyl chloride and rubber workers. Except for reductions over the shift of FEFso and FEE?? in nonsmoking rub ber workers, ventilatory capacity of nonsmokers did not change significantly from zero. Discussion This study was stimulated by the finding of an increased inci dence of liver angiosarcoma in vinyl chloride (VC) workers. Experi mental studies, case histories, and epidemiologic studies ail suggest VC acts as a lung irritant. This literature is briefly sum marized. Patty, Yant and Waite exposed guinea pigs for 30 minutes to 10-40% VC- The principle gross pathological effect was congestion and edema of the lungs, and hyperemia of the kidneys and lungs. Exposure to 10% VC resulted in slight signs of lung hyperemia while hemorrhages, edema, and severe damage to the tracheal epithelium were observed at the higher concentrations in animals dying from the initial exposure. Pulmonary congestion was present after two weeks in surviving animals. Lower exposures forlonger time periods (50-500 ppm for six months) resulted in no change in lung weight in rats, guinea pigs, rabbits and dogs. Expo-* sures at concentrations greater than 100 ppm produced reversible liver injury in the more sensitive species.7 Lester, et al.8 exposed rats to 20,000 ppm (2% VC) for three months, and 50,000 ppm (5% VC) for 19 days. They observed increases in liver weight and decreases in white blood cells, but concluded that the observed URL 18868 p = 0.05 "p = < 0.025 p 4 0.005 p ss < 0.005 tNS = wrameMr. tumdur p* or apt tmotar IS = win** ts A portno sn mticatti i ttaaono m pwtainn* n tfw (tier wort CV/VC rtiio a Hrttr (Aon the Mon wort CV/VC nta'Mmt % cMnp = rttr wort Pf Mm wort Pf . moon Pf * iftr + More 2 lung edema and congestion could not be attributed to tbe irritat ing effects of VC, and the pneumonia observed was probably due to secondary infection. Keubler'1' exposed rats, mice, and guinea pigs to 5,000, 15,000 and 25,000 ppm VC tor 2-hours a day for 100 days without observing any histological damage. In a report of a case of VC gassing in a plant in Creal Britain, a worker washing out a polymerization vessel with water collapsed and was given artifi cial respiratory. Subsequent symptoms included chest tightness. Boytsov9 reports ^without giving the number of species or animals, length or amount of exposure! that inhalation and intra tracheal administration of PVC resin produced a "development of peribronchitis in the lungs" and local thickening of interalveolar septa Two studies of PVC workers in Russia and Germany con clude that the respiratory system was affected.11 12 A ` consider able number"' of 96 PVC workers in Russia had a "changed bronchovascular pattern and increased pulmonary ventilation at rest".11 Eight of thirteen German PVC workers employed for 1.75 to 18 years showed partial pulmonary insufficiency with signs of predominantly restrictive changes of the lungs.12 This latter finding is in agreement with the case described by Szende, et al.1' A 31 year old male had shoveled PVC dust in a drafty place tor 1-year prior to being examined for severe dyspnea. The possibility of pneumoconiosis elicited by inhalation of PVC resin was suggested There are a number of problems with the experimental studies of animals. Only at very high VC concentrations are effects ob served, and these are anatomical changes from gross pathology or histology. More sensitive functional measurements at lower, more realistic, VC exposure concentrations have not been reported. The foreign studies are only suggestive, and do not provide pnough information to evaluate. There is only one reported epidemiologic study of VC and PVC workers in the U.S.M Cough and sputum production for most of 1year or more clinical signs of emphysema were reported in 31% of the 348 currently and previously employed VC and PVC workers. Smokers and nonsmokers over 40 years of age had similar rates of flow impairment. In the age group under 40 however, smokers had about twice the prevalence of flow impairment as nonsmokers. The higher prevalence of flow impairment in smokers compared to nonsmokers converged slightly with increasing duration of exposure; nonsmokers and smokers with more than 20 years exposure had the same prevalence of flow impairment The authors conclude that the etiologic agents produc mg the reduced air flow in workers over 40 years of age or with > 20 years expo sure is unclear but is not due to smoking. Exposure was to VC monomer, PVC resin dust, and pollution in a heavily industrialized city. The effect of smoking on respiratory symptoms has become common knowledge following the report of the Surgeon General on the Health Consequences of Smoking (19641 and its sub sequent supplements.'" The resuits of this study confirm these re ports, with the possible exception that light smokers (<1pack/day) in this study resembled exsmokers more than heavier smokers ( > 1 pack/dayl. Pipe or cigar smokers have been less studied, although it has been suggested that pipe smokers are to be distinguished from cigar smokers.1" Because of small num bers this has not always been done, and m this study these smokers were combined. They generally resembled light smokers and exsmokers in both symptoms and lung function. The effects of smoking on symptoms have generally shown cough and phlegm to be higher in smokers, with prevalence relal- IJRL18869 'Grate 2. 3. 4 breathlessness tGrade 3 and 4 bresthtesaress t Grate 2 ant 3 breathlessness 1 ed to the amount of smoking, a relationship supported by this Secondly, light smokers were obviously less affected than moder study. Breathlessness, while generally higher in smokers, is often ate and heavy smokers. In the analysis of smoking, one should ex not related to the amount of smoking. This is hypothesized as due to selective reduction of smoking in those developing symptoms amine this dose response relation, and not arbitrarily lump to gether all smokers. of breathlessness, a more severe and disabling condition than Careful consideration of possible smoking-exposure interactions "smokers cough".1'' Although movement of less fit heavy smokers is also required. For example, in a group of workers exposed to into lighter smoking or nonsmoking categories seems a plausible asbestos dust, smokers reported worse lung function than non- explanation, selection does not appear to be a strong factor in this population as there was a high prevalence of breathlessness cn the smokers when a single regression was used for height and age ad justment.29 The inference was that smoking increased the effect of heavy smoker category. dust exposure. Using only a single regression equation for stan A trend for breathlessness to increase with increasing age is dardization is misleading however, as smokers pulmonary function commonly observed.'9 and was apparent in this study. The associ decreases more rapidly than exsmokers and nonsmokers. Rossiter ation of cough and phlegm with increasing age has not been as and Weil29 point out that an apparent synergism between dust consistent. Some studies report an increase with age;19 20 others and age may instead be a smoking-age interaction. When separate (including this study) find little tendency for cough and phlegm to regression equations were used for standardization, i.e., one each increase with age.19 21 for nonsmokers, exsmokers, and smokers, they observed no dif Two studies in West Virginia have observed an association be ference in pulmonary function related to dust exposure. In fact, tween educational level and prevalence of respiratory symptoms the asbestos dust exposure had a more detrimental effect on non- and ventilatory lung function capacity, with the lower prevalence smokers than smokers. We know there was a smoking-age inter of symptoms partly attributed to fewer smokers in the higher action in this study. Since we are primarily interested in the effect educational categories." 23 Stebbings27 in 469 white male non- of VC exposure on pulmonary function, age and smoking ad smokers found no effect of social class (based on occupation, justments were made using the regression equations of the various education, and income) on most respiratory symptoms; only smoking categories (Figs 1-6). As there was no difference in dyspnea was significantly less common in the upper social class. pulmonary function between current job categories, and no non- Lower pulmonary function was also associated with lower socio exposed group of workers was available, the regression equations economic status in these nonsmoking males.27 Salaried chemical used for standardization were derived from the smoking cate workers in this study resemble the upper social class in that they gories of the total population. smoke less than hourly workers, and are probably more educated. After standardization, increasing duration of exposure in chemi Exposure to VC and PVC resin is generally less than that of the cals, plastics, and rubber was not associated with large changes in URL 18870 hourly workers. It is not known from this study whether social pulmonary function. Reductions in FEF75, the most sensitive mea class, exposure, smoking, or some combination of these factors sure of small airways obstruction, was consistently and significant can explain the lower symptom rates of salaried chemical workers ly associated with working in rubber; less significant reductions in when compared to hourly workers. The same argument applies to FEF75 were associated with hourly chemical and plastics workers, the slightly better pulmonary function of salaried chemical and FVC in hourly chemical workers. Other pulmonary function workers compared to hourly workers. Among hourly workers, no parameters generally remained the same or improved with in consistent or convincing differences in respiratory symptoms or creasing duration in chemicals or plastics, and with increasing function were observed. cumulative VC exposure. The reported case of pneumoconiosis When workers were divided into exposure categories of work on exposure to PVC resin11 is suggestive that high exposure to ing only in plastics or chemicals, the respiratory symptom rate was PVC resin may result in a restrictive type of respiratory disease. reduced considerably. Workers with more job mobility were less Baggers in the chemical plant are exposed to from 1-5 mg/m' of healthy in the sense that they reported more symptoms. This particulate (measured by area Hi-volume samplers); respiratory could be due to greater susceptibility, to having a more unfavor particulate is 1/10 to 1/20 these levels. Exposure to the PVC resin able job experience in terms of exposure, to self-propelled move in plastics appears higher for Banbury operators but lower for ment to less symptom producing jobs, or some combination of workers in milling, and negligible for other plastics workers. The these factors. Although the reasons are unknown, the fact remains low prevalence of FVC impairment in both chemical and plastics that the less mobile "pure'1 exposure groups had lower symptom workers does not support the hypothesis of PVC resin exposure rates than control rubber workers, maintenance, and the mixed producing volume restriction. exposure workers What difference is there between the workers in this study and Are the symptom rates observed in this population comparable those reported on by Miller, et al? Pulmonary function impairment to those in other populations? Table 10 compares the prevalence for FEV/FVC. FVC, and FEF75/FVC in this study employed the same of respiratory symptoms in this study with other studies. There is definition as Miller, et al.1* in their study of VC-PVC workers. The considerable variation between reported symptoms in the various prevalence of impairment for FEF25.75, using the same criteria, was populations. The workers in this study compare favorably except essentially zero, and thus far below the 53% they observed. The for perhaps breathlessness. cutoffs for impaired flow as measured by FEFso/FVC and FEF75/FVC The effects of age and smoking on pulmonary function ob have been shown to be reasonably effective in distinguishing nor served in this population are comparable to other reported mal subjects from those with obstructive lung disease.32 The high studies. There were no significant statistical differences in the in est prevalence (almost 50%) of impairment was for FEF?s/FVC and tercepts of the smoking regression equations. Heavier smokers in supports the observed association of reduced FEF75 with increas this cross-sectional analysis had significantly greater reductions of ing duration of work. All VC-PVC nonsmokers less than 40 years FEV, FEF2S-75, FEFso, and FEF75 with increasing age. Two points are of age, and all workers in this study less than 40 years of age, had of interest with regard to the effect of smoking. Before the age of about a 20% prevalence of impairment for FEV/FVC and 40, smoking had no apparent effect on pulmonary function. FEF?s/FVC. The prevalence of impairment increased for non- l1 URL 18871 smokers over 40 in both populations, but more so in the Miller study. The rates of flow impairment in both studies are the same for smokers over 40. In both studies the prevalence of impaired FVC I <10%) is considerably lower for flow impairment (18-75%) and much lower than expected from exposure to asbestos.1' The association of increased pulmonary function performance with increasing exposure and work duration, particularly in chemi cals, suggests that selection is occurring in this population. The reason tor the lack of association between VC exposure and respiratory effects may be due to differential migration of more susceptible workers out of chemicals. Note that the mixed expo sure group reported a higher symptom rate than chemical and rub ber workers who had never moved. The current plastics worker had symptom rates comparable to or higher than this mixed expo sure group, suggesting that net migration of ''susceptibies" may have been into plastics. The association of small airways ob struction with exposure may be more informative than symptoms because (1) small airways may be more sensitive to the effects of environmental agents; and (21 small airways obstruction is probably not noticed by the individual to the same extent as symptoms and FEV reductions, and would therefore not be af fected as much by selective migration. Several conclusions emerge concerning factors that affect base line pulmonary function in this population: ID age and smoking were the two most important factors reducing pulmonary func tion; smoking was associated with obstruction only (no significant effect on FVC). (2}'sma!l airways function as measured by FEFrs was affected by both smoking and work exposure, primarily rub ber worker exposure. 13) work, whether measured by duration or estimated cumulative exposure to VC and after adjustment for smoking effects, was not associated with reduced baseline pulmonary function (except for FEF75). (4) although prevalence rates of impairment are not often reported, the air flow im pairment (FEV/FVC, FEF7s), rates appeared elevated only in the older smoker. Respiratory symptoms and.measurement of pulmonary function are two somewhat independent and yet overlapping methods for investigating the risk of environmental exposures to the respira tory system. Symptomatology suffers from the disadvantage of be ing subjective, can show interobserver variation, and requires adequate recall on the part of the person being interviewed: symptomatology is useful in diagnosis, particularly chronic bron chitis. Independent validation of symptoms in practice turns out to be pulmonary function measures, despite the fact that different, albeit overlapping, parameters are being measured. The observed association of reduced pulmonary function with increasing severity of symptoms increased the validity of respiratory symp tom information. The conclusion that VC exposure did not consti tute a significant respiratory hazard in this population was strengthened by similar results for both symptoms and pulmonary function. While one methodologic strength of this present study is the ability to make internal comparisons between work groups, it must be stressed that none of the groups can be regarded as hav ing worked m hazard-free conditions. Both rubber and plastics workers are exposed to varied levels of dust and fume, and it could be that reduction in lung function among chemicals workers (whether due to vinyl chloride exposure, or to some other agent) are being obscured because of similar reductions in the other job groups. The results of FEF75/FVC impairment (pertaining to that part of the respiratory tract thought to be first affected by harmful exposure -- the small bronchioles) suggest increased abnormality rates in all workers. Indeed, for several of the pulmonary function tests the chemical workers had better results than rubber and plastics workers. To assess the potential respiratory hazard of each exposure category, before-and-after shift tests were carried out on some workers. Any reduction in respiratory function over the shift suggests the likelihood of some harmful exposure in the working environ ment.1* Smokers overall had statistically significant reductions for all ventilatory tests; nonsmokers showed little reduction. The most obvious reductions were flow rates at low lung volumes, i.e., mea sures of small airways performance. Chemical workers who were smokers had significant reductions for five of the seven tests, pre dominantly reflecting small airways obstructive change: both plastics and rubber workers who smoked showed reductions of similar magnitude. No clear picture emerges suggesting one job group faced greater daily respiratory insult than the others. Re duction in small airways performance among nonsmokers in each job group suggests that all workers were exposed to agents with adverse respiratory effects. In a clean environment the average re sult for these nonsmokers would have been a slight increase in test performance. Whether these acute reversible changes have any permanent effect is not known. The significance of these findings is unclear. The small airways obstruction associated with work in the rubber division, i.e., the control group, could be related to a number of environmental agents. The complexity of the industry and the heterogeneity of the control group does not allow for determining the etiologic agents. The rubber worker group was selected on the basis of lower particulate exposure and never hav ing worked in chemical or plastics. But they could have, and probably did, work in other jobs with higher exposures within the rubber division. A complete work history for each rubber worker would be required to investigate further the association of re duced small airways function with work in rubber Fundamental and more difficult questions remain. What is the biological significance of acute reversible reductions in small air ways function over a shift? What is the biological significance of statistically significant reductions in baseline pulmonary function over a longer time period? For FEFrs. the reduction is on the order of -1%/year. Over a 40 year period, this would amount to a 40% decrease. However, some individuals have a reversible change of 40% in the course of a day. The association of working in rubber with a downward slope of the percent predicted FEF75 is a finding that requires further investigation of workers in other areas of the rubber plant an investigation currently underway. The authors thank Mike Symons (or advise or statistical matters, NIOSH industrial hygienists lor assistance in deriving the VC exposure source. Mary lane Thomas lor manuscript preparation, and union and company officials for their assistance m the project. References 1. Patty FA: Industrial Hygiene and Toxicology, Vol 2, John Wiley and Sons. New York. 1963. 2. Wilson RH, McCormick WE. Tatum CF, and Creech JL: Occupational aeroosteolysis, report of 31 cases. )AMA 201:577-561. 1967 3 Dmman BO et at Occupational aeroosteolysis I An epidemiological Study. Arch Environ Hlth 22:61-73. 1971. 4. Cook WA et al: Occupational aeroosteolysis II. An industrial hygiene study Arch Environ Hlth 22'74-82. 1971. 5. Creech JL and lohnson MN: Angiosarcoma of the liver in the manufac ture of polyvinyl chloride. (OM 16:150-151, 1974. 6. Patty FA, Yam WP, and Waite CP: Acute responses of guinea pigs to vapors of some new commercial organic compounds V. vinyl chloride. U.5 Pub Hlth Reports 45:1963-1971. 1930. 7. Torkelson TR, Oyen F, and Rowe VK; The toxicity of vinyl chloride as determined by repeated exposure of laboratory animals Amer Indus Hyg Assoc I 22:354-361. 1961 8. Lester D, Greenberg LA, and Adams WR: Effects of single and repeated exposures of humans and rats to vinyl chloride Amer Indus Hyg Assoc I 24:265-275, 1963. 9. Boystov AN Problems of industrial hygiene in new plants for the manufacture of synthetic resins and plastics. Cigyena Truda 11 20-27. 1963. 10. Kuebler H The physiological properties of aerosol propellants. Aerosol Age 9 44 47-48. 50. 90-91. Abstract only. 1964. 11 Vertkm Yl and Mamontov YR: Condition of the bronchopulmonary system in workers manufacturing articles of polyvinylchloride. Cig Tr Pro! Zabol 14:29-32, 1970. 12 Lange CE. )uhe S, Stein G. and Veltman C Die Sogenante Vinylchloride-Krankheit-em berufsbedingte Systems Klerosef Int Arch Arbeils med 32:1-32, 1974. 13. Szende 8, Lapis K, Nemes A. and Pinter A. Pneumoconiosis caused b"{ the inhalation of polyvinylchloride dust. Med Lav 61:433-436. 1970. 14. Miller A et al: Changes in pulmonary function in workers exposed to vinyl chloride and polyvinyl chloride. Ann NY Acad Sci 246:42-52. 1975. 15. Suggested Standardized Procedures for Closing Volume Determina tions (Nitrogen Method), distributed by Division of Lung Diseases. National Heart and Lung Institute, luly (1973). 16. Annual Report of the Chief Inspector of Factories for the Year 1951. London: HMSO Cond 8772, March, 1953. 17. Smoking and Health, Report of the Advisory Committee to the Surgeon General of the PHS, USCPO. Washington, DC.. 1964 and sub sequent reports. 18 Seltzer CC: Morphologic constitution and smoking. IAMA 183:639645, 1963. 19. Higgins ITT: The epidemiology of chronic respiratory disease. Prev Med 2:14-33, 1973. 20. Higgins ITT, et al: Chronic respiratory disease in an industrial town: A nine-year follow-up study. (Preliminary report) Am / Pub Hlrh 58:1667, 1968. 21 Ferris BC |r and Anderson DO: The prevalence of chronic respiratory disease in a New Hampshire town. Am Rev fiesp Dis 86:165-177, 1962. 22. Higgins !TT. Higgins MW. Lockshin MD. and Canole N: Chronic respiratory disease tn mining communities in Marion County, West Virginia. Brir ) Ind Med 25:165. 1968 23. Enterline PE and Lainhart WS: The relationship between coal mining and chronic nonspecific respiratory disease. Am / Publ Hlth 57:484, 1967 24. Ferris BG Jr et al: Chronic nonspecific respiratory disease in Berlin, New Hampshire, 1961 to 1967. ARRD 107:110-122, 1973. 25. Higgins ITT: Occupational Factors in Chrome Bronchitis and Em physema, m Bronchitis III. Proc. of 3rd int. Conf. on Bronchitis at Groningen. The Netherlands, 23-26 Sept. 1960. 26. Stebbins |H: Chronic respiratory disease among nonsmokers in Hagerstown. Maryland -- III. Social class and chronic respiratory disease fnv Res 4:213-232. 1971. 27. Stebbings JH: Chronic respiratory disease among nonsmokers in Hagerstown. Maryland -- I. Design of the study and prevalence of symp toms. Env Res 4:146-162, 1971. 28. Denson PM et al: A survey of respiratory disease among New York City postal and transit workers. I. Prevalence of symptoms- Env Res 1-265< 286, 1967 1 29. Rossiter CE and Weil! H: Synergism between dust exposure and smoking: An artefact in the statistical analysis of lung function' Bull PhvstoPath Resp 10:717. 1974. 30. Morris JF, Koski A. and Johnson LC Spirometric standards for healthy nonsmoking adults. Amer Rev Resp Dis 103:57-67, 1971 31 Lapp NL and Hyatt RE: Some factors affecting the relationship of maximal expiratory flow to lung volume in health and disease. Dis Chest 51:475. 1967. 32. Bader ME et al: Pulmonary function and radiographic change in 598 workers with varying duration of exposure to asbestos. I Mi Smai Hosp 37:492, 1970. 33. Buist SA and Ross BB. Predicted values for dosing volumes using a modified single breath nitrogen test. Amer Rev Resp Dis 107:744-752 1973. 34. Gamble JF. McMichael A|, Williams T, and Battigeili MC: Respiratory function and symptoms: An environmental-epidemiologic study of rubber workers exposed to a phenol-formaldehyde type resin. Amer Ind Hyg Assoc / 37:1976. URL 1887 PO